IN RE: LIBOR-Based Financial Instruments Antitrust Litigation. (2025)

MEMORANDUM AND ORDER

LIBOR IX

Table of Contents

I. Introduction 4

1. Procedural History 6

2. Discovery 10

3. Pending Motions 14

II. General Legal Standards 16

1. Class Certification and Decertification 17

2. Expert Opinion 19

3. Summary Judgment 23

4. Local Civil Rule 56.1 26

III. Conspiracy 29

1. Applicable Law 29

2. The Asserted Agreement 33

3. Direct Evidence of Alleged Sixteen-Bank Conspiracy 43

4. Circumstantial Evidence of Alleged Sixteen-Bank Conspiracy 49

4.1. Daubert Motions Regarding Drs. Cragg and Marx 50

4.1.1. Dr. Cragg 51

4.1.2. Dr. Marx 67

4.1.3. Conclusion 83

4.2. Circumstantial Evidence of Parallel Conduct 85

4.3. Circumstantial Evidence of Plus Factors 93

4.3.1. Plus Factor #1: Common Motive to Conspire 95

4.3.2. Plus Factor #2: Actions Taken Against Self-Interest 106

4.3.3. Plus Factor #3: High Level of Interfirm Communications 111

4.3.4. Plus Factor #4: LIBOR's Structural Susceptibility to Collusion 162

4.3.5. Plus Factor #5: Concealment Efforts 166

4.3.6. Plus Factor #6: Econometric Models and Expert Evidence 169

4.3.7. Plus Factor #7: Government Investigations and Regulatory Settlements 171

4.3.8. Plus Factor #8: Other Evidence Refuting the Existence of Conspiracy 174

5. Conclusion 176

IV. Suppression 177

1. Connolly's Interpretation of the LIBOR Question 180

2. Daubert Motion Regarding Drs. Bernheim and Snow 185

2.1. Dr. Douglas Bernheim's Reports 187

2.1.1. Benchmark Analysis 188

2.1.2. Imputation Analysis 193

2.2. Dr. Karl Snow's Reports 198

2.2.1. Econometric Model 198

2.2.2. Alternative Model 200

2.3. Common Flaws: Dr. Bernheim's Benchmark Analysis and Dr. Snow's Econometric Model 201

2.3.1. Inapposite Benchmark Rates 202

2.3.2. Failure to Address an Obvious Alternative Explanation: the Financial Crisis 213

2.3.3. Improper Corroboration 229

2.4. Common Flaws: Dr. Bernheim's Imputation Analysis and Dr. Snow's Alternative Model 231

2.4.1. Ignores Relevant Data 231

2.4.2. Unreliable Methodology 234

2.5. Common Flaws: Panel Banks’ Borrowing Costs Provide Compelling Evidence that LIBOR Was Not Suppressed 237

3. Purported Borrowing Costs Requirement 238

4. Conclusion 241

V. Summary Judgment: Sherman Act Claims 244

VI. Summary Judgment: Common Law Claims 246

1. Fraud-Based Claims 246

1.1. Common Law Fraud 248

1.2. Fraudulent Inducement of a Contract 252

1.3. Aiding and Abetting Fraud 255

1.4. Conspiracy to Commit Fraud 257

2. Negligent Misrepresentation 259

3. Breach of Contract and Breach of the Implied Covenant of Good Faith and Fair Dealing 261

4. Tortious Interference with Contract 265

5. Unjust Enrichment 267

VII. Class Decertification 269

VIII. Conclusion 272

I. INTRODUCTION

This consolidated multidistrict litigation (“MDL”) commenced in 2011 and arises from allegations that sixteen major banks conspired to suppress the London Interbank Offer Rate (“LIBOR”), an interest-rate benchmark used in trillions of dollars’ worth of financial instruments.¹ More specifically, plaintiffs “contend that, beginning in 2007, the Banks engaged in a horizontal price-fixing conspiracy, with each submission reporting an artificially low cost of borrowing in order to drive LIBOR down.” Gelboim v. Bank of Am. Corp., 823 F.3d 759, 766 (2d Cir. 2016); see infra pp. 244-269 (summarizing plaintiffs’ claims).²

LIBOR was set as follows. Under the auspices of the British Bankers’ Association (“BBA”), sixteen major banks, referred to as “Panel Banks,” submitted an answer each day to the LIBOR question for “loans of various durations (called loan ‘tenors’)--typically ranging from overnight to one year--in each of the world's major currencies.” United States v. Connolly, 24 F.4th 821, 824 (2d Cir. 2022).³ The LIBOR question asked: “At what rate could you borrow funds, were you to do so by asking for and then accepting inter-bank offers in a reasonable market size just prior to 11 am” London time? LIBOR VII, 299 F. Supp. 3d 430, 522 (S.D.N.Y. 2018), ECF No. 2452 at *158. Thomson Reuters, an agent of the BBA, “received the [sixteen] panel banks’ rate submissions and eliminated from consideration, for each tenor, the four highest and the four lowest[,]” and published the “arithmetic average of the middle eight submitted rates,” which “constituted the LIBOR daily ‘fix’ ․ for that tenor.” Connolly, 24 F.4th at 825 (citations omitted).⁴

Now pending before the Court are six motions: (i) defendants’ motion for summary judgment with respect to all remaining claims; (ii) class plaintiffs’ motion to extend the class that was previously certified in LIBOR VII, 299 F. Supp. 3d 430 (S.D.N.Y. 2018), ECF No. 2452; (iii) class defendants’ motion to decertify the previously certified class; and (iv) three Daubert motions by defendants to exclude plaintiffs’ proposed experts.

1. Procedural History

A limited review of the litigation history of this MDL ⁵ is necessary to place the current motions in the proper context. At its numerical height, this MDL compromised approximately fifty individual cases and three putative classes. As this MDL has progressed, substantive orders from this Court and the Second Circuit, in combination with multiple settlements, have significantly reduced the number of parties remaining in this litigation. The citations to those decisions, which totaled in excess of 1,300 typed pages, are set out in the footnote below.⁶ Given the number of substantive decisions, for ease of reference, we adopted a Roman numerical shorthand to identify the lengthier opinions. Most of these opinions were reviewed by the Second Circuit contemporaneously.

Of particular significance to the contours of this case at present is LIBOR VII. In LIBOR VII, we granted class certification under Rule 23(b)(3) to a class of plaintiffs who bought over-the-counter instruments that paid interest indexed to the LIBOR rate and brought antitrust claims against Bank of America Corporation, Bank of America, N.A. (collectively “Bank of America”), JPMorgan Chase & Co. and JPMorgan Chase Bank, N.A (collectively “JPMorgan”). See 299 F. Supp. 3d at 581-608, slip op. at *298-362.⁷ To be more precise, the following class, which we refer to as the OTC class or OTC plaintiffs, was certified under Rule 23(b)(3):

All persons or entities residing in the United States that purchased, directly from a Panel Bank (or a Panel Bank's subsidiaries or affiliates), a LIBOR-Based Instrument that paid interest indexed to a U.S. dollar LIBOR rate set any time during the period August 2007 through August 2009 (“Class Period”) regardless of when the LIBOR-Based Instrument was purchased.

Id. at 608, slip op. at *361.⁸ The following plaintiffs were appointed as lead plaintiffs of the OTC class: the Mayor and City Council of Baltimore, Maryland; the City of New Britain, Connecticut; Vistra Energy Corp.; Yale University; and Jennie Stuart Medical Center, Inc. See id. at 608, slip op. at *361-62.

The OTC plaintiffs are central to the current motions. In addition to opposing the defendants’ motion for summary judgment, they seek to certify the class against additional defendants Credit Suisse AG, NatWest Markets plc (f/k/a Royal Bank of Scotland plc), NatWest Group plc (f/k/a The Royal Bank of Scotland Group plc), and UBS AG. ECF No. 4182.⁹ OTC plaintiffs also oppose OTC defendants’ Daubert motions to exclude the testimony of their two experts, Drs. Cragg and Bernheim.

The second category of remaining plaintiffs are direct action plaintiffs, or “DAPs,” who brought individual actions and did not seek class certification in LIBOR VII. Within this category, five plaintiffs, known as the “FFFP plaintiffs,” collectively submitted testimony from two experts about conspiracy and suppression. See ECF No. 4166 (“Defs. Br.”) at 18; ECF No. 4558 (“Defs. 56.1”) at 4-5. The FFFP plaintiffs are Federal National Mortgage Association (“Fannie Mae”), Federal Home Loan Mortgage Corporation (“Freddie Mac”), Federal Deposit Insurance Corporation (“FDIC”), Principal Financial Group, and Principal Funds, Inc. See Defs. Br. at xii. At the time that the parties submitted their motions, the DAPs included more plaintiffs than the FFFP plaintiffs. However, multiple DAPs settled after the parties completed their briefing, and the only DAPs remaining are the FFFP plaintiffs. See ECF No. 4583 at 1-2.¹⁰

2. Discovery

Discovery in this case may be described as proceeding in two phrases.¹¹ First, in 2016, the Court ordered defendants “to reproduce disclosures previously made to governmental authorities.” See ECF Nos. 1441, 1461. Those disclosures had been made to domestic and foreign regulatory agencies who had conducted investigations before this MDL was filed, as well as to domestic and foreign regulatory agencies who continued or commenced their parallel investigations into allegations of LIBOR manipulation. See In re LIBOR-Based Fin. Instruments Antitrust Litig., No. 11 MDL 2262 (NRB), 2023 WL 2871090, at *1 (S.D.N.Y. Apr. 10, 2023), ECF No. 3655 at 2. As of 2019, we observed that no settlement with these regulatory authorities “contain[ed] [an] allegation or finding that an inter-bank persistent suppression conspiracy existed.” LIBOR VIII, 2019 WL 1331830, at *18, slip op. at *46; infra pp. 171-174. As a result of the disclosure order, the Panel Banks produced to plaintiffs “over 3.4 million documents, spanning 15.9 million pages and 88,000 audio files, from over 500 custodians,” generally covering the LIBOR setting process in the “period of August 2007 to May 2010” that they had previously produced to regulators. Defs. 56.1 ¶ 5.¹²

The second phase of discovery followed the Second Circuit's decision in United States v. Connolly, 24 F.4th 821 (2d Cir. 2022), the first decision in which the Circuit interpreted the LIBOR question. Although the Connolly decision did not result from an appeal in this MDL, it has profound implications for both the substance and contours of this case. While the Second Circuit's interpretation of the LIBOR question will be discussed in greater detail below, infra pp. 79-82, 180-185, this Court and all parties are, of course, bound by the Circuit's understanding of LIBOR, and the instant motions “must be understood in light of Connolly,” see In re LIBOR-Based Fin. Instruments Antitrust Litig., No. 11 MDL 2262 (NRB), 2023 WL 2871090, at *8 n.12 (S.D.N.Y. Apr. 10, 2023), ECF No. 3655 at 18 n.12.

Accordingly, following the Circuit's decision in Connolly, this Court -- having requested and analyzed submissions from the parties as to next steps -- bifurcated future discovery. See ECF No. 3386. The first stage, which was intended to build upon the “extensive discovery” provided during the class certification process, LIBOR VII, 299 F. Supp. 3d at 531, slip op. at *180, was directed to the “Upstream Issues,” namely, (1) the alleged existence of a sixteen-bank conspiracy to suppress LIBOR and (2) the alleged existence of LIBOR suppression in light of the Second Circuit's decision in Connolly, see ECF Nos. 3386 at 1, 3425 at 1, 3687 at 1, 3921 at 9. Only once discovery and dispositive motions on the Upstream Issues were completed, would the parties, if necessary, focus discovery on the “Downstream Issues,” namely, (1) whether plaintiffs suffered injuries as a result of the alleged LIBOR suppression, (2) whether plaintiffs had notice of and relied on an allegedly inaccurate LIBOR, and (3) any other remaining issues. See ECF No. 3687 at 7.

Fact discovery on the Upstream Issues concluded on April 4, 2024. Id. at 5. “Plaintiffs served on Defendants more than 700 interrogatories (inclusive of subparts) and more than 300 requests for admission of facts.” Defs. 56.1 ¶ 6. The Panel Banks collectively produced an additional 400,000 documents. Id. ¶ 5. Plaintiffs also served “more than 100 ․ subpoenas” on third parties, including on interdealer brokerage firms, the BBA, and others. Id. ¶ 10. In addition, the parties agreed that plaintiffs could take up to 100 fact depositions of defendants’ current and former employees, ECF No. 3544 ¶ 5, and plaintiffs deposed 59 current or former Panel Bank employees as well as three former BBA employees, Defs. 56.1 ¶ 8. However, as discussed further below, see infra pp. 113-118, plaintiffs did not take all of the fact discovery to which they were entitled. For example, plaintiffs sought and were granted leave to depose eight individual interdealer brokers from entities that they assert were central to the alleged conspiracy, and yet plaintiffs did not avail themselves of the opportunity. Defs. 56.1 ¶ 11; ECF No. 3892. Similarly, plaintiffs never deposed any current or former employees from the Royal Bank of Canada (“RBC”), despite having permission to do so. Defs. 56.1 ¶ 9.

Expert discovery on the Upstream Issues concluded on September 13, 2024. ECF No. 3687 at 6. Eight experts provided opinions, and were deposed (some more than once), on issues relating to conspiracy and suppression. The OTC plaintiffs proffered testimony from two experts, Dr. Douglas Bernheim, who studied suppression, and Dr. Michael Cragg, who assessed conspiracy. See Defs. Br. at 18. The FFFP plaintiffs submitted affirmative, rebuttal, and reply expert reports from Dr. Karl Snow, who analyzed suppression, and Dr. Leslie Marx, who opined on conspiracy. Id. Defendants jointly engaged Dr. John Asker to opine on conspiracy and Dr. Glenn Hubbard to opine on suppression. Id. at 17. JPMorgan and Credit Suisse independently retained additional experts, Dr. Dennis Carlton and Dr. Divya Mathur respectively, to study both suppression and conspiracy. Id. at 17-18.

Altogether, the Panel Banks collectively produced nearly 3.8 million documents and over 93,000 audio files. Defs. 56.1 ¶ 5. Plaintiffs deposed 59 current or former Panel Bank employees and three former BBA employees. Id. ¶ 8. Eight experts opined on issues related to conspiracy and suppression. Defs. Br. at 17-18.

3. Pending Motions

There are six motions currently pending before the Court. First, defendants filed a joint motion for summary judgment on all remaining claims. See ECF No. 4164. Tellingly, plaintiffs did not cross-move for summary judgment. In connection with defendants’ motion, defendants jointly submitted an accompanying memorandum of law, see ECF No. 4166 (“Defs. Br.”), plaintiffs filed a joint opposition memorandum, see ECF No. 4335 (“Pls. Br.”), and defendants filed a joint reply brief, see ECF No. 4442 (“Defs. Reply Br.”).¹³ Additionally, in support of their respective memoranda, defendants jointly filed a Rule 56.1 Statement of Material Undisputed Facts, ECF No. 4558 (“Defs. 56.1”), and plaintiffs filed a Rule 56.1 Joint Counterstatement of Material Undisputed Facts, ECF No. 4557 (“Pls. 56.1”).¹⁴

Next, there are two motions related to class certification. As noted earlier, OTC plaintiffs moved to certify the OTC class against additional defendants. ECF No. 4182; supra pp. 8-9. OTC defendants oppose this motion and seek decertification of the OTC class. See ECF Nos. 4317, 4336.

Finally, defendants filed three motions to exclude the testimony of four of plaintiffs’ experts. See ECF Nos. 4147 (Dr. Marx), 4151 (Drs. Snow and Bernheim), and 4156 (Dr. Cragg).¹⁵

For the reasons set out at length below, the Court grants defendants’ motion for summary judgment on all remaining claims, denies class plaintiffs’ certification motion, and grants class defendants’ motion to decertify the previously certified class. As for the three pending motions to exclude plaintiffs’ expert testimony, one is granted, while two are granted in part and denied in part.

II. GENERAL LEGAL STANDARDS

In this section, we set forth the general legal standards applicable to our resolution of the pending motions. Because we exhaustively explicated the class certification and expert opinion standards in LIBOR VII, we summarize and describe those standards only to the extent that these standards have been further clarified since LIBOR VII or to the extent that an issue present here was not at-issue in LIBOR VII.

1. Class Certification and Decertification

As we exhaustively set forth in LIBOR VII, the “ ‘class action is an exception to the usual rule that litigation is conducted by and on behalf of the individual named parties only.’ ” LIBOR VII, 299 F. Supp. 3d at 458, slip op. at *4 (quoting Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 348 (2011) (internal citation omitted)). To proceed under that exception, a plaintiff “must affirmatively demonstrate [its] compliance with Rule 23.” Comcast Corp. v. Behrend, 569 U.S. 27, 33 (2013) (quoting Wal-Mart Stores, 564 U.S. at 350). “[T]he preponderance of the evidence standard applies to evidence proffered to establish Rule 23's requirements.” Teamsters Loc. 445 Freight Div. Pension Fund v. Bombardier Inc., 546 F.3d 196, 202 (2d Cir. 2008).

As noted earlier, in LIBOR VII, we certified a class limited to OTC plaintiffs’ antitrust claims against Bank of America and JPMorgan Chase. See 299 F. Supp. 3d at 581-608, slip op. at *298-362. We found that each of the requirements of Rule 23(a) were satisfied. Id. at 585-90, slip op. at *308-320. Rule 23(a) provides that “[o]ne or more members of a class may sue ․ as representative parties on behalf of all members only if: (1) the class is so numerous that joinder of all members is impracticable; (2) there are questions of law or fact common to the class; (3) the claims ․ of the representative parties are typical of the claims ․ of the class; and (4) the representative parties will fairly and adequately protect the interests of the class.” Fed. R. Civ. P. 23(a). These requirements are generally referred to as numerosity, commonality, typicality, and adequacy of representation.

In addition to satisfying each requirement of Rule 23(a), we found that the OTC class met one of the bases for certification under Rule 23(b). See LIBOR VII, 299 F. Supp. 3d at 590-608, slip op. at *320-362. As relevant here, Rule 23(b)(3) provides that a class action may proceed as such if “the court finds that the questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” Fed. R. Civ. P. 23(b)(3) (emphasis added). Predominance and superiority must each be satisfied. See Sykes v. Mel S. Harris & Assocs., 780 F.3d 70, 82 (2d Cir. 2015) (referring to Rule 23(b)(3) as a “disjunctive inquiry”).

However, class “certifications are not frozen once made,” Amgen Inc. v. Ct. Ret. Plans & Tr. Funds, 568 U.S. 455, 479 n.9 (2013), and a district court is “required to reassess [its] class rulings as the case develops,” Boucher v. Syracuse Univ., 164 F.3d 113, 118 (2d Cir. 1999) (citation and quotation marks omitted). Indeed, “district courts must ensure that a certified class satisfies Rule 23 throughout the litigation,” Jin v. Shanghai Original, Inc., 990 F.3d 251, 262 (2d Cir. 2021) (citation omitted), and should “decertify a class when the standards of Rule 23 have not been met,” In re KIND LLC “Healthy & All Natural” Litig., 627 F. Supp. 3d 269, 295 (S.D.N.Y. 2022) (citation and quotation marks omitted), aff'd, 100 F.4th 419 (2d Cir. 2024); see also Wu v. Pearson Educ. Inc., 2012 WL 6681701, at *5 (S.D.N.Y. Dec. 21, 2012) (“A district court may—and should—decertify a class when the standards of Rule 23 have not been met.” (citations omitted)); Monaco v. Hogan, 2016 WL 1322431, at *2 (E.D.N.Y. Mar. 31, 2016) (finding that the standard for decertification is the same as for certification), aff'd, 737 F. App'x 6 (2d Cir. 2018); Fed. R. Civ. P. 23(c)(1) Advisory Committee's notes to 2003 amendment (“A determination of liability after certification, however, may show a need to amend the class definition. Decertification may be warranted after further proceedings.”).

2. Expert Opinion

As we did in LIBOR VII, the Court must decide whether multiple experts’ opinions are admissible pursuant to the Federal Rules of Evidence.

“The law assigns district courts a ‘gatekeeping’ role in ensuring that expert testimony satisfies the requirements of Rule 702.” United States v. Farhane, 634 F.3d 127, 158 (2d Cir. 2011) (citation omitted); see also Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 589-97 (1993); accord Kumho Tire Co. v. Carmichael, 526 U.S. 137, 147-53 (1999).

Rule 702 of the Federal Rules of Evidence provides in full:

A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if the proponent demonstrates to the court that it is more likely than not that: (a) the expert's scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue; (b) the testimony is based on sufficient facts or data; (c) the testimony is the product of reliable principles and methods; and (d) the expert's opinion reflects a reliable application of the principles and methods to the facts of the case.

Courts in this Circuit frequently distill Rule 702 to a three-part test that requires the proponent of expert evidence to show “that (1) the expert is qualified; (2) the proposed opinion is based on reliable data and methodology; and (3) the proposed testimony would be helpful to the trier of fact.” Nike, Inc. v. StockX LLC, No. 22 Civ. 983, 2024 WL 3361411, at *2 (S.D.N.Y. July 10, 2024) (citing Nimely v. City of New York, 414 F.3d 381, 396 (2d Cir. 2005)); see also LIBOR VII, 299 F. Supp. 3d at 465-71, slip op. at *23-37 (outlining relevant standard).

Rule 702 has been amended in two key respects since we set forth the expert opinion standard in LIBOR VII. First, “Rule 702 was amended to ‘clarify and emphasize’ that proffered expert testimony must meet Rule 702’s admissibility requirements by a preponderance of the evidence.” In re Keurig Green Mountain Single-Serve Coffee Antitrust Litig., No. 14 MDL 2542 (VSB), 2025 WL 354671, at *1 (S.D.N.Y. Jan. 30, 2025) (quoting Fed. R. Evid. 702 advisory committee's note to 2023 amendments). This amendment confirms the long-standing practice in the Second Circuit. See, e.g., United States v. Williams, 506 F.3d 151, 160 (2d Cir. 2007) (“[T]he proponent of expert testimony has the burden of establishing by a preponderance of the evidence that the admissibility requirements of Rule 702 are satisfied[.]”).

Second, “Rule 702(d) has also been amended to emphasize that each expert opinion must stay within the bounds of what can be concluded from a reliable application of the expert's basis and methodology.” Fed. R. Evid. 702 Advisory Committee's note to 2023 amendments. As the Advisory Committee's note makes clear, “[j]udicial gatekeeping is essential because just as jurors may be unable, due to lack of specialized knowledge, to evaluate meaningfully the reliability of scientific and other methods underlying expert opinion, jurors may also lack the specialized knowledge to determine whether the conclusions of an expert go beyond what the expert's basis and methodology may reliably support.” Id.

In addition to Rule 702, “ ‘a judge assessing a proffer of expert scientific testimony ․ should also be mindful of other applicable [Federal Rules of Evidence],’ including Rule 403, which permits exclusion of relevant evidence ‘if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury.’ ” United States v. Pollok, No. 23-6114-CR, 2025 WL 1553327, at *7 (2d Cir. June 2, 2025) (quoting Daubert, 509 U.S. at 595).¹⁶ For instance, expert testimony “which merely ‘tell[s] the jury what result to reach’ creates a significant risk of unfair prejudice.” Scottsdale Ins. Co. v. McGrath, No. 19 Civ. 7477 (LJL), 2024 WL 4512210, at *6 (S.D.N.Y. Oct. 17, 2024) (citations omitted). Moreover, “expert testimony that seeks to address ‘lay matters which [the trier of fact] is capable of understanding and deciding without the expert's help’ is not relevant” in the first instance “and is therefore inadmissible.” United States v. Jiau, 734 F.3d 147, 154 (2d Cir. 2013) (citation omitted).

Similarly, while an expert may rely on hearsay in certain instances, an expert may only disclose the inadmissible facts or data “if their probative value in helping the jury evaluate the opinion substantially outweighs their prejudicial effect.” Fed. R. Evid. 703. Further, “a party cannot call an expert simply as a conduit for introducing hearsay under the guise that the testifying expert used the hearsay as the basis of his testimony.” Marvel Characters, Inc. v. Kirby, 726 F.3d 119, 136 (2d Cir. 2013) (citation and quotation marks omitted); cf. SEC v. Tourre, 950 F. Supp. 2d 666, 675 (S.D.N.Y. 2013) (finding that “mere narration” of facts “fails to fulfill Daubert’s most basic requirements” as it “does not convey opinions based on an expert's knowledge and expertise; nor is such a narration traceable to a reliable methodology”).

3. Summary Judgment

The current summary judgment motion is the first such motion in this case. Summary judgment “serves a vital function in the area of antitrust law” “[b]y avoiding wasteful trials and preventing lengthy litigation that may have a chilling effect on pro-competitive market forces.” Tops Markets, Inc. v. Quality Markets, Inc., 142 F.3d 90, 95 (2d Cir. 1998) (citations omitted).

Summary judgment must be granted where “there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed R. Civ. P. 56(a). “A fact is material if it ‘might affect the outcome of the suit’ ” and is genuinely in dispute if “the evidence is such that a reasonable jury could return a verdict for the nonmoving party.” Roe v. City of Waterbury, 542 F.3d 31, 35 (2d Cir. 2008) (citation omitted). “In ruling on a motion for summary judgment, the district court may rely on ‘any material that would be admissible or usable at trial.’ ” Major League Baseball Props., Inc. v. Salvino, Inc., 542 F.3d 290, 309 (2d Cir. 2008) (citations omitted).

The movant “always bears the initial responsibility of informing the district court of the basis for its motion,” as well as the basis for any absence of material fact in dispute. Celotex Corp. v. Catrett, 477 U.S. 317, 323 (1986). “[W]here the nonmovant will bear the ultimate burden of proof at trial on an issue, the moving party's burden under Rule 56 will be satisfied if he can point to an absence of evidence to support an essential element of the nonmoving party's claim.” Brady v. Town of Colchester, 863 F.2d 205, 210-11 (2d. Cir. 1988) (citing Celotex, 477 U.S. at 323-26). “In assessing the record to determine whether there is a genuine issue to be tried, we are required to resolve all ambiguities and draw all permissible factual inferences in favor of the party against whom summary judgment is sought.” Gorzynski v. JetBlue Airways Corp., 596 F.3d 93, 101 (2d Cir. 2010).

If the moving party makes a showing that it is entitled to summary judgment, “[t]hen the onus shifts to the party resisting summary judgment to present evidence sufficient to satisfy every element of the claim.” Holcomb v. Iona Coll., 521 F.3d 130, 137 (2d Cir. 2008). Unlike on a motion to dismiss or a motion for class certification, where the allegations in the plaintiffs’ complaint are accepted as true, at summary judgment, plaintiffs must provide “significant probative evidence,” which a reasonable factfinder could rely on to decide in their favor. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 249 (1986) (citation and quotation marks omitted). “The mere existence of a scintilla of evidence in support of the plaintiff's position will be insufficient[.]” Id. at 252. Plaintiffs may not rely upon “conclusory statements or mere allegations,” they must “go beyond the pleadings, and by ․ affidavits, or by the depositions, answers to interrogatories, and admissions on file, designate specific facts showing that there is a genuine issue for trial.” Davis v. New York, 316 F.3d 93, 100 (2d Cir. 2002) (citations and quotation marks omitted). “[A] complete failure of proof concerning an essential element of the nonmoving party's case necessarily renders all other facts immaterial.” Crawford v. Franklin Credit Mgmt., 758 F.3d 473, 486 (2d Cir. 2014) (quoting Celotex, 477 U.S. at 323 (alteration in original)). Thus, “a nonmoving party can defeat a summary judgment motion only ‘by coming forward with evidence that would be sufficient, if all reasonable inferences were drawn in [its] favor, to establish the existence of [an] element at trial.’ ” City of Waterbury, 542 F.3d at 36 (citations omitted).

4. Local Civil Rule 56.1

Summary judgment papers in this District must follow the strictures imposed by Local Civil Rule 56.1. Rule 56.1(a) requires the party moving for summary judgment to submit a “short[ ] and concise statement, in numbered paragraphs, of the material facts as to which the moving party contends there is no genuine issue to be tried.” Local Civ. R. 56.1(a) (emphases added). The nonmoving party, in turn, must submit “a correspondingly numbered paragraph admitting or denying, and otherwise responding to, each numbered paragraph in the statement of the moving party.” Local Civ. R. 56.1(b). “[I]f necessary,” the nonmoving party may submit “additional paragraphs containing a separate, short and concise statement of additional material facts as to which it is contended that there exists a genuine issue to be tried.” Id. (emphases added).

Rule 56.1 “does [not] contemplate creating more or less than an admission or a denial of the truth of the allegation for the purposes of the motion.” Emanuel v. Gap, Inc., No. 19 Civ. 3617 (PMH), 2022 WL 3084317, at *4 (S.D.N.Y. Aug. 3, 2022) (emphasis added). “In short, the purpose of a [Local Rule] 56.1 response is simply to advise the Court as to whether the specific fact asserted by the moving party is or is not disputed, and if it is disputed, to provide the Court with the evidence on which the non-moving party relies to dispute that particular fact. It presents no occasion for context, argument, semantic quibbles, opinions or conclusions.” Tripathy v. McCloskey, No. 21 Civ. 6584 (CS), 2024 WL 2135623, at *2 (S.D.N.Y. May 13, 2024).

Both parties have “flagrantly violated” Local Rule 56.1. Graves v. Deutsche Bank Sec., Inc., 548 F. App'x 654, 657 n.2 (2d Cir. 2013); see Defs. 56.1; Pls. 56.1. For one, the parties’ 56.1 statements and responses are not “short[ ] and concise,” as they are improperly filled with immaterial facts. Local Civ. R. 56.1(a) (emphasis added); see, e.g., Defs. 56.1 ¶¶ 18, 56, 278; Pls. 56.1 ¶¶ 34, 45, 96, 234; Torres v. Brend Restoration Servs. Inc., No. 22 Civ. 1191 (DEH), 2024 WL 4355030, at *1 n.2 (S.D.N.Y. Sept. 30, 2024) (“[B]oth parties improperly interjected ․ immaterial facts in response to facts.”).

Second, although Rule 56.1 requires non-argumentative statements, the parties here regularly include legal argument in their statements of fact or responses, quibbling over the materiality of asserted facts or the admissibility of the underlying evidence. See, e.g., Defs. 56.1 Resps. ¶¶ 17, 150, 158, 234; Pls. 56.1 Resps. ¶¶ 5, 27, 29, 227, 240. These disputes “do not create a genuinely disputed fact,” Tang Cap. Partners, LP v. BRC Inc., 757 F. Supp. 3d 363, 373 (S.D.N.Y. 2024) (collecting cases), and the Court sets aside “[t]he portions of [a responding party's] 56.1 Statement that contain legal argument bereft of factual matter,” Kesner v. Buhl, 590 F. Supp. 3d 680, 691 (S.D.N.Y. 2022), aff'd sub nom. Kesner v. Dow Jones & Co., Inc., No. 22-875, 2023 WL 4072929 (2d Cir. June 20, 2023); see also Graves, 548 F. App'x at 657 n.2 (“Local Civil Rule 56.1 ․ requires a ‘short and concise,’ non-argumentative response.” (emphasis added)); Smith v. City of New York, 697 F. App'x 88, 89 (2d Cir. 2017) (“[M]aterial relied on at summary judgment need not be admissible in the form presented to the district court. Rather, so long as the evidence in question will be presented in admissible form at trial, it may be considered on summary judgment.” (citations and internal quotation marks omitted)).

However, the parties “cannot simply dump papers on the court and expect the court to sift through them to determine if some nugget is buried somewhere in that mountain.” Emanuel, 2022 WL 3084317, at *5 (citation and quotation marks omitted). While we do not specifically rule on every single objection contained in the parties’ voluminous 56.1 statements and responses, the Court has conducted its own review to confirm that the evidentiary materials cited by the parties support the factual assertions set forth in their Rule 56.1 statements and that the evidence relied upon is material and admissible. See Vermont Teddy Bear Co. v. 1-800 Beargram Co., 373 F.3d 241, 244 (2d Cir. 2004) (“[I]n determining whether the moving party has met [its] burden of showing the absence of a genuine issue for trial, the district court may not rely solely on the statement of undisputed facts contained in the moving party's Rule 56.1 statement. It must be satisfied that the citation to evidence in the record supports the assertion.”).

III. CONSPIRACY

The first Upstream Issue focuses on whether a sixteen-bank, multi-year conspiracy exists. See ECF No. 3687 at 1. We first review the applicable legal standards before assessing the asserted agreement and the parties’ proffered evidence.

1. Applicable Law

Section 1 of the Sherman Act provides that “[e]very contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States, or with foreign nations, is declared to be illegal.” 15 U.S.C § 1. “In order to prove [a] claim that the defendants engaged in illegal price-fixing” under Section 1 of the Sherman Act,¹⁷ a plaintiff must prove each of the following four elements:

First, that the defendants entered into an agreement or conspiracy;

Second, that the purpose or predictable effect of this agreement or conspiracy was to fix prices;

Third, that the agreement or conspiracy restrained or affected interstate (or foreign) commerce; and

Fourth, that the plaintiff was injured in its business or property as a result of the defendants’ actions.

See 4 Modern Federal Jury Instructions-Civil P 79.02 (2025).¹⁸

Our discussion of conspiracy will focus on the first element of this test: whether “defendants entered into an agreement or conspiracy[.]” Id. To establish this element, “proof of joint or concerted action is required; proof of unilateral action does not suffice.” Anderson News, L.L.C. v. Am. Media, Inc. (“Anderson News I”), 680 F.3d 162, 183 (2d Cir. 2012) (citation omitted). “ ‘[T]he crucial question’ is whether the challenged anticompetitive conduct ‘stem[s] from [an] independent decision or from an agreement, tacit or express.’ ” Bell Atl. Corp. v. Twombly, 550 U.S. 544, 553 (2007) (quoting Theatre Enters., Inc. v. Paramount Film Distrib. Corp., 346 U.S. 537, 540 (1954)). “Indeed, parallel behavior that does not result from an agreement is not unlawful even if it is anticompetitive,” as such “behavior could be the result of ‘coincidence, independent responses to common stimuli, or mere interdependence unaided by an advance understanding among the parties.’ ” United States v. Apple, Inc., 791 F.3d 290, 315 (2d Cir. 2015) (citation omitted). Instead, “[c]ircumstances must reveal a unity of purpose or a common design and understanding, or a meeting of minds in an unlawful arrangement.” Gelboim, 823 F.3d at 781 (citations and quotation marks omitted).

Accordingly, “[t]o survive a motion for summary judgment,” a plaintiff “must present evidence ‘that tends to exclude the possibility’ that the alleged conspirators acted independently.” Matsushita Elec. Indus. Co. v. Zenith Radio Corp. (“Matsushita”), 475 U.S. 574, 588 (1986) (citation omitted). That evidence can be either “direct or circumstantial” in nature and it must “reasonably tend[ ] to prove that the [defendants] had a conscious commitment to a common scheme designed to achieve an unlawful objective.” Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752, 768 (1984). A court must “examine the existence of a conspiracy ‘as a whole’ taking into consideration the totality of the evidence, as opposed to ‘dismembering it and viewing its separate parts.’ ” Ross v. Am. Express Co., 35 F. Supp. 3d 407, 438 (S.D.N.Y. 2014) (citation omitted), aff'd sub nom. Ross v. Citigroup, Inc., 630 F. App'x 79 (2d Cir. 2015), as corrected (Nov. 24, 2015).¹⁹

Three aspects of the Matsushita standard are worthy of special mention. First, a plaintiff need not “disprove all nonconspiratorial explanations for the defendants’ conduct; rather, the evidence need be sufficient only to allow a reasonable fact finder to infer that the conspiratorial explanation is more likely than not.” Anderson News, L.L.C. v. Am. Media, Inc. (“Anderson News II”), 899 F.3d 87, 98 (2d Cir. 2018) (citations and quotation marks omitted). Second, “[w]here a plaintiff's theory of recovery is implausible, it takes ‘strong direct or circumstantial evidence’ to satisfy Matsushita’s ‘tends to exclude’ standard.” In re Publ'n Paper Antitrust Litig., 690 F.3d 51, 63 (2d Cir. 2012) (citation omitted). Thus, it is important to examine “the nature of the alleged agreement itself,” Anderson News II, 899 F.3d at 99, before evaluating the evidence in detail. Third, “antitrust law limits the range of permissible inferences from ambiguous evidence in a § 1 case.” Matsushita, 475 U.S. at 588. Accordingly, “[a]mbiguous evidence—evidence that is equally consistent with independent conduct as with illegal conspiracy—is not enough” to support a claim of conspiracy. SourceOne Dental, Inc. v. Patterson Companies, Inc., 310 F. Supp. 3d 346, 357 (E.D.N.Y. 2018) (citing Apple, Inc., 791 F.3d at 315).²⁰

2. The Asserted Agreement

We begin by “examin[ing] the nature of the [purported] agreement itself[,]” before considering the direct and circumstantial evidence offered by plaintiffs. Anderson News II, 899 F.3d at 99. If “the conspiracy claim is one ‘that simply makes no economic sense,’ ” plaintiffs “ ‘must come forward with more persuasive evidence to support [their] claim than would otherwise be necessary.’ ” AD/SAT, 181 F.3d at 235 (quoting Matsushita, 475 U.S. at 587 (alterations in original)); see also Eastman Kodak Co. v. Image Tech. Servs., Inc., 504 U.S. 451, 468–69 (1992) (“If the plaintiff's theory is economically senseless, no reasonable jury could find in its favor, and summary judgment should be granted.”).

