AMENDED OPINION & ORDER

PAUL A. ENGELMAYER, District Judge.

This decision resolves a motion by the putative class plaintiffs in this multi-district litigation for leave to file a proposed Fourth Amended Complaint ("PFAC"). The Court has sustained as well-pled plaintiffs' Sherman Act § 1 claims covering the years 2013-2016. Fact discovery as to these claims is set to close soon. The Court, however, has twice ruled against class plaintiffs on their bid to also pursue claims covering the five preceding years, 2008-2012. For the reasons that follow, the Court denies the motion to amend to the extent that plaintiffs again seek to add claims for 2008-2012. The Court, however, grants the motion to the extent that plaintiffs seek to amplify on their factual allegations bearing on the 2013-2016 claims.

I. Background

This case centers on claims of § 1 violations affecting the market for interest rate swaps ("IRS" or "IRSs"). Plaintiffs claim that the investment banks who dealt in IRSs (the "Dealer Defendants" or "Dealers") conspired to block the emergence and later the survival of electronic trading platforms that would have permitted IRSs to be traded on an anonymous, "all-to-all" basis. Such platforms, plaintiffs claim, would have provided price benefits to investors relative to the over-the counter ("OTC") model by which IRSs were historically traded, i.e., dealer-to-investor, with the investor's name disclosed to the dealer.

The Court has issued two decisions reviewing the claims and procedural history. On July 28, 2017, the Court issued a 108-page decision resolving motions to dismiss the Second Amended Complaints: one by the putative class of IRS investors ("SAC"); the other by non-class plaintiffs Javelin Capital Markets LLC ("Javelin") and Tera Group Inc. ("Tera"), each of which in or after 2013 opened electronic platforms for all-to-all trading of IRSs ("JTSAC"). See In re Interest Rate Swaps Antitrust Litigation, 261 F.Supp.3d 430 (S.D.N.Y. 2019) ("IRS I"). On May 23, 2018, the Court issued a 59-page decision resolving class plaintiffs' motion for leave to file a proposed Third Amended Complaint ("PTAC"). See In re Interest Rate Swaps Antitrust Litigation, Nos. 16 MD 2704 (PAE) & 16 MC 2704 (PAE), 2018 WL 2332069 (S.D.N.Y. May 23, 2018) ("IRS II").1 The Court incorporates those decisions by reference and, in summarizing them below, recites only necessary background.

A. IRS I

On November 25, 2015, the initial complaint before this Court was filed. On June 2, 2016, the United States Judicial Panel on Multi-District Litigation transferred all related matters to this Court for coordinated or consolidated pretrial proceedings. Dkt. 1.2 On December 9, 2016, after appointment of interim lead counsel for the class and the setting of a briefing schedule, class plaintiffs filed the SAC, Dkt. 142, and the Javelin/Tera plaintiffs filed the JTSAC, Dkt. 145.

On July 28, 2017, the Court, in IRS I, ruled on the motions to dismiss the SAC and JTSAC. The Court sustained the § 1 claims as to 2013-2016 (made by both sets of plaintiffs) but dismissed the § 1 claims as to 2008-2012 (made by class plaintiffs only). For each period, plaintiffs had pursued claims of per se illegal conduct.

As to the 2013-2016 period, the Court held that plaintiffs had plausibly pled a per se illegal group boycott among IRS Dealers. In this period, the Court held, plaintiffs had pled significant parallel conduct among the Dealers sufficient to give rise to an inference of an agreement among them aimed at boycotting and otherwise hobbling the platforms opened by Javelin, Tera, and a third entity, trueEX LLC ("trueEX"). In or after 2013, after the effective date of the Dodd-Frank Act, these three entities had each opened a "swap execution facility," or "SEF"—a platform on which IRSs could be electronically traded in an anonymous, "all-to-all" manner. Dodd-Frank, enacted July 21, 2010 and effective in 2013 after significant rulemaking and regulatory implementation, had anticipated, enabled, and mandated the opening of SEFs. See generally IRS I, 261 F. Supp. 3d at 472-81.

As to the 2008-2012 period, however, before Dodd-Frank's mandates had taken effect, the Court did not find plausible class plaintiffs' claims of a per se unlawful agreement among Dealers to prevent from coming into existence electronic trading platforms permitting all-to-all anonymous trading of IRSs. See id. at 463-72. Independently, the Court held, class plaintiffs lacked antitrust standing to pursue their pre-2013 claims. To pursue such claims, the Court noted, a plaintiff must be an "efficient enforcer" of the antitrust laws; a central factor in that inquiry is whether the plaintiff's claimed injury is speculative. The Court held that class plaintiffs' claims of injury from the failure of electronic trading to emerge in 2008-2012 were unduly speculative, as they required the finder of fact to postulate an "alternative history of IRS trading" that "require[d] too many leaps of imagination and guesswork for a claim of class injury to be viable." Id. at 494 (citation omitted). The Court, finally, held that plaintiffs' claims based on injuries incurred before November 25, 2011 fell outside the four-year statute of limitations and were time-barred. Id. at 487-90.

B. IRS II

After IRS I, the litigation proceeded to discovery, pursuant to a court-approved case management plan. See IRS II, 2018 WL 2332069, at *4. On February 21, 2018—the last day on which motions seeking leave to amend were authorized—class plaintiffs moved for leave to file the PTAC. Id. at *5. It sought to add a new class plaintiff, new allegations as to 2013-2016, and, most consequentially, to restore the putative class's claims as to 2008-2012. Id. at *6.

On May 23, 2018, the Court in IRS II resolved the motion for leave to amend. The Court authorized the addition of a new plaintiff and of new allegations as to 2013-2016. Id. at *8-9. The Court, however, denied the motion to restore the 2008-2012 claims, under Federal Rule of Civil Procedure 15(a), for the following reasons.

Futility: The PTAC did not remedy the deficiencies that had led to dismissal of the SAC's claims as to 2008-2012 and thus was futile. Id. at *9-19. This was so for multiple reasons, four of which the Court developed. First, the PTAC's theory of investor injury from the failure of electronic all-to-all IRS trading to then emerge remained "a product of speculation, imagination, and guesswork," id. at *10, making class plaintiffs' claims of injury "far too conjectural to survive," id. at *9. Second, the PTAC's allegations as to "Project Fusion," a joint venture among Dealers, did not make out per se unlawful conduct under § 1, and the PTAC's two paragraphs of allegations as to the rule of reason were too incomplete and spare to make out a rule of reason claim. Id. at *11-13. Third, the PTAC's allegations as to the Dealers' conduct towards Swapstream, a proposed platform of the Chicago Mercantile Exchange ("CME") for clearing transactions, also did not viably plead a § 1 violation, either per se (as the PTAC pled) or under the rule of reason (a theory the Court sua sponte considered). Id. at *13-17. Fourth, plaintiffs' claims predating November 25, 2011 remained time-barred. Id. at *18-19.

Independently, the Court held, considerations of delay, prejudice, and gamesmanship all required denial of the motion for leave to add claims for 2008-2012. Id. at *19-28.

Delay: The parties had organized and conducted discovery based on the 2013-2016 parameters. Discovery disputes had been negotiated and litigated based on the premise that these were the sole claims plaintiffs were pursuing. Permitting the 2008-2012 claims would have upended the negotiated and court-approved discovery schedule. Allowing the PTAC's pre-2013 claims "would [have been] functionally tantamount—or close to it—to allowing a new MDL-sized lawsuit to be hitched to the existing claims." Id. at *20; see also id. at *20-22 (listing discovery areas uniquely relevant to 2008-2012 claims and assessing implications of reopened discovery, as to parties, third parties, and data gathering).

Prejudice: Adding the 2008-2012 claims would have prejudiced defendants, who for seven months had expended time and money in discovery based on the "only rational assumption as to this case's temporal scope: that plaintiffs' surviving claims were limited to 2013-2016." Id. at *22; see also id. at *22-23.

Gamesmanship: Plaintiffs' counsel admitted that they had been intending and planning since the day IRS I issued to move to amend to revive the 2008-2012 claims. The Court chronicled plaintiffs' counsels' communications to the Court and the defense between IRS I and the PTAC's filing. Rather than disclose this intention, these communications instead repeatedly implied the opposite. Id. at *23-27. "In ways large and small," the Court stated, "plaintiffs' counsel throughout these seven months fed the false impression that the temporal scope of plaintiffs' claims was fixed at 2013-2016. The result was to mislead the Court and the defense and to occasion extensive work and expense that—had the retooled 2008-2012 claims proven viable—would have largely gone wasted." Id. at *24; see also id. at *24-26 (reviewing communications). The Court rejected plaintiffs' counsels' justifications for such "coy" and "tactical" behavior—that the work product doctrine or the need to avoid witness tampering necessitated it. Id. at *27. This "unacceptable gamesmanship," id. at *27, independently warranted denying the motion to add the 2008-2012 claims, id. at *28.

After the decision in IRS II, plaintiffs filed a Third Amended Complaint, consistent with the Court's rulings. Dkt. 398. Discovery continued, pursuant to the case management plan.

C. The PFAC

On October 26, 2018, class plaintiffs filed a motion for leave to amend via a proposed Fourth Amended Complaint, Dkt. 590, the PFAC, Dkt. 693 (Declaration of Daniel L. Brockett) Ex. A ("PFAC"),3 and a memorandum of law in support, Dkt. 591 ("Pl. Mem."). Plaintiffs had alerted the Court that a motion to amend was under consideration. See Dkt. 408 (June 19, 2018 letter from plaintiffs); Dkt. 430 (July 17, 2018 joint status letter).

The Court describes the PFAC more fully below. In brief, it principally seeks to restore plaintiffs' 2008-2012 claims. As to these, the PFAC reprises and recasts plaintiffs' earlier allegations regarding the two episodes on which plaintiffs' prior claims as to 2008-2012 have mainly been based: (1) Tradeweb/"Project Fusion," in which Dealers jointly acquired part of the business of trading-platform company Tradeweb; and (2) CME/"Project Magellan," in which the Dealers jointly negotiated and reached agreement with CME regarding its "Swapstream" product. The PFAC adds limited allegations as to these episodes. It also includes allegations as to several new areas of alleged Dealer collaboration. More than earlier complaints, the PFAC pleads, as an alternative to its theory of per se § 1 liability, a claim of § 1 liability under the rule of reason for 2008-2012. The PFAC also expands on its allegations as to 2013-2016. In particular, to a greater degree than before, it seeks to account for the failure to emerge, after Dodd-Frank became effective in 2013, of all-to-all platforms permitting anonymous trading of IRSs other than the allegedly boycotted platforms of Javelin, Tera, and trueEX.

On December 4, 2018, defendants filed an omnibus memorandum of law in opposition to the motion for leave to file the PFAC, Dkt. 638 ("Def. Mem."), and a declaration in support, attaching exhibits, Dkt. 637 ("Playforth Decl."). On December 27, 2018, class plaintiffs filed a reply brief, Dkt. 664 ("Pl. Reply"), and a declaration in support, Dkt. 665 ("Brockett Decl.").

II. Legal Standards Governing the Motion to Amend

Class plaintiffs' motion to amend is subject to two Federal Rules of Civil Procedure: 15(a) and 16(b).