As this case has advanced over the last fourteen years, plaintiffs’ theories of defendants’ alleged conspiracy have changed. At the outset of this litigation, plaintiffs alleged that the Panel Banks conspired in an effort to, among other things, “increase profits in individual financial transactions.” Gelboim, 823 F.3d at 766; see also LIBOR I, 935 F. Supp. 2d at 679, slip op. at *9 (similar). After the Second Circuit rejected this profit-based theory as “strange” given that “the Banks operated not just as borrowers but also as lenders in transactions that referenced LIBOR,” Gelboim, 823 F.3d at 783, plaintiffs then focused on proving that defendants possessed a “reputation-based motive to conspire, where increased profits followed from a positive reputation,” LIBOR VI, 2016 WL 7378980, at *5, slip op. at *11 (citation omitted).

Plaintiffs’ theories of the conspiracy have evolved further. At the close of discovery, OTC plaintiffs and DAPs separately responded to defendants’ contention interrogatories, which asked them to “detail the alleged conspiracy, e.g., what Defendants agreed to do, communications reflecting the purported agreement, and the degree to which LIBOR was suppressed.” Defs. 56.1 ¶ 7. Plaintiffs submitted these responses as exhibits to declarations of counsel, William Christopher Carmody, in opposition to defendants’ joint summary judgment motion: (1) OTC Plaintiffs’ Second Amended Responses to Defendants’ First Set of Interrogatories, ECF No. 4294-38 (Carmody Decl., Ex. 418); and (2) DAPs’ Amended Responses and Objections to Defendants’ First Set of Interrogatories, ECF No. 4527-49 (Carmody Decl., Ex. 1319). We refer to these as OTC Plaintiffs’ Interrogatory Responses and DAPs’ Interrogatory Responses, respectively.

Plaintiffs now contend that the “objects” of defendants’ conspiracy included the following: (1) “to project financial strength,” Pls. Br. at 82;²¹ (2) to avoid the “threat of a catastrophic domino or contagion effect across the Panel Banks” if any one bank appeared to be an outlier relative to its “pack” of peers, Pls. Br. at 1, 14 (citations omitted);²² and (3) “to conceal the fact that they were disregarding the LIBOR rules” so they could “remain on the LIBOR panel,” DAPs’ Interrogatory Responses, Resp. No. 9 at 144, 150.²³

In furtherance of these objectives, plaintiffs assert that defendants agreed to: (1) “persistently suppress[ ] LIBOR,” Pls. Br. at 127 (citations omitted);²⁴ (2) “set their LIBOR submissions together within a ‘pack,’ ” Pls. Br. at 1;²⁵ and (3) “disregard the LIBOR rules in order to make artificially low LIBOR submissions while coordinating to conceal an agreement to disregard the LIBOR rules[.]”²⁶ With respect to suppression, plaintiffs admit that the alleged conspiracy “did not involve an agreement on a specific level or magnitude of suppression.”²⁷

There are four main reasons why plaintiffs’ asserted conspiracy is economically senseless. First, in dollars and cents, the conspiracy makes no sense. While a persistent effort to suppress LIBOR may help defendants’ borrowing positions by reducing the rate of interest on payments to creditors, such an effort would also hurt defendants’ lending positions. See LIBOR VI, 2016 WL 7378980, at *3, slip op. at *6 (“If, as plaintiffs suggest, the conspiracy were profit-motivated, it would have required all of the sixteen panel banks to have made a parallel decision to be net borrowers of money over the suppression period in the LIBOR-based lending market.”), id. (“Contrary to Shakespeare's advice, ‘Neither a borrower nor a lender be,’ the defendant banks are both.”). Indeed, the Circuit in Gelboim wrote:

[C]ommon sense dictates that the Banks operated not just as borrowers but also as lenders in transactions that referenced LIBOR. Banks do not stockpile money, any more than bakers stockpile yeast. It seems strange that this or that bank (or any bank) would conspire to gain, as a borrower, profits that would be offset by a parity of losses it would suffer as a lender.

Gelboim, 823 F.3d at 783 (emphasis added). Despite its skepticism, the Circuit noted at the time that the “record [wa]s undeveloped,” and the “potential of a wash” could “only be properly analyzed at later stages of the litigation[.]” Id. In the nearly ten years since Gelboim, plaintiffs have had the virtually unfettered opportunity to develop a substantial record, see supra pp. 10-14, yet plaintiffs have presented no evidence to dispel the Circuit's skepticism. See Pls. Br. at 87 n.136.²⁸

Second, there is no reason for the Panel Banks to have conspired when they could have achieved the same benefits with unilateral action. As plaintiffs’ own experts concede, a panel bank could have realized the same credit-signaling benefits by unilaterally setting LIBOR at a level that conveyed their credit strength relative to peers. See Defs. 56.1 ¶ 348 (“Dr. Cragg [OTC plaintiffs’ expert] testified that Panel Banks had a unilateral incentive to signal financial strength.”); ECF No. 4160-4 (Cragg July 25, 2024 Dep. Tr.) at 66:22-24 (“Q. Could the banks have suppressed LIBOR and been in a pack without colluding? A. Yes.”); ECF No. 4304-24 (Marx June 25, 2024 Dep. Tr.) at 89:10–90:8 (“Q. An individual panel bank decides that it wants to make a submission that is not an outlier submission. Are you with me? A. Okay. Q. That individual panel bank could do so without forming an agreement with other banks not to make outlier submissions, correct? A. I think that's correct. There was a plethora of information that would allow the banks to make submissions within the pack.”); ECF No. 4304-26 (Marx June 25, 2024 Dep. Tr.) at 316:12-17 (A. “Here the -- the benefit of being able to use artificially low LIBOR submissions to signal financial strength in a time of high financial volatility and stress is something that would be benefitting each of the defendant banks, both individually and collectively.”).²⁹

Third, plaintiffs’ contentions regarding the Panel Banks’ means of furthering this alleged agreement are internally inconsistent, as a LIBOR submission in the “pack” is not necessarily “suppress[ed],” and vice versa. As one of plaintiffs’ experts opines, the Panel Banks were “concerned with staying ‘in the pack,’ ” as they were “motivated to make submissions that were neither too high nor too low relative to others.” ECF No. 4160-3 (Cragg Reply Report) ¶ 26 (emphasis added). At the same time, plaintiffs contend that the Panel Banks suppressed LIBOR by “underreporting the rate at which they could borrow funds.” OTC Plaintiffs’ Interrogatory Responses, Resp. No. 6 at 123; see also DAPs’ Interrogatory Responses, Resp. No. 6 at 120 (“DAPs contend that Defendants colluded to disregard the LIBOR rules in order to make artificially low LIBOR submissions and also coordinated to conceal that agreement to disregard the LIBOR rules.” (emphasis added)). This means that, if a Panel Bank artificially inflated its LIBOR submissions so it was neither too high nor too low relative to others, plaintiffs could use this submission as evidence of the alleged conspiracy. On the other hand, if a different Panel Bank, on the same day, artificially lowered its LIBOR submission under the rate at which it could borrow funds, plaintiffs could also offer this LIBOR submission as evidence of the very same conspiracy.³⁰ This heads-we-win, tails-you-lose approach “simply makes no economic sense,” Matsushita, 475 U.S. at 587, ex ante and at the very least, is puzzling, cf. LIBOR IV, 2015 WL 6243526, at *47, slip op. at *117-18 (“Indeed, it is puzzling how a conspiracy theory could be consistent with the attempts of some banks to be near the bottom of the pack, because not every bank could be below average.” (emphasis in original)).³¹

Finally, the alleged agreement is simply so vague that it would have been impossible to enforce and, hence, economically senseless. Plaintiffs admit that the alleged conspiracy “did not involve an agreement on a specific level or magnitude of suppression.” OTC Plaintiffs’ Interrogatory Responses, Resp. No. 6 at 123; see also DAPs’ Interrogatory Responses, Resp. No. 6 at 120 (similar). Similarly, as plaintiffs’ experts concede, there is no precise definition of the “pack[.]”³² So, even though each Panel Bank's LIBOR submission was transparent to the entire market, ECF No. 4160-1 (“Cragg Opening Report”) ¶ 126, there was no disciplinary mechanism in place if a bank suppressed too much or too little in violation of the sixteen-banks’ apparent agreement, see Kleen Prods. LLC v. Georgia-Pac. LLC, 910 F.3d 927, 937 (7th Cir. 2018) (“If this was a cartel, it would have tried to impose disciplinary measures on the ‘cheaters’ who did not go along with the price increases.”).³³

Ultimately, this case calls to mind the Second Circuit's decision in Anderson News II. There, the Second Circuit “observed at the pleading stage that the alleged conspiracy could be plausible” and “emphasized in that opinion the significance of [certain] allegation[s].” 899 F.3d at 100 (citation omitted). However, the Second Circuit later found that the agreement “ ‘makes no economic sense,’ ” given that “on summary judgment, evidence of key facts that would support this theory have not materialized,” and the Second Circuit's earlier determination at the pleading stage was “not determinative of the conspiracy's ultimate ‘plausibility.’ ” Id. at 99-102, 100 n.3 (citations omitted). The same result applies here.

Because plaintiffs’ asserted conspiracy is economically senseless, plaintiffs “ ‘must come forward with more persuasive evidence to support [their] claim[s] than would otherwise be necessary.’ ” Id. at 99 (quoting Matsushita, 475 U.S. at 597).³⁴ As stated earlier, this evidence may be either direct or circumstantial in nature. See Monsanto, 465 U.S. at 768. We begin with a review of the direct evidence underlying plaintiffs’ claims.

3. Direct Evidence of Alleged Sixteen-Bank Conspiracy

To establish a price-fixing conspiracy, a plaintiff may, of course, provide direct evidence of the existence of the conspiracy. See SourceOne Dental, Inc., 310 F. Supp. 3d at 357. “Direct” evidence is that which “is explicit and requires no inferences to establish” the existence of the conspiracy. In re Mylan N.V. Sec. Litig., 666 F. Supp. 3d 266, 319 (S.D.N.Y. 2023) (quoting In re Baby Food Antitrust Litig., 166 F.3d 112, 118 (3d Cir. 1999)), aff'd sub nom. Menorah Mivtachim Ins. Ltd. v. Sheehan, 23 Civ. 720, 2024 WL 1613907 (2d Cir. Apr. 15, 2024), cert. denied, 145 S. Ct. 436 (2024); see also B & R Supermarket, Inc. v. Visa Inc., No. 17 Civ. 2738 (MKB), 2024 WL 4334075, at *14 (E.D.N.Y. Sept. 27, 2024) (“[D]irect evidence requires an explicit showing of an agreement that requires no inferences to conclude that a conspiracy exists.”), reconsideration denied, No. 17 Civ. 2738 (MKB), 2025 WL 510041 (E.D.N.Y. Feb. 14, 2025), and motion to certify appeal denied, No. 17 Civ. 2738 (MKB), 2025 WL 845109 (E.D.N.Y. Mar. 18, 2025).

In this context, direct evidence of a price-fixing conspiracy might include “a recorded phone call in which two competitors agreed to fix prices,” Mayor & City Council of Balt. v. Citigroup, Inc., 709 F.3d 129, 136 (2d Cir. 2013), or documents or conversations “explicitly manifesting the existence of the agreement in question,” In re Ins. Brokerage Antitrust Litig., 618 F.3d 300, 324-25 n.23 (3d Cir. 2010). However, direct evidence need not be a “smoking gun” that conclusively establishes the existence of a conspiracy. Mayor & City Council, 709 F.3d at 136.

Here, the only direct evidence of a price-fixing conspiracy offered by plaintiffs are acontextual excerpts from the deposition testimony of two Panel Bank employees regarding the alleged sixteen-bank conspiracy. First, plaintiffs present an excerpt from the deposition testimony of Louisa Sans, Bank of America's LIBOR submitter during the relevant period. See Pls. Br. at 75. In that testimony, Ms. Sans, reviewing contemporaneous documents from her time at Bank of America, indicated that she was aware of a “market consensus” around the setting of LIBOR, which she clarified as a “general ‘understanding’ as to where LIBOR submissions would be made.” Id. (citing ECF No. 4527-25 (Sans Dep. Tr.) at 121:13–123:19; 127:17-128:2). Next, plaintiffs offer an excerpt from the deposition testimony of Tony Miller, Credit Suisse's LIBOR submitter during the relevant timeframe. See id. Plaintiffs highlight that, in response to a question about whether there was a “consensus ․ as to where LIBOR was going to fix that day,” Mr. Miller answered in the affirmative, stating that he believed there to be “an agreement as to where the brokers felt that the fix would be.” Id. (citing ECF No. 4527-21 (Miller Dep. Tr.) at 115:22-116:16).³⁵

These two pieces of evidence fall well short of persuading the Court that there is “sufficient evidence to allow a reasonable fact finder to infer that the conspiratorial explanation is more likely than not.” Publ'n Paper, 690 F.3d at 63 (citation and quotation marks omitted).

First, plaintiffs’ proffered testimony, from only two individuals, does not unambiguously establish the existence of a sixteen-bank, multi-year conspiracy. Plaintiffs suggest that use of the term “consensus” by two LIBOR setters is somehow talismanic, such that its mere invocation offers unequivocal evidence of an antitrust conspiracy. Pls. Br. at 75. However, plaintiffs have improperly conflated “consensus” and conspiracy, for there was nothing inherently unlawful with a LIBOR submitter maintaining a view, perhaps informed by broker estimates, research, and the bank's financial metrics at any given point, as to where the market was trending that was similar to that of his or her peers. See, e.g., Defs. 56.1 ¶¶ 36-40, 43, 51, 52, 60, 69, 81, 85, 88, 100, 216 (collecting testimony from various submitters as to the factors they used to form their submissions).

Further, plaintiffs do not reference critical portions from each witness's deposition testimony in their briefing, which, when properly included, place the surgically selected quotes from Ms. Sans and Mr. Miller in a wholly different light. For example, plaintiffs fail to note that Ms. Sans testified that she “never spoke to other LIBOR panel banks,” ECF No. 4175-66 (Sans Dep. Tr.) at 182:15–183:7, which significantly undermines plaintiffs’ assertion that a “reasonable jury could credit these admissions as direct evidence of the Panel Banks’ conspiracy,” Pls. Br. at 75. Similarly, plaintiffs fail to offer a complete representation of Mr. Miller's testimony. Later in his testimony, Mr. Miller explained that this “agreement as to where the brokers felt that the fix would be” referred to the brokers’ process of “averaging ․ where the fix looks likely” not “where it should fix” which depended on “the banks’ rules for each individual submission.” ECF No. 4527-21 (Miller Dep. Tr.) at 116:14-117:13.

Simply put, these excerpted portions of Ms. Sans's and Mr. Miller's testimony do not satisfy the threshold for establishing an antitrust conspiracy on direct evidence. These selected quotes do not constitute “an explicit showing of an agreement that requires no inferences to conclude that a conspiracy exists.” B & R Supermarket, Inc. v. Visa Inc., 2024 WL 4334075, at *14. At best, drawing all inferences in plaintiffs’ favor, the selected portions of deposition testimony constitute ambiguous “[e]vidence that is equally suggestive of competition as collusion,” which “is insufficient to survive summary judgment.” In re Payment Card Interchange Fee & Merch. Disc. Antitrust Litig., 714 F. Supp. 3d 65, 81 (E.D.N.Y. 2024) (collecting cases); see, e.g., In re Text Messaging Antitrust Litig., 782 F.3d 867, 879 (7th Cir. 2015) (“We can, ․ without suspecting illegal collusion, expect competing firms to keep close track of each other's pricing and other market behavior and often to find it in their self-interest to imitate that behavior rather than try to undermine it[.]”); Michelman v. Clark-Schwebel Fiber Glass Corp., 534 F.2d 1036, 1048 (2d Cir. 1976) (“Given the legitimate function of [creditworthiness] data [of customers, to protect sellers from risk exposure], it is not a violation of [Sherman Act §] 1 to exchange such information, provided that any action taken in reliance upon it is the result of each firm's independent judgment, and not of agreement.”).

It is revealing that, after years of developing a record -- including the production of millions of documents, tens of thousands of recorded phone calls, more than one thousand written discovery requests, deposition testimony from 59 current or former Panel Bank employees and three former BBA employees, Defs. 56.1 ¶¶ 5-10; see also supra pp. 10-14 -- plaintiffs offer nothing that approaches direct evidence of a conspiracy. See Monsanto, 465 U.S. at 768 (holding plaintiffs must present “direct or circumstantial evidence that reasonably tends to prove that the manufacturer and others had a conscious commitment to a common scheme designed to achieve an unlawful objective”).

Plaintiffs’ asserted conspiracy would have required significant coordination and communication throughout the relevant period. See ECF No. 4321-1 (“Marx Opening Report”) ¶ 118 (“regular meetings or communications can contribute to a cartel's success” (emphasis in original)), Cragg Opening Report ¶ 114 (“To collude, firms need to coordinate, and they most easily do so by explicitly communicating with one another.”). Given that, the dearth of inculpatory evidence here is all the more stark. See Apex Oil Co. v. DiMauro, 822 F.2d 246, 253 (2d Cir. 1987) (noting that conspiracies which “concern[ ] long-term complex relationships among competitors” are typically “more susceptible of direct proof.” (citations omitted)).

Based on the record submitted by plaintiffs, we have no difficulty concluding that plaintiffs have not offered direct evidence remotely sufficient to raise a genuine question of material fact as to the existence of a conspiracy. Accordingly, we now turn to an evaluation of the circumstantial evidence.

4. Circumstantial Evidence of Alleged Sixteen-Bank Conspiracy

Having found that “the scheme alleged is implausible” and there is insufficient direct evidence, plaintiffs here must present “strong circumstantial evidence” of the asserted conspiracy to sustain their antitrust claims. Apex Oil Co., 822 F.2d at 253 (emphasis added).

Strong circumstantial evidence of a conspiracy must consist of: (1) evidence of defendants’ parallel conduct; and (2) the presence of certain “plus factors,” i.e., the existence of additional circumstances that could allow a factfinder to infer a conspiracy. Id. Importantly, these two prongs are conjunctive, meaning that proof of either parallel conduct or the presence of “plus factors” is insufficient on its own to support an antitrust conspiracy. Id.; see also Apple, Inc., 791 F.3d at 315 (“Parallel action is not, by itself, sufficient to prove the existence of a conspiracy.”); Mylan N.V., 666 F. Supp. 3d at 318 (“The plus factors are necessary, not sufficient, and are still subject to case-specific assessment by the court since ‘such plus factors may not necessarily lead to an inference of conspiracy.’ ” (citation omitted)).

However, before we can analyze plaintiff's circumstantial evidence, we must first address the pending Daubert motions directed to plaintiffs’ experts. See, e.g., Colon ex rel. Molina v. BIC USA, Inc., 199 F. Supp. 2d 53, 68 (S.D.N.Y. 2001) (“[T]he court must evaluate evidence for admissibility before it considers that evidence in ruling on a summary judgment motion.” (citing Fed. R. Evid. 104(a)). This is critical, since plaintiffs rely heavily on the testimony of their experts to provide circumstantial evidence to support their conspiracy allegations.

4.1. Daubert Motions Regarding Drs. Cragg and Marx

Plaintiffs offer two experts to bolster their circumstantial evidence of the alleged conspiracy. OTC plaintiffs rely on the testimony of Dr. Michael I. Cragg, while FFFP plaintiffs rely on the testimony of Dr. Leslie Marx. In opposition, defendants collectively rely on the expert testimony of Dr. John Asker, and individual defendants JPMorgan and Credit Suisse independently retained experts Dr. Dennis Carlton and Dr. Divya Mathur, respectively.

Defendants moved to exclude the opinions of Drs. Cragg and Marx, ECF Nos. 4156, 4147, in separate sets of briefing, see ECF Nos. 4158 (“Cragg – Defs. Br.”), 4314 (“Cragg – Pls. Br.”), 4440 (“Cragg – Defs. Reply Br.”) and Nos. 4148 (“Marx – Defs. Br.”), 4320 (“Marx – Pls. Br.”), 4437 (“Marx – Defs. Reply Br.”). Plaintiffs have not moved to exclude defendants’ experts.³⁶

We begin by describing Dr. Cragg's opinions and challenges thereto. We then utilize the same procedure as it relates to Dr. Marx.

4.1.1. Dr. Cragg

OTC plaintiffs’ expert, Dr. Cragg, issued three reports: (1) an initial report, dated April 19, 2024 (Decl. of Jason M. Hall, Ex. 1, October 4, 2024, ECF No. 4160); (2) a rebuttal report, dated June 18, 2024 (Decl. of Jason M. Hall, Ex. 2, October 4, 2024, ECF No. 4160); and (3) a reply report, dated August 16, 2024 (Decl. of Jason M. Hall, Ex. 3, October 4, 2024, ECF No. 4160).³⁷ We refer to these as the Cragg Opening Report, Cragg Rebuttal Report, and Cragg Reply Report, respectively.

Dr. Cragg principally offers four opinions across his three reports. First, Dr. Cragg opines that the LIBOR-setting process was “susceptible to collusion” during the class period. See Cragg Opening Report ¶¶ 98-105, 124-36; Cragg Reply Report ¶¶ 122, 126-30. Second, he states that the Panel Banks had incentives to collude. See Cragg Opening Report ¶¶ 106-12, 137; Cragg Reply Report ¶¶ 7, 15-26. Third, Dr. Cragg asserts that the Panel Banks’ actions were consistent with collusion and inconsistent with independent action. See Cragg Opening Report ¶¶ 162-238; Cragg Rebuttal Report ¶¶ 46-111; Cragg Reply Report ¶¶ 27-30, 66-68. He bases this third opinion on his review of select items of record evidence and his econometric collusion screen. See id. Lastly, Dr. Cragg rebuts the analysis proffered by OTC Defendants’ experts regarding the possibility of collusion. See generally Cragg Rebuttal Report; Cragg Reply Report.

Defendants do not challenge Dr. Cragg's qualifications, and we agree that Dr. Cragg is generally qualified to offer his opinions.³⁸ We review each opinion to determine its admissibility.

4.1.1.1. LIBOR Structure's Susceptibility to Collusion

Dr. Cragg first opines that the “structure of the LIBOR setting process during the Class Period exhibited many of the features economics literature describes as characteristics of markets susceptible to collusion.” Cragg Opening Report ¶ 8(b). To this end, Dr. Cragg identifies several features, including pricing structures, homogeneity in economic incentives, communication channels, barriers to entry, and regulatory enforcement, that he stresses are generally indicative of a market that is susceptible to cartel-like conduct. Id. ¶ 105.

All these factors, Dr. Cragg opines, were present among the Panel Banks in the LIBOR-setting context. See id. To demonstrate the presence of these factors, Dr. Cragg relies heavily on a review of the record. He cites over fifty examples from the record of communications among the Panel Banks, between the Panel Banks and the FXMMC and BBA, and between the Panel Banks and interdealer brokers to support the notion that defendants had the ability to communicate freely and frequently, which is a “common feature” of collusion. Id. ¶¶ 114-23.

The parties’ briefing does not address this particular opinion of Dr. Cragg's in detail. See, e.g, Cragg – Pls. Br. at 11 n.10. However, in keeping with this Court's gatekeeping role, we may still assess the opinion's admissibility pursuant to Rule 702. See Keurig Green Mountain Single-Serve Coffee, 2025 WL 354671, at *3 n.2 (collecting cases).

Expert testimony must not “usurp either the role of the trial judge in instructing the jury as to the applicable law or the role of the jury in applying that law to the facts before it.” United States v. Bilzerian, 926 F.2d 1285, 1294 (2d Cir. 1991) (citations omitted). “The distinction between a legal conclusion and appropriate expert testimony is ‘extremely fine.’ ” U.S. Info. Sys., Inc. v. Int'l Bhd. of Elec. Workers Loc. Union No. 3, AFL-CIO, 313 F. Supp. 2d 213, 240 (S.D.N.Y. 2004) (citation omitted).

Applying this standard to expert testimony concerning a conspiracy, courts have generally allowed experts to “testify that the ‘climate’ of a specific market was consistent with a conspiracy” and “identif[y] factors, the existence of which would tend to indicate a conspiracy.” Id. (collecting cases). “These subjects are appropriate for an economist to opine on, given that economists regularly study the conditions that give rise to collusion. Further, this [type of] testimony would be helpful to the finder of fact, since the conditions of a market that make collusion possible are outside the layperson's body of knowledge and expertise.” In re Processed Egg Prods. Antitrust Litig., 81 F. Supp. 3d 412, 422 (E.D. Pa. 2015). However, an expert cannot “state that the defendants did or did not engage in anticompetitive conduct.” U.S. Info. Sys., 313 F. Supp. at 240. As a result, an expert “can aid the jury with his opinions on hypothetical facts, from which plaintiff can argue that the jury should draw certain inferences; but [an expert] cannot inform the jury that based on the evidence defendants likely engaged in anticompetitive conduct.” DPWN Holdings (USA), Inc. v. United Air Lines, Inc., No. 11 Civ. 0564 (BMC) (PK), 2019 WL 1515231, at *7 (E.D.N.Y. Feb. 21, 2019).

Thus, Dr. Cragg could, at a trial, offer opinions as to “whether market conditions reflected what economists might have expected to see as a result of anticompetitive conduct.” B & R Supermarket, Inc. v. Visa Inc., No. 17 Civ. 2738 (MKB), 2024 WL 4252031, at *14 (E.D.N.Y. Sept. 20, 2024) (citation omitted). However, Dr. Cragg could not “tell the jury ․ what weight, if any, to give them, ․ [or] what result to reach.” Id. (citations and quotation marks omitted).³⁹

4.1.1.2. Panel Banks’ Incentives to Collude

Second, Dr. Cragg opines that the Panel Banks had incentives to collude. See Cragg Opening Report ¶¶ 106-112, 137; Cragg Reply Report ¶¶ 7, 15-26.

Dr. Cragg opines that the Panel Banks had an “individual and collective” “incentive to suppress LIBOR to prop up their reputation and perceived creditworthiness” in the midst of the financial crisis. ECF No. 4160-4 (Cragg July 25, 2024 Dep. Tr.) at 107:16-20; Cragg – Pls. Br. at 3 (citation omitted). However, Dr. Cragg opines that the Panel Banks “were incentivized to suppress LIBOR collectively, rather than independently,” because: (1) “outlier” submissions outside of the “pack” could draw negative attention to a bank or banks that were comparatively “too high” or “too low,” Cragg Opening Report ¶ 109, 111; Cragg Reply Report ¶ 26; and (2) the downfall of one financial institution could have a “contagion” or domino effect, spreading across financial institutions, Cragg Opening Report ¶¶ 110, 111; Cragg Rebuttal Report ¶¶ 9, 14; Cragg Reply Report ¶ 15.

Dr. Cragg purports to corroborate this opinion with a review of the record. For example, his report includes thirty excerpted communications “between Panel Bank employees including FXMMC representatives and LIBOR submitters, and BBA employees,” which he concludes “corroborate the [collective] incentive [for defendants] to remain ‘within the pack.’ ” Cragg Opening Report ¶ 113.⁴⁰

We faced a similar issue at the class-certification stage in LIBOR VII. There, Exchange plaintiffs’ expert, Dr. Janet Netz, “interpret[ed] trader communications to opine on the traders’ actual motives and states of mind,” eventually concluding that “certain challenged conduct in fact occurred and Exchange plaintiffs were in fact impacted.” LIBOR VII, 299 F. Supp. 3d at 490, slip op. at *79, 81. We noted then that: “[w]hile she may opine on traders[’] and LIBOR submitters’ economic incentives as a general matter, those opinions may not extend to the traders[’] and submitters’ specific states of mind.” Id. at 490, slip op. at *81.

The same principle applies here. While Dr. Cragg may opine generally on LIBOR submitters’ economic incentives based on his economic expertise, review of applicable academic literature, and assessment of the climate of the LIBOR market, Dr. Cragg may not opine on “submitters’ specific states of mind.” LIBOR VII, 299 F. Supp. 3d at 490, slip op. at *81; see also Anderson News, L.L.C. v. Am. Media, Inc., No. 09 Civ. 2227 (PAC), 2015 WL 5003528, at *4 (S.D.N.Y. Aug. 20, 2015) (excluding opinions which referenced defendants’ “motivations, thought processes, and understanding”), aff'd, 899 F.3d 87 (2d Cir. 2018).⁴¹

4.1.1.3. Consistency of Panel Banks’ Actions with Collusion

Next, Dr. Cragg opines that the Panel Banks’ actions were consistent with their engaging in collusion during the class period. See Cragg Opening Report ¶¶ 162-238; Cragg Rebuttal Report ¶¶ 46-111; Cragg Reply Report ¶¶ 27-30, 66-68.

To begin, we reiterate, as a general rule, that while an expert may offer “opinions as to whether certain conduct was indicative of collusion,” he may not “tell the jury what result to reach.” B & R Supermarket, 2024 WL 4252031, at *14 (citations and quotation marks omitted); cf. ECF No. 4465 ¶¶ 1, 2 (Stipulation and Order stating that Dr. Carlton, JPMorgan's expert, may testify as to whether certain facts were “consistent with unilateral action” but barring him from testifying that “JPMorgan ․ did not enter into an agreement or conspiracy”).

Additionally, the bases for Dr. Cragg's opinion must be reliable. Dr. Cragg arrives at this opinion based on (1) his review of select pieces in the vast discovery record and (2) his econometric collusion screen. We review each basis of Dr. Cragg's opinion separately.

4.1.1.3.1. Record Evidence

Dr. Cragg first “evaluate[s] whether the ․ [record] evidence is consistent with the Panel Banks engaging in collusion during the Class Period.” Cragg Opening Report ¶ 137. In arriving at his opinion, Dr. Cragg “relied on 18 out of 60 witness deposition transcripts ․ and 170 documents, chiefly exhibits used during those depositions.” Cragg – Pls. Br. at 10 (citations omitted). Indeed, even a cursory review of the first seventy-one pages of Dr. Cragg's opening report, id. ¶¶ 12-136, makes it plainly obvious that his analysis is non-economic in nature, for he merely recites cherry-picked individual communications from the record that would be readily understood by any lay person without assistance.

“[W]hile an expert must of course rely on facts or data in formulating an expert opinion, see Fed. R. Evid. 703, an expert cannot be presented to the jury solely for the purpose of constructing a factual narrative based upon record evidence.” Highland Cap. Mgmt., L.P. v. Schneider, 379 F. Supp. 2d 461, 469 (S.D.N.Y. 2005). Such “testimony addressing lay matters is not based on an ‘expert's scientific, technical, or other specialized knowledge,’ and therefore fails to satisfy the first part of Rule 702(a).” LIBOR VII, 299 F. Supp. 3d at 469, slip op. at *32. “[N]or is such a narration traceable to a reliable methodology.” Tourre, 950 F. Supp. 2d at 675.

Thus, to the extent that Dr. Cragg's opinion on this issue recites “a factual narrative of the case and addresses ‘lay matters which a jury is capable of understanding and deciding without the expert's help,’ it is inadmissible.” Highland Capital, 379 F. Supp. 2d at 469 (quoting Andrews v. Metro N. Commuter R.R. Co., 882 F.2d 705, 708 (2d Cir. 1989)); see also Processed Egg, 81 F. Supp. 3d at 421 (excluding testimony where expert “simply read[ ] and interpret[ed] evidence of collusion as any juror might”).⁴²

As we noted above, supra pp. 52-55, Dr. Cragg, at a trial, could “explain whether conduct is indicative of collusion,” B & R Supermarket, 2024 WL 4252031, at *14 (quoting U.S. Info. Sys., 313 F. Supp. at 240). However, he could not “tell the jury what result to reach.” Id. (citations and quotation marks omitted). And, of course, Dr. Cragg could not “state that the defendants did or did not engage in anticompetitive conduct.” U.S. Info. Sys., 313 F. Supp. at 240.

4.1.1.3.2. Econometric Collusion Screen

Dr. Cragg also attempts to create an econometric model to exclude the possibility that observed patterns in defendants’ LIBOR submissions resulted from “unilateral independent action,” or, stated another way, to determine whether defendants’ actions were consistent with collusion. See Cragg Opening Report ¶¶ 163, 180-238; Cragg Reply Report ¶¶ 27-30, 66-68. His model tests whether it would have been possible for the Panel Banks to provide submissions within a pack “without advance knowledge of each other's LIBOR submissions or a planned LIBOR fix (but with access to information that would have been available at the time without collusion or foreknowledge of future events).” Cragg Opening Report ¶ 8(d); see also Cragg – Pls. Br. at 4 (similar). Dr. Cragg first finds that the “Panel Banks consistently ma[de] LIBOR submissions that [were] flat to a suppressed LIBOR value in a volatile market.” Cragg Reply Report ¶ 30.⁴³ Such activity, he then determines, “is consistent with [the Panel Banks’] knowing in advance what the LIBOR fix is likely to be.” Id. He arrives at this opinion despite his admission that he does not “have a view” about “the exact agreement” that may have been made between the defendants. See ECF No. 4160-4 (Cragg July 25, 2024 Dep Tr.) at 78:2-17, 79:6-8.

Dr. Cragg's econometric collusion screen is irrelevant and unreliable. First, Dr. Cragg's model is irrelevant because it is based on a view of defendants’ conspiracy which differs from plaintiffs’ allegations of conspiracy. Dr. Cragg's econometric model assumes that each Panel Bank sought to target a LIBOR submission that was a “fixed distance” from the LIBOR fix to avoid being perceived as an outlier. Cragg – Pls. Br. at 16; see also Cragg Opening Report ¶¶ 186-188 (stating it is “statistically anomalous” and “highly improbable without collusion” that the Panel Banks “could maintain a stable distance to the LIBOR fix”); Cragg Reply Report ¶¶ 30, 30 n.53 (similar). However, plaintiffs do not claim that the conspiracy worked this way. Rather, plaintiffs contend that defendants conspired “to suppress LIBOR by underreporting the rate at which they could borrow funds,” which “did not involve an agreement on a specific level or magnitude of suppression.” See OTC Plaintiffs’ Interrogatory Responses, Resp. No. 6 at 123. There is thus no suggestion of a “fixed and stable” LIBOR submission scheme, as Dr. Cragg suggests.

Moreover, this self-invented theory of conspiracy was the only theory tested by Dr. Cragg, thereby rendering the results from his analysis wholly irrelevant. See Cragg Reply Report ¶ 113 (“If a bank is colluding but not submitting flat to LIBOR then my test would not identify such collusion.”).⁴⁴ Accordingly, since Dr. Cragg's analysis generates conclusions about a conspiracy that differs from OTC plaintiffs’ position and one which is not at issue in this case, Dr. Cragg's testimony must be excluded as irrelevant, misleading, and unhelpful to the fact finder. See Nimely, 414 F.3d at 397; Raskin v. Wyatt Co., 125 F.3d 55, 66 (2d Cir. 1997) (noting that “expert testimony which does not relate to” the specific issues “in the case are not relevant and, ergo, non-helpful”); Franklin v. Consol. Edison Co. of N.Y., Inc., 37 F. App'x 12, 14 (2d Cir. 2002) (excluding expert opinion as unreliable because “it did not ‘fit’ with the facts of plaintiff's case”).

Second, Dr. Cragg's analysis is also unreliable. As mentioned before, Dr. Cragg's model tests “whether it would have been possible for the Panel Banks to provide such clustered and suppressed submissions without advance knowledge of each other's LIBOR submissions or a [presumed] LIBOR fix (but with access to information that would have been available at the time without collusion or foreknowledge of future events).” Cragg Opening Report ¶ 8(d). To test for this question, Dr. Cragg constructs an econometric model -- which he calls the “Independent Forecast” -- that he claims is capable of “forecast[ing] the LIBOR fix” based on “information that the Panel Banks would have had available at the time ․ without collusion.” Id. ¶ 166.⁴⁵ He then compares the “Independent Forecast to the actual actions of the Panel Banks” during the relevant period, and concludes that “if Panel Banks were able to significantly outperform the Independent Forecast for some parts” of the relevant period, “that would be ․ consistent with collusion.” Id. ¶ 167.

Key to this analysis is whether Dr. Cragg's model accurately captured the “information that the Panel Banks would have had available at the time ․ without collusion.” Cragg Opening Report ¶ 166. The model, however, omitted key variables. For example, Dr. Cragg admitted in his deposition that his model was built on “day-old,” publicly available information, see ECF No. 4160-5 (Cragg Sept. 11, 2024 Dep. Tr.) at 229:8-14. Dr. Cragg thus did not consider other variables that a bank would likely rely upon in setting LIBOR, such as “bank-specific borrowing data from the same day,” “offers that [we]re received by each bank prior to 11 a.m.,” “market color from brokers just before 11 a.m.,” “movements in derivative prices from just before 11 a.m. on the same day,” and the “banks’ liquidity positions on the same day,” which could all materially change between market close the prior day and the time at which a defendant bank would prepare to make its LIBOR submission. ECF No. 4160-5 (Cragg Sept. 11, 2024 Dep. Tr.) at 234:15-243:9; see also ECF No. 4160-6 (Asker OTC Rebuttal Report) ¶ 32. After JPMorgan's expert, Dr. Carlton, observed that “many of the days on which” Dr. Cragg's model was “least accurate” “correspond to days with significant market news” in the morning, Dr. Cragg admitted that his model includes “forecasting errors” which “arise around significant unforeseen events like the failure of Lehman Brothers, the high-water mark of the 2007-2009 Financial Crisis.” Cragg Reply Report ¶ 58 (quotation marks omitted).

Thus, by excluding these clearly relevant variables from his model -- information available to the banks between day-prior market close and the next morning's LIBOR submission, ECF No. 4160-5 (Cragg Sept. 11, 2024 Dep. Tr.) at 229:8-14, 234:15-243:9 -- Dr. Cragg's model is based on insufficient facts and data that renders it inadmissible at trial. See Laumann v. Nat'l Hockey League, 117 F. Supp. 3d 299, 303, 315 (S.D.N.Y. 2015) (“[T]he law is clear: expert opinions are inadmissible if they are not ‘based on sufficient facts or data,’ or on a reliable application of scientific methods to those facts or data. This is true no matter how burdensome or difficult collecting relevant data or devising methods to apply to that data may be.”).⁴⁶

Dr. Cragg's reliance on a conspiracy theory of his own invention and stale data is sufficient to preclude Dr. Cragg's collusion screen.

4.1.1.4. Rebuttal Opinions

Lastly, Dr. Cragg endeavors to rebut defense experts’ analyses of the possibility of collusion. See generally Cragg Rebuttal Report; Cragg Reply Report. For example, Dr. Cragg opines that the Panel Banks’ LIBOR submissions “exhibit lower dispersion in the [c]lass [p]eriod as compared to before or after, not higher as Prof. Asker claims” and Prof. Asker's analysis is “flawed because it fails to account for the impact of the volatility of the LIBOR fix.” Cragg Rebuttal Report ¶ 5(a).⁴⁷

Though defendants move to exclude all of Dr. Cragg's opinions, defendants do not specifically refer to Dr. Cragg's rebuttal opinions in their briefing. See Cragg – Defs. Br.; Cragg – Defs. Reply Br.; see also Cragg – Pls. Br. at 2 (noting that defendants do not move to exclude these opinions).

Having reviewed Dr. Cragg's rebuttal opinions, the Court finds them to be admissible to the extent that they do not run afoul of the parameters listed above. Supra pp. 52-65; see, e.g., U.S. Small Bus. Admin. v. Feinsod, No. 17 Civ. 3586 (JS) (JMW), 2025 WL 1677409, at *22-23 (E.D.N.Y. June 13, 2025)(granting in part plaintiff's motion to preclude defense expert's opening report while denying plaintiff's motion to preclude defense expert's rebuttal report).⁴⁸

4.1.2. Dr. Marx

We turn next to assess the admissibility of Dr. Marx's opinions. Dr. Marx, as FFFP plaintiffs’ expert, issued three reports: (1) an initial report, dated May 28, 2024 (Decl. of James R. Martin, Ex. 1, November 22, 2024, ECF No. 4321); (2) a rebuttal report, dated June 18, 2024 (Decl. of James R. Martin, Ex. 2, November 22, 2024, ECF No. 4321); and (3) a reply report, dated August 16, 2024 (Decl. of James R. Martin, Ex. 3, November 22, 2024, ECF No. 4321).⁴⁹ We refer to these as the Marx Opening Report, Marx Rebuttal Report, and Marx Reply Report, respectively.

Like Dr. Cragg, Dr. Marx principally opines that: (1) LIBOR's fundamental features -- when viewed in light of the basic economic principles of costs and benefits -- created a system that was conducive to collusion, see Marx Opening Report ¶¶ 91, 117-62, 222-36; (2) Panel Banks had incentives to collude, see Marx Opening Report ¶¶ 92-116; and (3) Panel Banks’ actions were consistent with collusion and inconsistent with independent action, see Marx Opening Report ¶¶ 163-221, 237-349.⁵⁰ Like Dr. Cragg, Dr. Marx endeavors to rebut opinions authored by defendants’ experts. See generally Marx Rebuttal Report; Marx Reply Report.

However, there are critical differences between the opinions of Dr. Cragg and Dr. Marx. Dr. Marx uses a dispersion analysis, as opposed to an econometric collusion screen, to support her assessment that the Panel Banks’ actions were consistent with collusion. See Marx Opening Report ¶¶ 168-88. In addition, unlike Dr. Cragg, Dr. Marx shares opinions on a wide variety of issues. For example, Dr. Marx opines that evidence of concealment is a “super plus factor.” Marx Opening Report ¶ 242. Dr. Marx also opines that “broker indications were predictive of future LIBOR fixes and thus could facilitate exchange of information in support of collusive conduct.” Id. ¶ 257. She further opines that, “absent collusion, Defendants would have been disincentivized from unilaterally disregarding the LIBOR rules to make artificially suppressed LIBOR submissions ‘due to a credible threat of discipline from the [FXMCC], the BBA, or other Defendant Banks.’ ” Marx – Pls. Br. at 18 (quoting Marx Rebuttal Report ¶ 54).

Defendants do not challenge Dr. Marx's qualifications, and we agree that Dr. Marx is well qualified to offer her opinions.⁵¹ We now review Dr. Marx's opinions individually to assess whether each is admissible.

4.1.2.1. LIBOR Structure's Susceptibility to Collusion

First, Dr. Marx offers a general description of the history, structure, and setting process associated with LIBOR, ultimately opining that, when viewed in light of the basic economic principles of costs and benefits, LIBOR's fundamental features created a system conducive to collusion. See Marx Opening Report ¶¶ 14-91, 117-62, 222-36. For example, Dr. Marx cites the price transparency in the setting of LIBOR -- where each bank's submission was public -- along with other factors, including the number of banks involved, high barriers to entry, ability to avoid detection, and opportunities to communicate, as elements typically found in cartels. See id. ¶¶ 117-58. To further corroborate this view, Dr. Marx reviews various record entries purportedly supporting her view that other features of LIBOR, including the role of monitoring and enforcement, were conducive to collusion. See id. ¶¶ 222-36.

Like Dr. Cragg, supra pp. 52-55, Dr. Marx may offer her opinion as to “whether market conditions reflected what economists might have expected to see as a result of anticompetitive conduct.” B & R Supermarket, 2024 WL 4252031, at *14 (citation omitted). However, Dr. Marx may not “tell the jury how ‘to determine the facts and what weight, if any, to give them,’ ․ nor can she ‘undertake[ ] to tell the jury what result to reach.’ ” Id. (citations omitted). Nor may Dr. Marx recite “a factual narrative of the case and address[ ] ‘lay matters which a jury is capable of understanding and deciding without the [her] help[.]’ ” Highland Capital, 379 F. Supp. 2d at 469 (citation omitted). “To the extent [defendants] believe[ ] that any part of the testimony by Dr. [Marx] falls outside of these permissible bounds, [they] may object to such opinions as they are offered at trial.” B & R Supermarket, 2024 WL 4252031, at *14.⁵²

4.1.2.2. Panel Banks’ Economic Incentives

Dr. Marx also assesses the Panel Banks’ economic incentives. Her conclusion, which is largely based on a review of select items in the evidentiary record, is that LIBOR's features supplied defendants with sufficient economic incentives to collude, and further, that defendants could have, in fact, colluded to take advantage of these features in order to achieve reputational strength and the economic benefits derived therefrom. See Marx Opening Report ¶¶ 10, 92-116.

As noted above with respect to Dr. Cragg, Dr. Marx may opine generally on LIBOR submitters’ economic incentives but may not opine on “submitters’ specific states of mind.” LIBOR VII, 299 F. Supp. 3d at 490, slip op. at *81.⁵³

However, unlike Dr. Cragg, we must limit Dr. Marx's opinion even further. Dr. Marx acknowledges that she formulated her opinion that the Panel Banks had an incentive to collude based on her opinion that the Panel Banks did collude. Specifically, she contends that, “when you see the [Panel Banks] choose to engage in collusive conduct,” it suggests that those Panel Banks “believed the expected benefits to exceed the expected cost,” see ECF No. 4175-24 (Marx June 25, 2024 Dep. Tr.) at 95:20-96:8, 100:4-6, and that, as a result, the Panel Banks had sufficient “economic incentives to collude,” see Marx Opening Report ¶ 92; see also Marx Reply Report ¶¶ 158, 202 (similar). “Opinions that assume a conclusion and reverse-engineer[ ] a theory to fit that conclusion are ․ inadmissible.” See In re Mirena Ius Levonorgestrel-Related Prods. Liab. Litig. (No. II), 341 F. Supp. 3d 213, 241 (S.D.N.Y. 2018), aff'd sub nom. In re Mirena IUS Levonorgestrel-Related Prods. Liab. Litig. (No. II), 982 F.3d 113 (2d Cir. 2020) (citations and quotation marks omitted). “[S]uch ‘circular reasoning fails to reveal a sufficiently rigorous analytical connection between [Dr. Marx's] methodology and [Dr. Marx's] opinion,’ ” and, thus, Dr. Marx's opinion is excluded to the extent it rests on this basis. Bocoum v. Daimler Trucks N. Am. LLC, No. 17 Civ. 7636 (JPC) (BCM), 2022 WL 902465, at *11 (S.D.N.Y. Mar. 28, 2022) (citation omitted).⁵⁴

4.1.2.3. Consistency of Panel Banks’ Actions with Collusion

Dr. Marx also opines that “the economic evidence ․ is consistent with the Defendant Banks acting collusively during the Conduct Period to disregard the LIBOR rules in order to make artificially low USD LIBOR submissions while coordinating to conceal an agreement to disregard the LIBOR rules and is inconsistent with independent action.” Marx Opening Report ¶ 13 (emphasis omitted); see also id. ¶¶ 163-221, 237-349.⁵⁵ Dr. Marx bases this opinion on her review of certain portions of the documentary and testimonial record, as well as her dispersion analysis.

Again, as we previously observed, an expert may offer “opinions as to whether certain conduct was indicative of collusion,” but may not “tell the jury what result to reach.” B & R Supermarket, 2024 WL 4252031, at *14 (citations and quotation marks omitted). We thus review each basis of Dr. Marx's opinion separately.

4.1.2.3.1. Record Evidence

Critical to Dr. Marx's opinion that defendants acted collusively is her review of the “documentary record, deposition testimony, [and] data.” Marx Opening Report ¶ 13 (emphasis omitted).⁵⁶ As part of this review, Dr. Marx performs an “economic analysis” of various “plus factors” that may be indicative of collusive conduct. See Marx Opening Report ¶¶ 237-340. While this “economic analysis” is largely devoid of quantitative data, Dr. Marx reviews many plus factors that track the legal standard, including channels of communication, concealment efforts, actions against self-interest, compliance enforcement and monitoring, and suggests that the evidentiary record supports her overarching conclusion that defendants’ observable conduct is more likely to be the result of collusion than independent action. See id.

As noted above with respect to Dr. Cragg, Dr. Marx cannot merely recite “a factual narrative of the case” that a jury is equally capable of understanding without the expert's help. Highland Capital, 379 F. Supp. 2d at 469 (citation omitted). Moreover, at a trial, Dr. Marx could not “tell the jury what result to reach,” B & R Supermarket, 2024 WL 4252031, at *14 (citations and quotation marks omitted), and Dr. Marx could not “state that the defendants did or did not engage in anticompetitive conduct,” U.S. Info. Sys., 313 F. Supp. at 240. If asked by counsel at a trial, however, Dr. Marx could “explain whether conduct is indicative of collusion[.]” B & R Supermarket, 2024 WL 4252031, at *14 (quoting U.S. Info. Sys., 313 F. Supp. at 240).⁵⁷

4.1.2.3.2. Dispersion Analysis

Dr. Marx also uses the results of her dispersion analysis to opine that the defendants’ actions were consistent with collusion. See Marx Opening Report ¶¶ 168-88. In conducting this analysis, she compares data on each bank's LIBOR submission to determine whether LIBOR submissions were more concentrated and less dispersed during the conduct period than the non-conduct, or “benchmark” period. Id. Controlling for certain variables like the risk-free rate, market illiquidity, and other factors, Dr. Marx finds that there is a relatively tighter dispersion of LIBOR rates during the conduct period than the benchmark period, which she contends is suggestive of non-independent activity. Id. This result, according to Dr. Marx, is consistent with collusion, but not necessarily the result of a specific conspiracy. Id.

While defendants raise issues relating to Dr. Marx's controls for volatility, Marx – Defs. Br. at 21; Marx – Defs. Reply Br. at 9-10, these issues go towards the weight of her testimony rather than its admissibility.⁵⁸ See In re Digital Music Antitrust Litig., 321 F.R.D. 64, 75 (S.D.N.Y. 2017) (“If an expert's testimony lies within ‘the range where experts might reasonably differ,’ the jury, and not the trial court, should ‘decide among the conflicting views of different experts.’ ” (citation omitted)).

4.1.2.4. Concealment as a “Super Plus Factor”

Next, Dr. Marx opines that evidence of concealment is a “super plus factor” because, “[i]ntuitively, concealment of collusive conduct would not be expected without collusion.” Marx Opening Report ¶ 242.

Dr. Marx's opinions on super plus factors have been excluded once before in this District. In Anderson News, Judge Crotty rejected Dr. Marx's testimony regarding super plus factors, noting that “it appear[ed] to be a label conjured up for litigation rather than the ‘product of reliable principles and methods.’ ” Anderson News, L.L.C. v. Am. Media, Inc., No. 09 Civ. 2227 (PAC), 2015 WL 5003528, at *3 (S.D.N.Y. Aug. 20, 2015) (citation omitted), aff'd, 899 F.3d 87 (2d Cir. 2018). We similarly reject Dr. Marx's use of the term, although we will permit her to opine on “the existence of plus factors, and whether she believes that those factors are consistent with Defendants’ actions in this case.” Id.

We note, however, that Dr. Marx's opinion is circular: “concealment of collusive conduct” presumes the existence of “collusive conduct” and cannot be a basis for inferring the existence of “collusive conduct.” Marx Opening Report ¶ 242. While we allow Dr. Marx to “identif[y] factors, the existence of which would tend to indicate a conspiracy,” U.S. Info. Sys., 313 F. Supp. 2d at 240 (citation omitted), this circularity will reduce the weight to which this opinion might otherwise be entitled as the Court considers the pending summary judgment motion.

4.1.2.5. Broker Indications

In addition, Dr. Marx opines that “broker indications were predictive of future LIBOR fixes and thus could facilitate exchange of information in support of collusive conduct.” Marx Opening Report ¶ 257. As with Dr. Marx's other opinions, she bases this opinion on her review of the record, see, e.g., Marx Opening ¶ 120; Marx Reply Report ¶ 75, as well as a data-driven analysis, see e.g., Marx Reply Report ¶¶ 76–80, 106–12, 113, 230, 277, 279, 286, 291, 317–22.

Unsurprisingly, given our previous holdings, to the extent that Dr. Marx's opinion addresses “ ‘lay matters which a jury is capable of understanding and deciding without the expert's help,’ it is inadmissible.” Highland Capital, 379 F. Supp. 2d at 469 (citation omitted). That is because Dr. Marx's review of the record means that she “simply read[ ] and interpret[ed] evidence of collusion as any juror might,” Processed Egg, 81 F. Supp. 3d at 421, and any opinions based on such a review are inadmissible.⁵⁹

As for Dr. Marx's data-driven analysis, she “conducted statistical tests to investigate whether LIBOR submissions indicated patterns consistent with frequent and meaningful communication among Defendants.” Marx – Pls. Br. at 21 (citations omitted). She ultimately concluded that “broker indications were generally good predictors of the LIBOR fix and, hence, panel bank submissions.” Marx Opening Report ¶ 170.

Defendants contend that Dr. Marx's data-driven analysis should be excluded as unreliable speculation. See Marx – Defs Br. at 19-21; Marx – Defs Reply Br. at 8-9. The crux of their argument is that Dr. Marx's analysis, at most, show that panel bank-broker communications “contained information about the future LIBOR fix beyond publicly available information[,]” which is expected given the function of brokers and “consistent with independent action to suppress LIBOR submissions.” Marx – Defs Br. at 20-21 (citation, quotation marks, and emphasis omitted). Thus, according to defendants, Dr. Marx cannot make the leap from her data-driven analyses to her conclusion that broker indications could have been a “focal point.” Id. at 19-21 (citation and quotation marks omitted).

This argument directed to Dr. Marx's conclusions goes to weight, as opposed to admissibility. See In re Mirena IUD Prods. Liab. Litig., 169 F. Supp. 3d 396, 421 (S.D.N.Y. 2016) (“Although Plaintiffs and their experts may take issue with Dr. Goldberg's conclusions, these criticisms go to the weight, not the admissibility, of his testimony and are best addressed on cross-examination.”). While defendants are correct that “ ‘conscious parallelism,’ as lawyers call it, ‘tacit collusion’ as economists prefer to call it[ ]—which means coordinating their pricing without an actual agreement to do so”— “does not violate section 1 of the Sherman Act,” In re Text Messaging Antitrust Litig., 782 F.3d 867, 871, 879 (7th Cir. 2015), this is a determination to be made at the summary judgment or trial stages rather than within the confines of a Daubert analysis.⁶⁰

Thus, to the extent that Dr. Marx opines that “broker indications were predictive of future LIBOR fixes and ․ could facilitate exchange of information in support of collusive conduct,” Marx Opening Report ¶ 257, based on her data-driven analysis, that testimony is allowed. However, any testimony reciting her review of the documentary record is barred.

4.1.2.6. Discipline Disincentive

Dr. Marx opines that defendants were incentivized to conspire because there existed “an economic disincentive [to] unilaterally mak[e] LIBOR submissions that were artificially low and in violation of the LIBOR rules due to a credible threat of discipline from the [FXMMC], the BBA, or other [Panel] Banks.” Marx Rebuttal Report ¶ 54.

Dr. Marx's opinion is inadmissible. It lacks a “reliable foundation,” see Daubert, 509 U.S. at 597, for it is based upon an understanding of the LIBOR rules that conflicts with the Second Circuit's decision in Connolly. For example, Connolly rejected the “one-true-interest rate theory” and held “there was no one true rate that [a Panel Bank] was required to submit” in response to the LIBOR question. 24 F.4th at 837, 839. Dr. Marx, however, insists that there is only “one answer” to the LIBOR question that would be “rule-abiding.” ECF No. 4154-5 (Marx Aug. 26, 2024 Dep. Tr.) at 266:2-13.

The implication of Dr. Marx's misperception is evident in her response to the following hypothetical posed during her deposition: “If the panel bank submitter reasonably believed that the range of rates at which that bank could borrow, in reasonable market size, was between 3.05 and 3.15 percent that day, then an honest belief as to any number within that range would be an accurate submission, correct?” ECF No. 4154-5 (Marx Aug. 26, 2024 Dep. Tr.) at 266:15-20. In response, Dr. Marx said she “disagree[d],” emphasizing that “the correct answer to the LIBOR question” was that which “the defendant bank ․ truthfully” believed reflected their borrowing costs. Id. at 266:22-267:4 (emphasis added). This hypothetical accurately captures Connolly’s holding, and Dr. Marx's unwillingness to accept it shows that the conduct she identifies as being non-compliant with the LIBOR rules differs from what the Circuit proscribes.

Since Connolly is controlling precedent, an understanding of the LIBOR question that is contrary to Connolly’s holding and any conclusions based thereon cannot provide a valid framework for assessing the alleged conspiracy. For example, a proper understanding of the LIBOR rules is a necessary pre-condition for Dr. Marx's opinion that there existed “an economic disincentive [to] unilaterally mak[e] LIBOR submissions that were ․ in violation of the LIBOR rules due to a credible threat of discipline from the [FXMMC], the BBA, or other [Panel] Banks.” Marx Rebuttal Report ¶ 54 (emphasis added).

Plaintiffs try to explain away this fundamental problem in two ways. First, they suggest that Connolly was factually distinct from this case and its holding does not control our analysis. We reject this argument elsewhere. See infra pp. 180-185, 244-269.⁶¹ Second, plaintiffs suggest that it does not matter how Dr. Marx understands the LIBOR definition, because “the vast majority of [her] analyses do not depend on [that] understanding” and her “empirical analyses and study of collusive structures stand on their own.” Marx – Pls. Br. at 9. However, this argument misses the point. Dr. Marx's understanding of permissible conduct under the LIBOR rules is inextricably intertwined with her opinion concerning defendants’ economic disincentive to unilaterally suppress in violation of the LIBOR rules. Thus, we exclude an expert opinion that is founded on an understanding of LIBOR that differs from the binding legal standard outlined in Connolly.

4.1.2.7. Rebuttal Opinions

Similar to Dr. Cragg, Dr. Marx responds to FFFP defendants’ experts’ analysis regarding the possibility of collusion. See generally Marx Rebuttal Report; Marx Reply Report.

Having reviewed Dr. Marx's rebuttal opinions, the Court finds them admissible to the extent they do not run afoul of the parameters stated above. See, e.g., Feinsod, 2025 WL 1677409, at *16 (granting in part plaintiff's motion to preclude defendant's expert's opening report while denying plaintiff's motion to preclude defendant's expert's rebuttal report).⁶²

4.1.3. Conclusion

For these reasons, we grant in part, and deny in part, defendants’ motions to exclude the testimony of Drs. Cragg and Marx. The foregoing determinations are made in the context of defendants’ Daubert motions, separate from defendants’ summary judgment motion which will be addressed below. Obviously, if the granting of summary judgment is affirmed by the Circuit, the discussion of the parameters of admissible opinions at a possible trial would be moot.

To summarize our determinations, Dr. Cragg would be permitted at a trial to opine as to whether LIBOR's structure was susceptible to collusion and whether the Panel Banks were incentivized to collude. However, he could not offer an opinion concluding that these structural factors or incentives in fact resulted in a conspiracy. Nor could he opine on the Panel Banks’ specific motivations, thought processes, and understanding. Similarly, Dr. Cragg could offer opinions as to whether the banks’ observed conduct was indicative of collusion, but he would be precluded from stating whether or not collusion in fact occurred. Moreover, Dr. Cragg would be precluded from relying on his collusion screen, as it is irrelevant and unreliable. Finally, Dr. Cragg would be permitted to respond to defendants’ experts so long as he would not be restating a previously-excluded opinion.

Like Dr. Cragg, Dr. Marx could offer opinions at a trial as to whether LIBOR's structure made it susceptible to collusion, but she could not conclude whether these factors resulted in collusion. Additionally, Dr. Marx could opine on the Panel Banks’ incentives to collude, although her opinion on this issue is limited due to the circularity of her incentive analysis. Furthermore, Dr. Marx could testify about the observed conduct among the Panel Banks and whether such conduct is hypothetically consistent with conspiracy, but she could not share a conclusion as to whether this behavior was, in fact, collusive. Dr. Marx would also be permitted to rely on her dispersion analysis and testify regarding broker indications. She could testify as to whether concealment of a conspiracy is indicative of a conspiracy, though she could not refer to it as a “super plus factor.” Moreover, Dr. Marx could not testify as to disincentives for non-compliance and discipline because they rest on an unsound foundation: namely, her misunderstanding of the LIBOR rules. Finally, Dr. Marx would be permitted to respond to defendants’ experts to the extent that she was not repeating a previously-excluded opinion.

Put simply, if Drs. Cragg and Marx testified as to what a conspiracy could look like based on reliable methods and data, their testimony would be admissible. In contrast, if the experts provided testimony about whether the conspiracy actually occurred based on their review of the record and unreliable econometric methods, this testimony would be inadmissible. In any event, the limited evidentiary basis underlying the opinions of Drs. Cragg and Marx that remain admissible reduces the weight accorded to those opinions as the Court considers the pending summary judgment motion.

4.2. Circumstantial Evidence of Parallel Conduct

Having resolved the pending Daubert motions, we will now assess the sufficiency of plaintiffs’ circumstantial evidence. Again, to support a conspiracy claim, plaintiffs must present evidence of (1) “parallel conduct” and (2) various asserted “plus factors.” Anderson News II, 899 F.3d at 104.

There is no dispute between the parties that the first necessary element to establish an antitrust conspiracy based on circumstantial evidence is a showing of “parallel conduct” among the alleged co-conspirators. See Defs. Br. at 75; Pls. Br. at 73; see also Apex Oil Co., 822 F.2d at 253 (same). There is a dispute, however, as to the level of “parallel conduct” that must be shown for a conspiracy claim to survive summary judgment. See Defs. Br. at 75; Pls. Br. at 76. According to defendants, “parallel conduct” requires either “proof that defendants took identical actions within a time period suggestive of prearrangement,” or conduct that was “uniform” in nature. Defs. Br. at 75 (citing, in part, Anderson News II, 899 F.3d at 104, 105). Plaintiffs, on the other hand, contend that a less stringent standard applies, namely, that parallel conduct does not require proof of “identical actions,” and that an agreement to conspire can be inferred without a showing of “exactly simultaneous and identical” conduct. Pls. Br. at 76 (citing, in part, LaFlamme v. Societe Air France, 702 F. Supp. 2d 136, 151 (E.D.N.Y. 2010)).

As observed by other district courts, “because parallel conduct alone is not sufficient to establish a conspiracy[,] ․ caselaw explicating the minimum degree of parallelism required for a conspiracy is not easily distilled, as may be expected with such a fact intensive inquiry.” In re Rail Freight Fuel Surcharge Antitrust Litig. (No. I), No. Civ. 11-1049, 2025 WL 1784519, at *19 (D.D.C. June 27, 2025) (collecting cases; emphasis in original). However, based on our review of the relevant caselaw, we conclude that plaintiffs have the better argument. “Generally, parallelism is shown by the same or very similar behavior over the same general timeframe.” Id. (citing Anderson News II, 899 F.3d at 104); see also Baby Food, 166 F.3d at 132 (“[P]arallel pricing does not require ‘uniform prices,’ and permits prices within an agreed upon range.” (quoting United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 222 (1940)). In addition, the Second Circuit has previously provided “[e]xamples of parallel conduct allegations that might be sufficient,” which do not include references to identical or uniform actions. See Mayor & City Council, 709 F.3d at 137. These examples include “parallel behavior that would probably not result from chance, coincidence, independent responses to common stimuli, or mere interdependence unaided by an advance understanding among the parties,” and “complex and historically unprecedented changes in pricing structure made at the very same time by multiple competitors, and made for no other discernible reason.” Id. (citation and quotation marks omitted). Thus, for purposes of this motion, we apply a less demanding standard than defendants urge.⁶³

We now turn to an assessment of the parties’ proffered evidence. In their effort to show parallel conduct, plaintiffs largely rely on charts and expert-derived econometric models, which they contend supports the proposition that the limited observable dispersion in LIBOR rates can only be explained by collusion. See, e.g., Pls. 56.1 ¶¶ 301-05; Pls. Br. at 76. For example, plaintiffs present a line chart depicting the Panel Banks’ 3M LIBOR submissions from August 2007 to May 2010, which they claim offers “plainly visible” evidence of parallel action. See Pls. Br. at 76.⁶⁴ Additionally, plaintiffs assert that, when controlling for market volatility and differences in credit default risk, their experts’ models show that defendants’ LIBOR submissions were less dispersed than they would have been absent collusion. Id. at 78–79. Specifically, plaintiffs contend that an “empirical analysis shows that lawful, independent responses to market events cannot fully explain” the consistent concentration of LIBOR submissions. Id. at 80.⁶⁵

Plaintiffs’ concessions undermine their assertion of parallel conduct. First, while “parallel pricing does not require ‘uniform prices,’ ” it does require defendants’ LIBOR submissions to be, at least, “within an agreed[-]upon range.” Baby Food, 166 F.3d at 132 (quoting Socony-Vacuum Oil Co., 310 U.S. at 222). However, plaintiffs concede that there is no agreed-upon range, as the alleged conspiracy “did not involve an agreement on a specific level or magnitude of suppression.” OTC Plaintiffs’ Interrogatory Responses, Resp. No. 6 at 123 (“[T]he Conspiracy did not involve an agreement on a specific level or magnitude of suppression.”); DAPs’ Interrogatory Responses, Resp. No. 6 at 120 (“DAPs do not contend that Defendants entered into an agreement to suppress USD LIBOR by specific percentages, basis points, or other specific quantitative measures․ [T]he Conspiracy did not involve an agreement on a specific level or magnitude of suppression.”).

Next, plaintiffs endeavor to rely on the theory that “parallelism” can be “shown by the same or very similar behavior over the same general timeframe.” Rail Freight Fuel Surcharge, 2025 WL 1784519, at *19 (citing Anderson News II, 899 F.3d at 104). Plaintiffs contend that this requirement is satisfied because “Defendants’ [LIBOR] submissions were less dispersed” when “properly accounting for increased market volatility and dispersion in Defendants’ default risk,” “[a]s [their] experts show.” Pl. Br. at 78 (citations omitted). Dispersion in this context refers to whether “submissions were similar.” Id.

Plaintiffs’ dispersion analysis does not support a finding of parallel conduct for two reasons. First, even if we assume that the dispersion calculations of plaintiffs’ experts are correct, this does not prove that defendants engaged in “the same or very similar behavior.” Rail Freight Fuel Surcharge, 2025 WL 1784519, at *19 (citation omitted). As defendants point out in their opening memorandum, for submissions of all sixteen banks to be tightly concentrated would have required that some of the banks “suppress[ed] their submissions to a lesser degree than others (or even inflate[d] them)” because all sixteen banks had different levels of creditworthiness and did not have the same borrowing costs. Defs. Br. at 82. This argument exposes a fundamental flaw in plaintiffs’ theory of conspiracy, which we addressed earlier: namely, that a LIBOR submission in the “pack” is not necessarily “suppress[ed],” and vice versa. See supra pp. 33-43; see also LIBOR IV, 2015 WL 6243526, at *47, slip op. at *117-18 (“Indeed, it is puzzling how a conspiracy theory could be consistent with the attempts of some banks to be near the bottom of the pack, because not every bank could be below average.”).⁶⁶

Second, plaintiffs’ experts acknowledged that dispersion increased in absolute terms during the relevant conduct period. See Marx Rebuttal Report ¶ 97 (“I agree with Professor Asker that, in the absence of controls for changes in the economic environment, there was greater dispersion in Defendant Banks’ LIBOR submissions during the Conduct Period.”); Cragg Rebuttal Report ¶ 59 (similar). Defendants contend that dispersion should be evaluated on an absolute basis, and thus, plaintiffs cannot prove parallel conduct. Plaintiffs, however, assert that dispersion estimates must “account for the effects of the Financial Crisis” because, otherwise, there would be “an apples-to-oranges comparison of a period of significant volatility, the Financial Crisis, to times of less volatility.” Pls. Br. at 78 (citations omitted). Plaintiffs also claim that this issue requires a “battle of the experts” which should be fought before a jury. Pls. Br. at 78 (citations omitted).⁶⁷

The only issue then is whether evaluating parallel conduct pursuant to plaintiffs’ asserted conspiracy depends on controlling for volatility. The answer is it does not. Fundamental to plaintiffs’ theory of conspiracy would be a finding of less dispersion in the face of great volatility. As plaintiffs argue:

• “Because of the volatility in the market and day-to-day swings of the Financial Crisis, no Defendant could predict where the other Defendants would set their LIBORs each day in order to align their submissions. The only way to suppress and stay within the pack was to coordinate their submissions.” Pls. Br. at 1-2.

• “[S]uch variety in volatile market circumstances should have resulted in far more dispersion than observed, especially because the rules prevented Defendants from sharing submissions in advance.” Id. at 79.

• “[T]he raw LIBOR submission patterns are themselves stunningly uniform[.]” Id. at 109.

Eliminating this volatility as a factor would permit plaintiffs to prove a theory contrary to historical fact, given that dispersion increased, as would be expected, in volatile circumstances. Thus, given the undisputed evidence that dispersion increased in absolute terms during the relevant period, see Marx Rebuttal Report ¶ 97; Cragg Rebuttal Report ¶ 59,⁶⁸ plaintiffs’ proffered evidence does not show that defendants engaged in “the same or very similar behavior over the same general timeframe,” Rail Freight Fuel Surcharge, 2025 WL 1784519, at *19 (citation omitted).⁶⁹

In sum, upon careful review, plaintiffs’ proffered evidence fails to support a reasonable inference of parallel conduct. See, e.g., Anderson News II, 899 F.3d at 105 (finding defendants “reacted in different ways” to similar economic stimuli); Rail Freight Fuel Surcharge, 2025 WL 1784519, at *20 (finding “defendants did not take parallel action during the alleged conspiracy period” because their “formulas prior to the start of the alleged conspiracy period were more uniform than their formulas after defendants made changes in 2003[,]” and defendants’ actions “lacked uniformity throughout the alleged conspiracy period”).

4.3. Circumstantial Evidence of Plus Factors

As set out earlier, a plaintiff endeavoring to establish a conspiracy by circumstantial evidence must meet a conjunctive standard, i.e., a plaintiff must show both (1) a reasonable inference of “parallel conduct” and (2) “plus factors” that lead to an inference of collusion. Anderson News II, 899 F.3d at 104; see also supra pp. 49-50 (same). Because we have found plaintiffs’ evidence of parallel conduct lacking, we need not engage in an analysis of potential “plus factors.” See supra pp. 49-50; see also Apex Oil Co., 822 F.2d at 253 (same); Anderson News II, 899 F.3d at 112 (“Without parallel acts to be reviewed in conjunction with the circumstantial evidence, ․ evidence supporting the presence of certain plus factors ․ can provide little support for a finding of unlawful conspiracy.” (citations and quotation marks omitted; emphasis added)).

Nevertheless, for the sake of completeness, we will evaluate the parties’ evidence regarding plus factors below. We note that, even if the plus factors are established, the presence of “such plus factors may not necessarily lead to an inference of conspiracy,” as they could “lead to an equally plausible inference of mere interdependent behavior, i.e., actions taken by market actors who are aware of and anticipate similar actions taken by competitors, but which fall short of a tacit agreement.” Apex Oil Co., 822 F.2d at 254.

In accordance with the Second Circuit's guidance in Gelboim, we evaluate the following plus factors: “(1) a common motive to conspire; (2) evidence that shows that the parallel acts were against the apparent individual economic self-interest of the alleged conspirators; and (3) evidence of a high level of interfirm communications.” Gelboim, 823 F.3d at 781 (citations and quotation marks omitted). As the Second Circuit noted, “[t]hese plus factors are neither exhaustive nor exclusive, but rather illustrative of the type of circumstances which, when combined with parallel behavior, might permit a jury to infer the existence of an agreement.” Id. (citations and quotation marks omitted).

The parties present additional plus factors specific to the LIBOR context, which we also evaluate. Specifically, plaintiffs submit evidence purportedly establishing that: (4) the LIBOR-setting process was conducive to collusion, see Pls. Br. at 104-05; (5) defendants engaged in “efforts to conceal the existence of a conspiracy,” see id. at 105-06; (6) LIBOR was disconnected from market fundamentals according to plaintiffs’ experts’ econometric models, see id. at 107-10; and (7) government investigations and regulatory settlements support plaintiffs’ theories, see id. at 110-11. Defendants also present evidence, which we review as plus factor (8), tending to refute the existence of a conspiracy.⁷⁰ See, e.g., Defs. Br. at 19-29.

4.3.1. Plus Factor #1: Common Motive to Conspire

The first plus factor we assess is whether defendants shared “a common motive to conspire.” Gelboim, 823 F.3d at 781 (citations and quotation marks omitted).⁷¹

“To begin, we note that the defendants had an unlikely motive to conspire, given our conclusion that the alleged conspiracy is economically implausible.” Anderson News II, 899 F.3d at 112; see supra pp. 33-43 (finding purported agreement to be economically senseless). Still, we evaluate the viability of each asserted motive, as a “weak motive to conspire does not ‘save defendants who have clearly, though foolishly conspired.’ ” Anderson News II, 899 F.3d at 112 (citation omitted). A “[m]otive to conspire tends to be negated [1] when a defendant shows that the alleged agreement would harm the alleged conspirators; or [2] when the defendant shows a plausible and justifiable reason for its conduct that is consistent with proper business practice.” Id. (citation and quotation marks omitted).

As stated previously, supra pp. 33-43, plaintiffs assert that defendants shared the following common motives to conspire: (1) “to project financial strength,” Pls. Br. at 82; (2) to avoid the “threat of a catastrophic domino or contagion effect across the Panel Banks” if any one bank appeared as an outlier relative to a “pack” of its peers, Pls. Br. at 1, 14 (citations omitted); and (3) “to conceal the fact that they were disregarding the LIBOR rules,” DAPs’ Interrogatory Responses, Resp. No. 9 at 150.⁷² We review each alleged motive below.

The Second Circuit held that the first asserted motive -- the desire to project financial strength -- was plausible at the motion to dismiss stage. See Gelboim, 823 F.3d at 781-82 (finding plaintiffs’ “allegations evince a common motive to conspire—increased profits and the projection of financial soundness”); see also LIBOR VI, 2016 WL 7378980, at *3, slip op. at *6 (“[T]he object of the conspiracy that the Circuit recognized and which meets the plausibility test is the projection of financial soundness.”), id. at *5, slip op. at *11 (“The only conclusion to be drawn is that the Circuit meant ‘increased profits and the projection of financial soundness’ to describe collectively a single, reputation-based motive to conspire, where increased profits followed from a positive reputation.”).

At the summary judgment stage, however, plaintiffs are wholly unable to explain why a defendant bank would be more incentivized to collude when, as their experts concede, that bank also had a plausible and justifiable reason for its conduct, namely a unilateral incentive to signal financial strength. See Defs. 56.1 ¶ 348 (“Dr. Cragg testified that Panel Banks had a unilateral incentive to signal financial strength.”) (citing ECF No. 4330-7 (Cragg July 25, 2024 Dep. Tr.) at 106:7-11); ECF No. 4330-17 (Cragg Sept. 11, 2024 Dep. Tr.) at 74:17-20 (“Q: [W]e agree that banks have a unilateral incentive to – to portray their financial health, right? A: Yes.”); ECF No. 4304-26 (Marx June 25, 2024 Dep. Tr.) at 316:12-17 (“Here the -- the benefit of being able to use artificially low LIBOR submissions to signal financial strength in a time of high financial volatility and stress is something that would be benefitting each of the defendant banks, both individually and collectively.”). Plaintiffs’ experts’ concessions are supported by statements from Panel Banks’ employees. Rabobank's submitter, for example, acknowledged the “potential negative stigma” associated with LIBOR submissions because “[n]o bank wants to be seen as in need for cash,” ECF No. 4327-32 (Jongmans Dep. Tr.) at 45:12-15, and another Rabobank employee wrote internally that “we need to protect our name” because “I wouldn't want any customer thinking Rabo can't fund itself at or below Libor,” ECF No. 4332-25 (PX1823) at 2. Similarly, “as part of its resolution with the DOJ, Barclays admitted ․ that ‘on some occasions’ between August 2007 and January 2009 Barclays unilaterally made USD LIBOR submissions that were ‘lower than Barclays otherwise would have submitted’ because Barclays ‘sought to avoid inaccurate negative media attention about Barclays’ financial health as a result of its high LIBOR submission relative to other banks.” Defs. 56.1 ¶ 284.