Rule 15(a) provides that a court "should freely give leave [to amend] when justice so requires." Fed. R. Civ. P. 15(a)(2). "[I]t is within the sound discretion of the district court to grant or deny leave to amend" under Rule 15. McCarthy v. Dun & Bradstreet Corp., 482 F.3d 184, 200 (2d Cir. 2007). "A district court has discretion to deny leave for good reason, including futility, bad faith, undue delay, or undue prejudice to the opposing party." Id. (citing Foman v. Davis, 371 U.S. 178, 182 (1962)). As noted in IRS II, 2018 WL 2332069, at *7, leave to amend under Rule 15(a) is properly denied upon a showing of prejudice or bad faith; "[a]mendment may be prejudicial when, among other things, it would `require the opponent to expend significant additional resources to conduct discovery and prepare for trial' or `significantly delay the resolution of the dispute.'" AEP Energy Servs. Gas Holding Corp. v. Bank of Am., N.A., 626 F.3d 699, 725-26 (2d Cir. 2010) (quoting State Teachers Ret. Bd. v. Fluor Corp., 654 F.2d 843, 856 (2d Cir. 1981)); see also Cresswell v. Sullivan & Cromwell, 922 F.2d 60, 72 (2d Cir. 1990) ("The court plainly has discretion . . . to deny leave to amend where the motion is made after an inordinate delay, no satisfactory explanation is offered for the delay, and the amendment would prejudice the defendant.").

Because the present motion to amend was filed after the deadline set in the court-ordered case management plan, it—unlike the motion addressed in IRS II—is also subject to Rule 16(b). See Kassner v. 2nd Ave. Delicatessen Inc., 496 F.3d 229, 243-45 (2d Cir. 2007) ("Rule 16(b) also may limit the ability of a party to amend a pleading if the deadline specified in the scheduling order for amendment of the pleadings has passed."); Fresh Del Monte Produce, Inc. v. Del Monte Foods, Inc., 304 F.R.D. 170, 174-76 (S.D.N.Y. 2014) (discussing interplay between Rules 15 and 16 in motions to amend). Rule 16(b) requires that the Court set a schedule, which "must limit the time to join other parties, amend the pleadings, complete discovery, and file motions." Fed. R. Civ. P. 16(b)(3)(A). Relevant here, Rule 16(b)(4) provides that such a schedule "may be modified only for good cause." Id. 16(b)(4). Where Rule 16(b) applies, it informs the Court's exercise of discretion under Rule 15(a): Although Rule 15(a) provides that leave to amend is to be freely given "when justice so requires," "a district court does not abuse its discretion in denying leave to amend the pleadings after the deadline set in the scheduling order where the moving party has failed to establish good cause [pursuant to Rule 16]." Parker v. Columbia Pictures Indus., 204 F.3d 326, 340 (2d Cir. 2000); see also Ramsay-Nobles v. Keyser, No. 16 Civ. 5778 (CM), 2018 WL 6985228, at *7 (S.D.N.Y. Dec. 18, 2018) ("[T]he `good cause' standard is permitted but not mandated when a party seeks to amend its pleadings after the deadline set in the scheduling order." (citation omitted)).

As the Second Circuit has explained, the scheduling order for which Rule 16 provides serves an important purpose: "By limiting the time for amendments, [Rule 16] is designed to offer a measure of certainty in pretrial proceedings, ensuring that at some point both the parties and the pleadings will be fixed." Parker, 204 F.3d at 340 (quotation marks and citation omitted). Therefore, "a finding of `good cause' depends on the diligence of the moving party" in complying with the deadline for amendments. Id. (citation omitted). In addition, a court applying Rule 16 has discretion to examine the Rule 15(a) factors, "in particular, whether allowing the amendment of the pleading at this stage of the litigation will prejudice defendants," Kassner, 496 F.3d at 244, as well as "futility, bad faith, [and] undue delay," McCarthy, 482 F.3d at 200 (citation omitted).

Practical considerations pertinent to the above multi-factor analysis include the duration of the case and the stage of discovery. In McCarthy, for example, the Second Circuit affirmed the denial of leave to amend under Rule 15 where the motion came "more than two months after discovery was completed and more than a year and a half after the filing of the original complaint." 482 F.3d at 201 (citation omitted). The district court had noted that "if the amendment is allowed, merits discovery will need to be reopened and the litigation will, in essence, start over—the same experts will likely need to produce new reports and be redeposed." Id. (quotation marks, alteration, and citation omitted). Similarly, in Ramsay-Nobles, Chief Judge McMahon noted when applying Rules 15 and 16 that the proposed counterclaims had come less than a month before fact discovery's close and "would have required the reopening of a case for which extensive discovery has been conducted and one that has been proceeding for far too long." 2018 WL 6985228, at *9 (quotation marks and citations omitted). Other decisions have denied motions to amend on similar grounds. See, e.g., Grochowski v. Phoenix Const., 318 F.3d 80, 86 (2d Cir. 2003) (upholding denial of motion to amend under Rules 15 and 16 where "plaintiffs delayed more than one year before seeking to amend their complaint [and] discovery had been completed"); Abdullah James George Wilson v. Corelogic Saferent, LCC, No. 14 Civ. 2477 (JPO), 2016 WL 482985, at *1-3 (S.D.N.Y. Feb. 8, 2016) (denying under Rules 15 and 16 motion for leave to amend filed six months after deadline for amendment where motion sought to add to class claim).

III. The PFAC's Claims as to 2008-2012

The Court first addresses plaintiffs' motion to amend to add claims as to 2008-2012—the primary focus of the new allegations in the PFAC.

The PFAC's core allegation as to this period remains that the Dealers collaborated to impede the emergence of electronic platforms permitting all-to-all IRS trading on an anonymous basis. To this end, the PFAC amplifies on the two episodes at the heart of the 2008-2012 allegations in earlier complaints. These involve (1) "Project Fusion," the joint venture by which the Dealers in 2007 and 2008 invested in and took control of Tradeweb, and allegedly induced it not to pursue all-to-all electronic IRS trading, PFAC ¶¶ 133-204; and (2) CME, whose Swapstream trading platform the Dealers allegedly collectively shunned in 2007-2008 and whose launch of an IRS clearinghouse usable to the buy-side the Dealers allegedly acted to stall in 2008-2012 in "Project Magellan," id. ¶¶ 205-96. These two episodes remain the centerpieces of plaintiffs' theories of how the Dealers, in 2008-2012, violated § 1.

The PFAC includes claims about two other joint ventures, in which Dealers participated with inter-dealer brokers ("IDBs")—ICAP and Tradition. The Dealers allegedly dissuaded these IDBs from supporting all-to-all anonymous trading. See id. ¶¶ 341-73 (Tradition); ¶¶ 394-413 (ICAP). Plaintiffs' earlier amended complaints contained allegations about ICAP and Tradition, but the PFAC's claims as to these entities are mostly new. The PFAC also newly alleges that the Dealers, in and after 2010, declined to support IDCG, a startup IRS clearinghouse. Id. ¶¶ 297-323. Finally, the PFAC newly alleges that the Dealers boycotted two trading platforms of nonparty Bloomberg (ALLQ and BSEF), with the intention to pressure Bloomberg not to open, following Dodd-Frank, a SEF permitting all-to-all IRS trading. Id. ¶¶ 324-40.

Addressing these actions thematically, the PFAC alleges that the Dealers over time pursued two main strategies. Id. ¶ 8. First, they made "`consortium investments' in critical IRS infrastructure," id. ¶ 8, such as Tradeweb and CME, with the goal of "mak[ing] sure those providers acted only to protect the dealers' interests," id. ¶ 9. Second, when that strategy proved unavailable or unsuccessful, the Dealers engaged in a group boycott of platforms such as Javelin, Tera, and/or trueEX, with the goal that the platform changed course or "withered on the vine." Id. ¶ 10.

A. Futility

Defendants argue that, for various reasons, the PFAC's 2008-2012 claims fail to state a claim and hence are futile. The Court here finds these claims futile in two respects. First, as before, class plaintiffs lack antitrust standing. Second, with one very narrow, arguable exception, the PFAC's 2008-2012 claims do not allege conduct that is per se illegal under § 1. The PFAC's principal theory, of per se liability, is thus not viable. In the interest of efficiency, the Court does not reach or resolve defendants' alternative arguments why these claims are futile. These include that the PFAC's claims as to Tradeweb, CME, and ICAP alleging conduct before November 25, 2011—and its claims as to IDCG, Tradition, and Bloomberg alleging conduct before October 25, 2014—are time-barred. In discussing, infra, the lack of good cause for amendment, the Court does, however, comment on the PFAC's new allegations of § 1 liability under the rule of reason.

1. Antitrust Standing

In both IRS I and IRS II, the Court held that class plaintiffs' theory of injury from the absence of electronic fora offering all-to-all anonymous trading was too speculative to make them "efficient enforcers" of the antitrust laws and thus give rise to antitrust standing, under cases including Associated General Contractors of California, Inc. v. California State Council of Carpenters, 459 U.S. 519 (1983), and Gatt Communications, Inc. v. PMC Associates, LLC, 711 F.3d 68 (2d Cir. 2013).

In IRS I, the Court explained:

IRS I, 261 F. Supp. 3d at 494. In IRS II, the Court rejected the PTAC's attempts to cure this problem, developing at greater length why plaintiffs' theory of investor injury was speculative:

IRS II, 2018 WL 2332069, at *10-11 (footnotes, including as to CFTC implementing rules for Dodd-Frank, omitted).

The PFAC tries again to fill this void. However, the Court's judgment, on close review of the new allegations and the documents from discovery on which these are based, is that the analysis above remains correct. The PFAC's theory of investor injury from the fact that an all-to-all marketplace permitting anonymous trading did not evolve to fruition during 2008-2012 still requires piling too many contingencies and assumptions. It remains far too conjectural to establish antitrust standing for class plaintiffs as efficient enforcers of the antitrust laws. This theory continues to stand in sharp contrast to class plaintiffs' allegations for 2013-2016, which involve a plot to boycott extant platforms, which have joined the investor plaintiffs in bringing this litigation against the Dealers.

The Court therefore affirms the analyses in IRS I and IRS II. The Court adds the following points, relevant to new material in the PFAC or arguments plaintiffs now make:

First, to a large extent, documents the PFAC newly cites reflect the unremarkable fact that, after Dodd-Frank, efforts were underway in the industry to build out the infrastructure necessary to support trading on SEFs consistent with Dodd-Frank's mandates. These included developing IRS clearing solutions accessible to the buy-side. See, e.g., Playforth Decl. Ex. 12 (email draft of CME clearing solution press release, Oct. 18, 2010) at 2 (identifying advisory group of 5 buy-side and 10 dealer participants in CME's clearing solution development, announcing its launch and "live test[ing]" that day with participation by PIMCO "and perhaps one other client," but adding: "[n]ote however there is material additional work required by the CME and the dealers before the platform is fully scalable"); id. Ex. 11 (Bank of America PowerPoint presentation, Dec. 6, 2010) (noting "only approximately 200 days before vanilla CDS and IRS will be mandated to cleared [sic] at an eligible clearinghouse," id. at 2, and noting CME's October 18, 2010 rollout, id. at 6); id. Ex. 10 (Barclays PowerPoint presentation, Dec. 6, 2011) at 5 (reporting the build-out of "BarCap's clearing platform" and "[c]urrently experiencing client ramp-up in OTC clearing ahead of mandatory dates" in a "Clearing update"); see also PFAC ¶ 446 & nn.535-36, 541 (citing all three documents).

Further, these documents, as defendants observe, reflect only the introduction of such clearing products and their usage on starter trades. See also, e.g., id. Ex. 9 (Citi OTC Derivatives Clearing Response to Caisse De Depot RFI, Dec. 6, 2011) at 19 ("Citi's current OTC Derivatives Clearing business . . . reflect[s] our leadership position, albeit in a market with largely ceremonial client activity. Citi expect [sic] this to increase significantly, given the levels of client engagement and mandates won, as clearing becomes mandatory."); PFAC ¶ 446 & nn.537-38 (citing same). They do not reveal the existence of meaningful buy-side clearing infrastructure before Dodd-Frank's mandates. Still less do they reveal the industry's practical readiness for all-to-all anonymous IRS trading independent of Dodd-Frank or before 2013.4 The fact that some buy-side clearing had begun as the industry moved to implement Dodd-Frank, see id. ¶ 446, similarly does not prove such readiness.

Second, the CFTC's rules and implementing regulations for Dodd-Frank underscore that major technological and infrastructural changes were needed to ready the industry for broad buy-side IRS clearing, that the industry was not ready for all-to-all anonymous IRS trading until long after Dodd-Frank's passage, and that Dodd-Frank's regulatory mandates were integral to achieving that readiness in 2013. In assessing whether plaintiffs' claim of injury is speculative, the Court takes notice of these historical events and the explanations given by the CFTC for its regulatory actions as reflecting the publicized condition of the market. See, e.g., Staer v. Hartford Fin. Servs. Grp., Inc., 547 F.3d 406, 425 (2d Cir. 2008) ("[I]t is proper to take judicial notice of the fact that . . . regulatory filings contained certain information [including] the publicized condition of the junk bond market during the relevant time period (citations omitted)); Cunningham v. Cornell Univ., 16 Civ. 6525 (PKC), 2017 WL 4358769, at *3 (S.D.N.Y. Sept. 29, 2017) ("Courts regularly take notice of publicly available documents during regulatory filings." (citations omitted)). Such is all the more appropriate here insofar as the PFAC draws upon CFTC rule-making and implementing regulations for Dodd-Frank, see, e.g., PFAC ¶¶ 449-50 & nn.547-53, and upon a buy-side commenter's submissions to the CFTC, see id. ¶ 442 & n.527; see also PTAC ¶ 333 & nn.124, 127-28 (citing buy-side comment letters on proposed CFTC rules).5

The CFTC's pronouncements reflect its determination that, even with Dodd-Frank's impetus, market infrastructure and buy-side participants were not yet ready for all-to-all SEF trading. These led to the CFTC's deferring until March 2013 the effective date of Dodd-Frank's mandate. See, e.g., Clearing Requirement Determination Under Section 2(h) of the CEA, 77 Fed. Reg. 74,284, 74,323 (Dec. 13, 2012) (IRS clearing requires "startup and ongoing costs relating to technology and infrastructure, new or updated legal agreements, . . . and costs related to collateralization of their positions."); id. at 74,324 ("In response to . . . concerns about legal documentation and operational readiness, the Commission has clarified that compliance with the clearing requirement will not be required for any swaps until March 11, 2013 . . . .").6 That there was "[r]obust [d]emand" from the buy-side for central clearing, PFAC ¶¶ 440-46, does not make less speculative the claim that, absent alleged misconduct, this costly infrastructure would have grown up in 2008, or 2009, or 2010, or any time before Dodd-Frank compelled it. Tasking a jury to gauge the stack of many unknowable questions embedded in the PFAC's alternative history— for example, whether a sufficient number of buy-side entities would have invested in such infrastructure and undertaken such collateralization to make all-to-all anonymous trading of IRSs a reality as of a particular date—would unavoidably invite an exercise in guesswork, not reliable fact-finding.

Third, in claiming to have cured the problem of speculative injury noted in the SAC and PTAC, plaintiffs emphasize the PFAC's allegations that Dealers acted with the intent to slow the evolution of certain offerings (of Tradeweb, CME, Tradition, Bloomberg, and ICAP) which had the potential to lead towards all-to-all trading. See, e.g., PFAC ¶¶ 19, 153, 230-39, 249, 325-27, 352-53, 392-93. But these intentions do not establish the market's readiness for such trading at any time, including 2008-2012. Some PFAC allegations, in fact, describe Dealer actions aimed at post-Dodd-Frank (and indeed post-2012) platforms. See, e.g., id. ¶¶ 325, 331-40 (allegations that Dealers, after Dodd-Frank, were concerned that Bloomberg would open a SEF permitting all-to-all IRS trading, and since 2012 have boycotted certain Bloomberg platforms); id. ¶¶ 344-73 (allegations that after Tradition announced in 2010 its intention to open a SEF-compliant platform, Dealers acted to take control over Tradition's SEF, did so by September 2013, and today govern it so as to frustrate all-to-all anonymous trading).

To be sure, the PFAC (even more than prior complaints) alleges instances in which Dealer personnel expressed hostility to all-to-all trading or acted with intent to discourage the market's evolution in that direction. But these do not show that, before 2013, the market was ready for this transformation. As the implementation of Dodd-Frank demonstrated, even when backed by regulatory mandates, the buy-side did not adapt on the CFTC's required timetable, necessitating an extension of the mandates until March 2013. Statements by Dealers in 2008-2012 fearing the eventual emergence of all-to-all trading do not mean that this market structure was then imminently viable.

The PFAC's claims as to 2008-2012 are therefore futile for lack of antitrust standing.

2. Claims of a Per Se § 1 Violation in 2008-2012

The PFAC, like plaintiffs' earlier complaints, dominantly pursues a § 1 theory of per se illegality. The Court first reviews its rulings as to the SAC's and PTAC's allegations in support of a per se theory, before assessing those in the PFAC.

a. The SAC

In IRS I, the Court sustained plaintiffs' per se theory as to 2013-2016, in that the SAC and JTSAC alleged "a recognized type of per se unlawful § 1 conspiracy: a group boycott," 261 F. Supp. 3d at 472, and parallel conduct and circumstantial evidence that made plausible the inference of a Dealer boycott aimed at the three new SEF platforms that offered all-to-all IRS trading, id. at 472-77. As to 2008-2012, however, the SAC's theory was based not on a boycott—for no all-to-all IRS trading platforms then existed—nor on meaningful parallel conduct, but largely on the Dealers' acquisition in 2007 alongside Thomson Corporation of a controlling stake in Tradeweb, allegedly in order to terminate a plan there to open such a platform.

The Court found the SAC's claim of a per se § 1 violation deficient for two reasons. First, its factual allegations as to the existence of such a plan at Tradeweb and of the Dealers' agreement to terminate it were hazy and conclusory. Id. at 466. Second, even assuming a well-pled agreement among Tradeweb's Dealer owners to terminate such a plan, the SAC did not plead facts under which an agreement would be per se unlawful:

Id. at 467 (citations and footnote omitted). The Court added:

Id. at 467-68 (citation and footnote omitted).

b. The PTAC

In the PTAC, plaintiffs based their theory of a per se violation of § 1 during 2008-2012 on two collaborations among Dealers.

The first was, again, "Project Fusion." But, the Court held in IRS II, the PTAC failed to rehabilitate the claims as to that arrangement. The PTAC had abandoned the SAC's conclusory claim that "`Tradeweb was planning to introduce electronic all-to-all trading to the IRS market,'" IRS II, 2018 WL 2332069, at *11 (quoting the SAC), replacing it with the claim that Tradeweb was gradually upgrading its swap trading capabilities and was "`primed to adopt all-to-all trading,'" id. (quoting PTAC ¶ 102), and that unspecified persons at Tradeweb had discussed doing so, id. (quoting PTAC ¶ 223). However, the Court held that "[t]hese reconfigured allegations about amorphous `discuss[ions]' and Tradeweb's being `primed' to adopt all-to-all trading, [PTAC] ¶ 102, again fall far short of plausibly alleging plaintiffs' `critical background premise': that Tradeweb was planning in 2007 to launch all-to-all IRS trading." Id. at *12 (quoting IRS I, 261 F. Supp. 3d at 466); see also id. ("The PTAC's theory of a plot to block a Tradeweb plan to introduce all-to-all IRS trading thus starts with the ill-pled premise that there ever was such a plan.").

In any event, the Court held, the PTAC failed to rectify the other pleading deficiencies in the SAC's "Project Fusion" allegations. The PTAC dubbed the "Project Fusion" joint venture a "sham" and a "formalistic shell." But the facts pled were inconsistent with these labels, including:

Id. at *12 (quoting IRS I, 261 F. Supp. 3d at 468 n.17). Accordingly, the Court held, "even assuming a well-pled agreement among owners to terminate a Tradeweb plan to open an all-to-all IRS trading platform, the Project Fusion joint venture as alleged remains subject to review only under the Rule of Reason." Id. (citation omitted).

The second basis for the PTAC's claim of per se illegal conduct in 2008-2012 involved CME. The PTAC alleged that, in 2006, CME had acquired Swapstream, a trading platform focused on a limited number of dealer-to-dealer IRS swaps in Europe; that CME had hoped to transform Swapstream into a worldwide electronic platform for all-to-all IRS trading; but that CME could not secure Dealer agreement to participate. Id. at *13. CME then began joint negotiations with the Dealers, who were collectively represented by Edward Rosen of the Cleary Gottlieb law firm. These negotiations, in which the Dealers sought "`governance'" over the platform, did not bear fruit. See id. (quoting PTAC ¶ 155). Eventually, CME initiated joint negotiations with the Dealers, again represented jointly by Rosen, regarding a "`clearing-only solution'" that would be "`open to the buy-side.'" See id. (quoting PTAC ¶¶ 160-61). These "Project Magellan" negotiations led to a written agreement in which Dealers agreed to provide minimum clearing volume to CME in exchange for revenue sharing. The PTAC alleged that the agreement contained a "`poison pill'" that effectively disabled CME from launching an execution platform, by allowing the Dealers to "`capture nearly all the profits.'" Id. (quoting PTAC ¶ 179).

In IRS II, the Court rejected this set of allegations, too, as a basis for a claim of a per se violation of § 1 during 2008-2012. Various PTAC factual allegations were contradicted by pled or cognizable facts. See id. at *14. These included the PTAC's implication that Swapstream had launched a trading platform, which a close reading revealed had not happened, and its errant characterization of the "Project Magellan" agreement to contain a "poison pill" provision to neuter any CME plan to launch an IRS trading platform. See id. ("By its plain terms, the Magellan Agreement did not `poison' CME's plans to launch an IRS clearinghouse."). The PTAC's discussion of CME also contained a significant number of conclusory or general allegations, including as to central allegations for the period before the Dealers' joint, counseled negotiations with CME had begun. Id. at *15.