At summary judgment, “antitrust law limits the range of permissible inferences from ambiguous evidence in a § 1 case.” Matsushita, 475 U.S. at 588. Applying this teaching here, this alleged motive “is equally suggestive of competition as collusion [and] ․ is insufficient [for plaintiffs’ claim] to survive summary judgment,” even if it was sufficient at the motion to dismiss stage. Payment Card Interchange Fee & Merch. Disc., 714 F. Supp. 3d at 81 (citations omitted).

We next address the second alleged motive, the “threat of a catastrophic domino or contagion effect across the Panel Banks” if any one bank appeared as an outlier relative to its “pack” of peers. See Pls. Br. at 1, 14. Colluding rather than competing, plaintiffs contend, was critical in enabling defendants to avoid being perceived as “outlier[s],” which “could lead to financial harm, reputational damage, and even a ‘panic’ that could imperil other banks.” Pls. Br. at 83. In the context of the financial crisis, plaintiffs assert that defendants’ economic motivations were ultimately rooted in staving off a “contagion” that sought to endanger the banking sector and beyond. Id.

To support this alleged common motive to conspire, plaintiffs largely rely on selected portions of testimony from its own experts, Drs. Cragg and Marx, and certain panel bank employees. See, e.g., Pls. Br. at 14, 83-86; Pls. 56.1 ¶ 30. For example, Dr. Cragg stated that there may have been “a collective incentive” for defendants to collude in order to avoid the potential reputational consequences of being perceived as an outlier and avert fears of potentially sparking a cascading reaction across the financial sector. ECF No. 4315-14 (Cragg Sept. 11, 2024 Dep. Tr.) at 75:3–20. While Dr. Cragg's testimony concerning whether the Panel Banks could have had this collective incentive in the abstract is admissible, see supra pp. 55-57, 83-85, the existence of such a motive in reality is purely speculative. Dr. Cragg did not investigate whether each submitter had this incentive on a day-to-day basis, could not recall seeing any evidence that a LIBOR submitter was “worried about the risk of detection from submitting a low outlier[,]” and could not recall whether he observed any evidence that the Panel Banks were concerned about “contagion[.]” ECF No. 4175-31 (Cragg July 25, 2024 Dep. Tr.) at 41:5-11, 42:10-23, 50:1-5, 104:9-105:6.

The limited other evidence that plaintiffs submit fares no better. Plaintiffs cite the deposition testimony of a JPMorgan executive who recalled telling a representative from FXMMC that increased LIBOR rates could trigger “a whole new round of LIBOR concern/fear/pressure.” ECF No. 4527-23 (Paterson Dep. Tr.) at 273:9-276:4. Plaintiffs also cite documentary evidence from Panel Bank employees: one UBS employee stated the following on August 10, 2007, “[we] dont want to be on the highs of libors” because “we just don[’]t want to give the market a wrong impression[,]” ECF No. 4289-10 (PX336) at 2. Similarly, following a call with certain Panel Banks, the BBA Director wrote to the HSBC submitter, among others, that “All also observed that right now, posting a high libor rate is interpreted by the market as meaning that you have a funding problem[,] so no one will do so as it's a self-fulfilling prophecy.” ECF No. 4281-13 (PX61) at 1-2.⁷³

Simply stated, this evidence does not support an inference that defendants possessed a motive to conspire based on a fear of potentially sparking a cascading reaction across the financial sector. Instead, these communications evince a self-interested motive to avoid making a high LIBOR submission. However, even assuming that it could support an inference of collusion, it is quintessential “[a]mbiguous evidence—evidence that is equally consistent with independent conduct as with illegal conspiracy[,]” which “is not enough” to support a claim of conspiracy. SourceOne Dental, 310 F. Supp. 3d at 357 (citation omitted). Because plaintiffs offer no unambiguous evidence to support this common motive to conspire and there is no such evidence in the record, there exists “a plausible and justifiable reason for [defendants’] conduct that is consistent with proper business practice.” Anderson News II, 899 F.3d at 112 (citation and quotation marks omitted).⁷⁴ Accordingly, this theory of motivation is rejected as unsupported by the extensive record developed by the parties.

Turning to the third asserted motive, plaintiffs argue that defendants had a motive “to remain on the LIBOR panel” and avoid penalties for disregarding LIBOR rules because “[i]n the absence of an agreement, a bank that submitted artificially low LIBORs would have risked discipline or removal from the USD LIBOR panel.” DAPs’ Interrogatory Responses, Resp. No. 9 at 144; see also Pls. Br. at 83 (similar).

This argument fails for two reasons. First, this motive is circular. Plaintiffs presume that defendants were violating LIBOR rules by making artificially low submissions, then deduce that defendants had a motive to do so, based on the presumption that they did so. A “litigant may not proceed by first assuming a conspiracy and then explaining the evidence accordingly.” Blomkest Fertilizer, Inc. v. Potash Corp. of Saskatchewan, 203 F.3d 1028, 1033 (8th Cir. 2000). In addition, plaintiffs’ argument is internally inconsistent. In one breath, plaintiffs contend that defendants “were also the police” and responsible for enforcing “the LIBOR rules.” Pls. Br. at 82.⁷⁵ In the next, they argue that the FXMMC could “ferret out suppressed submissions” because it was “comprised of experienced market practitioners who were well positioned to exercise an oversight role.” Id. at 92 n.144. If defendants were “the police,” then they had no motive to conceal any misconduct because there was little chance of blowback. Cf. Valspar Corp. v. E.I. Du Pont De Nemours & Co., 873 F.3d 185, 198 (3d Cir. 2017) (finding this type of argument “suffers from the loaded question fallacy” as plaintiffs set out to answer “How did [defendants] further the conspiracy” as opposed to “Does the [evidence] show that the conspiracy existed?”).

Second, plaintiffs have not provided any unambiguous evidence to support a conclusion that the Panel Banks possessed this common motive to conspire. To support such a conclusion, FFFP plaintiffs rely on an inadmissible opinion from Dr. Marx which is based on an understanding of the LIBOR rules that directly conflicts with the Second Circuit's decision in Connolly. See supra pp. 79-82.⁷⁶ Moreover, Dr. Marx's discipline theory is entirely speculative, as she is unable to point to a single instance of a panel bank being disciplined or removed from the panel, and is unable to describe the protocol and/or mechanism by which a panel bank would be removed. See ECF No. 4304-29 (Marx August 26, 2024 Dep. Tr.) at 176:10-22, 178:7-11. Additionally, Dr. Marx undertook no effort to “quantify the expected benefits or the expected costs to a Panel Bank of entering into a collusive agreement.” Defs. 56.1 ¶ 352; see also id. ¶ 353 (similar); ECF No. 4175-24 (Marx June 25, 2024 Dep. Tr.) at 95:9-14 (“Q: Is there a unit of measurement by which you can quantify the expected benefit associated with the course of action to engage in the alleged conduct? A: There might be. I didn't attempt to do that.”). Mere abstract assertions that the “reputational benefits of being on the LIBOR panel and the reputational harm of being removed from the LIBOR panel created an economic incentive for the Defendant Banks to coordinate with the BBA” are simply not sufficient to support a finding that the banks actually possessed this incentive. Marx Opening Report ¶ 105.

Accordingly, absent any substantiating facts or analysis, we cannot give credence to DAPs’ asserted discipline motive. See AD/SAT, 181 F.3d at 233 (“[I]f the defendants have ‘no rational economic motive to conspire, and if their conduct is consistent with other, equally plausible explanations, the conduct does not give rise to an inference of conspiracy.’ ” (quoting Matsushita, 475 U.S. at 596-97)).

* * *

Overall, all three of plaintiffs’ asserted motives to conspire are equally susceptible to “a plausible and justifiable reason for [defendants’] conduct that is consistent with proper business practice.” Anderson News II, 899 F.3d at 112 (citation omitted).

To put a finer point on this, we are skeptical as to why the Panel Banks would collude when they could achieve the same benefits – i.e., reputational strength, increased profits, and signaling financial stability to the market -- via unilateral action. See supra pp. 33-43. Although this Court and the Circuit in Gelboim previously determined that plaintiffs’ common motive to project financial strength “clear[ed] the bar of plausibility” at the motion to dismiss stage, Gelboim, 823 F.3d at 781-82; see also LIBOR VI, 2016 WL 7378980, at *3, slip op. at *6 (similar), plaintiffs at the summary judgment stage must present admissible evidence that tends to exclude “the possibility of independent action,” Anderson News LLC, 680 F.3d at 184. Here, despite having years to bolster the record, plaintiffs have failed to offer evidence sufficient to establish a triable issue as to whether defendants possessed a “rational economic motive to conspire,” given that defendants’ “conduct is consistent with other, equally plausible explanations,” that “do[ ] not give rise to an inference of conspiracy.” Matsushita, 475 U.S. at 596-97.

4.3.2. Plus Factor #2: Actions Taken Against Self-Interest

The second plus factor considers whether there is circumstantial evidence “show[ing] that the parallel acts were against the apparent individual economic self-interest of the alleged conspirators.” Gelboim, 823 F.3d at 781 (citations and quotation marks omitted).

“An action that is contrary to a firm's apparent economic self-interest is one that is implausible if undertaken alone, without the guarantee of cooperation by competitors.” PharmacyChecker.com, LLC v. Nat'l Ass'n of Boards of Pharmacy, 530 F. Supp. 3d 301, 335 (S.D.N.Y. 2021) (citation and quotation marks omitted); see also 1 ABA Antitrust Law Section, Antitrust Law Developments § 1B-1-b-(2) (9th ed. 2022) (“Among the most important plus factors are those that tend to show that the conduct would be in the parties’ self-interests if they all agreed to act in the same way, but would be contrary to their self-interests if they acted alone․”).

As support for this plus factor, plaintiffs first assert that it would be against a bank's self-interest to unilaterally suppress its LIBOR submission because the bank would be exposing itself to a “risk of media and regulatory scrutiny and reputational harm.” Pls. Br. at 91. In support of this assertion, plaintiffs cite a few selected intrabank chats and testimony by Dr. Cragg stating his opinion regarding the riskiness of unilaterally depressing one's LIBOR submission. See, e.g., ECF No. 4289-2 (PX327); ECF No. 4327-18 (Cragg Sept. 11, 2024 Dep. Tr.) at 75:3-20. Second, plaintiffs contend that a bank acting unilaterally to suppress LIBOR might have faced enforcement or punishment by the FXMMC, thereby suggesting that group cooperation was the only way for a bank to justify taking action against its own self-interest. Pls. Br. at 92. Finally, plaintiffs argue that a non-complying bank would have been acting against its own self-interest if it tolerated its competitors’ artificially depressed LIBOR submissions. Id.

As we have stated throughout, defendants’ purportedly packlike conduct is not necessarily contrary to each bank's self-interest. In fact, it may be just the opposite. As OTC plaintiffs’ expert Dr. Cragg conceded, it was entirely consistent with each bank's self-interest to act unilaterally and submit a LIBOR rate that was within the “pack,” so as to avoid the reputational and potential regulatory ramifications associated with outlier submissions. Defs. 56.1 ¶¶ 348, 351; see also ECF No. 4327-18, (Cragg Sept. 11, 2024 Dep. Tr.) at 76:5-16 (“A. Ultimately, the -- the banks, individually and collectively, have an incentive to maintain credibility.”). And, apart from their experts’ concessions, the other evidence plaintiffs offer to support this contention falls flat because plaintiffs ignore the possibility that avoiding regulatory and media scrutiny would be a wholly rational motivation irrespective of collusion. For example, plaintiffs cite a chat between two UBS traders in which they discuss the desirability of remaining in the “middle of the pack,” to avoid “get[ting] written about[.]” ECF No. 4289-2 (PX327) at 3-4. While plaintiffs assert that this discussion indicates collusion, it can just as reasonably be interpreted as an example of a non-collusive motivation to avoid attention.

The same analysis applies to other limited evidence cited by plaintiffs in support of this plus factor. Pls. Br. at 33-34, 92; Pls. 56.1 § V. We have carefully considered all evidence and focus on the arguably strongest citations. For example, plaintiffs cite the following communication sent by Natwest's submitter's supervisor: “Public contribution incentivises so-called ‘herd mentality’. Panel banks do not want to post rates that are far from the average for fear of potentially creating market perceptions that they are unable to fund themselves at levels similar to other participants.” Pls. Br. at 34 (quoting ECF No. 4299-45 (PX608) at 1). Notably, this chat actually undermines plaintiffs’ collective motive to conspire theory, as it states that defendants’ fear revolves around the “perception[ ] that they are unable to fund themselves at levels similar to other participants.” Id. (emphasis added). This document is, at best, ambiguous evidence, as it is equally consistent with Natwest having an independent interest to avoid “post[ing] rates that are far from the average.” ECF No. 4299-45 (PX608) at 1.

The final argument plaintiffs offer in support of this plus factor is that a non-complying bank would have been acting against its self-interest if it had tolerated its competitors’ artificially depressed LIBOR submissions. Pls. Br. at 92. However, defendants submitted unrebutted evidence that the Panel Banks did report their suspicions regarding their competitors’ low LIBOR submissions to the BBA, other regulators, central banks, see Defs. 56.1 ¶¶ 367, 372-73, 375, 381-95, 406-07, 414-17, journalists, see id. ¶¶ 371, 379, and even to plaintiff Freddie Mac, see id. ¶ 378.

It is also worth noting that plaintiffs’ experts failed to conduct a cost-benefit analysis of whether the costs of joining the conspiracy would have outweighed the benefits of joining. Defs. 56.1 ¶¶ 352-53; see also ECF No. 4175-25 (Asker FFFP Rebuttal) ¶ 15; ECF No. 4175-31 (Cragg July 25, 2024 Dep. Tr.) at 144:1-5 (“Q. Did you do anything to weigh the potential costs versus the potential benefits of entering into the alleged agreement? A. I haven't conducted a cost-benefit analysis of this.”), ECF No. 4175-46 (Asker Opening Report) ¶¶ 27-28, ECF No. 4175-51 (Asker OTC Rebuttal Report) ¶ 13. Simply put, “[a]bsent evidence regarding the long-term costs or benefits of acceding to the” conspiracy, “the inferential gap between the evidence presented to the conclusion that” the defendants’ actions were “against [their] economic interests is simply too great.” Anderson News II, 899 F.3d at 113. Nor have plaintiffs here “offered any evidence to bridge that gap.” Id.

* * *

Ultimately, taken as a whole, plaintiffs have failed to offer sufficient evidence of defendants acting against their self-interest. The evidence highlighted by plaintiffs ⁷⁷ falls well short of the standard required at the summary judgment stage. See Anderson News II, 899 F.3d at 98 (“[I]f the evidence is in equipoise, then summary judgment must be granted against the plaintiff.”); Kleen Prods. LLC, 910 F.3d at 934 (“The [plaintiffs] need evidence that would allow a trier of fact to nudge the ball over the 50-yard line and rationally ․ say that the existence of an agreement is more likely than not.”).

4.3.3. Plus Factor #3: High Level of Interfirm Communications

We next address the third plus factor, which considers “evidence of a high level of interfirm communications.” Gelboim, 823 F.3d at 781 (citations and quotation marks omitted).⁷⁸

When analyzing this plus factor, we must “[b]ear[ ] in mind that evidence of the mere exchange of information by competitors cannot establish a conspiracy.” Apex Oil Co., 822 F.2d at 257-58.⁷⁹ Similarly, a “mere showing of close relations or frequent meetings between the alleged conspirators ․ will not sustain a plaintiff's burden absent evidence which would permit the inference that these close ties led to an illegal agreement.” H.L. Moore Drug Exch. v. Eli Lilly & Co., 662 F.2d 935, 941 (2d Cir. 1981) (citation omitted). As such, our “[a]nalysis of inter-firm communications is not mechanical, [as] the probative value of such evidence depends on the participants, the information exchanged, and the context—specifically, the connection between the content and the price-fixing conspiracy alleged.” In re Commodity Exch., Inc., Gold Futures & Options Trading Litig., 328 F. Supp. 3d 217, 228 (S.D.N.Y. 2018).

“Information sharing among competitors without a legitimate purpose is ․ good evidence of a conspiracy.” Id. (citation omitted); see also Todd v. Exxon Corp., 275 F.3d 191, 198 (2d Cir. 2001) (“Information exchange is an example of a facilitating practice that can help support an inference of a price-fixing agreement.”).⁸⁰ Particularly probative communications are those that “ ‘represent[ ] a departure from the ordinary pattern’ of communications between defendants.” Anderson News, L.L.C. v. Am. Media, Inc., 123 F. Supp. 3d 478, 504 (S.D.N.Y. 2015) (citation omitted), aff'd, 899 F.3d 87 (2d Cir. 2018). Other evidence may involve “proof that the defendants got together and exchanged assurances of common action or otherwise adopted a common plan even though no meetings, conversations, or exchanged documents are shown.” In re Flat Glass Antitrust Litig., 385 F.3d 350, 361 (3d Cir. 2004) (citations and quotation marks omitted). Ultimately, the communications must reveal that defendants “had a conscious commitment to a common scheme designed to achieve an unlawful objective.” Monsanto, 465 U.S. at 764 (citation and quotation marks omitted).

In support of this factor, plaintiffs offer two categories of communications: (1) communications between interbank broker/dealers and employees at the Panel Banks; and (2) communications directly between employees at the different Panel Banks, including those within the BBA, FXMMC, and the LIBOR Steering Group. We review each category below.

4.3.3.1. Broker Communications

To support this plus factor, plaintiffs principally rely on communications between members of the Panel Banks and third-party intermediary brokers. See Pls. Br. at 26-37, 95-99; Pls. 56.1 § VI.

“Courts may consider co-conspirators’ use of third-party intermediaries, such as market analysts, as ‘conduits.’ ” In re Domestic Airline Travel Antitrust Litig., 691 F. Supp. 3d 175, 208 (D.D.C. 2023) (citation omitted).⁸¹ However, a “trader from one defendant exchanging market chatter with one non-defendant on a regular basis does not constitute an antitrust violation.” City of Pontiac Police & Fire Ret. Sys. v. BNP Paribas Sec. Corp., 92 F.4th 381, 399 n.9 (2d Cir. 2024) (citation omitted). In fact, communications with a third party “tend to suggest the absence of such [interfirm] communications” because “if, for example, [alleged co-conspirator 1] and [alleged co-conspirator 2] were talking, the former would not have had to rely on third parties to confirm the latter's strategy.” Mayor & City Council, 709 F.3d at 139 (emphasis in original).

Plaintiffs assert that brokers served a central role in the alleged conspiracy, for they would speak with LIBOR submitters across the Panel Banks, receive information about the Panel Banks’ planned submissions, and then share that information with the other bankers with whom they would speak. See Pls. 56.1 § VI.⁸² Brokers then “refin[ed] the indications until [d]efendants reached a ‘consensus’ around which they formed their submissions.” Pls. Br. at 96. Ultimately, plaintiffs argue that it is this dynamic between brokers and bankers that allowed the defendants to effectuate the alleged conspiracy without extensive direct collusion.

At the outset, we note that two concessions by plaintiffs significantly limit their broker-banker theory. First, plaintiffs acknowledge that there were legitimate, pro-competitive reasons for defendants to speak with brokers, as “interdealer brokers served as intermediaries for clients, including Panel Banks, to identify borrowers or lenders and to provide information regarding ․ bid and ask quotes in the market.” Pls. 56.1 ¶ 159.⁸³ Moreover, plaintiffs do not supply any evidence that broker-banker communications in the relevant period “ ‘represent[ ] a departure from the ordinary pattern’ of communications between defendants.” Anderson News, L.L.C. v. Am. Media, Inc., 123 F. Supp. 3d 478, 504 (S.D.N.Y. 2015) (citation omitted), aff'd, 899 F.3d 87 (2d Cir. 2018).⁸⁴ Thus, standing alone, the fact that Panel Banks communicated with brokers tells us very little about whether an illegal agreement was entered into by the defendants. Such an inference can only be made based on the content of the broker-banker communications. See In re Commodity Exch., 328 F. Supp. 3d at 228 (stating “the probative value of such evidence depends on the participants, the information exchanged, and the context—specifically, the connection between the content and the price-fixing conspiracy alleged”).

Second, despite their allegations that brokers were critical to the asserted conspiracy, plaintiffs concede that they “did not depose” a single broker. Defs. 56.1 ¶ 11 (emphasis added). Plaintiffs’ failure to depose any broker is highly revealing, if not devastating, given that they admittedly “applied for and obtained letters of request from this Court pursuant to the Hague Convention ․ to take depositions abroad of eight individual interdealer brokers formerly employed by BGC, Tullett Prebon, Tradition, and ICAP, which entities, Plaintiffs allege, participated in or facilitated the conspiracy to suppress LIBOR.” Id. (citations omitted); see also ECF No. 3892 (granting plaintiffs leave to depose these individuals).⁸⁵ Having failed to depose any broker, plaintiffs instead rely on their self-serving interpretation of broker-banker communications, deposition testimony from Panel Banks’ employees, and their experts’ opinions concerning the broker-banker communications. See Pls. Br. at 26-37, 95-99; Pls. 56.1 § VI.

For plaintiffs’ broker-banker theory to be correct, three assertions must be proven true: (1) the Panel Banks provided their planned submissions to brokers for the purpose of aligning all banks’ submissions in a pack; (2) brokers then disseminated this information (as opposed to information about market activity) to other Panel Banks consistently throughout the relevant period; and (3) Panel Banks relied on information concerning others banks’ planned submissions (as opposed to information about market activity) when making their own submissions, as part of a common understanding previously reached with other Panel Banks. See Pls. 56.1 § VI.

Accordingly, to evaluate plaintiffs’ evidence in support of each assertion, we have divided the parties’ voluminous evidence on this issue into four categories: (1) the Panel Banks’ communications to brokers; (2) brokers’ communications to the Panel Banks; (3) the Panel Banks’ bases for their LIBOR submissions; and (4) evidence which undermines plaintiffs’ theory. These categories are merely illustrative, as we base our conclusions on an “examin[ation] [of] the existence of a conspiracy ‘as a whole[,]’ taking into consideration the totality of the evidence, as opposed to ‘dismembering it and viewing its separate parts.’ ” Ross, 35 F. Supp. 3d at 438 (citation omitted). We review each category below.

4.3.3.1.1. Category #1: Panel Banks’ Communications to Brokers About Planned Submissions

As noted above, the first proposition that must be proven true to establish plaintiffs’ broker-banker theory is that the Panel Banks provided their planned submissions to brokers for the purpose of aligning all banks’ submissions in a pack. See Pls. 56.1 § VI.

In Appendix A to plaintiffs’ Joint Counterstatement, plaintiffs supply perhaps their strongest evidence of the broker-banker theory: 367 instances in which Panel Banks purportedly spoke to brokers about what LIBOR rates the Banks planned to submit. See ECF No. 4337 (“App. A”) (listing 367 entries); Pls. 56.1 ¶ 177 (citing App. A).

Before commenting on the substance of the listed communications, we observe that these communications were sporadic, considering the number of days in the relevant period and the fact that sixteen Panel Banks were allegedly involved in this purported conspiracy. As defendants note, “Appendix A catalogs 367 communications[ ] and relates to less than 150 of the more than 700 submission days during the Relevant Period.” Pls. 56.1 ¶ 177 Resp.⁸⁶ Given that sixteen Panel Banks made submissions each day throughout the Relevant Period, these 367 communications reflect less than 3.2% of all possible communications that the Panel Banks could have sent to brokers during the relevant FFFP time period and 4.4% of all possible communications that the Panel Banks could have sent to brokers during the OTC class period.⁸⁷ Thus, even if we assume that defendants shared planned submissions “in order to align their submissions” on some days, Pls. 56.1 § VI.C (capitalizations omitted), plaintiffs have not offered significant probative evidence tending to exclude the possibility of independent action for a substantial portion of the multi-year conspiracy, as required at this stage of the litigation. See Anderson News II, 899 F.3d at 104 (“[T]he picture that emerges is too murky for us to conclude that evidence is anything other than ambiguous.”).

Nevertheless, for the sake of completeness, we turn to the substance of the cited communications. We provide a representative sample here, taken directly from plaintiffs’ Appendix:

• Johnson (BGC) asked Parker (Lloyds) in a chat where he was going 3s LIBOR and Parker responded: “28.” App. A (entry 366).

• Bridgewater (ICAP) wrote Curtler (Deutsche Bank): “what will you go first 3 libors ․ then I can push people into giving something because of lower libors” and Curtler responded: “58 62 65.” App. A (entry 34).

• In a chat with Walker (Natwest), Gordon (Tradition) asked: “where are you going 1s libor??” Walker responded: “20.” App. A (entry 117).

• As Barclays’ submitter Peter Johnson explained “I have conversations with the money brokers, who then disseminate what I think to all the other banks[.]” ECF No. 4310-30 (PX975-A) at 4; see Pls. Br. at 27; App. A (entry 20) (citing this evidence).88

• Yexley (Tullett Prebon) wrote King (Deutsche Bank): “What you going 3mnt libor tday mate?” King responded: “there's no real mkt mate which doesn't help. i'm being made 88/85 so i'll stay at 88 where i went ydy.” App. A (entry 47).

• James Fay (BGC) asked Andrew Parker (HBOS) in a chat: “what do you think for libor today” and Parker responded: “88 3s 79 6s.” App. A (entry 48).

• Johnson (Barclays) called Storey (Barclays) to go through his LIBORs and proposed, based on conversations with brokers, to go up to 40 after going 30 the day before. Storey asked if Johnson had spoken to Dearlove (Barclays), and then Johnson interrupted the conversation to ask an unidentified individual offline to find out “where HBOS is gonna go, because he was the one who sort of was out with us yesterday.” Storey then interrupted the conversation to talk to an unidentified individual off line, and came back and reported that the “guidance is to stick within the bounds[,] so no head above [the] parapet,89 no partial bleeding etcetera ․ Mr. Lucas doesn't want us to be outside the top end ․ and that they apparently chatted on the whole of the 31st floor, by the way.” Johnson responded to this with an expletive and then complained about how he “can't find out what else is doing anything at the moment ․ so I am literally going blind.” Johnson then spoke offline to Freeman (Tradition) to ask if he knew “who's going what in ones twos and threes.” Johnson asked Freeman “Did you talk to HBOS ․ Find out where he's going. Freeman replied: “Just asking, just asking” and Johnson responded: “I need to know urgently.” Johnson and Storey continued to converse, and then Freeman returned and said “30, 20, 18 mate ․ [Natwest] 25, 17, 17.” App. A (entry 76).

• Richard Fenn (Tradition) asked Parker (Lloyds) in a chat: “MAY I ASK WHAT UR GOING FOR 1M LIBOR TODAY MATE?” and Parker replied: “25.” App. A (entry 82).

Though this may be plaintiffs’ strongest evidence, it is still extraordinarily weak. As an initial matter, some of the entries in plaintiffs’ Appendix are plainly too vague to be considered instances in which defendants provided their submissions to brokers for transmittal to other banks. For example:

• Johnson (Barclays) emailed a large internal group and reported: “Hearing a libor contributor is paying 4.50 for 6 month USD (2 others are paying 3.97). This particular contributor will be putting his libor substantially below the 4.5 level.” App. A (entry 282).

• Doherty (UBS) had a document titled “Things to do when I am away” which listed a number of tasks, including in a section “11.45 am Libor Fixings,” in which she wrote: “Get run through from both ICAP and Prebons. We want our fixings to be roughly in the middle of the pack.” She then went on to provide instructions for the mechanics of submitting the rates. App. A (entry 348).90

These types of communications do not reveal a “meeting of minds in an unlawful arrangement.” In re Foreign Exch. Benchmark Rates Antitrust Litig., No. 13 Civ. 7789 (LGS), 2022 WL 294118, at *4 (S.D.N.Y. Feb. 1, 2022) (citation and quotation marks omitted).

Moreover, in many Appendix entries, Panel Bank employees ⁹¹ are complaining (using colorful language) about other Panel Banks’ submissions:

• Barclays’ primary USD LIBOR submitter told a Tullett Prebon broker: “I'm very disappointed in Credit Suisse being such a fucking tool․ He must be in the shit, he's been told by his management to fucking go low.” App. A (entry 240).

• RBC's primary USD LIBOR submitter wrote a BGC broker in a chat: “0X1's and 0X3's todays fix?” The BGC broker replied: 94-90 ․ 95-92 ․??!!!․ [Natwest] going 88 & 91 ․ Lloyds goes 86 & 88 ․ but I'm sure they[’]ll see the lite.” Later, the BGC broker gave his 1 and 3m LIBOR runs as 93 and 94. App. A (entry 170).

• A Bank of America employee wrote to another Bank of America employee: “[Natwest's] libor calls 1m 65 2m 60 3m 57 ․ HSBC libors 65 60 57 1S to 3S.” The second Bank of America employee responds, “Thats what we set yesterday – surely they are higher again today?” The first Bank of America employee replies, “dreamer.” App. A (entry 11).

• A Deutsche Bank employee sent both an ICAP broker and a Tradition broker the same message separately, writing: “You guys are the london brokers and you are making a joke of this. At the moment LIBOR is NOT a reflection of where cash is traded. The NYC desks are loving this because they hate everything setting off a london index but unless you sort it out it will start to shift to another index.” Bridgewater responded: “mike it is not the brokers setting libors. it is something that should be taken up with the other banks direct and bba. as i already said there is probably [sic] half a dozen banks on the panel that shoudnt be, because there input into the cash mkt is negligible to zero.” The Tradition broker responded: “As brokers I'm sure we would unanimously agree that libors a being set too low Unfortunatley [sic] the libor is an average of the contributing banks to the bba ․ as much as we disagree we have no control over the fixing of libors. App. A (entry 177).

• A Deutsche Bank employee told a Tullett Prebon broker, “You have to talk to these idiots at [Natwest] and fucking Lloyd's ․ If the cash is 290 bid, you should not but a 275 fucking fixing.” App. A (entry 179).

• Barclays’ primary USD LIBOR submitter complained to a Tradition broker about getting “pilloried” for being honest and about [Natwest] “setting LIBOR at sixty two when through the arb his other branch is paying seventy one,” and asked where the Tradition broker would go for LIBORs that day. The Tradition broker shared Prebon's LIBOR predictions, which Freeman thought were not high enough and so gave his higher predictions. Johnson agreed, asking, “Could you make sure the market knows that's what” and Freeman replied: “I will do my best to influence but ․ it seems the people at Prebon/ICAP keep low balling it.” App. A (entry 21).

• A Deutsche Bank broker messaged both an ICAP broker and a Tullett Prebon broker at the same time (but as bccs), stating: “any ideas why 2 people left their 3m libor [unchanged] today and a couple of others are so low ․? hopefully they'll see the light [tomorrow].” App. A (entry 44).

These complaints from alleged co-conspirators about other alleged co-conspirators’ actions do not “reveal a unity of purpose,” Gelboim, 823 F.3d at 781 (citation and quotation marks omitted), and in fact undermine plaintiffs’ claimed conspiratorial agreement that defendants shared this information “in order to align their submissions,” Pls. 56.1 § VI.C (capitalizations omitted).

Lastly, while plaintiffs argue that a bank would not disclose “such competitively sensitive information (its anticipated submissions) in the absence of a conspiracy,” Pls. Br. at 97, plaintiffs ignore the purpose of the LIBOR question, which concerned at “what interest rate ‘could’ [a Panel Bank] borrow a typical amount of cash if it were to seek interbank offers and were to accept,” Connolly, 24 F.4th at 835. “After all, [P]anel [B]anks were chosen to be [P]anel [B]anks because, as leaders in the inter-bank lending market, they normally had information about inter-bank lending.” LIBOR IV, 2015 WL 6243526, at *46, slip op. at *117; cf. Gelboim, 823 F.3d at 779 (referring to the Panel Banks as “16 of the world's most important financial institutions”).

4.3.3.1.2. Category #2: Brokers’ Communications to Panel Banks

The second proposition that must be proven true to establish plaintiffs’ broker-banker theory is that the brokers disseminated information about Panel Banks’ planned submissions (as opposed to information about market activity) to other Panel Banks consistently throughout the relevant period. See Pls. 56.1 § VI.

As noted earlier, even if we assume that the brokers spoke to the Panel Banks about the brokers’ “indications” or “run throughs,” i.e., estimates of where USD Libor would be set, “in the morning every business day,” Pls. 56.1 ¶¶ 161, 162, 164,⁹² the simple fact that this occurred would not allow a reasonable fact finder to infer a conspiracy; the factfinder would still need to evaluate the content of the brokers’ “indications” or “run throughs.”

Accordingly, we address the content of these communications. Plaintiffs contend in a section header of their 56.1 statement that these broker indications “represented defendants’ planned submissions rather than permissible information about market activity.” Pls. 56.1 § VI.B (capitalizations omitted). However, despite this broad claim, plaintiffs subsequently rely on far narrower assertions about these indications and with multiple caveats:

• “Brokers, at times, told Defendants that their LIBOR indications were based, at least in part, on information about where Panel Banks were planning to submit their LIBORs, rather than where the brokers were observing activity (if any) in the London interbank market.” Pls. 56.1 ¶ 174 (emphasis added).

• “Defendants understood that broker indications were based, at least in part, on information brokers had about where Panel Banks were planning to submit their LIBORs, rather than where the brokers were observing activity (if any) in the London interbank market.” Id. ¶ 175 (emphasis added).

An examination of the citations underlying these claims, as well as the vast discovery record as a whole, makes it clear why plaintiffs had to include such caveats. Significantly, in many of plaintiffs’ cited communications, the brokers and banks explicitly state that they are “guessing” at where they “think” LIBOR will be set. See, e.g., ECF No. 4288-10; ECF No. 4323-37. A guess from defendants and non-defendants alike as to what others are doing is not evidence that defendants “had a conscious commitment to a common scheme designed to achieve an unlawful objective.” Monsanto, 465 U.S. at 764 (citation and quotation marks omitted).

A few examples of such communications from plaintiff's 56.1 statement illustrate the point:

• An Unidentified broker tells Lloyds back-up submitter: “[B]ut we are not helping the situation, brokers together generally, by walking in the same, all the LIBORs, we're guessing, not where we think LIBORs are, we're guessing what we think the fixers will put them at today, that's what we're doing.” ECF No. 4288-10 at 9; see also Pls. 56.1 ¶ 174 n.252 (citing this evidence).

• Tullett Prebon broker tells Bank of America submitter: “People are asking me what they think the banks are going to set one's LIBOR, that's the thing. It's not what one's LIBOR. What are banks going to set it? And we think that's what they're going to set it at.” Submitter responds, “But I think a lot of people just follow what you say.” ECF No. 4318-25 at 4-5; see also Pls. 56.1 ¶ 174 n.252 (citing this evidence).

• Tradition broker tells Deutsche Bank: “When I send you these libor levels each morning mate I must point out that its what we think they will be ․ not what we thibk [sic] they should be.” ECF No. 4323-37 at 1; see also Pls. 56.1 ¶ 174 n.252 (citing this evidence).

• Norinchukin's submitter asks Tullett Prebon broker where he thinks one month LIBOR was “going to come in” and broker responds: “About 85, mate ․ It's too low, but I think that's where it's going to come in.” ECF No. 4294-27 at 2:1-6; see also Pls. 56.1 ¶ 174 n.252 (citing this evidence).

• Natwest's submitter asks: “[W]hat you got, LIBORs?” and ICAP broker responds: “I'm going to go 20, 25, 30. Only ‘cause I think that's where they're going to be fixed. I mean, I'm not actually thinking that's where the LIBORs are.” ECF No. 4302-32; see also Pls. 56.1 ¶ 174 n.252 (citing this evidence).

• Tullett Prebon broker writes UBS’ submitter: “Libors are an estimation based on where they are currently being set in London, which seems to have almost no correlation to where any real offers are in the market.” ECF No. 4289-6 at 1; see also Pls. 56.1 ¶ 174 n.252 (citing this evidence).

• Barclays’ submitter: “when I sort of ask the brokers where is LIBOR gonna set, they're not telling me where they think the market is, they're telling me where they think LIBOR's gonna set.” ECF No. 4310-14 at 8; see also Pls. 56.1 ¶ 175 n.253 (citing this evidence).

• BTMU's primary submitter: “Q. Every morning, more generally, brokers gave you estimates of where US dollar LIBOR would set? A. Some brokers, correct, yes.” “As part of the run through, yes.” ECF No. 4327-30 (Hoskings Dep. Tr.) at 95:4-9; see Pls. 56.1 ¶ 164 n.237 (citing this evidence).

• Norinchukin's primary submitter: Q. “Okay. Do you remember receiving LIBOR indications from brokers?” A. “Yes.” Q. “And those were indications of what LIBOR was expected to set at before the fixed time?” A. “Yes.” ECF No. 4527-27 (Sheehan Dep. Tr.) at 55:15-20; see Pls. 56.1 ¶ 164 n.237 (citing this evidence).93

• In a November 29, 2007 recorded call with the BBA's Ewan, Barclays’ FXMMC representative Storey said: “You know how LIBOR fixings work: half the time, banks actually ask brokers[.]” “Yeah, banks ask the broker where the market is,” Ewan replied. “As against a bank working out where the market is,” Storey said. ECF No. 4310-36 at ‘777; see Pls. Br. at 31 n.49 (citing this evidence).