As to the PTAC's well-pled allegations of collective action among Dealers regarding Swapstream, the Court noted, these "are centered on the period after joint negotiations between CME and the Dealers are alleged to have begun." Id. This activity, the Court held, did not fit within a category that courts applying § 1 had held susceptible to per se condemnation:

Id. at *15-16 (footnote omitted).

c. The PFAC

In pursuing a theory of per se § 1 illegality as to 2008-2012, the PFAC again principally relies on the Tradeweb and CME episodes, while also adding allegations about episodes relating to several other entities, discussed below.

As to Tradeweb and CME, the PFAC reorders and re-packages the PTAC's allegations— e.g., referring to the two episodes as embodying a "consortium strategy" across Dealers, see, e.g., PFAC ¶¶ 8, 141-43—but adds limited new factual allegations. These do not disturb the Court's earlier assessment that these complex dealings do not fall into a recognized category of conduct which the courts have condemned per se. The Court holds, as before, that any § 1 assessment of the balance between the pro-competitive purposes and anti-competitive effects of these ventures must instead be under the rule of reason.

Tradeweb: The PFAC claims the "Project Fusion" joint venture was illegitimate because participating Dealers were motivated, in gaining control of Tradeweb and in setting its course, to shape IRS market developments so as to reduce the possibility that exchange-like trading would emerge. See, e.g., PFAC ¶¶ 15-17, 142, 154-55; see also Pl. Mem. at 8 (collecting PFAC allegations as to Dealers' aim to, inter alia, "`negate[] the threat'" of IRS exchange trading, id. (quoting PFAC ¶ 155 (alteration in original)) and "`slow the inevitable slide to all-to-all,'" id. at 9 (quoting PFAC ¶ 191)). The PFAC claims that Dealers later ran Tradeweb "as a utility," preferring the interests of its Dealer majority owners over profit-maximization. PFAC ¶ 195. These allegations that the Dealers harbored such sectarian intentions track allegations in earlier complaints. See, e.g., IRS I, 261 F. Supp. 3d at 447-49 (discussing SAC and JTSAC).

The Court credits as well-pled that the Dealer personnel quoted in the PFAC had the intentions alleged: to steer Tradeweb's evolution, and to influence the evolution of the market, in a direction that maximized Dealers' economic interests, including to preserve trading margins. But whether per se condemnation under § 1 is available does not turn on subjective intent. The availability of the per se rule does not turn on case-specific determinations about the quality of evidence of (or pleadings as to) intent. It does not turn on the adjectives or metaphors used by competitors in emails. It is a categorical doctrine. Per se condemnation is available only when courts'"considerable experience" teaches that the category of restraint at issue has "manifestly anticompetitive effects" and "lack[s] any redeeming virtue." Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 551 U.S. 877, 886 (2007) (quotation marks and citations omitted); see also NYNEX Corp, v. Discon, Inc., 525 U.S. 128, 137-38 (1998) (declining to apply per se rule to alleged boycott claim despite allegation that defendants acted with a "special anti-competitive motive" to force the plaintiff out of business; Court did not "find a convincing reason why the presence of this special motive should lead to the application of the per se rule"); cf. Conn. Ironworkers Emp'rs' Ass'n v. New England Reg'l Council of Carpenters, 324 F.Supp.3d 293, 305 (D. Conn 2018) ("[A]llegedly pernicious motives on their own do not suffice to overcome the `presumption in favor of [the] rule-of-reason standard.'" (quoting Bus. Elecs. Corp. v. Sharp Elecs. Corp., 485 U.S. 717, 726 (1988)).

Here, for the reasons given in IRS I and IRS II, the collaboration alleged in "Project Fusion" does not correspond to any category of activity among horizontal competitors that courts to date have condemned per se. It cannot be condemned on its face as an act of as price-fixing, market allocation, or—in contrast to the Dealers' alleged actions in 2013-2016 towards the new opened SEFs—as a group boycott. Rather, its structure is that of a capitalized and operational joint venture. It cannot facially be condemned as an illegitimate sham.

The Court of course does not and cannot on the pleadings assess the net competitive impact of the "Project Fusion" joint venture. But whatever the evidence adduced on a well-pled rule of reason claim might have shown as to the balance of pro- and anti-competitive effects, plaintiffs' pleadings themselves reveal the venture's potential pro-competitive effects. These include, as noted, that after participants invested $280 million in Tradeweb's fledgling platform, Tradeweb expanded aspects of its trading operations. See IRS II, 2018 WL 2332069, at *12 (quoting IRS I, 261 F. Supp. 3d at 468 n.17).7 IRS I invited plaintiffs to "cite any case in which a decision by competitors to invest in or acquire control over a business, or to direct the activities of the business, in the context of a legitimate joint venture, has been evaluated under the per se standard." IRS I, 261 F. Supp. 3d at 467-68. Plaintiffs have not done so.8

CME: The PFAC's allegations, though re-sequenced, are in substance quite similar to those in the PTAC, including as to the successive rounds of joint negotiations between Dealers and CME on which the Court focused in IRS II. See PFAC ¶¶ 223-56 (first round followed by abandonment of Swapstream plan to open all-to-all platform); id. ¶¶ 257-93 (second round followed by execution of "Project Magellan" agreement, in which Dealers agreed to provide minimum clearing volume to CME in exchange for revenue sharing and opening of CME clearinghouse for buy-side IRS trades).

For the same reasons noted in IRS II, these business arrangements also do not resemble any that courts, based on considerable experience, have condemned as per se anti-competitive. See IRS II, 2018 WL 2332069, at *15 (collecting cases). Plaintiffs do not cite any case in which a court has per se condemned a horizontal collaboration that is structurally or thematically similar to that pled. Nor do they cite any case so condemning a business agreement meaningfully akin to the written, counseled agreement with CME that resulted from these negotiations. Further making per se condemnation unavailable is the presence, on the pleadings, of the potential pro-competitive justifications noted in IRS II for these joint negotiations and outcomes. See id.

To be sure, the PFAC adds concrete allegations as to the motives of participating Dealer personnel preceding and during the sequential joint discussions with CME. See, e.g., PFAC ¶¶ 230, 237, 239-43, 256, 293. As with Tradeweb, it is, in general, well-pled that these persons aspired to delay evolution of exchange trading, in favor of existing trading norms. But, again, as with Tradeweb, whether per se condemnation under § 1 is available turns on judicial experience with a category of business arrangement, not on participants' mentes reae. Otherwise, business arrangements that the courts have not condemned as per se and that may after discovery prove net pro-competitive could be held illegal at the threshold. This could deter economically worthwhile conduct. See, e.g., NYNEX, 525 U.S. at 137.

Otherwise, the main substantive change from the PTAC as to the CME narrative is that the PFAC drops the allegation that a "Project Magellan" agreement provision (§ 5.05) was a "poison pill"—an allegation the Court found inconsistent with the agreement's text. See IRS II, 2018 WL 2332069, at *14. The PFAC now assails § 5.05 as "so onerous in practice that it effectively prevents CME from launching an IRS trading platform," PFAC ¶ 279, on account of its requirement of consultation with Dealer founding members over trading protocols, id., and its "most favored nation" and "right of first refusal" clauses, id. ¶ 280; see also id. ¶¶ 281-82 (discussing the effect of these clauses). The parties debate whether these terms have potential pro-competitive justifications. See Def. Mem. at 32 n.26 ("obvious" justifications include "incentiviz[ing] the Magellan dealers to support the new IRS clearing house by granting them a potential right to participate in the upside if CME successfully leveraged the new clearinghouse into a vertically integrated trading platform" (citing federal antitrust guidelines)); Pl. Reply at 22 n.43 (distinguishing guidelines as inapposite). In assessing the availability of per se condemnation of "Project Magellan," the Court, however, takes a holistic focus. For the reasons stated in IRS II, the overall agreement negotiated among the Dealers and CME is properly evaluated under the rule of reason. Whether any provision of that agreement, including § 5.05, is pro- or anti-competitive would form part of the rule of reason inquiry.

"A unitary conspiracy": Plaintiffs separately argue that they have alleged a "unitary conspiracy" that predated 2013 and should not be "forced to prosecute a truncated conspiracy." Pl. Reply at 1, 3. The import of this argument is not clear. The Court construes plaintiffs to argue that because they have alleged a viable claim of per se unlawful conduct for 2013-2016, and because they have also alleged collaboration among Dealers in the preceding five years, they are entitled to application of the per se rule for their allegations as to collaboration in 2008-2012.

To the extent this argument seeks to relieve plaintiffs of the pleading burdens imposed by the rule of reason for the Dealers' pre-2013 business arrangements as alleged, it is wrong. The character of the Dealer collaboration alleged in 2008-2012 is fundamentally different from that alleged in 2013-2016. For 2013-2016, plaintiffs have alleged conduct among horizontal competitors (a group boycott aimed at extant platforms) of a kind that is quintessentially unlawful per se. It does not follow, however, that earlier agreements among Dealers of a different character (e.g., the "Project Fusion" and "Project Magellan" joint ventures)—even if motivated by a common desire to have the market maintain name-disclosed RFQ trading and not evolve towards anonymous all-to-all trading—were therefore per se illegal. Plaintiffs have not shown that those earlier joint arrangements are of a kind that judicial experience has condemned as per se unlawful. To be sure, pre-2013 dealings among Dealers may prove admissible at a trial to shed light, inter alia, on the existence or nature of agreements among them during 2013-2016. But plaintiffs have not offered any authority why the fact of viably pled claims of per se illegal conduct in 2013-2016 alters the analysis that antitrust law uses to assess earlier agreements of a different character. See, e.g., City of Groton v. Conn. Light & Power Co., 662 F.2d 921, 928 (2d Cir. 1981) (even where alleged conduct is "interrelated and interdependent," courts must "analyze the various issues individually").

ICAP, Tradition, Bloomberg, and IDCG: Finally, although the Tradeweb and CME episodes remain the heart of plaintiffs' pre-2013 claims, the PFAC alleges joint Dealer conduct towards four other entities: ICAP, Tradition, Bloomberg, and IDCG. The Court elaborates on these allegations later, in the course of discussing delay and prejudice. The PFAC's allegations as to these episodes do not, however, warrant sustaining the claim of a per se violation of § 1 spanning 2008-2012.

As pled, two of the four episodes—those involving ICAP and Tradition—centrally involve a facially legitimate joint venture between Dealers and an inter-dealer broker. As such, these arrangements cannot be condemned per se, even if the motivations of participants were, as plaintiffs claim, to "[m]aintain the [s]tatus [q]uo." Pl. Mem. at 26 (quotation marks omitted). Under the authorities above, they are to be evaluated under the rule of reason. As to ICAP, the PFAC alleges that ICAP developed an electronic trading platform (i-Swap) for European dealerto-dealer trading. PFAC ¶ 397. It initially achieved "no traction" but, in 2011, ICAP, in "Project Snowflake," recruited three Dealers (Barclays, Deutsche Bank, JP Morgan) to support (by streaming prices to it) a revamped i-Swap trading platform in exchange for an equity stake and governance rights. Id. ¶ 402. i-Swap launched in the United States when the new SEF rules took effect in 2013. See id. ¶ 410. Plaintiffs allege that these Dealers have used their influence to induce ICAP not to open an all-to-all trading platform. See id. ¶¶ 394-413. As to Tradition, the PFAC alleges that other Dealers entered into a "Trad-X" joint venture with Tradition, in which they too agreed to support a new Trad-X trading platform in exchange for equity and governance rights. See id. ¶¶ 341-73. Trad-X launched in the United States in 2013. See id. ¶ 357.9

The PFAC's new allegations about a plot against non-party Bloomberg also do not support a per se claim for 2008-2012. The PFAC's allegations as to Bloomberg overwhelmingly relate to 2013-2016. Specifically, the PFAC alleges that Dealer personnel met with Bloomberg in late 2011 and early 2012 to discuss Bloomberg's forthcoming SEF and conveyed that they might support Tradeweb if a partnership between Bloomberg with the Dealers were not arranged. Id. ¶¶ 325-36. It alleges that, in 2013, Bloomberg decided to launch an anonymous CLOB, known as BSEF, which allowed RFQ trading and received CFTC approval, but that the Dealers, viewing BSEF as a threat to develop into an all-to-all trading platform, today "refuse to provide liquidity" to BSEF. Id. ¶ 340; see id. ¶¶ 338-40. The PFAC's only allegation as to Bloomberg services offered before 2013 is that, in 2011, Bloomberg extended an OTC trading screen called ALLQ to IRSs, id. ¶ 326, but that Goldman Sachs, "`along with several other banks,'" withheld quote streams from ALLQ, id. ¶ 330 (quoting Playforth Decl. Ex. 43 (Goldman Sachs offsite management committee meeting slide notes) at 4) (emphasis omitted). The PFAC does not, however, allege any other parallel conduct or plus factors indicating that the decision by Goldman Sachs and other unspecified Dealers not to do business with ALLQ was coordinated. And, as defendants note, a document quoted in the PFAC appears to contradict this claim of Dealer non-participation, stating that "`7 of the top 10 [dealers] "opted in'" to ALLQ at its launch, and by 2013, `[a]ll dealers [were] live' on ALLQ." Def. Mem. at 42 & n.39 (quoting Playforth Decl. Exs. 40, 41).10

Of the four newly alleged episodes, the PFAC's claims alleging a boycott are strongest as to IDCG. The PFAC alleges that IDCG was a small clearinghouse which in 2007 developed a promising clearing solution but which by 2011 had not "gain[ed] any traction." PFAC ¶ 314 (quotation marks and footnote omitted). The PFAC does not plead any non-conclusory claim of concerted Dealer activity towards IDCG until July 2010. That Dealers did not use IDCG before Dodd-Frank, despite pitches IDCG had made to them, id. ¶ 305, does not support inferring collective action. See IRS II, 2018 WL 2332069, at *17 ("[P]arallel lack of interest is not indicative of an inter-Dealer agreement" because, "[a]s pled, such conduct was consistent with any Dealer's independent self-interest." (citation omitted)). However, the PFAC claims, in July 2010, after IDCG cleared a starter trade with emerging SEF Javelin, PFAC ¶ 308, Dealers exchanged emails "trying to figure out who was involved in Javelin and IDCG," id. ¶ 309-10. Thereafter, on August 11, 2010, a "bank group" meeting was held at Deutsche Bank involving representatives of six Dealers. Id. ¶ 311. The notes from the meeting state, inter alia, that "`banks will oppose moves which move significant volumes to exchanges.'" Id. (quoting Playforth Decl. Ex. 38 at 2). Thereafter, the PFAC alleges, Dealers "kept up their engagement with IDCG on the surface," id. ¶ 315, but continued to give it limited support until its sale in 2012, id. ¶¶ 322-23.

Whether this plausibly pleads a per se agreement among the six Dealers thereafter to boycott IDCG presents a close question. On the one hand, the PFAC does not allege that these Dealers changed their behavior towards IDCG: both before and afterwards, as alleged, they gave limited support to IDCG. And, as defendants note, the notes from the August 11, 2010 meeting, in the main, address a different matter entirely from IDCG. They describe a meeting between LCH and its board-member banks whose purpose was to consider "how LCH could succeed in the US market." Playforth Decl. Ex. 38 at 1. They mention IDCG solely as one of a number of potential U.S. "partners" for LCH; the notes' reference to opposing "mov[ing] significant volumes to exchanges" appears in a list of general topics as to which "[t]here was broad agreement," id. at 2; see also id. at 2-3 (stating that the "discussion focused on the need for and attractiveness of US partners" for LCH; listing points of "agreement" among LCH and its board about potential partners; identifying five "potential partners," including IDCG, and indicating that one "[n]ext step[]" was to "investigate" IDCG). No aspect of the meeting notes other than the excerpt quoted in the PFAC supports the PFAC's theory. On the other hand, that excerpt is certainly suggestive. Significant, too, is that IDCG had just provided clearing assistance on a maiden trade the month earlier to Javelin. The Court has held, in IRS I, that plaintiffs have adequately pled a plot among Dealers to boycott and destroy Javelin's post-Dodd-Frank SEF. The PFAC's claim that certain Dealers exchanged emails about Javelin and IDCG after IDCG assisted with Javelin's starter trade makes more plausible that the Dealers that met thereafter and discussed IDCG agreed to shun it on account of its assistance to Javelin. The Court thus will assume arguendo that a plausible agreement has been pled among these Dealers effective August 2010 and lasting until IDCG's purchase in 2012 to continue not to utilize (i.e., to boycott) IDCG's services.11

However, this strand is alone among the many facts that the PFAC pleads as to 2008-2012 in (arguably) plausibly pleading an instance of a per se unlawful agreement. No others do so. The IDCG episode as pled involves an agreement among a subset of Dealers from August 2010 until 2012 to continue to make limited use of a single clearinghouse that had theretofore been largely in a state of disuse. It does not and cannot support the sweeping claim of per se illegality pursued by the PFAC as to 2008-2012: of a five-year agreement among all Dealer Defendants to boycott a host of concerns with some potential to one day facilitate all-to-all anonymous trading.

The Court, therefore, holds that to the extent that the PFAC claims that defendants' conduct during 2008-2012 was per se illegal, it is futile.

B. Delay, Prejudice, and Absence of Good Cause

The Court next considers the related questions whether granting leave to add the PFAC's claims covering 2008 through 2012 would (1) unduly delay this litigation and (2) cause prejudice to the defense. After considered review, the Court holds yes as to both. The scale of delay and prejudice would far exceed that which permitting the PTAC's claims as to these five years would have caused. The Court further holds that plaintiffs have not shown good cause for pursuing these claims at this late date.

1. Delay

The issue of delay is framed by the chronology of this litigation.

Fact discovery began upon the July 28, 2017 decision in IRS I. Today, the deadline for fact discovery is April 10, 2019, some four weeks from now. Dkt. 549.

Throughout, fact discovery has entailed extended negotiations among the parties, and occasionally with third parties, as to scope and parameters. It has entailed resolution by the Court of intermittent disputes regarding such discovery. See, e.g., Dkt. 266 (October 10, 2017 decision); Dkt. 296 (October 24, 2017 decision); Dkt. 322 (January 8, 2018 decision); Dkt. 492 (August 22, 2018 decision); Dkt. 494 (August 22, 2018 decision); Dkt. 550 (September 21, 2018 decision); Dkt. 583 (October 19, 2018 decision); Dkt. 618 (November 13, 2018 decision); Dkt. 686 (January 23, 2019 decision). The Court has held monthly status calls with counsel, beginning in November 2017. Dkt. 268. These have enabled the Court informally to resolve, or guide the resolution of, many other discovery disputes (some concrete and some inchoate). These calls have kept the Court abreast of the discovery process.

Early in the fact discovery process, the Court resolved a dispute particularly important here: as to the start date of document discovery for the surviving (2013-2016) claims. Dkt. 266 (October 10, 2017 decision). Class plaintiffs had sought an order requiring defendants to produce documents created on or after January 1, 2009; defendants had proposed a start date of January 1, 2012. See Dkt. 265 (defendants' letter). The Court chose an intermediate start date: July 21, 2010, the date of Dodd-Frank's passage. Dkt. 266 at 2. This date, the Court held, balanced plaintiffs' interest in discovering defendants'"pre-conspiracy dealings" against the burdens of obliging the 11 defendant groups to make pre-conspiracy discovery. Id. Explaining why documents created after July 21, 2010 were more likely relevant than those created before, the Court stated: "[A]fter Dodd-Frank was enacted, it became materially more foreseeable that electronic all-to-all trading platforms for IRSs would one day develop." Id. The July 21, 2010 start date framed the parties' ensuing negotiations and many ensuing rulings on fact discovery disputes. These included an omnibus ruling on January 8, 2018, which resolved whether, and to what extent, a host of topics pertinent to events in 2007-2012 that plaintiffs sought to explore were relevant to the surviving claims. Dkt. 322.

Under the court-approved case management plan, the deadline for substantial completion of document discovery, as extended, was June 29, 2018.12 As of December 2018, the parties and third parties had produced more than four million documents. See Brockett Decl. ¶ 6 & n.1. In addition, substantial data compilations, largely capturing the terms of IRS trades between 2013 and 2016, have been produced from multiple sources.

On October 26, 2018, plaintiffs filed the pending motion for leave to file the PFAC. Dkts. 590-92. The motion was filed more than eight months after the February 21, 2018 deadline set for amended pleadings in the case management plan. Briefing on the motion was completed in late December 2018, 10 months after that deadline.

Fact depositions are nearing completion. The Court set by order a cap on each side's fact depositions. The Court, inter alia, authorized plaintiffs to take 110 depositions of fact witnesses across the 11 defendant groups and an additional five Rule 30(b)(6) depositions on data issues; and authorized defendants to take 30 depositions of fact witnesses across the three platform plaintiffs. Dkt. 583. As of the parties' February 20, 2019 joint status report, plaintiffs had taken or noticed 86 depositions of witnesses affiliated with defendants, defendants had taken 25 depositions of witnesses affiliated with plaintiffs, and an additional 25 depositions had been taken or noticed of third-party witnesses.

Also underway, consistent with the case management plan, is briefing on plaintiffs' motion for class certification. See Dkt. 522 (briefing schedule). On February 20, 2019, plaintiffs submitted their opening motion and expert reports. See Dkts. 722-25. Defendants are to depose plaintiffs' class certification experts by April 10, 2019; defendants are to submit opposition briefs and supporting expert reports by May 31, 2019; plaintiffs are to depose defendants' experts by June 28, 2019, and to submit replies in support of their motion by August 16, 2019. See Dkt. 522.

Particularly viewed in light of the mature stage of this litigation, permitting the PFAC to supplant the current pleadings would upend the schedule and introduce enormous delay. The PFAC would more than double the temporal duration covered by plaintiffs' claims, adding five years to the current four. It would introduce a vast array of events, entities, concepts, witnesses, custodians, trades, and data not presently in the case. Given the nature of the PFAC's claims as to defendants' alleged conduct during 2008-2012, it would also require the testing of a complex and layered theory of § 1 liability that could proceed at most under the rule of reason, markedly different from the far simpler per se theory of § 1 liability, involving a boycott of three new trading platforms, that plaintiffs have pursued as to 2013-2016. For these reasons, adding the PFAC would effectively re-start the discovery process in this long-running litigation.