• Barclays’ submitter said on an April 2008 recorded call: “I just set my LIBORs where, basically at the moment I'm told to set them by the brokers. I don't dare get out of line with what the brokers say or what the market goes[.]” ECF No. 4308-20 at ‘988.46; see Pls. Br. at 30 (citing this evidence).

• As another example, on August 29, 2007 Credit Suisse submitter Miller wrote: “Morning, there is no cash market essentially, USD libors are set where the brokers suggest as at least this gives us a consensus.” ECF No. 4322-45; see Pls. Br. at 30 n.46 (citing this evidence).

• WestLB submitter asks RP Martin broker for 1-3 month LIBOR indications and he responds: “You tell me what you are going to do. You set them, you tell me.” ECF No. 4296-7 at 1; see also Pls. 56.1 ¶ 174 n.252 (citing this evidence).

• Tullett Prebon broker tells Norinchukin submitter: “I had some arguments this morning. First thing, Barclays ․ I went, okay, whatever, I said, I don't set them. I'm going 85, ones, twos, threes․ And then spoke to [Natwest]․ it was 76, 78, 80. So, I sort of pitched them in the middle․ I'm just putting them in--I'm putting them in the middle of what everyone's going.” ECF No. 4294-13 at 1; see also Pls. 56.1 ¶ 174 n.252 (citing this evidence).

• Citi: “they come into the bank ․ speak to the brokers, and the brokers say, well [Natwest] is setting it here, Citibank is setting it here, and they just go right, go right down the middle all the time.” ECF No. 4322-22 at 4:18-5:3; see also Pls. 56.1 ¶ 175 n.253 (citing this evidence).

At most, these communications provide a sense of whether the LIBOR fix will fall, but are neither precise nor definite, “show[ing] little more than industry participants endeavoring to gather information regarding how their competitors were reacting to the” market. Anderson News, L.L.C. v. Am. Media, Inc., 123 F. Supp. 3d 478, 507 (S.D.N.Y. 2015), aff'd, 899 F.3d 87 (2d Cir. 2018).

These communications, and others like it are, at best, ambiguous. Stated otherwise, plaintiffs’ proffered evidence does not “tend to rule out the possibility that the defendants were acting independently.” Twombly, 550 U.S. at 554 (citation omitted). Accordingly, to deny defendants’ summary judgment motion, the Court would be required to draw impermissible inferences in plaintiffs’ favor. We decline to do so. At summary judgment, plaintiffs must demonstrate “significant probative evidence,” which a reasonable factfinder could rely on to decide in their favor, and cannot rely upon the “mere existence of a scintilla of evidence in support of the plaintiff's position.” Liberty Lobby, 477 U.S. at 249, 252 (quotation marks and citation omitted).⁹⁴ Plaintiffs have not cleared this basic evidentiary standard.

4.3.3.1.3. Category #3: Panel Banks’ Basis for Their Submissions

The third proposition that must be proven true to establish plaintiffs’ broker-banker theory is that Panel Banks relied on information concerning other Panel Banks’ planned submissions when making their own submissions, as opposed to information about market activity, as part of a common understanding reached with other Panel Banks. See Pls. 56.1 § VI. Stated otherwise, plaintiffs’ theory is that brokers “refin[ed] the indications until [d]efendants reached a ‘consensus’ around which they formed their submissions.” Pls. Br. at 96.

To support this proposition, plaintiffs rely on: (1) expert opinions, (2) record evidence referencing the words “pack” and “consensus,” and (3) arguments concerning purported violations of the LIBOR rules. We review each in turn.

4.3.3.1.3.1. Subcategory #1: Expert Evidence

Plaintiffs argue that their experts establish that Panel Banks’ submissions were based on broker indications, and not market factors. See Pls. Br. at 98.⁹⁵ However, their experts acknowledge that each bank had access to “a plethora of information that would allow the banks to make submissions within the pack,” “close to the other submissions,” without “forming an agreement with other banks not to make outlier submissions.” See ECF No. 4304-24 (Marx June 25, 2024 Dep. Tr.) at 89:10–90:8; see also ECF No. 4160-4 (Cragg July 25, 2024 Dep. Tr.) at 66:22-24 (“Q. Could the banks have suppressed LIBOR and been in a pack without colluding? A. Yes.”). Accordingly, plaintiffs’ experts’ opinions are equally consistent with “ ‘conscious parallelism,’ as lawyers call it, ‘tacit collusion’ as economists prefer to call it[ ]—which means coordinating [defendants’] pricing without an actual agreement to do so.” In re Text Messaging Antitrust Litig., 782 F.3d 867, 871, 879 (7th Cir. 2015). Simply stated, “conscious parallelism[ ] does not violate section 1 of the Sherman Act.” Id. at 879.

4.3.3.1.3.2. Subcategory #2: Consensus

We next address plaintiffs’ suggestion that any reference by the Panel Banks to the words “pack” and “consensus,” see, e.g., Pls. 56.1 ¶¶ 168, 171,⁹⁶ serves as clear evidence of a conspiracy. Examples of defendants’ use of these words include:

• UBS submitter instructions for “11.45 am Libor Fixings,” included direction: “Get run through from both ICAP and Prebons. We want our fixings to be roughly in the middle of the pack.” ECF No. 4288-56 at 5; see Pls. 56.1 ¶ 164 n.237 (citing this evidence).

• Credit Suisse's submitter testified at a deposition that “brokers talked to each other,” talked to other Panel Banks and then, “over the course of the morning, consensus would develop as to where LIBOR was going to fix that day.” ECF No. 4527-21 (Miller Dep. Tr.) at 115:22-116:16. “Yes,” he testified, “I think there was an agreement as to where the brokers felt that the fix would be.” Id. at 116:14-16; see Pls. Br. at 29-30 (citing this evidence).

• Bank of America's primary USD LIBOR Submitter, Ms. Sans, similarly viewed LIBOR as a “market consensus” and testified that by “market consensus” she meant “[a]n understanding” as to LIBOR submissions. ECF No. 4304-41 (Sans Dep. Tr.) at 121:13-123:19; see Pls. Br. at 30, 30 n.47 (citing this evidence).

• Additionally, Deutsche Bank trader Christian Bittar told his boss, Nicholls, in an internal communication: “[T]he consensus [is] built around whatever the broker thinks ․ and that consensus is built around what the broker says, and obviously, he has no fucking clue, and although people know it's wrong, understandably, you don't want to be out the crowd, so you want to be around the consensus, so if the consensus is about 2.90, well, you're going to put something around 2.90, but if the consensus is 3.20, then you're going to put 3.20.” Nicholls agreed and said: “[Y]ou don't want to be putting 3.20, when everyone else [is] 2.90. It's going to panic.” PX26 (DB-USDMDL 01360367) (audio) & ECF No. 4280-21 (PX26-A (transcript)) at 16; ECF No. 4304-31 (Nicholls Dep. Tr.) at 123:23-126:20. See Pls. Br. at 44 n.52 (citing this evidence).

• The notes of an April 1, 2008 meeting between JPMorgan and the BBA reflect that primary submitter Amanda Adams (née Sudworth) expressed that “Libor has become more of a consensus,” a point that submitter supervisor Frederic Mouchel agreed with. ECF No. 4310-52 (PX1014) at 1.

• In a December 30, 2008 email from Credit Suisse's Tony Miller to Andrew Reid, Miller described Credit Suisse's process for setting LIBORs as follows: “At 10.50am CS talks to the brokers to see what the consensus feeling is for Libor ․” ECF No. 4323-4 (PX1294) at 2. See Pls. Br. at 44 n.52 (citing this evidence).

• Credit Suisse submitter Tony Miller stated regarding where to set LIBOR: “1.27 seems consensus, I am torn on our setting as the rules, as you know, say where funds are offered at 11.00 am in London, and in reality they arent! But it seems stupid to set a high Libor when we could have got funds at 0.75 in NY y'day. I'll porb [sic] go concensus [sic] in the end[.]” ECF No. 4328 (PX1591). Miller also provided LIBOR indications as follows: “USD Libors, morning, 1m 5.25 2m 5.16 3m 5.14 6m 4.90 9m 4.63 & 12m 4.44. No volume but these are the consensus thoughts.” ECF No. 4322-48 (PX1279) at 2. See Pls. Br. at 44 n.52 (citing this evidence).

• Deutsche Bank's Christian Bittar observed that “these LIBORs are way too low ․ if you say, ‘I see it at 85,’ then every bank, they don't want to be, there is a consensus building around that, and no one wants to be out of the consensus.” PX5 (DB-USDMDL 01212599) (audio) & ECF No. 4280-5 (PX5-A (transcript)) at 3, 5. See Pls. Br. at 44 n.52 (citing this evidence).

• JPMorgan submitter Amanda Adams sent a colleague the “CONCENSUS FIX”. ECF No. 4282-13 (PX109) at 4. UBS submitter Joachim Ruh stated on December 10, 2007: “[A]gain beeing [sic] abotu [sic] 3 points below consensus in 1-3 month[.]” ECF No. 4289-13 (PX340) at 2. See Pls. Br. at 44 n.52 (citing this evidence).

• On September 24, 2008, WestLB's submitter Joyce Taylor responded to a question as to where to set LIBOR: “Well, consensus seems to be it's going to be around 40 somewhere today.” PX434 (PORTIGON0042111) (audio) & ECF No. 4296-8 (PX434-A (transcript)) at 2. See Pls. Br. at 44 n.52 (citing this evidence).

Far from implying a conspiracy, this evidence demonstrates defendants’ purported desire to be in the “pack” or near the “consensus,” which is more consistent with a Panel Bank's individual incentive to set LIBOR at a level that conveyed their credit strength relative to their peers.⁹⁷ Moreover, an “agreement as to where the brokers felt that the fix would be,” ECF No. 4527-21 (Miller Dep. Tr.) at 115:22–116:16, if it existed, is not the same as plaintiffs’ asserted conspiracy, which is that the Panel Banks consciously coordinated to set LIBOR together.

This evidence also tends to show that defendants put their submissions in the middle of where the brokers estimated the fix would be in an effort to avoid outlier status. Such evidence “tend[s] to suggest the absence of [a high level of interfirm] communications” because, if the Panel Banks had an illicit agreement, they “would not have had to rely on third parties to confirm [each other's] strategy” between themselves. Mayor & City Council, 709 F.3d at 139 (emphasis in original). Rather, these communications indicate that defendants “did not know what their competitors were doing. In particular, the communications detail Defendants’ numerous attempts to determine how other Defendants were reacting to the [market]—communications that would be unnecessary if Defendants had reached a ‘common ․ understanding.’ ” Anderson News, L.L.C. v. Am. Media, Inc., 123 F. Supp. 3d 478, 507 (S.D.N.Y. 2015), aff'd, 899 F.3d 87 (2d Cir. 2018) (citation omitted; emphasis in original).

Thus, the repeated mention of “consensus” and “pack” is hardly sufficient, without the support of stronger circumstantial evidence, “to allow a reasonable fact finder to infer that the conspiratorial explanation is more likely than not.” Publ'n Paper, 690 F.3d at 63 (citation omitted).

4.3.3.1.3.3. Subcategory #3: Violation of LIBOR Rules

Plaintiffs argue that the record evidence shows defendants made their submissions based on the anticipated LIBOR submissions of other Panel Banks in violation of LIBOR rules. See, e.g., Pls. Br. at 30-31, 94-95. There are two flaws to this theory.

First, plaintiffs presume, rather than prove, a violation of the LIBOR rules. While the LIBOR rules provided that “Contributor Banks shall input their rate without reference to rates contributed by other Contributor Banks,” Defs. 56.1 ¶ 15, as the quoted examples demonstrate, broker indications are not synonymous with the planned submissions of other Panel Banks. Thus, to the extent that defendants’ submissions were based on brokers’ estimates of where the LIBOR fix would be, those submissions did not necessarily violate the LIBOR rules.⁹⁸ Indeed, anything a broker predicted or understood was not binding on either the bank who spoke to the broker or the broker who received the prediction from the bank.

Second, even if defendants had made their submissions based on the anticipated LIBOR submissions of other Panel Banks, any alleged violation of the LIBOR rules does not necessarily establish a conspiracy, as “[e]vidence that a business consciously met the pricing of its competitors does not prove a violation of the antitrust laws.” Blomkest, 203 F.3d at 1032–33. Plaintiffs assert that any citation to a conscious parallelism price-fixing case is inapposite because a violation of the LIBOR rules “could only be accomplished by [defendants’] jointly agreeing to disregard those rules.” Pls. Br. at 94 n.147. This assertion is unfounded: a Panel Bank could unilaterally violate the LIBOR rules by setting LIBOR at a level designed to convey its credit strength relative to its peers.⁹⁹

One of our colleagues in this District, the Honorable Lorna G. Schofield, supplied a useful analogy in a different antitrust litigation concerning price-fixing of a benchmark rate. See In re Foreign Exch. Benchmark Rates Antitrust Litig., No. 13 Civ. 7789 (LGS), 2022 WL 294118, at *9 (S.D.N.Y. Feb. 1, 2022). The court there wrote:

[I]t would be wrong to say that many drivers exceed the speed limit, use common means to do so and have a common goal for traffic to move more quickly and therefore are engaged in a single global conspiracy. Although a common goal is essential, and parallel conduct is probative, the evidence must provide something more to bind the participants and their conduct together to show a collective venture -- and hence an illegal conspiracy -- rather than independent unlawful conduct.

Id. The same analogy applies here: even if the Panel Banks all violated the LIBOR rules, the evidence must establish something more to bind the participants and their conduct together.¹⁰⁰

4.3.3.1.4. Category # 4: Evidence Which Undermine Plaintiffs’ Broker-Banker Theory

Finally, we turn to our last category of evidence relevant to our review of the broker-bank communications submitted by the parties: the voluminous evidence which undermines plaintiffs’ broker-banker theory. We highlight three particular pieces of the record: (1) uncontradicted testimony by Panel Banks that they did not use brokers to facilitate a conspiracy; (2) testimony by James King, Deutsche Bank's principal USD LIBOR submitter; and (3) evidence related to Credit Suisse.

4.3.3.1.4.1. Subcategory #1: Panel Banks’ Uncontradicted Testimony

While plaintiffs cite ambiguous quotations from recorded chats to support their theory of the conspiracy, defendants cite unambiguous denials by bank submitters that they used brokers as conduits to communicate their planned submissions to other Panel Banks. We provide a few examples here from defendants’ 56.1 statement, which plaintiffs do not dispute:

• “To the extent that brokers on occasion purported to communicate another Panel Bank's anticipated LIBOR submissions to a Bank of America employee, members of Bank of America's LIBOR Submission Team testified that they understood the broker to be making a prediction or providing an estimate, rather than sharing information that actually derived from another Panel Bank.” Defs. 56.1 ¶ 209.101

• “Mr. Miller confirmed that he (i) never entered into an understanding or agreement with any other USD LIBOR Panel Bank to submit artificially low LIBOR rates at any time; and (ii) was not aware of anyone else at Credit Suisse entering into such an understanding or agreement.” Defs. 56.1 ¶ 218.

• “Mr. White also confirmed that Credit Suisse ‘didn't know what the other banks were submitting. We wouldn't ask. We had our own view and our own submission process.’ ” Defs. 56.1 ¶ 219.

• “Mr. White further testified that he spoke to brokers to get a market benchmark, not to receive other Panel Banks’ LIBOR settings or to share Credit Suisse's with other banks.” Defs. 56.1 ¶ 221.

• “Andrew Parker, the primary USD LIBOR submitter for HBOS from August 2007 until HBOS was acquired by LBG, and who also served as backup USD LIBOR submitter for Lloyds Bank as of August 25, 2009, testified that he (i) did not believe he ever entered into, and did not intend to enter into, any agreement with any other USD LIBOR Panel Bank to try to suppress USD LIBOR, or about where USD LIBOR would be set on any day; (ii) never asked anybody at any other USD LIBOR Panel Bank to enter into an agreement to try to suppress USD LIBOR, and was never asked by anyone at any other USD LIBOR Panel Bank to enter into an agreement to suppress USD LIBOR; and (iii) was never asked by anyone at any other USD LIBOR Panel Bank to enter into an agreement about where USD LIBOR would be set on any day.” Defs. 56.1 ¶ 229.

• “Mr. Bellinger and Mr. Collinson testified that they communicated with brokers for the purpose of gaining ‘market color’ to inform their USD LIBOR submissions; they did not communicate with brokers for the purpose of exchanging information with other Panel Banks regarding anticipated USD LIBOR submissions.” Defs. 56.1 ¶ 233.

• “Ms. Adams testified that she did not rely on or use as a reference other Panel Banks’ submissions when setting JPMorgan's submission. She testified that she understood that the LIBOR rules did not permit her to reference other Panel Banks’ rates in inputting JPMorgan's rate, and so she did not.” Defs. 56.1 ¶ 237.

Although these quotations only constitute a small portion of extensive evidence submitted by defendants, see infra pp. 174-176; Defs. 56.1 § IV.B.2, the uncontradicted testimony from at least six different witnesses that only informational exchanges occurred deals a serious blow to plaintiffs’ case, see Baby Food, 166 F.3d at 133 (“Courts generally reject conspiracy claims that seek to infer an agreement from communications despite a lack of independent evidence tending to show an agreement and in the face of uncontradicted testimony that only informational exchanges took place.” (cleaned up)); Weit v. Cont'l Ill. Nat'l Bank and Tr. Co. of Chicago, 641 F.2d 457, 462-63 (7th Cir. 1981) (“Given the need for some degree of cooperation in a venture of this nature, the opportunity to conspire evidence lacks significant probative value. Of greater significance is the sworn testimony compiled during eight years of depositions which uniformly denies discussion of any agreement or understanding as to the interest rate to be charged.”).

4.3.3.1.4.2. Subcategory #2: James King's Communications and Testimony

We next turn to the communications and testimony of Deutsche Bank's principal USD LIBOR submitter, James King. At various points in their briefing and 56.1 statement, plaintiffs rely on King's communications and trial testimony to support their conspiratorial narrative. However, properly understood, these select quotations demonstrate that plaintiffs’ pejorative characterization is misplaced. Two examples follow. First, plaintiffs assert that “Deutsche Bank's submitter James King was ․ caught on a recorded call explaining how” the broker-bank communications worked. Pls. Br. at 96; see also id. at 29 (similar). Mr. King reportedly explained: “[E]veryone goes to a broker” and asks “Where do you think LIBOR should be, given that you guys talk to all the [submitters] and therefore you can get to know the flavor that Deutsche Bank is going in down here; Barclays is going in up there, UBS going there or whatever, West LB is going in there, you should be around here.” Id. at 96 (quoting PX4 (DB-USDMDL 01212566) (audio) & ECF No. 4280-4 (PX4-A) (transcript) at 19). “So of course, you then become the central point for it all.” Id. (citation and quotation marks omitted). Second, plaintiffs cite Mr. King's testimony in Connolly, in which he testified that “it was logical to ‘change the [LIBOR submission] rates ․ so that they lined up with what the brokers were predicting’ because ‘the brokers have access to all the banks, they know where we can borrow money or they think they know where we can borrow money.’ ” Connolly, 24 F.4th at 828 (citation omitted); see also Pls. Br. at 96 (similar).

Even looking at these pieces of the record in isolation, they are plainly consistent with a Deutsche Bank employee ascertaining the range of “what interest rate ‘could’ DB borrow a typical amount of cash if it were to seek interbank offers and were to accept” in compliance with the LIBOR question. Connolly, 24 F.4th at 835. This is not “evidence that would allow a trier of fact to nudge the ball over the 50-yard line and rationally ․ say that the existence of an agreement is more likely than not.” Kleen Prods. LLC, 910 F.3d at 934.

Moreover, plaintiffs’ reliance on James King is risible when taking into account other portions of his testimony in Connolly. The trial in Connolly focused on whether two former Deutsche Bank employees manipulated LIBOR to benefit the bank's trading positions. 24 F.4th at 823. At trial, the government relied on testimony from Mr. King, an “alleged coconspirator[ ] who had entered into [a] cooperation agreement[ ] with the government.” Id. at 824. King crucially testified that “Gavin Black occasionally asked me to manipulate the rates or to put in a submission that was higher or lower than I would have done to benefit his trading position,” and that “Matthew Connolly on occasion made requests for me to change our LIBOR rates and to benefit the trader's position.” Id. at 828 (emphasis altered). He also testified “that while sometimes the rates the brokers were suggesting were all similar, ‘there [we]re periods when the rates are vastly different’; and then ‘[we] have to come up with something that we feel is a fair reflection of our rate.’ ” Id. (citation omitted; alterations in original).

Clearly, Mr. King's testimony concerning an intrabank conspiracy to manipulate rates by putting in submissions that were higher or lower to benefit the bank's own trading position is not evidence of an interbank conspiracy to suppress submissions to be in a pack with fifteen other banks. See LIBOR VI, 2016 WL 7378980, at *5 & n.8, slip op. at *11 & n.8 (“Profit-motivated trader-based manipulation, which was sporadic and would result in both the inflation and deflation of LIBOR submissions, has nothing to do with the persistent suppression conspiracy that is at issue in the antitrust claims.” (citations omitted)). Also, Mr. King's testimony that he did not always follow the brokers’ suggested rates because those estimates varied significantly is also inconsistent with plaintiffs’ asserted conspiracy. Cf. Commodity Exch., Inc., Gold Futures & Options Trading Litig., 328 F. Supp. 3d at 230 (dismissing complaint where “chats describe coordinated trading and sharing of order flow information that is either inapposite to or inconsistent with a fix-suppression scheme”).

4.3.3.1.4.3. Subcategory #3: Evidence Relating to Credit Suisse

Finally, defendants have provided evidence challenging the likelihood of any connection between Credit Suisse and a conspiracy facilitated by brokers.¹⁰² For example, testimony from Credit Suisse's USD LIBOR submitter and backup submitter significantly undermines multiple aspects of plaintiffs’ broker-communications theory, namely that: (1) Credit Suisse did not provide brokers its planned submissions for the purpose of aligning all banks’ submissions in a pack;¹⁰³ (2) Credit Suisse did not understand brokers to be disseminating Panel Banks’ submissions (as opposed to information about market activity);¹⁰⁴ and (3) Credit Suisse did not rely on information concerning others banks’ planned submissions (as opposed to information about market activity) when making their own submissions in compliance with a common understanding previously reached with the other banks.¹⁰⁵

By contrast, plaintiffs cherry-pick from the record to connect Credit Suisse to the alleged conspiracy. For instance, plaintiffs assert that Credit Suisse “asked those brokers for the ‘consensus’ and formed its submissions based on a ‘desire to be within some sort of consensus’ ” based on a single email and the deposition testimony of Credit Suisse's LIBOR submitter, Mr. Miller. Pls. Br. at 115-116, 116 n.181 (citing ECF No. 4323-1; ECF No. 4527-21 (Miller Dep. Tr.) at 137:17-21, 138:4-9).¹⁰⁶ However, plaintiffs conveniently omit the portion of Mr. Miller's deposition testimony in which he states that, while he “would take the broker information to inform the LIBOR[,]” it would be “a bit of a jump” to suggest that he “relied on those brokers’ suggestions of LIBOR to determine Credit Suisse's LIBOR submissions.” ECF No. 4453-1 (Miller Dep. Tr.) at 112:22-113:5. In fact, Miller unequivocally stated that “the brokers’ suggestions had little merit in where Credit Suisse actually ended up submitting their fix.” Id.

Because Credit Suisse's testimony flatly contradicts plaintiffs’ broker-communications theory and plaintiffs offer only cherry-picked evidence, a finding of conspiracy would not be well-supported. “When opposing parties tell two different stories, one of which is blatantly contradicted by the record, so that no reasonable jury could believe it, a court should not adopt that version of the facts for purposes of ruling on a motion for summary judgment.” Scott v. Harris, 550 U.S. 372, 380 (2007).

4.3.3.1.5. Conclusion

Taking into consideration the totality of the evidence submitted by the parties, the broker-bank communications tend to disprove the existence of conspiracy, and at best, constitute “ambiguous evidence—that is, evidence that is equally consistent with independent conduct as with illegal conspiracy.” Anderson News II, 899 F.3d at 98 (citations and quotation marks omitted).

As we stated at the beginning of this section, for plaintiffs’ broker-banker theory to be correct, three specific assertions must be proven true: (1) the Panel Banks provided their planned submissions to brokers for the purpose of aligning all banks’ submissions in a pack; (2) brokers then disseminated this information (as opposed to information about market activity) to other Panel Banks consistently throughout the relevant period; and (3) Panel Banks relied on information concerning others banks’ planned submissions (as opposed to information about market activity) when making their own submissions, as part of a common understanding previously reached with other Panel Banks. See Pls. 56.1 § VI. Notably, plaintiffs have failed to provide unambiguous evidence proving any of these assertions. Put simply, to read the evidence as plaintiffs do would require the Court to make multiple unreasonable factual inferences in their favor, which is not required of any Court at the summary judgment stage. Accordingly, plaintiffs have failed to present a genuine issue as to any material fact relevant to their broker-banker theory of conspiracy. See Gorzynski, 596 F.3d at 101; see also Apex Oil Co., 822 F.2d at 254 (“[I]nferring a conspiracy from the existence of these communications would amount to speculation under the circumstances herein.”).

4.3.3.2. Direct Communications

In support of the third plus factor, “evidence of a high level of interfirm communications,” Gelboim, 823 F.3d at 781 (citations and quotation marks omitted), plaintiffs also rely on communications between members of the Panel Banks. See, e.g., Pls. Br. at 16-22, 37-38, 99-104; Pls. 56.1 §§ III, IV, VII. Plaintiffs argue that these communications occurred both “under the auspices of the [British Bankers’ Association,] BBA, [Foreign Exchange & Money Markets Committee,] FXMMC, and LIBOR Steering Group” and outside of these groups. Pls. Br. at 99. We begin by reviewing the communications purportedly made “under the auspices” of these three groups, then address other communications.

4.3.3.2.1. BBA, FXMMC, LIBOR Steering Group

We begin with a short summary of the three relevant players. The BBA “was a trade association for the U.K. banking sector that introduced LIBOR in 1986” and “administered USD LIBOR” during the relevant period. Defs. 56.1 ¶¶ 12, 13(a).¹⁰⁷ According to John Ewan, the BBA's LIBOR manager and director during the relevant period, the FXMMC was “a decision making body” whose “main role ․ involved oversight of the membership of the submitter panels, and the currencies that would make up the rate.” Defs. 56.1 ¶¶ 24, 26. The FXMMC also oversaw “issues relating to governance, scrutiny and surveillance of LIBOR.” Id. ¶ 26. Finally, the LIBOR Steering Group was composed of “leading market practitioners active in the inter-bank money markets in London.” Marx Opening Report ¶ 26 (citation and quotation marks omitted).

“[A]lthough the nature of trade associations is such that they are frequently the object of antitrust scrutiny, every action by a trade association is not concerted action by the association's members.” AD/SAT, 181 F.3d at 234 (citation omitted). Accordingly, “an antitrust plaintiff must present evidence tending to show that association members, in their individual capacities, consciously committed themselves to a common scheme designed to achieve an unlawful objective.” Id. Otherwise, “[w]hile meetings among competitors undoubtedly provide opportunities to conspire, deeming those opportunities as proof of a conspiracy would condemn independent professional associations.” Ross, 35 F. Supp. 3d at 444 (citation omitted); cf. PharmacyChecker.com, 530 F. Supp. 3d at 335 (“[M]embership in common trade associations and common attendance of trade shows indicates an opportunity to conspire, but it is not enough, standing alone, to provide the ‘plus factors’ needed to allege a conspiracy.” (collecting cases)).

Although this Court dismissed the BBA as a defendant because it is not a member of the conspiracy upheld in Gelboim, see LIBOR VIII, 2019 WL 1331830, at *15, slip op. at *39, plaintiffs continue to assert that the BBA and FXMMC were part of the conspiracy, see OTC Plaintiffs’ Interrogatory Responses, Resp. No. 1 at 21 (“The BBA, both directly and with the [FXMMC], also acted as a co-conspirator, agent, facilitator, conduit, and/or intermediary to effectuate the Panel Banks’ Conspiracy.”); DAPs’ Interrogatory Responses, Resp. No. 2 at 66 (“The BBA, both by itself—through John Ewan (the BBA Director responsible for LIBOR and coordinating, scheduling, and attending FXMMC meetings between Panel Banks) and ․ Angela Knight (the BBA's CEO)—and through the FXMMC, acted as an agent, facilitator, conduit, and/or intermediary to effectuate the Panel Banks’ Conspiracy.”).¹⁰⁸

In any event, plaintiffs contend that communications made under the “auspices” of the BBA, FXMMC, and LIBOR Steering Group support an inference of a conspiracy, despite: (1) not identifying any purportedly nefarious LIBOR Steering Group communications; (2) not every defendant attending the BBA and FXMMC meetings; (3) non-defendants attending the BBA and FXMMC meetings; and (4) the communications being equally consistent with independent action, if not disproving the conspiracy entirely. We expand on these deficiencies below.

First, although plaintiffs repeatedly mention the LIBOR Steering Group in their briefing, plaintiffs only reference the group in three paragraphs of their 56.1 statement and fail to cite a single communication occurring at a meeting of the LIBOR Steering Group. See, e.g., Pls. 56.1 ¶¶ 32 n.33, 55 n.56, 240.¹⁰⁹

Second, plaintiffs concede that “all [d]efendants did not attend the FXMMC and LIBOR Steering Group meetings.” Pls. Br. at 99. “No Norinchukin, Portigon, Rabobank, or RBC employees were members of the FXMMC” or even “participated in FXMMC meetings” during the relevant period. Defs. 56.1 ¶ 27(b), (c). Indeed, Dr. Marx, FFFP plaintiffs’ expert, compiled a useful chart depicting the Panel Banks who were invited and attended FXMMC and LIBOR panel bank meetings during the relevant period. See Marx Opening Report Fig. 7. We copy it below:

Tabular or graphical material not displayable at this time.

Id. Once again, attendance at these meetings, without more, does not establish an inference of a conspiracy even for those defendants who did attend. Cf. In re Graphics Processing Units Antitrust Litig., 527 F.Supp.2d 1011, 1023 (N.D. Cal. 2007) (“Attendance at industry trade shows and events is presumed legitimate and is not a basis from which to infer a conspiracy, without more.”).¹¹⁰

Third, certain FXMMC meetings included participants who were not Panel Bank employees.¹¹¹ Notably, SocGen participated in FXMMC meetings before it joined the USD LIBOR panel in February 2009, see Marx Opening Report Fig. 7; Defs. 56.1 ¶ 3, the Association of Corporate Treasurers attended an FXMMC meeting on March 23, 2009, see Defs. 56.1 ¶ 28 Resp. (citing ECF No. 4318-20 (PX1113)), and members of the Abbey National attended multiple FXMMC meetings, see id.; ECF No. 4445-7, Blakemore Reply Decl., Ex. 7 (Marx Opening Back-up Materials, “Marx-FXMMC meeting summary table.xlxs”); ECF No. 4323-16 (minutes from an October 29, 2008 FXMMC meeting listing a non-Panel Bank attendee from Abbey National). “The presence of numerous uninvolved observers at ․ meetings tends to dispel any specter of illegality.” Ross, 35 F. Supp. 3d at 445 (citation and quotation omitted).

Fourth, the communications cited by plaintiffs in their 56.1 Counterstatement are equally consistent with independent action. See Pls. 56.1 §§ III.A, III.C, III.D, VIII.A, IX.A. Plaintiffs make the following broad assertions, among others: (1) during meetings of the FXMMC with the BBA between August and December 2007, defendants “informally discuss[ed] the challenge of setting LIBORs in a non-functioning interbank market but agreed to take no corrective action[ ] and agreed to revisit the issue in the new year,” Pls. 56.1 § III.A (capitalizations omitted); (2) in 2008, defendants “participated in the BBA's months-long formal ‘consultation’ process, led by the FXMMC, to review the LIBOR definition and instructions,” Pls. 56.1 § III.C (capitalizations omitted); (3) the LIBOR “ ‘consultation’ process led by the FXMMC resulted in no changes to the LIBOR definition and instructions,” Pls. 56.1 § III.D (capitalizations omitted); (4) the BBA “orchestrate[d] co-ordinated movement” in mid-April 2008 when defendants “jointly moved LIBOR up to conceal their suppression,” Pls. 56.1 § VIII.A (capitalizations omitted), ¶ 195; and (5) “Thomson Reuters conducted tolerance checks of submissions on behalf of the FXMMC,” Pls. 56.1 § IX.A (capitalizations omitted).

Defendants vigorously dispute the truth of these assertions, point out that plaintiffs often stretch their cited exhibits beyond any reasonable inferences, and observe that plaintiffs’ citations often lack context. See, e.g., Pls. 56.1 §§ III, VIII, IX; Pls. 102-06. Having read the record fully, the Court agrees with defendants.¹¹²

Nonetheless, even assuming there was adequate support for these assertions, plaintiffs fail to provide “evidence of a high level of interfirm communications ․ [that] might permit a jury to infer the existence of an agreement.” Gelboim, 823 F.3d at 781 (citations and quotation marks omitted). Plaintiffs’ assertions are equally consistent with independent action, i.e., that the Panel Banks were unilaterally making low LIBOR submissions to signal the bank's creditworthiness to investors. Thus, plaintiffs have not presented communications made “under the auspices of the BBA, FXMMC, and LIBOR Steering Group,” Pls. Br. at 99, that “tend[ ] to show that [defendants], in their individual capacities, consciously committed themselves to a common scheme designed to achieve an unlawful objective,” AD/SAT, 181 F.3d at 234.

4.3.3.2.2. Other Communications

Plaintiffs also rely upon direct communications between defendants outside the “auspices” of the BBA, FXMMC, and LIBOR Steering Group, though plaintiffs assert that these “are the communications about which [p]laintiffs know the least.” Pls. Br. at 99-102; see also Pls. 56.1 § VII.¹¹³

Plaintiffs’ arguments and proffered evidence are unconvincing. For example, plaintiffs cite one email in which HSBC FXMMC Representative Wood asked his manager, John Flint, “Shud we host informal after skool clearing bnks mon mkts meeting to discuss current mkt conditions and share mkt intell.” Pls. Br. at 100 (quoting ECF No. 4281-20 (PX71)). After Flint responds “Yes – a good idea” and “Make sure the drinks are cheap though – we can't afford much,” Wood emails back “Will set bar tab up ․ Will call the suspects on mon am․ Will come back with some more whacky ideas when beer starts taking effect. Won't be long.” ECF No. 4281-20 (PX71). From this conversation, plaintiffs argue that it is a “fair inference that the banks at the bar discussed the lack of interbank offers and how to coordinate their settings.” Pls. Br. at 100.

It goes even without saying that one email exchange discussing a potential happy hour with peers and colleagues and containing a vague reference to “shar[ing] m[ar]k[e]t intel[,]” ECF No. 4281-20 (PX71), does not establish an antitrust conspiracy. Given that plaintiffs failed to show that this meeting even occurred, Pls. 56.1 ¶ 60 Resp. -- let alone what was said at the bar once the drinks started flowing -- we reject plaintiffs’ argument that it's a “fair inference that the banks at the bar discussed the lack of interbank offers and how to coordinate their settings,” Pls. Br. at 100. These types of cherrypicked examples are insufficient to show clandestine conduct by some individual bankers, let alone a far-ranging, sixteen-bank conspiracy operating in the shadows for years without detection.

Further, despite having access to millions of documents and tens of thousands of audio files, see supra pp. 10-14, plaintiffs have pointed to a scant number of allegedly inculpatory communications. That failure is striking in light of their experts’ opinions that “[t]o collude, firms need to coordinate, and they most easily do so by explicitly communicating with one another.” Cragg Opening Report ¶ 114; see also Marx Opening Report ¶ 118 (similar); Apex Oil Co., 822 F.2d at 253 (noting conspiracies which “concern[ ] long-term complex relationships among competitors” are typically “more susceptible of direct proof”).

4.3.3.3. Conclusion

Ultimately, plaintiffs have not presented “evidence of a high level of interfirm communications,” Gelboim, 823 F.3d at 781 (citations and quotation marks omitted), and none of the communications cited in the parties’ voluminous papers—viewed separately or as a whole—provide “sufficient evidence to allow a reasonable fact finder to infer that the conspiratorial explanation is more likely than not,” Publ'n Paper, 690 F.3d at 63 (citation omitted). In sum, plaintiffs have not advanced the ball.

4.3.4. Plus Factor #4: LIBOR's Structural Susceptibility to Collusion

Next, plaintiffs endeavor to rely on a fourth plus factor that considers whether an “industry[’s] structure” is conducive to collusion. See Pls. Br. at 104-05 (citing Publ'n Paper, 690 F.3d at 65; In re High Fructose Corn Syrup Antitrust Litig., 295 F.3d 651, 656-57 (7th Cir. 2002)).

There are two issues with plaintiffs’ argument. First, plaintiffs’ effort does not get off the ground because in the Second Circuit, an industry's conduciveness to collusion is not a plus factor at the summary judgment stage. See, e.g., Venture Tech., Inc. v. Nat'l Fuel Gas Co., 685 F.2d 41, 47 (2d Cir. 1982) (“[O]ne who alleges that he is a victim of an antitrust conspiracy and seeks to impose the heavy sanctions of the Sherman Act upon the accused, must show more than the existence of a climate in which such a conspiracy may have been formed.”); Gamm v. Sanderson Farms, Inc., 944 F.3d 455, 466 (2d Cir. 2019) (“The law is clear that ‘[t]he mere opportunity to conspire does not by itself support the inference that such an illegal combination actually occurred.’ ” (citations omitted)).¹¹⁴

In arguing otherwise, plaintiffs misconstrue the Second Circuit's decision in Publ'n Paper Antitrust Litig., 690 F.3d 51 (2d Cir. 2012). In that case, the Circuit listed “additional evidence” that plaintiffs provided which “support a reasonable finding that [defendants] engaged in price fixing.” Id. at 65. The Circuit noted, but did not rest its holding on, that “the district court found, and defendants do not contest, that the publication paper industry is conducive to collusion.” Id. Crucially, however, the Circuit held that “there is ample evidence of conspiratorial behavior” since it found that “[m]ost notably, it is undisputed that in private phone calls and meetings—for which no social or personal purpose has been persuasively identified—[an employee] shared [his company's] pricing strategies with [another company's employee] and both men disclosed to each other their companies’ intentions to increase prices before those decisions had been publicly announced.” Id.¹¹⁵ The second deficiency with plaintiffs’ argument is simple: the evidence they cite does not tend to exclude the possibility that the alleged conspirators acted independently. Plaintiffs cite testimony ¹¹⁶ from the BBA director, Mr. Ewan, and a report conducted in 2012 by the director of England's Financial Services Authority ¹¹⁷ to argue that the LIBOR-setting process itself was “unusually susceptible to collusion” because “the fact that Defendants were all members of the BBA put [the BBA] in a difficult position in terms of enforcing adherence to the LIBOR definition.” Pls. Br. at 47-48, 104-105 (citing Pls. 56.1 ¶ 154, § V.A; Wheatley Report). However, to argue that this evidence implies the existence of a massive, industry-wide conspiracy requires a major leap in logic, as it fails to establish: (a) that the Panel Banks were not adhering to the LIBOR definition on a systemic basis; and (b) that the BBA would not have been put in a difficult position if the banks were independently ignoring the LIBOR definition. Moreover, a conflict of interest premised on the BBA's difficulty policing its own members does not tend to exclude the possibility of independent action. Cf. N. Am. Soccer League, LLC v. United States Soccer Fed'n, Inc., 883 F.3d 32, 40 (2d Cir. 2018) (affirming district court's ruling that “there was insufficient evidence of concerted action” even though it found evidence of a conflict of interest).

Additionally, plaintiffs cite the testimony of Drs. Cragg and Marx regarding the factors that make the LIBOR-setting process conducive to collusion. See Pls. Br. at 105, 105 n.163 (citations omitted). However, while we found the opinions of Drs. Cragg and Marx on this topic to be admissible, supra pp. 83-85, we also ruled that the experts could not “inform the jury that based on the evidence defendants likely engaged in anticompetitive conduct,” DPWN Holdings, 2019 WL 1515231, at *7. Therefore, their testimony does not move the needle for plaintiffs.

Thus, even assuming “the fact that Defendants were all members of the BBA put [the BBA] in a difficult position in terms of enforcing adherence to the LIBOR definition,” Pls. Br. at 104-05 (citations omitted), none of the submitted evidence “would allow a trier of fact to nudge the ball over the 50-yard line and rationally ․ say that the existence of an agreement is more likely than not,” Kleen Prods. LLC, 910 F.3d at 934; see also id. at 935 (noting that “[b]ecause of the competing inferences that can be drawn from this market structure, the district court properly found that the economic evidence did not tend to exclude the possibility of independent action”).

4.3.5. Plus Factor #5: Concealment Efforts

Plaintiffs also endeavor to present “evidence of efforts to conceal the existence of a conspiracy” as a plus factor supporting an inference of a conspiracy. See Pls. Br. at 105-06; Pls. 56.1 § VIII. Plaintiffs levy several assertions, including a “ ‘CYA’ (‘cover your ass’)”¹¹⁸ theory whereby defendants, apparently “expect[ing] that regulators might investigate their [LIBOR] submissions,” limited written and recorded communications, destroyed audio files, set up separate ledgers that logged transactions above the LIBOR rate, and met in private to discuss LIBOR. See Pls. Br. at 105-106.

At bottom, plaintiffs are making much ado about very little. First, plaintiffs rely heavily on the two-week period in mid-April 2008 during which the BBA purportedly “orchestrate[d] co-ordinated movement” so defendants could “jointly move[ ] LIBOR up to conceal their suppression.” Pls. 56.1 § VIII.A (capitalizations omitted), ¶ 195. This two-week inflationary period, however, is inconsistent with the alleged multi-year suppression conspiracy. In addition, plaintiffs’ reliance on the BBA's responses to various press inquiries is not supportive of the alleged conspiracy. “The BBA is not a financial institution whose main concern is to project financial soundness, and any act of assurance that the BBA allegedly took did not further the alleged conspiracy.” LIBOR VIII, 2019 WL 1331830, at *15, slip op. at *39.¹¹⁹

Next, plaintiffs try to support their contentions of a cover-up by purporting to offer clear evidence of concealment through one-off statements from individual bank employees and speculative assertions about why the banks might have undertaken supposedly clandestine actions. See Pls. 56.1 § VIII.C. However, plaintiffs largely present isolated examples that, when viewed in context, fail to support the conclusion that the defendant banks sought to cover up a sixteen-bank conspiracy. See id.

Moreover, even assuming that defendants made efforts to conceal their suppression by not publicly borrowing at rates above LIBOR, see Pls. 56.1 § VIII.D, which defendants vigorously dispute,¹²⁰ this evidence is equally consistent with unilateral action by defendants to suppress. In short, it does not establish “unity of purpose with respect to fixing” LIBOR. United States v. Aiyer, 470 F. Supp. 3d 383, 405 (S.D.N.Y. 2020), aff'd, 33 F.4th 97 (2d Cir. 2022).

Perhaps recognizing that their evidence of concealment is weak, plaintiffs turn this factor on its head and suggest that defendants’ active efforts to cover their tracks prevented plaintiffs from finding more robust evidence of concealment. See, e.g., Pls. Br. at 100. This argument cannot stand. As the notable astronomer Carl Sagan once famously stated: “Absence of Evidence is not Evidence of Absence.” Carl Sagan, The Demon-Hunted World 200, 210 (2d ed. 1997); see also Saccenti v. Target Corp., No. 20-CV-4098 (BMC), 2021 WL 2716644, at *2 (E.D.N.Y. July 1, 2021) (noting “the absence of evidence is not evidence of absence” in the summary judgment context).

Stated otherwise, the dearth of evidence supporting a conspiracy is not evidence that concealment, in fact, occurred. And the assumption of concealment cannot circularly prove conspiracy. This is common sense, but it is also imperative at the summary judgment stage, which requires plaintiffs to come forward with “evidence that would be sufficient, if all reasonable inferences were drawn in [plaintiffs’] favor, to establish the existence of [an] element at trial.” City of Waterbury, 542 F.3d at 36 (citation and quotation marks omitted). A “litigant may not proceed by first assuming a conspiracy and then explaining the evidence accordingly.” Blomkest, 203 F.3d at 1033; see also Valspar Corp, 873 F.3d at 198 (same).

4.3.6. Plus Factor #6: Econometric Models and Expert Evidence

Plaintiffs briefly address as another “plus factor” econometric models from their suppression experts, Drs. Bernheim and Snow, as well as opinions of their collusion experts, Drs. Cragg and Marx, that they claim further support the inference of a conspiracy. See Pls. Br. at 107-110.

Given that we exclude the models from Drs. Bernheim and Snow, as discussed infra pp. 201-238, we need not expand on that analysis with much additional detail. However, even assuming, arguendo, that plaintiffs’ experts’ models could be relied upon, the conclusion that plaintiffs seek to extract from those models -- “LIBOR was lower[ ] than it otherwise would have been absent the alleged conspiracy.” Pls. Br. at 58 (citation and quotation marks omitted); see also Pls. Br. at 107 (similar) -- says nothing about collusion. While it may speak to potential suppression -- which we independently reject below, see infra pp. 201-238 -- it does not speak to whether the Panel Banks colluded in their LIBOR submissions. Accordingly, this evidence does not influence our analysis with respect to plaintiffs’ conspiracy claims.

As for Drs. Cragg and Marx, we ruled that neither expert could “state that the defendants did or did not engage in anticompetitive conduct.” Supra pp. 83-85; see also U.S. Info. Sys., 313 F. Supp. at 240 (similar). Thus, we excluded much of what plaintiffs seek to rely on here. See Pls. Br. at 108-10; supra pp. 83-85. Consequently, given our rulings on the experts, there is no evidentiary underpinning of this asserted plus factor.

4.3.7. Plus Factor #7: Government Investigations and Regulatory Settlements

For the seventh plus factor, we are called upon once again, to evaluate whether the findings and allegations resulting from government investigations and regulatory settlements may be used to support plaintiffs’ theories. See, e.g., Defs. Br. at 25-27; Pls. Br. at 110-111; Defs. Reply Br. at 31-33.

As we noted in LIBOR I, “domestic and foreign regulatory agencies have ․ reached settlements with several banks involved in the LIBOR-setting process, with penalties reaching into the billions of dollars.” LIBOR I, 935 F. Supp. 2d at 676, slip op. at *1. In 2015, we reviewed certain settlements and concluded “the public settlement agreements do not reveal any broad pattern of collusion in USD LIBOR.” LIBOR IV, 2015 WL 6243526, at *45, slip op. at *112. In 2019, we reviewed more recent settlements and similarly concluded that they “contain no allegation or finding that an inter-bank persistent suppression conspiracy existed.” LIBOR VIII, 2019 WL 1331830, at *18, slip op. at *46.

These conclusions remain unchanged. Regulators’ investigations of the Panel Banks generally fall into two categories. For seven Panel Banks, no government regulator made any allegation or finding as to USD LIBOR manipulation. See Defs. 56.1 ¶¶ 282, 290-91, 297, 298, 309, 310, 315, 575, 578-79, 583-84, 587-89, 591 (concerning Bank of America, Credit Suisse, HSBC, JPMorgan, Norinchukin, Portigon, and RBC).¹²¹ For the remaining Panel Banks, no government regulator made any allegation or finding that an interbank persistent suppression conspiracy existed. Id. ¶¶ 287,¹²² 288-89, 294, 301, 304, 306, 308, 313, 318, 593. It is noteworthy that despite serious investigations conducted by government regulators, none of them report findings that support plaintiffs’ theory in this case.

Moreover, some findings by regulators and law enforcement agencies conflict with plaintiffs’ asserted conspiracy, as they show Panel Banks independently making decisions to benefit themselves as opposed to a sixteen-bank conspiracy. For example, included in the record are findings and consent orders pertaining to Deutsche Bank in which “regulatory and law enforcement agencies concluded that Deutsche Bank traders engaged in opportunistic, up-and-down manipulation and attempted manipulation intended to benefit trading positions.” Defs. 56.1 ¶ 581. According to these agencies, “ ‘[o]n almost every day’ ” in 2008 to 2009, Deutsche Bank's USD LIBOR submitter “ ‘altered [Deutsche Bank]’s USD LIBOR submissions to align with the needs of this trading strategy, i.e. persistently low 1 month and high 3 and 6 month USD LIBOR submissions.’ ” Id. ¶ 582 (citations omitted; emphasis added). Similarly, “Lloyds Bank and HBOS's resolution with the DOJ described three isolated requests of USD LIBOR submitters to adjust submissions on a particular day and in a particular direction to benefit the banks’ own trading position.” Id. ¶ 300 (emphasis added).¹²³

These findings concerning intrabank actions to manipulate rates by putting in submissions to benefit the banks’ own trading positions are inconsistent with an interbank suppression conspiracy. See LIBOR VI, 2016 WL 7378980, at *5 & n.8, slip op. at *11 & n.8 (“Profit-motivated trader-based manipulation, which was sporadic and would result in both the inflation and deflation of LIBOR submissions, has nothing to do with the persistent suppression conspiracy that is at issue in the antitrust claims.” (citations omitted)).¹²⁴

4.3.8. Plus Factor #8: Other Evidence Refuting the Existence of Conspiracy

Defendants provide extensive evidence that they contend refutes the alleged conspiracy.¹²⁵ Having reviewed the record, we agree. We provide a few illustrative examples below.

One compelling type of evidence is that Panel Banks often complained about their fellow banks’ submissions to the BBA. For example, in September 2007, “Miles Storey, Barclays’ then Head of Liquidity and Balance Sheet Management, told Mr. Ewan that ‘a lot of people are posting LIBORs significantly lower from where they would be if’ their LIBOR submissions ‘reflected] closer [to] where they actually could get money,’ including ‘examples of banks who are posting LIBORs ․ [t]ens of basis points under where they're effectively paying.’ ” Defs. 56.1 ¶ 367. In April 2008, Mr. Knight of Bank of America “expressed concern to John Ewan that, in Mr. Knight's view, some other Panel Banks were not making their LIBOR submission in line with where they were borrowing USD.” Id. ¶ 362. Representatives from JPMorgan also communicated with the BBA during the relevant period regarding the accuracy of LIBOR. See id. ¶ 407. “RBC raised concerns about the submissions of other Panel Banks to John Ewan.” Id. ¶ 417.

Similarly, the Panel Banks often brought their complaints to journalists and regulators. In September 2007, a Barclays employee told a journalist with the Financial Times that “while Barclays could be ‘extremely honest’ about LIBOR because its ‘liquidity [wa]s robust,’ other Panel Banks ‘that are indeed facing a funding crisis are hiding in the pack and playing down the issue’ by making artificially low USD LIBOR submissions.” Id. ¶ 371. “The Federal Reserve Bank of New York acknowledged that between ‘fall of 2007 and early 2008,’ Barclays regularly communicated their concerns that ‘some banks could be underreporting their LIBOR’ and that ‘reported ‘Libors’ appeared unrealistically low.’ ” Id. ¶ 373. “Rabobank employees at times questioned the publicly reported LIBOR submissions of other Panel Banks ․ with regulators.” Id. ¶ 414. Representatives from JPMorgan communicated with “officials at the Bank of England, and the Federal Reserve Bank of New York on several occasions during the [r]elevant [p]eriod regarding the accuracy of LIBOR.” Id. ¶ 407. Taken together, these complaints are inconsistent with the possibility that defendants “had a conscious commitment to a common scheme designed to achieve an unlawful objective.” Monsanto, 465 U.S. at 768.

The absence of evidence relating to SocGen is equally compelling. Plaintiffs do not explain how SocGen was supposedly “initiated” into the alleged conspiracy -- as would have been necessary for the alleged sixteen-bank conspiracy to continue -- when it replaced HBOS on the LIBOR panel in February 2009. See ECF Nos. 4175-25 (Asker FFFP Rebuttal Report) ¶ 75, 4175-35 (Asker FFFP Reply Report) ¶ 11. Here, the absence of specific evidence showing how SocGen became “a party to the general conspiratorial agreement,” Sitts v. Dairy Farmers of Am., Inc., 417 F. Supp. 3d 433, 469 (D. Vt. 2019) (quoting United States v. Avery, 128 F.3d 966, 971 (6th Cir. 1997)), further undermines plaintiffs’ theory of collusion.

5. Conclusion

In sum, plaintiffs have failed to provide “ ‘strong direct or circumstantial evidence’ sufficient to satisfy Matsushita’s ‘tends to exclude’ standard.” Publ'n Paper, 690 F.3d at 63. Plaintiffs’ direct evidence is extremely weak. Even when viewed through the most charitable lens, plaintiffs’ circumstantial evidence is entirely speculative, requires several inferential leaps, or relies on the fundamentally flawed analyses of their experts Drs. Cragg and Marx.

Ultimately, in consideration of the entire record, “we conclude that a factfinder could not reasonably infer that the conspiratorial explanation is more likely than not.” Anderson News II, 899 F.3d at 113. Accordingly, despite nearly a decade spent building an evidentiary record, plaintiffs have not proffered evidence sufficient to create any dispute of material fact as to the existence of a multi-year, sixteen-bank conspiracy to suppress LIBOR.

IV. SUPPRESSION

The second Upstream Issue concerns “the alleged suppression of LIBOR.” See ECF No. 3687 at 1. Stated otherwise, this “familiar factual question [refers to] whether the plaintiff has indeed suffered harm, or ‘injury-in-fact.’ ” LIBOR VII, 299 F. Supp. 3d at 592, slip op. at *324 (quoting Cordes & Co. Fin. Servs. v. A.G. Edwards & Sons, Inc., 502 F.3d 91, 106 (2d Cir. 2007)). As the Second Circuit has explained, “the injury-in-fact question” considers “whether a plaintiff was harmed.” Cordes & Co., 502 F.3d at 107 n.11 (emphasis in original).

Whether a plaintiff was harmed should not be confused with the separate inquiry into damages, which considers “by how much a plaintiff was harmed.” Id. (emphasis in original); see also In re Visa Check/MasterMoney Antitrust Litig., 192 F.R.D. 68, 82 (E.D.N.Y. 2000) (“The fact of injury, which is required as an element of the plaintiff's claim, should not be confused with the extent of injury (as reflected by the amount of damages).” (emphasis in original)), aff'd, 280 F.3d 124 (2d Cir. 2001).¹²⁶

To establish antitrust injury, a plaintiff “must show that a defendant's anticompetitive act was a ‘material’ and ‘but-for’ cause of plaintiff's injury, although not necessarily the sole cause.” In re Actos End-Payor Antitrust Litig., 848 F.3d 89, 97 (2d Cir. 2017) (quoting Publ'n Paper, 690 F.3d at 65-66); see also P. Areeda & H Hovenkamp, Antitrust Law: An Analysis of Antitrust Principles and Their Application § 392 LexisNexis (database updated Nov. 2024) [hereinafter “Areeda & Hovenkamp”] (“[P]roving the fact of injury, [a] plaintiff must prove with reasonable certainty that the defendant's antitrust violation caused the harm suffered by the plaintiff.”). Of course, a lack of causation in fact “is fatal to the merits of any antitrust claim.” Lotes Co. v. Hon Hai Precision Indus. Co., 753 F.3d 395, 415 n.8 (2d Cir. 2014) (citation and quotation marks omitted).

Plaintiffs here claim that “LIBOR was lower[ ] than it otherwise would have been absent the alleged conspiracy.” Pls. Br. at 58 (citation and quotation marks omitted); see Gelboim, 823 F.3d at 775 (finding that plaintiffs “claimed an actual injury placing [plaintiffs] in a ‘worse position’ as a consequence’ of the Banks’ conduct” (citation omitted)).¹²⁷

Thus, suppression, or injury, in the LIBOR context is only possible if the following proposition is true throughout the relevant period:

But-For LIBOR > Published LIBOR

See, e.g., LIBOR VII, 299 F. Supp. 3d at 595, slip op. at *330-31 (“To show injury, OTC plaintiffs will need to offer classwide evidence that actual published LIBOR was suppressed (i.e., below but-for LIBOR).”).

Defendants contend that plaintiffs have failed to reliably establish “But-For LIBOR,” and therefore, cannot show injury-in-fact. See Defs. Br. at 2. Plaintiffs disagree, of course, and instead argue that “[a]t trial, a jury may find ‘as a matter of just and reasonable inference’ that Defendants’ conduct ‘caused damage to the plaintiffs.’ ” Pls. Br. at 55 (citation omitted).

Before we address the parties’ experts, we first address the Second Circuit's decision in Connolly, which interpreted the LIBOR question and controls how “But-For LIBOR” must be understood.

1. Connolly’s Interpretation of the LIBOR Question

The Second Circuit's decision in United States v. Connolly, 24 F.4th 821 (2d Cir. 2022) followed the conviction of two former Deutsche Bank employees for allegedly manipulating LIBOR to benefit the bank's trading positions. Id. at 823-24. The Second Circuit reversed defendants’ convictions for conspiracy to commit wire fraud and bank fraud “given that the government failed to present evidence to show falsity in the trader-influenced submissions.” Id. at 843.

Two aspects of the Connolly decision are relevant here. First, the Second Circuit interpreted the LIBOR question. Id. at 824-825, 834-835. The LIBOR question asks: “At what rate could you borrow funds, were you to do so by asking for and then accepting inter-bank offers in a reasonable market size just prior to 11 am” London time? LIBOR VII, 299 F. Supp. 3d at 522, slip op. at *158. In relevant part, the Second Circuit's decision states:

• “LIBOR is meant to reflect, on a given day, the rates at which one bank can borrow money from other banks.” Connolly, 24 F.4th at 824.

• The BBA LIBOR Instruction constituted “a series of hypotheticals.” Id. at 835.

• “The BBA LIBOR Instruction did not ask about an actual loan. Rather, it asked a question that was ‘hypothetical.’ ” Id. (citation omitted).

• “The precise hypothetical question to which the LIBOR submitters were responding was at what interest rate ‘could’ [Deutsche Bank (“DB”)] borrow a typical amount of cash if it were to seek interbank offers and were to accept.” Id.

Second, having interpreted the LIBOR question, the Second Circuit then evaluated the government's theory of fraud and the sufficiency of the evidence in support thereof. Id. at 835-44. In relevant part, the Second Circuit's decision states:

• “[W]here, as here a[n alleged] scheme to defraud is premised on an instruction--in this case from the BBA to the submitting bank--the government has the burden to negate any reasonable interpretation of the instruction that would make Deutsche Bank's submission responsive.” Id. at 834 (citation omitted; emphasis omitted).

• “If the rate submitted is one that the bank could request, be offered, and accept, the submission, irrespective of its motivation, would not be false.” Id. at 835.

• “[N]one of the witnesses testified that DB could not have borrowed a typical amount of cash at the rate stated in any of DB's LIBOR submissions ․ [a]nd ․ whether DB ‘could’ do so was the precise question to which the LIBOR submissions were to respond, and was thus the key to whether a given submission was false.” Id. at 835-836.

• “There are two principal respects in which the trial evidence, viewed as a whole, fails to support the foundations of the government's theory of falsity, i.e., that there was (a) one true interest rate, (b) automatically generated by the pricer, (c) which was DB's LIBOR submission as generated except when there was a request from a trader. First, the testimony of the government's witnesses revealed that there were many factors other than the data automatically received by the pricer that informed DB's final LIBOR submission. Second, there were many loans available to DB, with varying interest rates; and as DB could agree to such rates, there was no one true rate that it was required to submit.” Id. at 837.

• “If the submissions did reflect rates at which DB could have borrowed, they complied with the BBA LIBOR Instruction, and the LIBOR submissions were not false.” Id. at 839.

• “Regardless of the motive for severely limiting the size of the trader-influenced move--or for making any move at all--the record remains bereft of any evidence that the trader-influenced rates that [DB's submitters] submitted were rates that DB could not have requested, been offered, and accepted.” Id. at 840.

• “While [an expert] testified to ‘an understanding that [banks] would submit the one best estimate of the true borrowing costs they had’ ․ he did not link that understanding to the BBA LIBOR Instruction, which contains no similar qualification.” Id. at 841 (citation omitted).

There is one overarching takeaway from Connolly for this MDL. LIBOR, as the Second Circuit has interpreted it, is a “hypothetical” rate because the LIBOR question “did not ask about an actual loan.” Id. at 835. Instead, the “precise hypothetical question to which the LIBOR submitters were responding was at what interest rate ‘could’ [the panel bank] borrow a typical amount of cash if it were to seek interbank offers and were to accept.” Id. Thus, “there was no one true rate that [a Panel Bank] was required to submit” in response to the LIBOR question. Id. at 837.¹²⁸

According to plaintiffs, Connolly does not “control” here because Connolly is about the criminal wire fraud statute, which is irrelevant to plaintiffs’ claims. Pls. Br. at 56-58, 68-73. Plaintiffs are wrong: the Second Circuit's interpretation of the LIBOR question is controlling for the construction of but-for LIBOR on all parties to this litigation and their experts. For example, FFFP's suppression expert, “Dr. Snow[,] recognized that the LIBOR Question ․ requires [Panel Banks] to submit their single best estimate of the rates at which they could attract offers consistent with the LIBOR Question,” Pls. Br. at 70, a principle squarely rejected by Connolly: “While [an expert] testified to ‘an understanding that [banks] would submit the one best estimate of the true borrowing costs they had’ ․ he did not link that understanding to the BBA LIBOR Instruction, which contains no similar qualification,” Connolly, 24 F.4th at 841 (citation omitted); see also id. at 837 (“[T]here was no one true rate that [DB] was required to submit.”). In addition, to the extent that plaintiffs’ allegations rested on the assumption that there was one true LIBOR,¹²⁹ the “concept of truthfulness must be understood in light of Connolly,” see In re LIBOR-Based Fin. Instruments Antitrust Litig., No. 11 MDL 2262 (NRB), 2023 WL 2871090, at *8 n.12 (S.D.N.Y. Apr. 10, 2023), ECF No. 3655 at 18 n.12.

2. Daubert Motion Regarding Drs. Bernheim and Snow

Defendants challenge the suppression opinions of plaintiffs’ experts. Given the Second Circuit's holding in Connolly that “[t]he BBA LIBOR Instruction did not ask about an actual loan[;] [r]ather, it asked a question that was ‘hypothetical,’ ” Connolly, 24 F.4th at 835 (citation omitted), a proper but-for LIBOR submission, i.e., what the LIBOR defendants would have submitted had they not engaged in allegedly collusive behavior, constitutes a range as opposed to an exact figure.¹³⁰

Thus, to prove suppression for plaintiffs’ claims, plaintiffs must show that actual published LIBOR was below a range of but-for LIBOR. See Gelboim, 823 F.3d at 775 (finding that plaintiffs “claimed an actual injury placing [them] in a ‘worse position’ as a consequence’ of the Banks’ conduct” (citation omitted)).

Of course, to be admissible, any expert's model that calculates but-for LIBOR must be reliable and account for variables unassociated with plaintiffs’ alleged suppression. See Wills v. Amerada Hess Corp., 379 F.3d 32, 50 (2d Cir. 2004) (“[F]ailure to account for [variables] as possible causes of [injury], strongly indicated that [expert's] conclusions were not grounded in reliable [methods]”). And to prove plaintiffs’ theory of persistent suppression, evidence of suppression must be more than “sporadic.” LIBOR VI, 2016 WL 7378980, at *5 n.8, slip op. at *11 n.8 (finding that “sporadic” and trader-based LIBOR manipulation “has nothing to do with the persistent suppression conspiracy that is at issue”).

Plaintiffs offer two experts and a selection of evidence from the documentary record provided to those experts in their effort to prove suppression. OTC plaintiffs rely on expert testimony from Dr. Douglas Bernheim,¹³¹ while FFFP plaintiffs rely on expert testimony from Dr. Karl Snow. In opposition, defendants offer Dr. Glenn Hubbard's expert testimony.¹³²

While plaintiffs do not seek to exclude Dr. Hubbard's opinions, defendants move to entirely exclude the opinions of Dr. Bernheim and Dr. Snow in a consolidated set of briefs, ECF No. 4151, reserving the right to move to exclude Dr. Bernheim's opinions on downstream issues at that phase of the case, ECF No. 4152 (“Bernheim/Snow – Defs. Br.”) at 1, 4 n.6; see also ECF No. 4438 (“Bernheim/Snow – Defs. Reply Br.”). Plaintiffs oppose this motion. See ECF No. 4311 (“Bernheim/Snow – Pls. Br.”).

We describe the methodologies used by Dr. Bernheim and Dr. Snow before assessing common flaws in both experts’ opinions.

2.1. Dr. Douglas Bernheim's Reports

OTC plaintiffs offer three reports from Dr. Douglas Bernheim: (1) an initial report, dated April 19, 2024 (Decl. of Patrick W. Blakemore, Ex. 1, October 4, 2024, ECF No. 4153); (2) a rebuttal report, dated June 18, 2024 (Decl. of Patrick W. Blakemore, Ex. 2, October 4, 2024, ECF No. 4153); and (3) a reply report, dated August 16, 2024 (Decl. of Patrick W. Blakemore, Ex. 3, October 4, 2024, ECF No. 4153). We refer to these as the Bernheim Initial Report, Bernheim Rebuttal Report, and Bernheim Reply Report, respectively. Defendants do not challenge Dr. Bernheim's qualifications, and we agree that Dr. Bernheim is well qualified to offer these opinions.¹³³

Across these reports, Dr. Bernheim uses two methods to establish suppression of the 1M and 3M tenors between August 2007 and August 2009. The first is a “benchmark” analysis, which relies upon various Eurodollar deposit offer rate series that purportedly mirror LIBOR. See Bernheim/Snow – Pls. Br. at 3-4 (citations omitted). The second is an “imputation” analysis, which relies upon interbank borrowing costs on consummated London interbank deposit transactions for certain days, and constructed estimates of average borrowing costs on other days. See id. at 4-6 (citations omitted).

2.1.1. Benchmark Analysis

Dr. Bernheim prefers his first method, the benchmark analysis, to model but-for LIBOR. For this method, Dr. Bernheim first identifies a benchmark period, i.e. a time when LIBOR was not allegedly suppressed, and a benchmark rate, i.e. a rate similar to LIBOR which was not allegedly suppressed. Bernheim Initial Report ¶¶ 63, 94-96; Bernheim Reply Report ¶ 225.

For a benchmark period, Dr. Bernheim selected two timeframes: January 2005 through July 2007 along with September 2009 through December 2014, both outside of the OTC Class Period, which is between August 2007 and August 2009. Bernheim Initial Report ¶¶ 94-96.

As for possible benchmark rates, he identified “Eurodollar deposit offer rate series from ICAP, Tullet Prebon, and Bloomberg” which he says “closely mirror offer rates that the LIBOR ․ submission process contemplated.” Bernheim/Snow – Pls. Br. at 3 (citations omitted); see also Bernheim Reply Report ¶ 225 (“In my Opening Report, I estimated LIBOR suppression using three Eurodollar offer rate series: the ICAP high, the Tullett Prebon ask, and the Bloomberg ask.”).¹³⁴

Dr. Bernheim then estimated the “spread between the LIBOR fix or individual submissions and the [selected] Eurodollar deposit rate” for the benchmark period and uses this spread, along with a collection of variables, to “predict the most likely outcome but for the alleged conduct” during the OTC Class Period. Bernheim Initial Report ¶¶ 64, 93, 96.¹³⁵

The “simplest approach” under this benchmark method uses the identified offer rates as “but-for LIBOR benchmarks” without adjustment “under the assumption that the economic factors that might cause the two to diverge are relatively unimportant.” Bernheim Initial Report ¶¶ 64, 81. For instance, Dr. Bernheim estimates: “[o]utside of the Class Period, LIBOR was, on average, 3.4 basis points lower than the [ICAP] Eurodollar rate for the 1M tenor․ In contrast, during the Class Period, LIBOR was, on average, 30.2 basis points lower than the [ICAP] Eurodollar rate for the 1M tenor ․ As a first pass, these comparisons suggest LIBOR suppression during the Class Period averaging 26.8 basis points (= 30.2 – 3.4) for the 1M tenor.” Id. ¶ 84.¹³⁶

A “more elaborate approach” under this benchmark method “uses econometric methods to adjust for factors that might cause [London interbank deposit (“LID”)] and Eurodollar deposit offer rates to diverge.” Bernheim Initial Report ¶ 64. Dr. Bernheim acknowledges that these rates may diverge because (a) a broader set of banks receive Eurodollar deposits than participate on a LIBOR panel and (b) depositors for Eurodollar transactions are not limited to banks like LIBOR. Id. ¶ 69. For example, Dr. Bernheim concedes that a “possible innocuous explanation for the” difference of 26.8 basis points calculated above for the 1M tenor “is that there may have been systematic differences in borrowing conditions during the Class Period between the LIBOR Panel Banks and other banks that participated in the Eurodollar market.” Id. ¶ 84.

As part of this “more elaborate approach,” which Dr. Bernheim refers to as his “econometric model” or “baseline model,” Dr. Bernheim contends that he controls for the following: “measures of overall market liquidity (round-trip cost of U.S. corporate bonds, bid-ask spread for E.U. corporate bonds), macroeconomic conditions (ADS business condition index), credit market conditions (OIS), expectations about market volatility (VIX), and the dispersion of credit risk across Prime Banks.” Bernheim Initial Report ¶¶ 64, 100, 105-107; Bernheim Reply Report ¶ 131.

Of the potential benchmark comparators, “Dr. Bernheim concludes that the ICAP Offer Series is the most reliable (but not exclusive) benchmark for but-for LIBOR as, ․ [it] most closely mirrors the defining characteristics of LIBOR.” Bernheim/Snow – Pls. Br. at 4 (citations omitted); see also Bernheim Initial Report ¶¶ 131-38. To be specific, “ICAP publishes two daily Eurodollar series, a ‘high’ and a ‘low’ series, representing their brokers’ best estimates of the range within which most Eurodollar transactions by upper tier banks would take place.” Bernheim Initial Report ¶ 73. Dr. Bernheim “adopt[s] Bloomberg's interpretation that the ICAP Eurodollar ‘high’ series is a representative ask/offer, while the ICAP Eurodollar ‘low’ series is a representative bid.” Id. ¶ 74. Thus, by “ICAP Offer Series,” plaintiffs and Dr. Bernheim are referring to “ICAP High.” Bernheim Reply Report ¶ 225; Bernheim/Snow – Pls. Br. at v (identifying “ICAP Offer Series” as “The high end of the ICAP Eurodollar series, which represents offers or asks”).

Ultimately, when using his unadjusted calculations for his preferred benchmark, ICAP High, Dr. Bernheim calculates suppression to be 27 basis points for the 1M LIBOR fix, and 30 basis points for the 3M LIBOR fix. See Bernheim Initial Report Fig. 42. When applying his baseline econometric model to ICAP High, Dr. Bernheim calculates average suppression estimates of “22 basis points for the 1M LIBOR fix and 18 basis points for the 3M LIBOR fix.” Bernheim Initial Report ¶ 118.¹³⁷ The table below, taken from Dr. Bernheim's initial report, reflects these estimates:

Tabular or graphical material not displayable at this time.

Bernheim Initial Report Fig. 42.

2.1.2. Imputation Analysis

Plaintiffs assert that “Dr. Bernheim's second method for modeling LIBOR suppression uses interbank borrowing costs on consummated London interbank deposit transactions.” Bernheim/Snow – Pls. Br. at 4 (citations omitted).

Dr. Bernheim acknowledges that “using borrowing cost data to identify suppression ․ is challenging.” Bernheim Initial Report ¶ 157. For example, Dr. Bernheim finds that “[o]n most days during the Class Period, most Panel Banks did not have [London interbank deposit] transactions of reasonable size.” Bernheim Rebuttal Report ¶ 30.

Specifically, Dr. Bernheim claims that there are three issues with this borrowing cost data: (1) “Sparsity: On a typical day during the Class Period, a substantial majority of Panel Banks did not engage in [London interbank deposit] borrowing transactions of reasonable size;” (2) “Suitability: Transactional borrowing costs are not offer rates and are on average lower than offer rates;” and (3) “Selectivity: Panel Banks chose to borrow through [London interbank deposit] transactions when the terms on these transactions were most favorable (both in absolute terms and compared to other types of transactions), and when they could do so at rates consistent with their LIBOR submissions.” Bernheim Reply Report ¶ 368 (emphases in original).

Dr. Bernheim thus opines that “[e]ach of these problems” requires “adjustments” to borrowing cost data before “constructing valid LIBOR benchmarks.” Id. ¶ 369. To address the sparsity issue, Dr. Bernheim uses a regression model ¹³⁸ to “construct imputed transactional borrowing costs for Panel Banks during the Class Period,” i.e. “average Panel Bank borrowing rates.” Bernheim Initial Report ¶¶ 181-83, 199, 202-04; Bernheim Rebuttal Report ¶ 98; Hubbard OTC Rebuttal Report ¶ 15. To address the suitability issue, Dr. Bernheim simply “add[ed] half of the spread between the ICAP Eurodollar offer and bid rates to the imputed borrowing costs and the transactional borrowing costs.” Bernheim Rebuttal Report ¶ 98; see also Bernheim Initial Report ¶ 199; Hubbard OTC Rebuttal Report ¶¶ 169-73, Ex. 48 (describing Dr. Bernheim's imputation model). To address the selectivity issue, Dr. Bernheim developed an “adjustment based on the observed differences between the observed transactional borrowing rates of Panel Banks and otherwise similar non-Panel Prime Banks that were unconcerned about LIBOR submissions.” Bernheim Reply Report ¶ 386.

Dr. Bernheim presents four sets of suppression estimates based on different adjustments to the Panel Banks’ actual borrowing costs. Dr. Bernheim's first two estimates rely upon imputed borrowing costs. The first suppression estimate simply uses imputed borrowing costs as but-for LIBOR on all days within the OTC Class Period, and “add[s] half of the spread between ICAP Eurodollar offer and bid rates,” ultimately calculating that average estimated suppression is “25 basis points for the 1M LIBOR fix and 26 basis points for the 3M LIBOR fix.” Bernheim Initial Report ¶¶ 199, 202. Dr. Bernheim's second estimate uses imputed borrowing costs as but-for LIBOR for the days in the OTC Class Period where he contends there is no appropriate London interbank deposit transaction of reasonable size and uses a Panel Bank's actual transactional borrowing costs, “adjusted to account for the suitability and selectivity issues,” as but-for LIBOR on the days in the OTC Class Period where there are these transactions. Id. ¶¶ 200, 201, 204. Of course, because Dr. Bernheim finds that “[o]n most days during the Class Period, most Panel Banks did not have [London interbank deposit] transactions of reasonable size,” Bernheim Rebuttal Report ¶ 30, that means that this estimate uses imputed borrowing costs for most days. In any event, this second estimate calculates an “[a]verage estimated suppression [of] 26 basis points for the 1M LIBOR fix and 27 basis points for the 3M LIBOR fix.” Bernheim Initial Report ¶ 204.