Briefly:

As to topics, the PFAC would inject into this litigation, among others, the following five episodes. The Court has referred to them earlier in the context of analyzing whether a per se violation of § 1 has been adequately pled. The following synopses drawn from the PFAC underscore that each is complex, involved numerous participants, including a large number of non-parties to this lawsuit, played out over years, and transpired largely if not exclusively before the passage of Dodd-Frank:

In addition to these episodes,13 the PFAC's claim of a plot to inhibit, pre-Dodd-Frank, the emergence of anonymous all-to-all electronic trading would inject the following three thematic topics into the case. Each, too, is complex, and implicates a large array of new concepts, events, persons (including third parties, and records).

As the above review makes abundantly clear, adequate exploration of these alleged pre-Dodd-Frank episodes and topics would entail a vast amount of new discovery—documentary, testimonial, and expert. The Court so observed in noting the substantial delay that the PTAC would cause. See IRS II, 2018 WL 2332069, at *20-22. The Court's assessment is that the delay the PFAC would occasion would be far greater, given the expanded number of complex episodes and the topics it implicates, and given that more than a year of fact discovery has passed since the decision denying leave to add the PTAC's claims as to 2008-2012.

As to document collection, review, and production, as the Court noted in IRS II, "the parties' agreements as to document production parameters in 2013-2016 would not, logically, control discovery relevant to the 2008-2012 claims." Id. at *20. Document production in this case is nearly complete. As to document custodians, given the PFAC's new claims, collection of electronic and hard-copy records would need to re-start, and would now need to reach back at least to the start of 2008 (if not earlier: the PFAC contains many allegations as to circumstances in 2006 and 2007, and some as to earlier years). Records retrieval would also assuredly be necessary for a very sizable new group of custodians. That is because, "at particular Dealers, given natural turnover and promotions and reassignments, there were assuredly different personnel responsible for IRS-related matters during the two periods," id. at 21, and because the many issues and incidents novel to the earlier period "likely involved substantially different personnel at particular Dealers from those as to whom discovery has been authorized as to 2013-2016," id.

As to the delineation of topics and search terms that would guide document collection and review of materials germane to the new five years of claims, much as the Court noted in IRS II:

Id.

As to fact depositions, they too are close to complete. That plaintiffs would seek substantial new and reopened depositions of past and present defense personnel is inevitable, given the multi-chapter, multi-dimensional nature of the § 1 scheme among Dealers that the PFAC postulates, and that the PFAC names many dozens of Dealer personnel, including in connection with the episodes and topic areas synopsized above. Unavoidably, adding claims as to 2008-2012 would occasion, and justify, a new round of depositions. The Court's expectation is that between new and reopened depositions, the PFAC's added claims would credibly support depositions of approximately the same number of defense deponents as has been authorized for the existing claims.

The PFAC's expanded claims also implicate third parties. The Court's description of the PTAC applies: "Those identified on the face of the [PFAC] include numerous clearinghouses, interdealer brokers, other service providers, and industry associations." Id. The number of third-party entities implicated for 2008-2012 is, in fact, greater in the PFAC than the PTAC, given its expanded claims for this period. The number of such third parties vastly outstrips those implicated by plaintiffs' 2013-2016 claims, consistent with plaintiffs' simpler theory of § 1 liability for the later period (a boycott of three existing SEFs). Extensive and protracted third-party discovery—both documentary and deposition—would certainly be sought by both sides were the PFAC's 2008-2012 claims permitted to go forward. There would also assuredly be no shortage of disputes about the scale of such discovery (and how the expense occasioned by it is to be borne). Even as to the 2013-2016 claims, the Court has already been called upon on multiple occasions to resolve or pretermit discovery disputes regarding third-party discovery. See, e.g., Dkts. 456 (Tradition), 550 (Tradition), 609 (ICAP and Tullett).

Finally, as to data, as the Court has previously noted:

IRS II, 2018 WL 2332069, at *21. The same need to identify, harvest, and review these data sets applies to the PFAC.

In IRS II, the Court's assessment was that

Id. *21 (footnotes omitted). Considering related factors including the need to elongate deposition discovery, the Court's conservative estimate was that permitting the PTAC's 2008-2012 claims to stand would cause a delay of, at the very least, 12 months for the completion of fact discovery. Id. ¶ 22.

The Court's assessment here is that, relative to the PTAC, permitting the PFAC's 2008-2012 claims to stand would occasion a materially longer delay of the case schedule. Unlike at the time of IRS II, fact discovery today is close to complete. Even were fact discovery on the 2008-2012 claims to commence immediately, it would virtually all occur after the current April 10, 2019 deadline for fact discovery. And with document production substantially complete and most depositions taken, there are limited opportunities to achieve efficiencies or economies of scale across discovery for the 2008-2012 and 2013-2016 events.

In effect, adding the 2008-2012 claims would again "allow[] a new MDL-sized lawsuit to be hitched to the existing claims." Id. at *20. And with fact discovery all but complete for 2013-2016, the processes of organizing and taking fact discovery for these two lawsuits would effectively proceed seriatim, with the 2013-2016 discovery wrapping up shortly after 2008-2012 fact discovery commenced. With limited efficiencies to be gained, fact discovery on the 2008-2012 claims would thus be tantamount to undertaking fact discovery in a new litigation.

In estimating the added length of fact discovery were the PFAC's 2008-2012 claims permitted, a fair starting point is the length of time such discovery will ultimately have taken as to plaintiffs' 2013-2016 claims. From the date of IRS I (July 28, 2017) to April 10, 2019 was 20.5 months. To more precisely calibrate the extent of the delay in fact discovery that the PFAC would occasion, certain adjustments would need to be made. On the one hand, two factors would tend to reduce (at the margin) the extent to which the 2008-2012 claims would extend the fact discovery deadline. First, four weeks remain before the current fact discovery deadline. Second, as a result of the initial (2013-2016) round of fact discovery, both sides' counsel are doubtless more familiar with the concepts and terminology in this area, and with one another. This would facilitate discovery requests and negotiations. On the other hand, two factors would significantly elongate 2008-2012 discovery relative to 2013-2016 discovery. First, the PFAC's claims as to 2008-2012 are materially more complex and more difficult to discover than those for 2013-2016. These claims implicate more events, older events, older and harder to secure evidence, and—most importantly—far more varied episodes and third parties than do the 2013-2016 claims. Second, assuming arguendo that plaintiffs' § 1 claims as to 2008-2012 were held viable, the Court's assessment is that these claims, which largely involve claims related to the Dealers'"consortium investments," could at most state a claim under the rule of reason, in contrast to plaintiffs' 2013-2016 claims of a per se unlawful boycott of the nascent SEF platforms. The issues implicated by a rule of reason claim (e.g., as to the product and geographic markets, as to product substitutes and barriers to entry, and as to the anti-competitive effects versus the pro-competitive benefits of the challenged practices) would complicate and measurably add to the duration of fact discovery. See discussion infra p. 61.

All in, the Court's best estimate—based on its experience overseeing this litigation to date, its broader experience involving complex civil litigation, and its assessment of the unique features of the 2008-2012 claims—is that fact discovery attributable to the PFAC's 2008-2012 claims would, conservatively, add approximately 18-21 months to the current fact discovery deadline, and quite possibly more. Adding these claims would also further require a re-setting of all ensuing deadlines in the case management plan (e.g., class certification), with these deadlines presumably keyed to the new end dates for fact discovery.

In arriving at this estimate, the Court notes that plaintiffs, strikingly, do not engage at all with the implications of their claims for relevant fact discovery. They do not argue, for example, that the discovery addressed above would or could be provided with materially quicker dispatch. Nor, given the instructive experience of this litigation to date, could they credibly do so. Instead, plaintiffs declare that—as long as the defendants are ordered to produce to them productions that they have made to the CFTC in connection with a regulatory investigation—plaintiffs will "agree[]" not to pursue further document discovery and to a fact discovery extension of just three months. Pl. Mem. at 6; see also Pl. Reply at 6.

For a host of reasons, this bid is a non-starter. The Court has rejected plaintiffs' attempt to gain defendants' document productions in the CFTC's regulatory investigation (and in another civil litigation). See Dkt. 266. In the interests of all parties, and consistent with the integrity of the truth-seeking process, the metes and bounds of document production here are to be measured against the viable claims pursued in this litigation. A party's regulatory production may overlap only slightly with pertinent records in a civil litigation. And it may include materials outside the proper scope of discovery here, which plaintiffs have no right to access.14

Plaintiffs' suggestion that a fact discovery extension for the 2008-2012 claims can be cabined to three months is also inconsistent with their acknowledgement (in a footnote) that, if leave to amend is granted, they intend to pursue third-party discovery (and presumably third-party trading data). See Pl. Mem. at 6 n.18. Notably, too, plaintiffs have not foresworn deposition discovery, of defendants or third parties. This, too, makes facetious the estimate that a bare three-month fact discovery extension would suffice. Plaintiffs' request that they be allowed "full discovery" into the viability of all-to-all IRS trading platforms before 2013, so as to meet the defense that the investor plaintiffs' injuries are too speculative to supply antitrust standing, id. at 21, is also inconsistent with the notion of a three-month extension.

Finally, plaintiffs ignore defendants' right to discovery. Even if plaintiffs were taken to have unilaterally disarmed from any further fact discovery, defendants would have the right to pursue, presumably via new or amended subpoenas, third-party document discovery, data, and depositions germane to the added claims. And, for the reasons addressed, third-party discovery, by whichever party sought, is apt to be substantial and time-consuming. The Court would expect substantial third-party discovery to be directed by defendants, at a minimum, to the entities above whom the Dealers are alleged to have impaired, pressured, or boycotted; and to the buy-side entities whom the PFAC claims were deprived of an otherwise available and desired opportunity to trade IRSs anonymously on an all-to-all electronic platform.15 See, e.g., PFAC ¶¶ 220-21, 303-04, 442 (listing such buy-side entities). Finally, to meet plaintiffs' broad claims as to 2008-2012, defendants would be expected to review and harvest their own records, including employee emails, covering the years of the new claims. Production of materials from among these records—e.g., in support of defense motions on summary judgment or opposing class certification or in connection with a joint pretrial order—would assuredly be met with demands for production of a fuller set.

Accordingly, plaintiffs' proposal to forgo certain document discovery is not a credible means of addressing the substantial delay that the PFAC would produce.

Plaintiffs, lastly, ask that, as an alternative to permitting the claims as to 2008-2012, the Court permit plaintiffs to pursue claims from the date of Dodd-Frank's enactment forward. This proposal is, presumably, keyed to the start date of document discovery. But this, too, is not a coherent means of redressing delay. As plaintiffs acknowledge, even if the start date of the class period were moved forward 30 months from January 1, 2008 to July 21, 2010, discovery would still be needed as to "the origins and impact of Defendants' conspiracy in the 2007-2010 period." Pl. Mem. at 6 n.19. That is because the alleged stratagems on which plaintiffs base their claims of a pre-2013 § 1 violation—e.g., Tradeweb/"Project Fusion," CME/"Project Magellan," ICAP/"Project Snowflake," IDCG, and Tradition—predate Dodd-Frank and often 2008. While plaintiffs' proposal would prune temporally the evidence bearing on investor damages, on plaintiffs' theory, substantially the same discovery would be needed to test issues bearing on § 1 liability, including the existence, origins, scope, and participation in the alleged scheme; whether investor plaintiffs' injuries are sufficiently non-speculative to support antitrust standing; and whether the elements of a rule of reason claim are met.

The Court, accordingly, finds that permitting plaintiffs' 2008-2012 claims to proceed would cause substantial delay, realistically 18-21 months or more, to this litigation.

2. Prejudice

Drawing on its experience closely supervising pretrial proceedings in this litigation, the Court also finds that the amendment to add the 2008-2012 claims, even more than the proposed amendment in the PTAC, would palpably prejudice the defendants. Cf. IRS II, 2018 WL 2332069, at *22-23.

In IRS II, the Court noted that as of the date of the PTAC, defendants had for seven months "expended time, money, and energy in the discovery phase of this litigation on the only rational assumption as to this case's temporal scope: that plaintiffs' surviving claims were limited to 2013-2016." Id. at *22. The Court noted that, based on that temporal premise, counsel had exchanged hundreds of discovery communications, held dozens of meet-andconfers, and "devoted enormous effort to reaching agreement on matters such as the as scope of discovery, search terms, and custodians." Id. (quotation marks and citation omitted). Counsel had also litigated a series of discovery disputes that turned on this time period, including those that involved the document-discovery start date and the proper subjects of document collection and review.

In finding prejudice, the Court noted:

Id. (internal citation and footnote omitted). However:

Id. (footnote omitted).

These considerations apply with even greater force more than a year after IRS II. As of February 2018, at least some efficiencies remained achievable via coordinating discovery (document and deposition) on plaintiffs' 2008-2012 claims with as-yet unfinished discovery on the 2013-2016 claims. But today, those efficiencies are realistically all but gone, with document discovery complete save for loose ends, and deposition discovery due to close in under a month.

Were the PFAC's claims to be permitted, the substantial costs and burdens borne by defendants would therefore include ones that would have not existed had plaintiffs filed the PFAC within—or even close to—the February 21, 2018 deadline set in the case management plan. For each of the 11 defendant groups, the cumbersome, time-intensive, and costly processes associated with modern document discovery would need to begin from even closer to a "standing start," id. at *20, than in the case of the PTAC.

There would first be the "laborious process of framing the document discovery to be produced for that period . . . ." Id. Much time and expense would assuredly be spent on new rounds of negotiating search terms and protocols keyed to the PFAC's new claims, including testing these terms and protocols based on hits yielded by experimental search terms. This process would be all the more involved and time-consuming given the complex, multi-faceted nature of the conduct that the PFAC claims was part of an overall § 1 conspiracy during the years before SEFs took root in 2013 pursuant to Dodd-Frank. Substantial time would inevitably be spent, too, litigating discovery disputes. Once parameters were in place, each defendant would then need to identify, find, and collect millions of electronic (and hard copy) records from or for numerous custodians; retain and train a new set of contract lawyers about this litigation and the review protocols specific to it; put in place the mechanical apparati for such review; and arrange and train teams for second-tier production tasks (e.g., privilege reviews).

As to depositions, plaintiffs have taken or noticed most allotted for defense personnel. And plaintiffs would undoubtedly demand new and reopened depositions to account for the five years of added claims. The Court would surely permit additional depositions, to allow for fair discovery of areas of conduct newly relevant in light of the PFAC. But the untimely date at which depositions as to events and concepts relevant to pre-2013 claims were decided upon too would prejudice defendants, relative to the scenario in which overall discovery was negotiated more holistically. Addressing all forms of discovery in IRS II, the Court noted that:

Id. at *21. Today, with deposition and document discovery largely complete as to 2013-2016, defendants have lost the ability to negotiate or move, in the wake of the added claims, for a reduced scope of discovery as to 2013-2016 in the interests of keeping manageable the overall discovery burden.

The Court therefore assesses that the tangible prejudice to defendants from plaintiffs' late amendment would substantially exceed the scale of prejudice estimated in IRS II from permitting the PTAC's similar amendment.

Plaintiffs, notably, do not challenge the account in IRS II as to the costs and burdens that adding the 2008-2012 claims would then have added. Nor do they dispute defendants' statement that adding such claims now would add to these costs and burdens. That is unavoidably correct. Plaintiffs instead argue that as a matter of law, the "time, effort and money" that a party's late amendment visits on its adversary cannot alone qualify as "prejudice." Pl. Mem. at 23-24 (quotation marks and citation omitted). Plaintiffs claim that only "legal prejudice" is cognizable, and that a defendant cannot show this where "the proposed amendment is simply variations on the original theme." Id. at 23 (quotation marks and citations omitted).

Plaintiffs incorrectly portray the law. As the authorities reviewed above recognize, an amendment may be prejudicial when it would require an opponent "`to expend significant additional resources to conduct discovery. . . .'" AEP Energy Servs., 626 F.3d at 725 (quoting State Teachers Ret. Bd., 654 F.2d at 856). One case on which plaintiffs rely says precisely that. See Pasternack v. Shrader, 863 F.3d 162, 174 (2d Cir. 2017) ("A litigant may be `prejudiced' within the meaning of [Rule 15(a)] if the new claim would: `(i) require the opponent to expend significant additional resources to conduct discovery and prepare for trial [or] (ii) significantly delay the resolution of the dispute. . .'").

To be sure, in the two Second Circuit cases on which plaintiffs rely, leave to amend was held wrongly denied based solely on delay and associated litigation expenses. But these are easily distinguished. Miller v. Selsky, 234 F.3d 1262 (2d Cir. 2000), an unpublished opinion, solely involved Rule 15. Id. at *2. It did not involve Rule 16. And no discovery had taken place as of the date of the requested amendment. Id. Pasternack, too, was resolved only under Rule 15. The Circuit's decision does not address Rule 16. On the contrary, it states: "[I]t does not appear that there is any allegation of untimeliness based on a scheduling order." 863 F.3d at 174. The Circuit held only that the trial court had abused its discretion when it denied leave to amend because the litigation had lasted years and defendants had spent "a vast amount of money litigating the sufficiency of various complaints in this case." Id. (quotation marks and citation omitted). These cases do not bear on the situation here, where—long after the lapse of the amendment deadline and with fact discovery near complete—plaintiffs seek to add claims that would more than double the case's scope, far more than double its complexity, and require the re-starting of discovery at great cost and inefficiency.16

In fact, courts in this District have repeatedly found prejudice where motions to amend were filed late and after extensive discovery. See, e.g., Levy v. Young Adult Inst., No. 13 Civ. 2861 (JPO) (SN), 2016 WL 3637109, at *3 (S.D.N.Y. June 29, 2016) (denying leave to amend under Rule 16 when deadline to amend complaint had passed 2 years before); Eastman Chem. Co. v. Nestle Waters Mgmt. & Tech., No. 11 Civ. 2589 (KPF), 2014 WL 1316772, at *5-10 (S.D.N.Y. Apr. 1, 2014) (finding prejudice in a motion to amend two-and-a-half months after the deadline where amendment would have required, inter alia, a whole new area of discovery on "commercial reasonableness"); Pereira v. Cogan, No. 00 Civ. 619 (RJS), 2002 WL 31496224, at *3-4 (S.D.N.Y. Nov. 8, 2002) (denying leave to amend after finding prejudice under Rule 15 where the motion to amend was filed 5 days before trial).

3. Good Cause

The Court, finally, addresses whether plaintiffs had good cause to move for leave to file the PFAC eight months after the deadline for amended pleadings. As briefed, this question turns on whether new discovery justified plaintiffs' late filing. Plaintiffs argue that the PFAC is based on "newly discovered evidence"—documents produced that revealed a viable § 1 claim. Pl. Mem. at 6; see also Pl. Reply at 3-5; Brockett Decl. ¶¶ 2-15 (describing counsels' pre-PFAC review and analysis of defense productions). Defendants counter that the PFAC essentially recycles the PTAC's factual allegations using different phrases and packaging, Def. Mem. at 45, and relies largely on public sources available before the PTAC's filing, id. at 5 & n.3, on statements of a consultant who has worked with plaintiffs since 2017, id. at 5-6, on unattributed assertions, id. at 6 & n.4, and on "new" documents mostly produced by the June 29, 2018 deadline for substantial completion of document production, id. at 6-7 (claiming over 75% of the documents cited in the PFAC were produced by that deadline). Defendants argue that, far from having good cause to file the PFAC out of time, plaintiffs failed to act with the required diligence.17

Were the question of "good cause" to turn on whether newly-found evidence justified the late filing, the Court's assessment is that defendants would have, on balance, the better of this argument. Having examined the series of amended complaints, actual and proposed, and the materials cited therein, the Court's judgment is that, as to the two episodes on which the PFAC overwhelmingly relies in claiming a 2008-2012 conspiracy—involving Tradeweb/"Project Fusion" and CME/Swapstream/"Project Magellan"—the PFAC's allegations are largely recycled from past complaints. These two episodes were the crux of the PTAC's claims as to 2008-2012. See IRS II, 2018 WL 2332069, at *11-13 (Tradeweb), *13-17 (CME). The Tradeweb episode had also been the crux of the 2008-2012 claims in the SAC, see IRS I, 261 F. Supp. 3d at 465-69. The PFAC newly packages these episodes as part of a "consortium strategy" among the Dealers during this period, aimed at investing in or otherwise coopting platforms or clearing products that had the potential to lead to all-to-all IRS trading. But a close examination reveals that, once such recasting efforts (and the PFAC's resequencing of allegations) are accounted for, the PFAC adds only minor, largely cosmetic details to these two central episodes.

In contrast, as to three secondary episodes alleged in connection with the 2008-2012 conspiracy, the PFAC alleges new facts. These are that Dealers entered into joint ventures with the goal of gaining control over ICAP's i-Swap and Tradition's Trad-X platforms, and denied support to IDCG's startup clearinghouse. On review, it appears that the documents from which plaintiffs derive their claims as to these episodes were largely obtained in discovery. However, the heart of the PFAC's § 1 claims as to 2008-2012 remains the Tradeweb and CME episodes. These are pled in far more detail than any others. And they far more plausibly allege significant collective action by Dealers with the aim of steering the industry's long-term trajectory away from a course leading eventually towards all-to-all anonymous trading. Even putting aside defendants' challenges to the plausibility of the PFAC's pleadings as to ICAP, Tradition, and IDCG, the new allegations regarding them would not provide good cause for such a late, consequential, and disruptive pleading.

There is, however, an additional and more fundamental impediment to finding good cause: plaintiffs' failure, until the PFAC, to seriously pursue a rule of reason theory of § 1 liability. As described further below, the one area in which the PFAC moves plaintiffs' claims materially forward towards viability is in its articulation of a rule of reason theory. As the Court's decisions in IRS I and IRS II have made clear, given the focus of plaintiffs' 2008-2012 claims on the Tradeweb/"Project Fusion" and CME/"Project Magellan" episodes, this route, while imposing demanding pleading and proof requirements, had more potential to state a § 1 claim than did a theory of per se illegal conduct during this period. Yet, until filing the PFAC in late October 2018, plaintiffs all but disdained a rule of reason claim.