For Dr. Bernheim's third set of estimates, he uses actual borrowing transactions, adjusted for suitability, to calculate but-for LIBOR for days where there is a qualifying transaction. Bernheim Initial Report ¶ 205, Fig. 88. On days when there are no qualifying transactions, Dr. Bernheim uses either imputed borrowing costs (adjusted for suitability) or predictions from his baseline model based on ICAP High. Bernheim Initial Report ¶¶ 201, 205, 206. As a result, Dr. Bernheim estimates suppression of 18-23 basis points for 1M LIBOR and 15-26 basis points for 3M LIBOR. Id. ¶ 206, Fig. 88.

For his fourth estimate, Dr. Bernheim uses borrowing transactions, without any adjustments, to calculate but-for LIBOR for days where there is a qualifying transaction. Id. ¶ 317, Fig. 244. For the days when there are no qualifying transactions, Dr. Bernheim uses either imputed borrowing costs (without any adjustments) or predictions from his baseline model based on ICAP High. Id. As a result, Dr. Bernheim estimates suppression of 10-22 basis points for 1M LIBOR and 8-18 basis points for 3M LIBOR. Id.

Simply stated, Dr. Bernheim does not rely upon actual “interbank borrowing costs on consummated London interbank deposit transactions,” as plaintiffs contend. Bernheim/Snow – Pls. Br. at 4 (citations omitted). In three of four estimates, Dr. Bernheim adjusts the transactional costs when he does use the data. Moreover, in all four estimates, Dr. Bernheim uses some form of constructed “estimate[ ] of LIBOR suppression,” whether that is imputed borrowing costs or predictions from his baseline model based on ICAP High, on days without any borrowing transactions. Bernheim Rebuttal Report ¶ 95. Again, because Dr. Bernheim finds that “[o]n most days during the Class Period, most Panel Banks did not have [London interbank deposit] transactions of reasonable size,” Bernheim Rebuttal Report ¶ 30, Dr. Bernheim's estimates primarily rely on these constructed metrics. Thus, plaintiffs’ summary of their own expert's methodology is highly misleading.

2.2. Dr. Karl Snow's Reports

FFFP Plaintiffs offer three reports from Dr. Karl Snow: (1) an initial report, dated April 19, 2024 (Decl. of Patrick W. Blakemore, Ex. 4, October 4, 2024, ECF No. 4153); (2) a rebuttal report, dated June 18, 2024 (Decl. of Patrick W. Blakemore, Ex. 5, October 4, 2024, ECF No. 4153); and (3) a reply report, dated August 27, 2024 (Decl. of Patrick W. Blakemore, Ex. 6, October 4, 2024, ECF No. 4153). We refer to these as the Snow Initial Report, Snow Rebuttal Report, and Snow Reply Report, respectively. Defendants do not challenge Dr. Snow's qualifications.¹³⁹

2.2.1. Econometric Model

Across these reports, Dr. Snow uses an econometric model to establish suppression of the 1M, 3M, 6M, and 12M tenors between August 2007 and May 2010. Snow Initial Report ¶¶ 14, 16, 17. This econometric model is similar to Dr. Bernheim's benchmark analysis. Supra pp. 188-193.

First, Dr. Snow's model uses data from the benchmark period of January 2005 through July 2007 and December 2011 through December 2014. Snow Initial Report ¶¶ 17, 222, 237. Dr. Snow disregards data from the period of June 2010 through November 2011, which he refers to as a “Potential Contamination Period,” as he states that there is “evidence that there may have been manipulation during this period.” Id. ¶¶ 212-216, 222, 237. He refers to the period between August 2007 and May 2010 as the “Conduct Period.” Id. ¶¶ 14, 237.¹⁴⁰

Second, Dr. Snow uses a regression analysis to “establish the relationship during the benchmark period between the LIBOR—ICAP Spread and other economically relevant variables that were unaffected by potential manipulation.” Snow Initial Report ¶ 222. Like Dr. Bernheim, Dr. Snow relies on Bloomberg's characterization of the “ICAP high series as an offer/ask rate, and the ICAP low series as a bid rate” and states that ICAP High is “a more appropriate benchmark for comparison to LIBOR than the ICAP [L]ow.” Id. ¶ 162.

Dr. Snow identifies several explanatory variables that may be “potentially relevant for modeling the LIBOR-ICAP Spread.” Id. ¶ 233, Fig. 93. Like Dr. Bernheim, Dr. Snow identifies “the relative credit default swap (CDS) spreads of the Panel Banks versus prime banks, the CBOE volatility index (VIX), and the overnight index swap (OIS) rate.” Bernheim/Snow – Defs. Br. at 6 (citations omitted). In addition, Dr. Snow “also includes the Bloomberg Eurodollar London composite bid-ask spread, the spread between the Bloomberg Eurodollar London composite ask and ICAP High, and a linear time trend.” Id. (citations omitted).

As the next step in his opinion, Dr. Snow “use[s] the observed movements of the untainted explanatory variables during the Conduct Period together with the relationship established [between the LIBOR-ICAP Spread and other economically relevant variables] to calculate the expected LIBOR-ICAP Spread during the Conduct Period.” Snow Initial Report ¶ 223. Dr. Snow opines that the “difference in the actual and expected spreads provides a reliable measure of potential manipulation that is isolated from—and therefore unexplained by—changes in otherwise economically relevant factors.” Id.

Ultimately, the figure below represents the average suppression Dr. Snow finds in his preferred model between August 2007 and May 2010:

Tabular or graphical material not displayable at this time.

Snow Initial Report Fig. 1.

2.2.2. Alternative Model

Separately, Dr. Snow opines that “Defendants’ borrowing transactions in the London interbank market during the Conduct Period do not, on their own, provide sufficient information to evaluate the alleged manipulation of LIBOR.” Snow Initial Report ¶ 17. In the alternative, he provides other specifications of his econometric model that “incorporates the relatively limited data available on relevant borrowing transactions by Defendants” and finds that his conclusions are not altered when including such data. Id.

Specifically, for the 1M, 3M, 6M, and 12M tenors, Dr. Snow incorporates transaction data on only 27%, 16%, 7%, and less than 2% of days, respectively, relying on his ICAP High-based “but for” submissions for the rest. Bernheim/Snow – Defs. Br. at 7 (citations omitted). We refer to this model as Dr. Snow's “alternative model.”

2.3. Common Flaws: Dr. Bernheim's Benchmark Analysis and Dr. Snow's Econometric Model

Dr. Bernheim's benchmark analysis and Dr. Snow's econometric model are similar. Both experts estimate but-for LIBOR within the alleged suppression period based on data from outside of the relevant period ¹⁴¹ concerning the spread between a Eurodollar deposit offer rate and LIBOR along with a collection of self-selected explanatory variables. See supra pp. 188-193, 198-200.

As a result, both analyses share many of the same flaws since each analysis relies on inapposite benchmark rates, fails to account for other possible causes of plaintiffs’ alleged injury, and is supported by improper corroboration. We address each of these flaws in turn.

2.3.1. Inapposite Benchmark Rates

2.3.1.1. Background

In an antitrust case, a benchmark or yardstick ¹⁴² model “us[es] data from analogous geographic markets or industries that are not affected by the conspiracy as a benchmark” “to estimate the but-for prices.” ABA Section of Antitrust Law, Proving Antitrust Damages: Legal and Economic Issues § III.8.C.2 (3d ed. 2017).

“The major difficulty encountered in the use of a yardstick is finding a suitable one.” Areeda & Hovenkamp § 392. “The selection of comparators will seldom approach the ‘Utopian ideal’ of identifying the perfect clone.” Celebrity Cruises Inc. v. Essef Corp., 434 F. Supp. 2d 169, 189 (S.D.N.Y. 2006) (citation omitted). After all, “[t]he but-for world is, by definition, hypothetical.” In re Restasis (Cyclosporine Ophthalmic Emulsion) Antitrust Litig., 335 F.R.D. 1, 19 (E.D.N.Y. 2020) (citation omitted).

“However, there comes a point where the proposed benchmark is so different ․ that it cannot serve as the ‘corresponding data for a firm or in a market unaffected by the violation’ necessary to support the yardstick analysis.” Keurig Green Mountain Single-Serve Coffee, 2025 WL 354671, at *11 (citation omitted). In other words, “[a]s one or more of these points of comparison become more distant, the yardstick is accordingly a poorer measure” and the “proposed yardstick” “must be reasonably comparable.” Areeda & Hovenkamp § 392 (citation omitted); see also Boucher v. U.S. Suzuki Motor Corp., 73 F.3d 18, 21 (2d Cir. 1996) (expert testimony should be excluded when it is “based on assumptions that are ‘so unrealistic and contradictory as’ ․ to be in essence an ‘apples and oranges comparison’ ” (citation omitted)).

As the Second Circuit has determined, “LIBOR is meant to reflect, on a given day, the rates at which one bank can borrow money from other banks.” Connolly, 24 F.4th at 824. The Panel Banks submitted an answer to the LIBOR question which was a “hypothetical” and “did not ask about an actual loan.” Id. at 835 (citation and quotation marks omitted). “The precise hypothetical question to which the LIBOR submitters were responding was at what interest rate ‘could’ [a Panel Bank] borrow a typical amount of cash if it were to seek interbank offers and were to accept.” Id.

Here, in Dr. Bernheim's benchmark analysis and Dr. Snow's econometric model, both experts use Eurodollar deposit offer rate series as benchmark rates contending that those rates are similar to LIBOR and thus appropriate for an analysis of LIBOR suppression. Bernheim Initial Report ¶ 63; Bernheim Reply Report ¶ 225; Snow Initial Report ¶¶ 156-68; Snow Rebuttal Report ¶ 83 (“I performed multiple comparisons between LIBOR and ․ the ICAP Eurodollar high rate, the Tullett Prebon Eurodollar ask rate, the Bloomberg Eurodollar London composite ask rate, Bloomberg Eurodollar best ask quotes, and the NYFR.”); Snow Reply Report ¶ 295 (similar). Moreover, out of the Eurodollar offer rate series, both experts prefer ICAP High. See Bernheim Initial Report ¶¶ 131-38; Snow Initial Report ¶¶ 158, 160; Snow Reply Report ¶ 180; ECF No. 4153-7 (Snow June 12, 2024 Dep. Tr.) at 176:5-7, 185:5-8 (Dr. Snow testifying that he uses ICAP High in his baseline model).

We address the experts’ choice of Eurodollar deposit series as benchmark rates first and then discuss the experts’ preference for ICAP High.

2.3.1.2. Choice of Benchmark Rates 143

Drs. Bernheim and Snow use three Eurodollar deposit rate series, ICAP High, Tullet Prebon, and Bloomberg, to model suppression. These rates differ from LIBOR in three critical respects.

First, unlike LIBOR, these Eurodollar deposit rates are indisputably not bank specific, and are merely indicative of where the bulk of market activity would take place. See Bernheim Opening Report ¶¶ 69, 73, 74; Hubbard OTC Rebuttal Report ¶¶ 67-70; Hubbard FFFP Rebuttal Report ¶¶ 69-72; Defs. 56.1 ¶ 636(d) (ICAP High and ICAP Low are “both essentially offered rates, but they were indicative offered rates designed to give a sense of where the bulk of activity would take place.”);¹⁴⁴ see also Defs. Br. at 49-50 (summarizing evidence).

Second, unlike LIBOR, these Eurodollar deposit rates are not restricted to the sixteen Panel Banks, as a broader set of banks receive Eurodollar deposits than the sixteen banks that participate on a LIBOR panel. See Bernheim Initial Report ¶ 69; Snow Initial Report ¶ 262. Thus, Eurodollar deposit offer rates may reflect rates of non-interbank transactions. See Defs. 56.1 ¶ 636(e) (ICAP High and ICAP Low “reflect rates of non-interbank transactions”); Hubbard OTC Rebuttal Report ¶¶ 67-70; Hubbard FFFP Rebuttal Report ¶¶ 69-72; see also Bernheim Opening Report ¶ 69 (similar). Notably, non-Panel Banks were not as creditworthy as Panel Banks,¹⁴⁵ and obviously, “Banks deemed relatively more creditworthy can generally access funds at lower rates than banks deemed to be less creditworthy,” Pls. 56.1 ¶ 6.

Third, unlike LIBOR, these Eurodollar deposit rates reflect initial offers, not offers that Panel Banks would necessarily “accept.” Hubbard OTC Rebuttal Report ¶¶ 56, 58; Hubbard FFFP Rebuttal Report ¶¶ 58, 60. As the Second Circuit has held, “[t]he precise hypothetical question to which the LIBOR submitters were responding was at what interest rate ‘could’ [a specific panel bank] borrow a typical amount of cash if it were to seek interbank offers and were to accept.” Connolly, 24 F.4th at 835. Moreover, as stated in LIBOR VII, the rates a Panel Bank would be offered “could be sufficiently high such that the panel bank would not ‘accept’ the offer, even as a starting point for negotiations.” LIBOR VII, 299 F. Supp. 3d at 522, slip op. at *158.

Without an appropriate mechanism that screens out offers that are “sufficiently high such that the panel bank would not ‘accept’ the offer,” the offer cannot be considered an appropriate but-for response to the LIBOR question. Id. Plaintiffs agree, asserting that they “do not take the extreme position that an unrealistic offer should be considered and the ICAP [High] excludes such outlier offers.” Bernheim/Snow – Pls. Br. at 10 (citations omitted). However, as explained by plaintiffs’ experts, ICAP High does not exclude all pertinent outlier offers. ICAP High “excludes any prime bank that appears to be trading as an outlier for credit reasons on a given day.” Bernheim Initial Report ¶ 73 (citation and quotation marks omitted); see also Snow Reply Report ¶ 140 (same). And of course, because Eurodollar deposit offer rates are not specific to the Panel Banks, that means on a given day, ICAP High may exclude unrepresentative trading from a specific bank, but still include unrepresentatively high offers to non-Panel Banks as well as unacceptably high offers to Panel Banks. See ECF No. 4153-9 (Snow Sept. 5, 2024 Dep. Tr.) at 111:25-112:15 (testifying that ICAP Eurodollar rate series has a wider range of rates due to increased variety of banks); Bernheim Rebuttal Report ¶ 21; Hubbard OTC Rebuttal Report ¶¶ 56, 58; Hubbard FFFP Rebuttal Report ¶¶ 58, 60.

In short, the combination of the three differences listed above means that plaintiffs cannot distinguish between acceptable initial offers to Panel Banks, unrepresentative initial offers to non-Panel Banks, and unacceptable initial offers to Panel Banks. The latter two have no relevance to the LIBOR question. See LIBOR VII, 299 F. Supp. 3d at 522, slip op. at *158.

2.3.1.3. Preferred Benchmark Rate

Out of the Eurodollar offer rate series, both experts prefer ICAP High.¹⁴⁶ Separate and apart from the issues related to using Eurodollar offer rate series as a comparable benchmark, there is yet another reason that the experts’ reliance on, and preference for, ICAP High is misplaced, and that is it depends upon an unreliable characterization of the record.

As we addressed above, “ICAP publishes two daily Eurodollar series, a ‘high’ and a ‘low’ series, representing their brokers’ best estimates of the range within which most Eurodollar transactions by upper tier banks would take place.” Bernheim Initial Report ¶ 73. In response to interrogatories in this action, ICAP defined both ICAP High and ICAP Low as “offer[ ] rates.”¹⁴⁷ Indeed, because Panel Banks were generally more creditworthy, one would expect them to receive offers near the low end of the range, see Hubbard FFFP Reply Report ¶¶ 8, 45; Hubbard FFFP Rebuttal Report ¶ 65; Hubbard OTC Rebuttal Report ¶ 63, and if one recalculates plaintiffs’ experts’ suppression estimates using ICAP Low instead of ICAP High, the suppression estimates approach “zero for the three-month tenor and nearly zero for the one-month tenor.” Hubbard OTC Rebuttal Report ¶ 81 (emphasis omitted).¹⁴⁸

However, in an effort to avoid the impact of ICAP's description of its own rates, Drs. Bernheim and Snow invent grossly misleading reasons to ignore ICAP's Interrogatory Responses and insist that ICAP Low is a “bid” rate. See Bernheim Initial Report ¶ 74; Snow Initial Report ¶ 162. Dr. Bernheim relies on a series of emails with ICAP from outside of the relevant time period and inaccurately asserts that they all support his interpretation of ICAP Low as a bid rate, even when they do not. For example, Dr. Bernheim quotes a 2013 email from ICAP, which largely reiterates that ICAP High is the “ ‘the high end’ of ‘the range within which most deposit activity at upper-tier banks would take place.’ ” Bernheim Reply Report ¶¶ 64-65 (citation omitted). Dr. Bernheim then quotes a 2017 email from ICAP, which states that ICAP chose to populate the “Bid” field with the ICAP Low and the “Ask” field with the ICAP High, yet also states that Bloomberg “can utilize the bid/offer fields on their feed but still intend them to mean something other than bid/offer.” Id. ¶¶ 66-67. This email is, at best, ambiguous. Significantly, while Dr. Snow relies on this same 2017 email, he concedes that “ICAP appears to have provided somewhat conflicting explanations of its rates.” Snow Reply Report ¶¶ 185, 187.

Dr. Bernheim must provide a reliable basis for his assumption that ICAP Low is not an offer rate. Yet, Dr. Bernheim's other citations fare no better at supporting his interpretation. He cites Bloomberg's “highly respected” reputation and states that “[i]t would be unusual for Bloomberg's labeling of its data to be unreliable.” Bernheim Reply Report ¶ 68. This citation does not contradict ICAP's answers which is that Bloomberg's methodology was reliable at a certain point, but eventually became inconsistent with the framework for the market. See ECF No. 4153-29 (ICAP Cap. Mkts. LLC's Second Resps.) at 2. Moreover, Dr. Bernheim cites a statement from one working paper, which does not support his position, and broadly asserts that “my interpretation of the ICAP High as an offer rate is consistent with its use in the peer-reviewed academic literature.” Bernheim Reply Report ¶¶ 70, 70 n.74.¹⁴⁹ Thus, even if we assume that his reading of the working paper is accurate (it is not), this citation merely establishes that ICAP High is an offer rate, not that ICAP Low is a bid rate.

Lastly, Dr. Bernheim contends that he has not ignored statements regarding whether ICAP Low is an offer rate but “evaluated them in the context of the broader record” and asserts that ICAP's explanation does not make sense, given that “it would have always expected substantial activity to take place below the Low.” Bernheim Reply Report ¶ 61. However, his own analysis indicates that ICAP's explanation does make sense, as there was substantial activity below ICAP Low: 74% of Defendants’ London Interbank Deposit transactions in the 1M tenor and 62% in the 3M tenor occurred below the ICAP Low rate during the OTC Class Period. Id. Fig. 4. Dr. Snow similarly confirms “it is true that a meaningful percentage of Defendant [London interbank deposit] transactions took place below the ICAP Eurodollar low rate.” Snow Reply Report ¶ 204; see also ECF No. 4153-9 (Snow Sept. 5, 2024 Dep. Tr.) at 111:25-112:15 (testifying that ICAP High has wider range of rates due to increased variety of banks and “it's not surprising that the panel banks might have rates that are below these general bid indices.”).¹⁵⁰

Indeed, Dr. Snow's explanation is not adequately reliable either. While Dr. Snow demonstrates that LIBOR “tracked more closely” to ICAP High than ICAP Low, he only demonstrates this for the period “outside” of the relevant period. Snow Reply Report ¶ 189, Fig. 40. And while Dr. Snow claims that ICAP Low “behaved akin to a bid rate” not an offer rate, Snow Reply Report ¶ 205, Dr. Snow concedes that the Panel Banks’ weighted average cost of funds and ICAP Low do not exhibit the expected relationship between a bid series and a transaction series during the relevant period, see ECF No. 4153-9 (Snow Sept. 5, 2024 Dep. Tr.) at 112:21-116:13.

Thus, Dr. Bernheim's benchmark analysis and Dr. Snow's econometric model are excludable as they have “failed to demonstrate that the benchmarks” chosen are sufficiently reliable. Fed. Hous. Fin. Agency v. Nomura Holding Am., Inc., No. 11 Civ. 6201 (DLC), 2015 WL 539489, at *11 (S.D.N.Y. Feb. 10, 2015) (citation omitted); see also Boucher, 73 F.3d at 21 (expert testimony should be excluded when it is “based on assumptions that are ‘so unrealistic and contradictory as’ ․ to be in essence an ‘apples and oranges comparison’ ” (citation omitted)).

2.3.2. Failure to Address an Obvious Alternative Explanation: the Financial Crisis

Having found that plaintiffs’ experts’ “comparable” benchmarks are not, in fact, “comparable,” these experts’ failure to account for an alternative explanation for the divergence between LIBOR and the Eurodollar deposit rate series during the relevant period is yet another sufficient reason to reject plaintiffs’ experts’ opinions.

“[A]n antitrust plaintiff must prove that its injury was, in fact, caused by the defendant's violation of the antitrust laws” and “not shown to be attributable to other causes.” U.S. Football League v. Nat'l Football League, 842 F.2d 1335, 1377, 1378 (2d Cir. 1988) (citation omitted; emphasis in original). Thus, “determination of the ‘but for’ prices ․ must take into account nonconspiratorial factors that may have caused prices to be different in the conspiracy period even if there had been no conspiracy.” Areeda & Hovenkamp § 399.

For an expert's testimony on this issue to be reliable, the expert “must demonstrate that [he] has adequately accounted for obvious alternative explanations.” Mirena Ius Levonorgestrel-Related Prods., 341 F. Supp. 3d at 262 (citation and quotation marks omitted). The “failure to test for ․ obvious and significant alternative explanations renders [an expert's] analysis ‘essentially worthless.’ ” In re Wireless Tel. Servs. Antitrust Litig., 385 F. Supp. 2d 403, 428 (S.D.N.Y. 2005) (citation omitted).

Here, there is an obvious, alternative explanation for why LIBOR diverged from the Eurodollar deposit offer rates during the alleged suppression period ¹⁵¹ : namely, the global financial crisis.¹⁵² However, despite acknowledging the impact of the financial crisis on the banking sector, Drs. Bernheim and Snow primarily attribute to suppression, not the financial crisis, the divergence between LIBOR and the Eurodollar deposit offer rates; and they further contend that their flawed models appropriately accounted for the effects of the financial crisis.¹⁵³

Thus, for the experts’ approach of estimating LIBOR suppression to be reliable, one of two assumptions must be reasonable: either, the relationship between LIBOR and Eurodollar deposit offer rates did not diverge significantly within the relevant period for reasons other than the alleged suppression (such as the financial crisis), or their models appropriately accounted for any significant differences between LIBOR and the Eurodollar deposit offer rate that occurred for reasons other than the alleged suppression (such as the financial crisis) within the relevant period.

Accordingly, to evaluate Drs. Bernheim and Snow's approach, we divide the discussion below into three parts: (1) Drs. Bernheim and Snow's acknowledgement that the financial crisis was a possible cause of the divergence between LIBOR and the Eurodollar deposit rate series; (2) an analysis of Drs. Bernheim and Snow's reasons for attributing to suppression, rather than the financial crisis, the divergence in the LIBOR-Eurodollar rate spread; and (3) an examination of whether Drs. Bernheim and Snow's models appropriately accounted for the effects of the financial crisis.

2.3.2.1. Impact of the Financial Crisis

Drs. Bernheim and Snow acknowledge the significant impact of the financial crisis on the banking sector. As Dr. Snow opines, the financial crisis which “began in August 2007 ․ caused major disruptions to financial markets in the US and around the world.” Snow Initial Report ¶ 47. Dr. Bernheim similarly states that the financial crisis “was in many ways an exceptional event.” Bernheim Initial Report ¶ 106. During this crisis, “many large financial institutions,” such as Lehman Brothers, failed, while others “were acquired by other firms,” such as “LIBOR Panel Bank HBOS, which was acquired by fellow panelist Lloyds.” Snow Initial Report ¶ 48. “The pressure that” the global financial crisis and the European sovereign debt crisis “put on the banking sector had implications for the Defendants.” Id. ¶ 53. “As a result of these widespread banking failures, banks began to question the solvency of other banks.” Id. ¶ 49.

Plaintiff's experts explicitly concede that the dislocation caused by the financial crisis may have had an impact on the divergence between LIBOR and the Eurodollar deposit rate series. Dr. Bernheim opines that “systematic differences in borrowing conditions during the Class Period between the LIBOR Panel Banks and other banks that participated in the Eurodollar market” “might cause [LIBOR and Eurodollar deposit offer rate series] to diverge.” Bernheim Initial Report ¶¶ 64, 84; see also id. ¶ 106 (similar). Similarly, Dr. Snow opines that “differences between LIBOR and the ICAP Eurodollar could reflect differences in the risks of these two sets of banks.” Snow Initial Report ¶ 262. Indeed, while the Panel Banks included many of the largest and most financially sound institutions in the world, both before and during the financial crisis, see Defs. 56.1 § IV.B.3, the broader set of banks receiving Eurodollar deposits were not as creditworthy, as plaintiffs’ own experts acknowledge.¹⁵⁴

Plaintiffs’ experts implicitly acknowledge the dislocation caused by the financial crisis when responding to the analysis of defendants’ expert. Dr. Hubbard finds that “all instances of significant fluctuations between LIBOR and the ICAP High rate during the period analyzed take place at times of intense financial market instability.” Hubbard OTC Rebuttal Report ¶ 109. While plaintiffs’ experts quibble with Dr. Hubbard's finding of “all instances,” their reply reports include graphs showing that certain fluctuations between LIBOR and the Eurodollar deposit rates correspond to certain specific events in the financial crisis. For instance, Dr. Snow's reply report has the following graph:

Tabular or graphical material not displayable at this time.

Snow Reply Report Fig. 69. Dr. Bernheim has a similar graph:

Tabular or graphical material not displayable at this time.

Bernheim Reply Report Fig. 14.

2.3.2.2. Attribution of the LIBOR-Eurodollar Rate Divergence to Suppression

Despite acknowledging the impact that the financial crisis had on the banking sector during the relevant period, both Dr. Bernheim and Dr. Snow primarily attribute the divergence between LIBOR and the Eurodollar deposit rate series to suppression, rather than the financial crisis. Drs. Bernheim and Snow wholly fail to explain away the facts supporting Dr. Hubbard's finding that “instances of significant fluctuations between LIBOR and the ICAP High rate during the period analyzed take place at times of intense financial market instability.” Hubbard OTC Rebuttal Report ¶ 109. Below we address plaintiffs’ experts’ three main attempts to dismiss the correlation between times of intense financial market instability and significant fluctuations in the LIBOR-Eurodollar rate spread.

First, plaintiffs’ experts adopt plaintiffs’ litigation posture and argue that it is unnecessary to conduct further analysis of alternative reasons for the LIBOR-Eurodollar rate divergence. For example, Dr. Bernheim claims that Dr. Hubbard's reasoning “ignores the nature of the Plaintiffs’ allegations [that] ․ the Panel Banks engaged in more aggressive and pervasive coordinated LIBOR suppression during the Crisis to avoid undermining the market's perception of their financial stability.” Bernheim Reply Report ¶ 169. Dr. Snow similarly states that he did not have to conduct further analysis of “the relationship between ICAP [H]igh and LIBOR during the conduct period” because “I'm assuming liability.” ECF No. 4153-9 (Snow Sept. 5, 2024 Dep. Tr.) at 159:17-160:12. Beyond the plaintiffs’ complaints, plaintiffs’ experts provide no basis in fact for their claims, and “[a]ssumptions based on conclusory statements of the expert's client, rather than on the expert's independent evaluation are not reasonable.” Supply & Bldg. Co. v. Estee Lauder Int'l, Inc., 95 Civ. 8136 (RCC), 2001 WL 1602976, at *4 (S.D.N.Y. Dec. 14, 2001) (citation omitted).¹⁵⁵ Moreover, as stated above, the “failure to test for ․ obvious and significant alternative explanations renders [an expert's] analysis ‘essentially worthless.’ ” Wireless Tel. Servs., 385 F. Supp. 2d at 428 (citation omitted).

Second, Dr. Snow and Dr. Bernheim both point out that of the four offer rate series Dr. Bernheim examined (LIBOR, ICAP, Tullett Prebon, and Bloomberg), only LIBOR diverged from the other three during the financial crisis. Bernheim Reply Report ¶ 167; Snow Reply Report ¶ 295. The other rates “continued to move together while, consistent with the existence of suppression, LIBOR fell well below the three benchmark series.” Bernheim Reply Report ¶ 167. This is not as persuasive as plaintiffs’ experts contend. Obviously, if there were “differences in offer rates between Panel Banks and other prime banks” during the financial crisis, Bernheim Reply Report ¶ 163, that would impact all three Eurodollar offer rate series differently than it would affect LIBOR, supra pp. 204-208.¹⁵⁶

Third, in an effort to dismiss the correlation between the fluctuations between LIBOR and the Eurodollar deposit rates and specific events in the financial crisis, Dr. Bernheim opines without evidence. He asserts that “the timing of the largest deviations does not provide a valid basis for distinguishing between competing explanations for their existence.” Bernheim Reply Report ¶ 169. Even assuming Dr. Bernheim's critique is correct, it is Dr. Bernheim's responsibility to provide “sufficient facts and data” to distinguish between competing explanations, rather than relying on ipse dixit. Fed. R. Evid. 702; LIBOR VII, 299 F. Supp. 3d at 467-68, slip op. at *28 (“[N]othing in either Daubert or the Federal Rules of Evidence requires a district court to admit opinion evidence that is connected to existing data only by the ipse dixit of the expert.” (quoting Gen. Elec. Co. v. Joiner, 522 U.S. 136, 146 (1997))); In re: NFL “Sunday Ticket” Antitrust Litig., No. ML 15-02668 (PSG) (SKX), 2024 WL 3628118, at *7 (C.D. Cal. Aug. 1, 2024) (excluding expert opinion of but-for world that “was developed based on speculation and ipse dixit opinion”); Cf. In re Live Concert Antitrust Litig., 863 F. Supp. 2d 966, 974–76 (C.D. Cal. 2012) (excluding testimony when the expert “simply assume[d]—without further examination—that the [yardstick damages were] due entirely to Defendants’ allegedly anticompetitive conduct”).

2.3.2.3. Controls for the Effects of the Financial Crisis

Despite acknowledging that the financial crisis may have had an impact on the divergence between LIBOR and the Eurodollar deposit rate series, plaintiffs’ experts fail to adequately control for the possible impact the financial crisis had in their models.

As we have previously discussed, Drs. Bernheim and Snow, in their attempts to estimate suppression, identify several variables that they contend explains any significant differences between LIBOR and the Eurodollar deposit offer rate for reasons other than the alleged suppression within the relevant period. See supra pp. 188-193, 198-200.¹⁵⁷ Drs. Bernheim and Snow then use data from outside the relevant period ¹⁵⁸ to estimate the relationship between these variables and the LIBOR-Eurodollar spread within the relevant period. See Bernheim/Snow – Defs. Reply Br. at 10 (citation omitted); ECF No. 4439-3 (Bernheim Aug. 30, 2024 Dep. Tr.) at 87:19-88:21 (“[M]y forecasting models do not use the data from the class period for the purposes of estimation.”); Bernheim Initial Report ¶¶ 92-94, 97.

There are two flaws with this approach taken by plaintiffs’ experts. First, the benchmark period is too different, for reasons other than the alleged suppression, to yield reliable out-of-sample estimates in the relevant period.¹⁵⁹ In fact, Dr. Hubbard finds that there are statistically significant differences for all of Dr. Bernheim's explanatory variables and for all but one of Dr. Snow's models, which conforms to common sense. Hubbard OTC Rebuttal Report ¶¶ 115, 117; Hubbard FFFP Rebuttal Report ¶¶ 116, 118.¹⁶⁰

A review of one example is appropriate to illustrate this point. As discussed, one of the differences between LIBOR and Eurodollar deposit offer rates is that a broader set of banks receive Eurodollar deposits than the sixteen banks that participate on a LIBOR panel. See Bernheim Initial Report ¶ 69; Snow Initial Report ¶ 262; supra pp. 204-208. Therefore, “differences between LIBOR and the ICAP Eurodollar could reflect differences in the risks of these two sets of banks” rather than suppression. Snow Initial Report ¶ 262; see also Bernheim Initial Report ¶¶ 64, 84, 106 (similar). Thus, Drs. Bernheim and Snow use two variables, the Relative CDS spread and the cross-sectional standard deviation of CDS spreads, to purportedly control for differences in the creditworthiness between the LIBOR Panel Banks and the full set of prime banks.¹⁶¹ However, the latter variable exhibits radically different patterns in the relevant period than in the benchmark period. During Dr. Bernheim's benchmark period, for instance, the “Cross-sectional Standard Deviation of CDS Spreads Across Prime Banks does not exceed 1.626.” Hubbard OTC Rebuttal ¶ 114. During the OTC class period, however, this same figure “reaches 3.414,” indicating a significant difference in the relationship between this variable and the LIBOR-Eurodollar spread during the relevant period as opposed to outside of it. Id.; see also Hubbard FFFP Rebuttal ¶ 115 (similar analysis for Dr. Snow's report).¹⁶²

Simply stated, a reliable model to estimate LIBOR suppression must appropriately account for “differences in offer rates between Panel Banks and other prime banks” during the financial crisis, Bernheim Reply Report ¶ 163; see also id. ¶ 16 (similar), and the models of Drs. Snow and Bernheim fail to do so. See LIBOR VII, 299 F. Supp. 3d at 552-54, slip op. at *230-32 (excluding expert who offered model which did “not sufficiently control for differences in credit risk across panel banks” during the relevant period); Boucher, 73 F.3d at 21 (expert testimony should be excluded when it is “based on assumptions that are ‘so unrealistic and contradictory as’ ․ to be in essence an ‘apples and oranges comparison’ ” (citation omitted)).

Second, neither Dr. Bernheim nor Dr. Snow tested whether these explanatory variables could reliably explain the differences between Eurodollar deposit rates and LIBOR during the relevant period. See Bernheim/Snow – Defs. Reply Br. at 8 n.8 (citing ECF No. 4153-9 (Snow Sept. 5, 2024 Dep. Tr.) at 159:17-24 (“Q. Have you tested to see whether there was a structural break between, for example, ICAP High and LIBOR as a result of the financial crisis? A. No ․”)); ECF No. 4439-2 (Bernheim Jul. 19, 2024 Dep. Tr.) at 219:3-16) (“Q. And so have you -- have you performed any test specifically to determine whether there was any structural breaks in this period? A. I think you asked that question before. I haven't and I don't deem that to be the appropriate procedure ․”)); ECF No. 4439-1 at 148:3-149:9 (Snow Sept. 5, 2024 Dep. Tr.) at 149:10-17 (“I don't rule out th[e] possibility” of a structural break).¹⁶³ As stated above, the “failure to test for ․ obvious and significant alternative explanations renders [an expert's] analysis ‘essentially worthless.’ ” Wireless Tel. Servs., 385 F. Supp. 2d at 428 (citation omitted).¹⁶⁴

2.3.2.4. Conclusion

In sum, for the experts’ preferred approach of estimating LIBOR suppression to be reliable, one of two assumptions must be reasonable: either, the relationship between LIBOR and Eurodollar deposit offer rates did not diverge significantly within the relevant period for reasons other than the alleged suppression, or their models appropriately accounted for any significant differences between LIBOR and the Eurodollar deposit offer rate that occurred for reasons other than the alleged suppression within the relevant period. Plaintiffs’ experts have not established that either assumption is reasonable.

To be clear, we are not deciding one way or the other that the financial crisis caused the divergence in the relationship between Eurodollar deposit offer rates and LIBOR. Rather, given that the financial crisis poses an “obvious” alternative explanation, we hold that the failure of plaintiffs’ experts to appropriately account for this possible cause of injury renders their analysis, on an issue which plaintiffs bear the burden of proving, unreliable and inadmissible. See Wills, 379 F.3d at 50 (“[F]ailure to account for [variables] as possible causes of [injury], strongly indicated that [expert's] conclusions were not grounded in reliable [methods]”); Reed Constr. Data Inc. v. McGraw-Hill Cos., 49 F. Supp. 3d 385, 400 (S.D.N.Y. 2014) (“[T]o be admissible,” an expert's methodology “must control for” the “major factors ․”), aff'd, 638 F. App'x 43 (2d Cir. 2016); Wireless Tel. Servs., 385 F. Supp. 2d at 427 (“Where an expert conducts a regression analysis and fails to incorporate major independent variables, such analysis may be excluded as irrelevant.”).

2.3.3. Improper Corroboration

Both Dr. Bernheim and Dr. Snow attempt to corroborate their estimates of suppression “through citation to contemporaneous admissions from the Panel Banks’ LIBOR Submitters [allegedly stating] that LIBOR was suppressed.” Bernheim/Snow – Pls. Br. at 6 n.13 (citing Bernheim Initial Report § VI.G.; Bernheim Reply Report § IV.A; Bernheim Initial Report ¶ 152); see also id. at 8 n.18 (citing Snow Reply Report § VI.F.2).

However, neither Dr. Bernheim nor Dr. Snow can “offer opinions that purport to interpret documents like trader communications, government reports, and orders issued by regulatory authorities” because these “opinions are not the product of [their] expertise, as the documents [they] purport[ ] to interpret are equally understandable by the trier of fact at this stage, and these opinions are therefore inadmissible.” LIBOR VII, 299 F. Supp. 3d at 584, slip op. at *304 (citing Jiau, 734 F.3d at 154).