In IRS I, the Court canvassed the law as to the limited categories of agreements among horizontal competitors that the law treats as per se illegal, i.e., "agreements whose nature and necessary effect are so plainly anticompetitive that no elaborate study of the industry is needed to establish their illegality. . . ." Nat'l Soc'y of Prof'l Eng'rs v. United States, 435 U.S. 679, 692 (1978). As to these categories of restraints, experience has enabled courts to "predict with confidence that the rule of reason will condemn it," and therefore permits courts to apply "a conclusive presumption that the restraint is unreasonable." Arizona v. Maricopa Cty. Med. Soc'y, 457 U.S.332, 344 & n.15. Paradigmatic examples, the Court noted, are horizontal agreements among competitors to fix prices or to divide markets, and group boycotts. See IRS I, 261 F. Supp. 3d at 467. The Court held plaintiffs' allegations regarding the 2013-2016 group boycott of Javelin, Tera, and trueEX to be of such a nature. Id. at 472.

But, as reviewed above, plaintiffs' flagship allegations as to 2008-2012, involving "Project Fusion," in which the Dealers jointly invested alongside Thomson in a controlling stake in Tradeweb, described an arrangement that was not susceptible to per se condemnation. The Court explained in detail why the SAC's allegations as to "Project Fusion" did not fit into any category of agreement recognized as per se illegal. See id. at 467-68. And the SAC had not seriously pursued any rule of reason claim as to "Project Fusion." See id. at 468 (outlining pleading requirements for rule of reason claim). It did not contain any section addressed to the rule of reason. And its factual allegations did not fill that void:

Id. at 469 (footnote omitted).

The decision and discussion in IRS I was an invitation for plaintiffs to pursue, in a timely amended complaint, a claim under § 1 for 2008-2012 based on the rule of reason, at least as a carefully-pled alternative to a theory of illegality per se.

Plaintiffs, however, chose in the PTAC, filed on the final day permitted, again to pursue—almost exclusively—a per se theory. That theory remained defective, for the reasons reviewed above. See IRS II, 2018 WL 2332069, at *12 (quoting IRS I, 261 F. Supp. 3d at 468 n. 17). And the PTAC, like the SAC, made only a nominal hat tip towards a rule of reason claim as to Tradeweb:

Id. at *12-13.

For similar reasons (and others), the Court in IRS II held the PTAC's allegations as to CME and "Project Magellan" defective. See id. at *13-17. That project also centrally involved joint, counseled business negotiations between CME and the Dealers which could not be condemned as per se anti-competitive under § 1. See, e.g., id. at *16 & n.11 (reviewing facts inconsistent with per se condemnation). The Court then added:

Id. at *17.

In the PFAC, plaintiffs for the first time make more than a token effort to plead a rule of reason claim, alongside their predominant claim of a per se violation of § 1. See PFAC ¶¶ 626-72. In contrast to the SAC, which lacked rule of reason allegations, and the PTAC, which contained just two paragraphs as to the rule of reason, the PFAC contains concrete allegations as to the foundational concepts for a rule of reason claim. It defines the defendants' product market as the IRS market, id. ¶¶ 628-38; defines the geographic market as the United States, id. ¶¶ 639-41; sets out the Dealers' market shares, individually and in the aggregate, id. ¶¶ 642-45; makes allegations as to Dealers' market power in light of the Justice Department's Merger Guidelines, id. ¶¶ 646-49; addresses why the alternative market definition (the market for trading platforms) posited by defendants is, ostensibly, not correct, id. ¶¶ 650-52; alleges why the "Project Fusion" and "Project Magellan" joint ventures had anti-competitive purposes and effects, id. ¶¶ 653-58; and alleges, for each joint venture, why the anti-competitive effects of the venture outweigh the pro-competitive justifications, id. ¶¶ 659-72.

However, plaintiffs nowhere explain why it took until October 26, 2018 to first meaningfully pursue a rule of reason claim. The subject is unaddressed in the PFAC and in the plaintiffs' memoranda of law. No valid reason is apparent for plaintiffs' failure to pursue such a claim long ago. Plaintiffs cannot pin the absence of an earlier rule of reason claim on recent discovery. On the contrary, the PFAC's rule of reason allegations center on the very two episodes (Tradeweb and CME) as to which its factual allegations are in substance the same as in the PTAC. Moreover, as pled in the PFAC, plaintiffs' rule of reason claim does not cite, or turn at all on, new document discovery. The same rule of reason claim could equally have been made in (if not before) the PTAC or at any time after IRS II. The Court is left to infer that plaintiffs held off until late October 2018 to plead a rule of reason claim not for any reason inherent to that claim. Instead, it appears that plaintiffs deferred pursuing this theory until more than eight months after the deadline for moving to amend had lapsed, in the hope that such would never prove necessary—that the basis for a viable theory of per se liability for 2008-2012 would emerge in discovery.

It is far too late for a complex legal claim of this nature to be added, particularly without a convincing justification. A late pleading of this nature is highly disruptive. The addition of a rule of reason claim, by its nature, would inject into the case, on the brink of the fact discovery deadlines, a host of new factual issues. These include issues relating to the definitions of the product and the geographic markets, defendants' market share(s), entry barriers, potential entrants, and potential substitutes. Relevant issues would also be generated based on the parties' articulations of the anti-competitive effects of, and the pro-competitive justifications for, the various agreement(s) alleged among defendants during 2008-2012. Although expert analysis would surely be brought to bear as to these issues, a consequential expansion of fact discovery— above and beyond that which a per se claim for these years would necessitate—would be necessary to elicit evidence and to develop the record relating to unique elements of a rule of reason claim. Plaintiffs have not articulated good cause for disrupting the case schedule by first pursuing a rule of reason theory so late in the day. The Court finds none.

In finding an absence of good cause for first pursuing a rule of reason claim so close to the end date for fact discovery, the Court has no occasion to definitively resolve whether the PFAC has adequately pled such a claim. Defendants have chronicled deficiencies in the PFAC's iteration of that claim. See generally Def. Mem. 24-34. A number of these critiques appear substantial. These include that because the PFAC tightly focuses its rule of reason allegations on the Dealers' attempts to control Tradeweb and CME through the two joint ventures, a rule of reason claim can lie only if these entities had market power.18

Notwithstanding the particular manner in which the PFAC is pled, the Court's judgment is that a viable rule of reason claim quite possibly could have been formulated for 2008-2012. Plaintiffs have amassed evidence of statements by officials at various Dealers preferring the traditional model of dealer-to-customer IRS trading and disfavoring the prospect of anonymous trading of IRSs on an exchange. Plaintiffs have also alleged various actions by Dealer personnel, including in connection with Projects "Fusion" and "Magellan," that made it less likely that exchange trading along the lines that plaintiffs preferred would take root. While the Dealers' complex business arrangements in 2008-2012—whether labeled as "consortium investments" or otherwise—do not implicate any category of conduct recognized as per se anti-competitive, a viable rule of reason claim potentially could have been pled for this period, alleging an overall agreement among Dealers to steer future IRS trading towards bilateral, party-disclosed trading and against the development of anonymous all-to-all exchange trading. Such a claim would, presumably, have alleged that the overall anti-competitive effects of this agreement outweighed the pro-competitive potential of such a course.

Had plaintiffs seriously pursued a rule of reason claim earlier in this litigation rather as an untimely after-thought, there would have been opportunity for considered briefing and decision, and to the extent found necessary, for a potentially curative amended complaint. The iterative process by which claims are assessed on a motion to dismiss and refined through an amended pleading is unusually well-tailored to complex claims such as under the rule of reason, which may require successive iterations to hone to viability. It is, however, far too late to embark on that journey now, with fact discovery due to close in under a month.

For the reasons above, the Court therefore does not find good cause to permit plaintiffs to file the PFAC more than eight months after the deadline for amended pleadings set in the case management plan. The episodes on which the PFAC's § 1 claim predominantly relies are not new. Indeed, they were articulated earlier in the SAC and/or the PTAC and do not substantially turn on new evidence. And plaintiffs failed to pursue earlier the one potentially viable theory of § 1 liability for 2008-2012, under the rule of reason, despite having ample opportunities to do so. Furthermore, that theory appears to require significant modification to approach viability and would significantly disrupt the existing schedule.

D. Conclusion

For the above reasons, the Court denies plaintiffs' motion for leave to amend, so as to add claims covering 2008-2012.

IV. The PFAC's Claims as to 2013-2016

The PFAC also adds a limited number of allegations bearing on the period 2013-2016. Most notably, it alleges why two entities—Tradeweb and Bloomberg—that potentially could have offered all-to-all anonymous trading of IRSs did not do so during this period.

As to Tradeweb, it alleges that Tradeweb could offer "an all-to-all RFQ platform," but that it chose instead to limit its RFQ "to facilitate dealer-to-client or dealer-to-dealer trading by allowing only dealers to respond to the quote requests submitted by the platform." PFAC ¶ 204. This, the PFAC alleges, "is the result of the Dealer Defendants' control of Tradeweb." Id.

As to Bloomberg, it alleges that Bloomberg's BSEF received regulatory approval from the CFTC on July 31, 2013 and was equipped to enable order book trading between any market participants. Id. ¶ 338. However, it alleges, the Dealers today refuse to provide liquidity to BSEF's CLOB except "on a name-disclosed basis," such that Bloomberg's CLOB therefore has "no volume on it." Id. ¶ 340 ("As a result of these tactics against Bloomberg's IRS trading platforms, the Dealer Defendants have succeeded in preventing Bloomberg from offering a meaningful all-to-all anonymous trading platform and have forced buy-side customers to trade IRS on Bloomberg via a non-anonymous RFQ protocol.").

The Court will permit the addition of these allegations. The status during 2013-2016 of other platforms for the trading of IRSs is, for a variety of reasons, relevant context for plaintiffs' well-pled claims of a contemporaneous boycott by Dealers of Javelin, Tera, and trueEX. Among other things, to the extent that no other platforms permitting the all-to-all anonymous trading of IRSs are shown to have been operational during this period, plaintiffs' claim that an effective boycott of Javelin, Tera, and trueEX affected all investors who wished to trade IRSs in such a manner on such an exchange is strengthened. Separately, Dealers' actions during this period towards other platforms that carried the promise of enabling anonymous all-to-all IRS trading have the potential to bear significantly on the factfinder's determination whether the Dealers did or did not boycott Javelin, Tera, and trueEX. Consistent with its discovery rulings, the Court has assumed throughout that the Dealers' conduct towards other platforms during 2013-2016 besides the three boycott targets is within the scope of matters to be discovered. The Court therefore authorizes plaintiffs to file a FAC incorporating the PFAC's allegations bearing on the manner in which IRSs were and were not traded on Tradeweb and Bloomberg during 2013-2016. As these matters have been within the scope of existing authorized discovery, this amendment will not result in any change to the existing discovery schedule.

CONCLUSION

For the reasons stated above, the Court denies plaintiffs' motion for leave to amend to add allegations regarding 2008-2012 but grants the motion for leave to amend to add allegations regarding 2013-2016 as described above.

SO ORDERED.