Moreover, the cited communications are too infrequent to support the experts’ claims of persistent suppression throughout the relevant period. There are 527 days in the Class Period and 715 days in the FFFP relevant period. Defs. 56.1 ¶ 13; Hubbard OTC Reply Report ¶ 39, Ex. C-1A. Yet the communications cited by OTC Plaintiffs’ expert, Dr. Bernheim, cover less than 10% of days during the Class Period, Hubbard OTC Reply ¶ 39, and the documents cited by FFFP Plaintiffs’ expert, Dr. Snow, yields a similarly limited percentage of days within the relevant period, see Defs. Br. at 52-53 (“Dr. Snow's analysis of documents that purportedly demonstrate LIBOR suppression covers an even smaller fraction of the Relevant Period.” (citations omitted)); see also Defs. Reply Br. at 21-22 (estimating that the total communications cited by plaintiffs and their experts cover “less than 14% of the Relevant Period and less than 17% of the Class Period” (citations omitted)).¹⁶⁵

The cited set of infrequent communications is simply too sporadic to prove plaintiffs’ theory of persistent suppression. See LIBOR VI, 2016 WL 7378980, at *5 n.8, slip op. at *11 & n.8 (finding that “sporadic” and trader-based LIBOR manipulation “has nothing to do with the persistent suppression conspiracy that is at issue”).

2.4. Common Flaws: Dr. Bernheim's Imputation Analysis and Dr. Snow's Alternative Model

Having excluded the benchmark analysis and econometric model submitted by plaintiffs’ experts, we now address two other models, the imputation analysis and alternative model, proffered by Drs. Bernheim and Snow. See supra 193-198, 200-201. The imputation analysis and alternative model are similar in that they both purport to use borrowing costs to model LIBOR suppression. See Bernheim/Snow – Pls. Br. at 4, 29 (citations omitted).

However, both analyses ignore large swathes of relevant data. Additionally, Dr. Bernheim's imputation analysis is separately subject to exclusion because it depends upon an unreliable methodology. We address each flaw in turn.

2.4.1. Ignores Relevant Data

As stated earlier, the Second Circuit held in Connolly that the BBA LIBOR Instruction constituted “a series of hypotheticals” and “LIBOR is meant to reflect, on a given day, the rates at which one bank can borrow money from other banks.” 24 F.4th at 824, 835. The Circuit further observed that a Panel Bank's “final LIBOR submission” was “informed” by “many factors” and that “there was no one true rate that [a Panel Bank] was required to submit” because “there were many loans available to [a Panel Bank], with varying interest rates; and [a Panel Bank] could agree to such rates.” Id. at 837.

Here, plaintiffs’ experts acknowledge that they disregard in their imputation analysis and alternative model: (1) all borrowing transactions that occurred outside of London; and (2) all transactions below a certain threshold ($10 million). See Bernheim Initial Report ¶ 162; Snow Initial Report ¶¶ 86, 92. As a result of these exclusions, defendants assert that plaintiffs’ experts “ignore at least 95% of [d]efendants’ borrowing transactions.” Defs. Reply Br. at 17 (citing Snow Initial Report ¶ 100, Fig. 18).¹⁶⁶

Neither exclusion is appropriate. First, in contravention of Connolly, Drs. Bernheim and Snow's analyses improperly ignores the “many loans available to [a Panel Bank], with varying interest rates” that a Panel Bank could permissibly consider when forming its LIBOR submission. 24 F.4th at 837. Similarly, in LIBOR VII (a pre-Connolly decision), this Court observed that “a panel bank's LIBOR submission should be based on its perception of the rates it would be offered” as well as “the ‘totality of the information’ available to it.” LIBOR VII, 299 F. Supp. 3d at 528, slip op. at *172. To this end, “a panel bank's submission should be informed by actual transaction rates, which unquestionably form part of the ‘totality of information available to [the] bank.’ ” Id. at 523, slip op. at *161 (citation omitted; alterations changed).

Second, the experts’ exclusions not only ignore the binding teachings of Connolly but are separately unsupportable. The first exclusion by both experts of non-London borrowing costs is hypocritical. The Eurodollar deposit rates, which these same experts assert are appropriate comparators to LIBOR, see supra pp. 204-208, “were not limited to the London interbank market.” Hubbard OTC Rebuttal ¶ 66; see also Bernheim Reply Report ¶ 167 n.157 (same). Besides pointing out that the “L” in “LIBOR” stands for “London,” plaintiffs’ experts provide no reason for why a proper but-for submission of LIBOR should be overly restricted to London-based interbank trading.

The second exclusion, both experts’ disregard of all transactions below a certain threshold ($10 million), is also unsupported. The Circuit in Connolly observed that the term “ ‘reasonable market size’ was one that the BBA had not defined.” 24 F.4th at 835. Nonetheless, both experts proceed to define this term, with Dr. Bernheim stating that $10 million is a “conservative estimate of ‘reasonable market size,’ ” Bernheim Initial Report ¶ 162 n.162, and Dr. Snow opining that “reasonable market size ․ was likely between $50 million and $200 million for much of the period at issue,” Snow Initial Report ¶ 90. These inconsistent estimates demonstrate the unreliability of the data used when defining a term that the Circuit in Connolly held was left undefined.

Thus, because experts’ exclusions ignore the binding teachings of Connolly and are unsupportable, the experts’ analyses are subject to exclusion as they are “based on data untethered to the facts of this case.” Nypl v. JP Morgan Chase & Co., No. 15 Civ. 9300 (LGS), 2022 WL 819771, at *3 (S.D.N.Y. Mar. 18, 2022) (citation omitted).

2.4.2. Unreliable Methodology

Dr. Bernheim's imputation analysis is separately subject to exclusion because it depends upon an unreliable methodology.

In his imputation model, Dr. Bernheim opines that three issues with the borrowing cost data, namely, “Sparsity,” “Suitability,” and “Selectivity,” require “adjustments” to this data before “constructing valid LIBOR benchmarks.” Bernheim Reply Report ¶¶ 368, 369; supra pp. 193-198. We address Dr. Bernheim's methodology regarding sparsity and selectivity in turn.¹⁶⁷

The first issue that Dr. Bernheim identifies with the borrowing cost data is “Sparsity,” i.e., “[o]n a typical day during the Class Period, a substantial majority of Panel Banks did not engage in [London interbank deposit] borrowing transactions of reasonable size.” Bernheim Reply Report ¶ 368. However, this issue is entirely self-invented, as Dr. Bernheim improperly excludes much of the borrowing cost transactions, supra pp. 231-234. Moreover, Dr. Bernheim's adjustments to the borrowing cost data based on this sparsity issue are equally unreliable. To address the sparsity issue, Dr. Bernheim uses some form of constructed “estimate[ ] of LIBOR suppression,” whether that is imputed borrowing costs ¹⁶⁸ or predictions from his baseline model based on ICAP High, on days without any borrowing transactions. Bernheim Rebuttal Report ¶¶ 95, 98; see also Bernheim Initial Report ¶¶ 181-183, 199, 202-204; Hubbard OTC Rebuttal Report ¶ 15. This constructed “estimate[ ] of LIBOR suppression,” which is based on ICAP High, is unreliable for the same reasons why ICAP High was an inappropriate comparator in Dr. Bernheim's benchmark analysis. See supra pp. 202-229.¹⁶⁹

The second purported issue that Dr. Bernheim identifies with the borrowing cost data is “Selectivity,” i.e., that “Panel Banks chose to borrow through [London interbank deposit] transactions when the terms on these transactions were most favorable (both in absolute terms and compared to other types of transactions), and when they could do so at rates consistent with their LIBOR submissions.” Bernheim Reply Report ¶¶ 368, 369. To address this issue, Dr. Bernheim developed an “adjustment based on the observed differences between the observed transactional borrowing rates of Panel Banks and otherwise similar non-Panel Prime Banks that were unconcerned about LIBOR submissions.” Id. ¶ 386.

However, Dr. Bernheim did not conduct an analysis of whether this selectivity issue even exists, and if so, how frequently it appeared. See Defs. Reply Br. at 18 (citations omitted). Moreover, any adjustment based on this purported issue is senseless. As defendants’ briefing points out, “no matter how many times a Defendant allegedly chose not to borrow funds in London, such decisions would have no relevance to that Defendant's cost of funds on the numerous days when it did borrow funds.” Id. at 18-19 (emphasis in original). Indeed, Dr. Bernheim concedes that “Panel Banks sometimes did engage in [London interbank deposit] borrowing at rates above LIBOR.” Bernheim Reply Report ¶ 395.

Each of these flaws in Dr. Bernheim's methodology constitute independent grounds for exclusion of his imputation analysis. See Amorgianos v. Nat'l R.R. Passenger Corp., 303 F.3d 256, 267 (2d Cir. 2002) (“To warrant admissibility, ․ it is critical that an expert's analysis be reliable at every step.”).

2.5. Common Flaws: Panel Banks’ Borrowing Costs Provide Compelling Evidence that LIBOR Was Not Suppressed

Neither of plaintiffs’ experts uses unadjusted borrowing costs to model but-for LIBOR, supra pp. 187-201, while Dr. Hubbard's “calculations establish that Defendants’ daily LIBOR submissions overwhelmingly were within or above the range of each Panel Bank's corresponding borrowing rates on submission days with associated borrowing transactions,” Defs. Br. at 20 (citation omitted).

The failure of plaintiffs’ experts to substantively address this compelling evidence confirms that exclusion of the experts’ models is warranted. See, e.g., Wireless Tel. Servs., 385 F. Supp. 2d at 428 (“[F]ailure to test for these obvious and significant alternative explanations renders [an expert's] analysis ‘essentially worthless.’ ” (citation omitted)).¹⁷⁰

3. Purported Borrowing Costs Requirement

Before addressing the ramifications of the exclusion of plaintiffs’ experts, we briefly discuss defendants’ argument that a “submission below a Panel Bank's [actual] borrowing costs is necessary but not sufficient to prove suppression.” Defs. Br. at 31 n.13. In support of this argument, defendants cite this Court's opinion in LIBOR VII, the Second Circuit's decision in Connolly, and various market actors’ understanding of the LIBOR question, as well as plaintiffs’ own allegations. See Bernheim/Snow – Defs. Br. at 9-14; Bernheim/Snow – Defs. Reply Br. at 1-7; Defs. Br. at 8-9, 31-33 (similar); Defs. Reply Br. at 7, 8 (similar); ECF No. 4336 (Class Decertification) at 11-12 (similar).

Defendants’ argument is rejected. This Court in LIBOR VII (a pre-Connolly opinion) held that “actual transactions are relevant in analyzing whether a panel bank made suppressed LIBOR submissions,” LIBOR VII, 299 F. Supp. 3d at 562, slip op. at *252 (emphasis added), not that actual transactions are necessary in analyzing whether a panel bank made suppressed LIBOR submissions. In fact, we cited a document explicitly disclaiming this view: “The rates are not necessarily based on actual transaction[s], indeed it would not be possible to create the suite of [LIBOR] rates if this was a requirement.” Id. at 523, slip op. at *159 (citation and quotation marks omitted; emphasis and alterations in LIBOR VII).

Likewise, the Second Circuit's decision in Connolly does not support defendants’ argument. The Second Circuit specifies that the “BBA LIBOR Instruction did not ask about an actual loan,” id. at 835, which conflicts with defendants’ position that the cost of funds is an essential component of determining but-for LIBOR, see, e.g., Bernheim/Snow – Defs. Br. at 9.

Defendants’ additional evidence, some of which we already reviewed at the class certification stage in LIBOR VII, does not change our conclusion. For example, although defendants quote BBA guidance accurately, defendants place their emphasis on the wrong phrase of the BBA publications. Defendants argue that “the BBA's interpretation of the LIBOR question ․ confirms that the ‘rate at which each bank submits must be formed from that bank's perception of its cost of funds in the interbank market.’ ” Bernheim/Snow – Defs. Br. at 11 (citations omitted; emphasis in defendants’ brief). Defendants’ emphasis is misplaced. It's more accurate to argue that “the BBA's interpretation of the LIBOR question ․ confirms that the ‘rate at which each bank submits must be formed from that bank's perception of its cost of funds in the interbank market.’ ” Id. (citations omitted; emphasis altered).

Similarly, defendants cite an exhibit which states that “LIBOR submissions should reflect ‘a fair and accurate reflection of that bank's opinion of its cost of funds.’ ” Bernheim/Snow – Defs. Br. at 11 n.36 (quoting ECF No. 4153-11 (Terms of Reference for LIBOR Contributor Banks, British Bankers’ Association, July 2009, RABO_METZLER_0078539 at -539)) (emphasis added). A bank's “opinion” or “perception” of its cost of funds is not the same as measuring the bank's “cost of funds.” Bernheim/Snow – Defs. Br. at 11 (citations and quotation marks omitted).¹⁷¹

Lastly, our review of plaintiffs’ complaints confirm that actual transactions are relevant, but not necessary, in analyzing whether a panel bank made suppressed LIBOR submissions. See, e.g., Defs. Br. at 8-9 (collecting citations to plaintiffs’ various complaints); Pls. Br. at 124-25 (summarizing claims).

4. Conclusion

As we addressed previously, to establish antitrust injury, a plaintiff “must show that a defendant's anticompetitive act was a ‘material’ and ‘but-for’ cause of plaintiff's injury, although not necessarily the sole cause.” Actos End-Payor, 848 F.3d at 97 (quoting Publ'n Paper, 690 F.3d at 66); see also Areeda & Hovenkamp § 392 (“[P]roving the fact of injury, [a] plaintiff must prove with reasonable certainty that the defendant's antitrust violation caused the harm suffered by the plaintiff.”). Of course, a lack of causation in fact “is fatal to the merits of any antitrust claim.” Lotes Co., 753 F.3d at 415 n.8 (citation and quotation marks omitted). Further, in determining whether defendants have met their burden, the Court “may only consider admissible evidence” in ruling on a motion for summary judgment. Humphrey v. Diamant Boart, Inc., 556 F. Supp. 2d 167, 173-74 (E.D.N.Y. 2008) (citations omitted); see also Lavoho, LLC v. Apple, Inc., 232 F. Supp. 3d 513, 525-26, 528 (S.D.N.Y. 2016) (granting defendant's summary judgment motion as plaintiff's antitrust injury expert offered a “theoretical argument” “untethered to any evidentiary record”), aff'd sub nom. Diesel eBooks, LLC v. Simon & Schuster, Inc., 869 F.3d 55 (2d Cir. 2017); In re Acetaminophen – ASD-ADHD Prods. Liab. Litig., No. 22 MC 3043 (DLC), 2024 WL 3874183, at *4 (S.D.N.Y. Aug. 20, 2024) (granting summary judgment on causation following the exclusion of plaintiffs’ experts).

Here, plaintiffs claim that “LIBOR was lower[ ] than it otherwise would have been absent the alleged conspiracy,” Pls. Br. at 58 (citation and quotation marks omitted); see Gelboim, 823 F.3d at 775 (same),¹⁷² and endeavored to establish a triable issue of fact that actual published LIBOR was below but-for LIBOR throughout the relevant period. Both OTC and FFFP plaintiffs rely upon expert analysis and documentary evidence to prove suppression. OTC plaintiffs cite Dr. Bernheim's suppression estimates for the 1M and 3M tenors, while FFFP plaintiffs rely upon Dr. Snow's analysis to estimate suppression for the 1M, 3M, 6M, and 12M tenors. Otherwise, to establish suppression, both sets of plaintiffs cite documentary and other record evidence, such as Panel Bank employees’ gripes that other Panel Banks’ LIBOR submissions were too low. Pls. Br. at 53-54, 71-73.

Plaintiffs’ proffered evidence does not come close to demonstrating that a reasonable jury could return a verdict in their favor on the issue of suppression. Both Drs. Snow and Bernheim's analyses are inadmissible because the data their models depended on was unreliable and incomplete; their methodologies improperly failed to account for other non-conspiratorial causes for the alleged suppression, namely, the financial crisis; and their opinions exceeded the bounds of what could be concluded from a reliable application of their methodologies. See supra pp. 201-238.

As for plaintiffs’ citation of documentary and record evidence, the communications they cite are too infrequent to support their claims of persistent suppression throughout the relevant period. Even assuming that plaintiffs’ self-serving spin of the documentary record is accurate (and there is no reason to do so),¹⁷³ these communications constitute “less than 14% of the Relevant Period and less than 17% of the Class Period,” Defs. Reply Br. at 21-22, which is simply too sporadic to prove plaintiffs’ theory of persistent suppression, supra pp. 229-231; LIBOR VI, 2016 WL 7378980, at *5 n.8, slip op. at *11 & n.8 (finding that “sporadic” and trader-based LIBOR manipulation “has nothing to do with the persistent suppression conspiracy that is at issue”). At summary judgment, plaintiffs can no longer rely upon the “mere existence of a scintilla of evidence in support of [their] position,” and must instead demonstrate “significant probative evidence,” which a reasonable factfinder could rely on to decide in their favor. Liberty Lobby, 477 U.S. at 249, 252 (quotation marks and citation omitted). Plaintiffs have failed to clear this basic evidentiary bar.

Accordingly, without a reliable estimate of but-for LIBOR throughout the relevant period, plaintiffs cannot establish that they suffered an injury as a result of defendants’ conduct. See LIBOR VII, 299 F. Supp. 3d at 595, slip op. at *330-331 (“To show injury, OTC plaintiffs will need to offer classwide evidence that actual published LIBOR was suppressed (i.e., below but-for LIBOR).”); Gelboim, 823 F.3d at 775 (plaintiffs must prove “actual injury” demonstrating that plaintiffs are “in a worse position as a consequence of the Banks’ conduct” (citations and quotation marks omitted)); LIBOR VI, 2016 WL 7378980, at *18, slip op. at *48 (“[E]stimation of but-for LIBOR is the job of the parties’ competing experts.”).

V. SUMMARY JUDGMENT: SHERMAN ACT CLAIMS

Having found that plaintiffs failed to provide evidence sufficient to raise a triable issue of fact as to conspiracy and suppression, see supra pp. 29-244, we now address defendants’ motion for summary judgment on plaintiffs’ Sherman Act claims.¹⁷⁴

As we set out above, the first element a plaintiff asserting a Section 1 violation must establish is “ ‘a combination or some form of concerted action between at least two legally distinct economic entities.’ ” Mylan N.V., 666 F. Supp. 3d at 300 (quoting Capital Imaging Assocs., P.C. v. Mohawk Valley Med Assocs., Inc., 996 F.2d 537, 542 (2d Cir. 1993)). For the reasons stated above, plaintiffs have failed to support a genuine issue of material fact as to a “horizontal price-fixing conspiracy” whereby the Panel Banks “colluded to depress LIBOR.” Gelboim, 823 F.3d at 771; see supra pp. 29-177.

The second element a plaintiff asserting a Section 1 violation must establish is antitrust injury. See Actos End-Payor, 848 F.3d at 97 (citation omitted).¹⁷⁵ Again, having failed to raise a triable issue of fact, plaintiffs cannot establish the second element of their claim of “actual injury” whereby the Panel Banks “plac[ed] [them] in a worse position as a consequence of the Banks’ [alleged price-fixing] conduct.” Gelboim, 823 F.3d at 775 (citations and quotation marks omitted); see supra pp. 177-244. Accordingly, defendants are granted summary judgment and plaintiffs’ claims under Section 1 of the Sherman Act are dismissed.

VI. SUMMARY JUDGMENT: COMMON LAW CLAIMS

In addition to their Sherman Act claims, plaintiffs assert several common law claims predicated on the same conduct that underlies their federal claims, i.e., conspiracy and suppression. We address these claims below.¹⁷⁶

1. Fraud-Based Claims

First, we address plaintiffs’ fraud-based claims. As an initial matter, we note that such claims generally require plaintiffs to establish the existence of a misrepresentation, which necessarily requires a showing of falsity. See, e.g., Ramiro Aviles v. S&P Global, Inc., 380 F. Supp. 3d 221, 281 (S.D.N.Y. 2019) (“Under New York law, a misrepresentation can typically form the basis for a fraud claim only if it was ‘known to be false’ by the speaker.” (citation omitted)).

As discussed previously, the Second Circuit held when dismissing fraud claims in Connolly that “there was no one true rate that [a Panel Bank] was required to submit” in response to the LIBOR question. 24 F.4th at 837. Accordingly, in order to prove falsity, plaintiffs must demonstrate that the LIBOR rates submitted by defendants were rates that defendants “could not have requested, been offered, and accepted.” Id. at 840.

While plaintiffs dispute that Connolly applies to their claims, Pls. Br. at 68-73, to the extent that plaintiffs assert any claim or theory that rests on whether the Panel Banks “set their LIBOR contributions at rates that were not truthful answers to the LIBOR question,” the “concept of truthfulness must be understood in light of Connolly,” see In re LIBOR-Based Fin. Instruments Antitrust Litig., No. 11 MDL 2262 (NRB), 2023 WL 2871090, at *8 n.12 (S.D.N.Y. Apr. 10, 2023), ECF No. 3655 at 18 n.12 (citation and quotation marks omitted; emphasis in original).¹⁷⁷ And, contrary to plaintiffs’ assertion that “the estimable record in this case shows that [d]efendants gave false answers to the LIBOR Question and knew that LIBOR was suppressed,” Pls. Br. at 3, plaintiffs have not put forward sufficient evidence that would allow a reasonable factfinder to infer that the LIBOR rates submitted by defendants were rates that defendants “could not have requested, been offered, and accepted,” Connolly, 24 F.4th at 840; supra pp. 231-238, 241-244 (excluding unreliable models from plaintiffs’ experts, Drs. Bernheim and Snow, which purported to exclude multiple categories of Panel Bank borrowing transactions).¹⁷⁸

With these principles in mind, we address plaintiffs’ claims below.

1.1. Common Law Fraud

First, all remaining DAPs assert common law fraud claims against defendants. See ECF No. 4583-1; Fannie Mae Compl. ¶¶ 130-41; FDIC Compl. ¶¶ 286-307; Freddie Mac Compl. ¶¶ 404-25; Principal Financial Compl. ¶¶ 302-13; Principal Funds Compl. ¶¶ 299-310.¹⁷⁹

A fraud claim generally requires a plaintiff to prove: (1) a misrepresentation or omission by the defendant; (2) that the defendant “intend[ed] or ha[d] reason to expect” that a “person or [a] class of persons ․ [would] act or ․ refrain from action in reliance upon the misrepresentation;” (3) that “the plaintiff actually relied on the misrepresentation;” (4) that “the plaintiff's reliance was reasonable or justifiable;” and (5) that the “plaintiff's reliance ․ proximately caused the plaintiff's injury.” LIBOR IV, 2015 WL 6243526, at *51, slip op. at *127-28 (citation and quotation marks omitted).

The parties do not dispute that two categories of fraud claims remain live at this stage of the litigation, namely, those relating to “LIBOR quality fraud” and “false data fraud.” LIBOR IV, 2015 WL 6243526, at *50, slip op. at *125. Broadly stated, the former encompasses claims that plaintiffs’ “counterparties omitted (either fraudulently or negligently) to disclose information about LIBOR manipulation when plaintiffs and the counterparties entered into swap contracts, even though the counterparties possessed superior knowledge of the manipulation.” Id. at *71, slip op. at *176-77. The latter involves claims that “LIBOR panel banks committed fraud by submitt[ing] false quotes to the BBA.” Id.

Defendants assert that they are entitled to summary judgment dismissing both categories of fraud claims because plaintiffs have failed to raise a genuine issue of material fact with respect to whether defendants were submitting false LIBOR rates. See Defs. Br. at 104-05. We agree.

The first element of any fraud claim is a misrepresentation or omission by the defendant. See LIBOR IV, 2015 WL 6243526, at *51, slip op. at *127-28. As noted above, the Second Circuit determined in Connolly that the LIBOR question is “hypothetical,” meaning that a bank's LIBOR submission “would not be false” if the “rate submitted [wa]s one that the bank could request, be offered, and accept[.]” 24 F.4th at 835. In light of our determination that plaintiffs have not presented sufficient evidence that would allow a reasonable factfinder to infer that the LIBOR rates submitted by defendants were rates that defendants “could not have requested, been offered, and accepted,” Connolly, 24 F.4th at 840; supra pp. 231-238, 241-244 (excluding unreliable models from plaintiffs’ experts, Drs. Bernheim and Snow, which purported to exclude multiple categories of Panel Bank borrowing transactions), plaintiffs have failed to provide evidence demonstrating the existence of any misrepresentation or omission by defendants relating to LIBOR that is sufficient to support a claim for fraud.

A similar analysis applies with respect to plaintiffs’ false data fraud claims. Again, in light of the Circuit's decision in Connolly, plaintiffs have failed to identify sufficient evidence to demonstrate that the Panel Banks “set their LIBOR contributions at rates that were not truthful answers to the LIBOR question.” see In re LIBOR-Based Fin. Instruments Antitrust Litig., No. 11 MDL 2262 (NRB), 2023 WL 2871090, at *8 n.12 (S.D.N.Y. Apr. 10, 2023), ECF No. 3655 at 18 n.12 (citation and quotation marks omitted; emphasis in original). As such, having failed to provide sufficient evidence to establish the existence of a misrepresentation or omission by defendants in their LIBOR submissions, plaintiffs’ false data fraud claims must also fail.

The only argument raised by plaintiffs in their brief is that defendants “focus[ ] exclusively on the first element” of the fraud standard and “do not move to dismiss on the other elements of [the] fraud claims.” Pls. Br. at 124. Plaintiffs’ argument is patently frivolous. An omission or “[m]isrepresentation is an essential element of a claim for common law fraud.” Heil v. Lebow, No. 91 Civ. 8656 (JFK), 1994 WL 637686, at *8 (S.D.N.Y. Nov. 14, 1994) (citations omitted) (granting summary judgment on fraud claim where plaintiff “failed to produce any evidence of a misrepresentation”); see also Shaoxing Daqin Imp. & Exp. Co., Ltd. v. Notations, Inc., No. 19 Civ. 2732 (JSR), 2019 WL 6498397, at *5-6 (S.D.N.Y. Dec. 3, 2019) (granting summary judgment on fraud claim where “plaintiff has not provided clear or convincing evidence of any intentional misrepresentation on the part of [defendants]”).

Accordingly, as both remaining categories of fraud require evidence sufficient to show a misrepresentation or omission by defendants, which plaintiffs have failed to identify, the Court grants summary judgment in defendants’ favor and dismisses plaintiffs’ remaining common law fraud claims.

1.2. Fraudulent Inducement of a Contract

In LIBOR IV, we denied defendants’ motion to dismiss a claim by one DAP, Fannie Mae, alleging fraud by affirmative misrepresentation in the inducement of a contract against certain counterparties in the context of swap agreements. 2015 WL 6243526, at *56-57, slip op. at *138-41. Specifically, Fannie Mae alleged that, “each time that a counterparty traded a swap pursuant to an ISDA agreement, [the] counterparty thereby renewed a promise that [it] was not in breach of any previous swap agreement.” Id.; see also Fannie Mae Compl. ¶¶ 134-38.

To state a claim for fraudulent inducement, a plaintiff must show that: “ ‘(1) the defendant made a material false representation, (2) the defendant intended to defraud the plaintiff thereby, (3) the plaintiff reasonably relied upon the representation, and (4) the plaintiff suffered damage as a result of such reliance.’ ” Axginc Corp. v. Plaza Automall, Ltd., 759 Fed. App'x 26, 30 (2d Cir. 2018) (quoting Bridgestone/Firestone, Inc. v. Recovery Credit Servs., Inc., 98 F.3d 13, 19 (2d Cir. 1996)).¹⁸⁰ A fraudulent inducement claim “can be supported by a false statement of present fact, or by a false statement of future intent which concerns a matter collateral to a contract between the parties.” Four Finger Art Factory, Inc. v. Dinicola, No. 99 Civ. 1259, 2001 WL 21248, at *3 (S.D.N.Y. Jan. 9, 2001) (citation and quotation marks omitted).

Here, Fannie Mae's fraudulent inducement claim is premised on the assertion that “[a] breach of good faith on one swap was a breach of contract, and thus a Potential Event of Default.” LIBOR IV, 2015 WL 6243526, at *56, slip op. at *139-40. Fannie Mae contends that, when defendants traded subsequent swaps with Fannie Mae, they “represented that no Potential Event of Default” existed and, in doing so, made a materially false misrepresentation to plaintiffs, upon which they reasonably relied.” Id.

Plaintiffs assert that Fannie Mae's claim survives summary judgment “for the same reasons as [the] DAPs’ other” fraud claims. Pls. Br. at 125 n.201.¹⁸¹ However, Fannie Mae's claim fails for the same reasons that plaintiffs’ other fraud claims fail. Namely, given our determination that plaintiffs have failed to put forth sufficient evidence to demonstrate that the Panel Banks “set their LIBOR contributions at rates that were not truthful answers to the LIBOR question,” in light of the Circuit's decision in Connolly, see In re LIBOR-Based Fin. Instruments Antitrust Litig., No. 11 MDL 2262 (NRB), 2023 WL 2871090, at *8 n.12 (S.D.N.Y. Apr. 10, 2023), ECF No. 3655 at 18 n.12 (citation and quotation marks omitted; emphasis in original), Fannie Mae has failed to establish that their counterparties materially misrepresented or omitted information about their alleged manipulation of LIBOR when engaging in swap trading pursuant to an ISDA Agreement. Without having established the existence of a material misrepresentation or omission – an “essential element” of its fraudulent inducement claim – Fannie Mae's claim may not proceed. See Ithaca Capital Investments I S.A. v. Trump Panama Hotel Management LLC, 450 F. Supp. 3d 358, 369 (S.D.N.Y. 2020) (“A claim for fraudulent inducement must satisfy the same elements as a claim for common law fraud.”); see also Coughlan v. Jachney, 473 F. Supp. 3d 166, 198 (E.D.N.Y. 2020) (granting summary judgment where plaintiffs failed to establish “the essential elements” of their fraudulent inducement claims); Allen v. JP Morgan Chase & Co., No. 06 Civ. 8712 (JGK), 2009 WL 857555, at *14 (S.D.N.Y Mar. 31, 2009) (similar).

1.3. Aiding and Abetting Fraud

Next, certain plaintiffs bring claims against defendants for aiding and abetting fraud. See ECF No. 4583-1; Fannie Mae Compl. ¶¶ 142-49; FDIC Compl. ¶¶ 308-12; Principal Financial Compl. ¶¶ 314-18; Principal Funds Compl. ¶¶ 311-15.¹⁸²

In LIBOR IV, this Court determined that plaintiffs had adequately alleged three theories of aiding and abetting liability: (i) Barclays’ liability for a conspiracy involving its traders and their counterparts at other Panel Banks for its trader-based conspiracy, 2015 WL 6243526, at *49, slip op. at *121-24; see also LIBOR I, 935 F. Supp. 2d at 723, slip op. at *118-22; (ii) that “each bank, by allegedly submitting artificial LIBOR quotes, furthered other banks’ manipulation” by making it more difficult to detect; and (iii) that multiple Panel Banks’ false submissions jointly influenced and, as a result, artificially suppressed the final LIBOR fix. LIBOR IV, 2015 WL 6243526, at *49, *112, slip op. at *121-24, 270; see also Pls. Br. at 125.

Plaintiffs assert that, “[i]n light of the substantial evidence of suppression[,]” all three theories underlying plaintiffs’ aiding and abetting claims remain viable. Pls. Br. at 125. Defendants, however, maintain that each of plaintiffs’ aiding and abetting claims must fail because “no [d]efendant can be found to have aided and abetted a fraud that did not exist.” Defs. Br. at 105.¹⁸³

“[T]he existence of a fraud” is the first element necessary to establish aiding and abetting liability in connection with common law fraud. Abu Dhabi Com. Bank v. Morgan Stanley & Co., Inc., 651 F. Supp. 2d 155, 186 (S.D.N.Y. 2009). As stated supra pp. 246-252, plaintiffs have failed to raise a genuine issue of material fact establishing suppression by the defendants. When considered in conjunction with the lack of evidence establishing the falsity of defendants’ submissions and the Second Circuit's decision in Connolly, it becomes clear that plaintiffs’ claims of aiding and abetting liability, each of which is premised on the manipulation and suppression of LIBOR, must fail, as well. See Loreley Financing (Jersey) No. 3 Ltd. v. Wells Fargo Sec., 412 F. Supp. 3d 392, 412 (S.D.N.Y. 2019) (granting summary judgment for defendants on plaintiffs’ fraud claims, noting “[p]laintiffs cannot sustain claims for aiding and abetting, since they have failed to establish fraud”) (citing Dodona I v. Goldman Sachs & Co., 132 F. Supp. 3d 505, 517 (S.D.N.Y. 2015) (granting summary judgment on aiding and abetting fraud claim for failure to show sufficient evidence of underlying fraud)). Accordingly, defendants’ motion for summary judgment on plaintiffs’ aiding and abetting fraud claims is granted.

1.4. Conspiracy to Commit Fraud

Certain of the remaining plaintiffs also bring claims for conspiracy to commit fraud. See ECF No. 4583-1; Fannie Mae Compl. ¶¶ 150-60; FDIC Compl. ¶¶ 313-18.¹⁸⁴

“ ‘It is well settled under New York law that there is no substantive tort of conspiracy.’ ” Cortes v. Twenty-First Century Fox America, Inc., 285 F. Supp. 3d 629, 630 (S.D.N.Y. 2018) (quoting Antonios A. Alevizopoulos & Assocs., Inc. v. Comcast Int'l Holdings, Inc., 100 F. Supp. 2d 178, 187 (S.D.N.Y. 2000) (citation omitted)). Accordingly, to make out a civil conspiracy claim, plaintiffs must first establish the existence of an underlying tort. See Reich v. Lopez, No. 13 Civ. 5307 (JPO), 2014 WL 4067179, at *17 (S.D.N.Y. Aug. 18, 2014) (quoting Treppel v. Biovail Corp., No. 03 Civ. 3002 (PKL), 2005 WL 2086339, at *5 (S.D.N.Y. Aug. 30, 2005)). As set forth above, plaintiffs are unable to support a genuine issue of material fact sufficient to support a finding that defendants engaged in any underlying acts of fraud related to the persistent manipulation or suppression of the LIBOR rate. Having found that plaintiffs have failed to sustain their underlying fraud claim, defendants’ motion for summary judgment on plaintiffs’ civil conspiracy claim must be granted. See Bandler v. BPCM NYC, Ltd., No. 12 Civ. 3512 (PGG), 2014 WL 5038407, at *15 (S.D.N.Y. Sept. 29, 2014) (granting summary judgment with respect to plaintiffs’ civil conspiracy claim where plaintiffs failed to sustain their underlying tort claim) (citing Nissan Motor Acceptance Corp. v. Scialpi, 94 A.D.3d 1067, 1069 (2d Dep't 2012) (granting summary judgment and dismissing conspiracy to commit fraud claim “since a cause of action sounding in civil conspiracy cannot stand alone, but stands or falls with the underlying torts”)).¹⁸⁵

2. Negligent Misrepresentation

Next, we turn to the DAPs’ negligent misrepresentation claims, which are centered on defendants’ alleged omissions “ ‘in the course of offering or transacting LIBOR-based securities.’ ” Pls. Br. at 125-26 (quoting LIBOR IV, 2015 WL 6243526, at *69, slip op. at *170-71); see ECF No. 4583-1; FDIC Compl. ¶¶ 319-41; Principal Financial Compl. ¶¶ 319-28; Principal Funds Compl. ¶¶ 316-25.¹⁸⁶

Under New York law, the following elements are required to establish a negligent misrepresentation claim: “ ‘(1) the defendant had a duty, as a result of a special relationship, to give correct information; (2) the defendant made a false representation that he or she should have known was incorrect; (3) the information supplied in the representation was known by the defendant to be desired by the plaintiff for a serious purpose; (4) the plaintiff intended to rely and act upon it; and (5) the plaintiff reasonably relied on it to his or her detriment.’ ” Elliot v. Nelson, 301 F. Supp. 2d 284, 287 (S.D.N.Y. 2004) (quoting Hydro Investors, Inc. v. Trafalgar Power Inc., 227 F.3d 8, 20 (2d Cir. 2000)).

In LIBOR IV, we noted that “[t]he general rule is that negligent misrepresentation is actionable as to omissions, so long as a defendant had a special duty to communicate information[,]” and determined that the DAPs could bring negligent misrepresentation “claims based on [defendants’] failure to communicate th[eir] knowledge” of LIBOR suppression or manipulation before transacting. LIBOR IV, 2015 WL 6243526, at *69, slip op. at *170-72. Plaintiffs now contend that “evidence establishing [d]efendants’ suppression of LIBOR is overwhelming” and, at a minimum, presents a disputed issue of material fact as to whether defendants knew they were misstating LIBOR and failed to communicate that to the DAPs. Pls. Br. at 126.

To the contrary, plaintiffs have not presented credible evidence of LIBOR suppression or manipulation, nor have they proffered sufficient evidence to raise a disputed issue of material fact as to a crucial element of these claims: whether defendants knew or should have known that they were misstating LIBOR and failed to communicate that information to plaintiffs. See Elliot, 301 F. Supp. 2d at 287.

In light of plaintiffs’ failure to establish the existence of a misrepresentation or omission by defendants regarding their alleged manipulation of LIBOR, defendants’ motion for summary judgment is granted with respect to plaintiffs’ claims for negligent misrepresentation. See id. (granting summary judgment on negligent misrepresentation claim where defendant's “alleged representations were not false when made”); see also Sita v. Danek Med., Inc., 43 F. Supp. 2d 245, 260 (E.D.N.Y. 1999) (granting summary judgment on negligent misrepresentation claim where plaintiffs failed to provide proof of “a specific false statement, much less an intentionally false statement”).

3. Breach of Contract and Breach of the Implied Covenant of Good Faith and Fair Dealing

Plaintiffs also bring breach of contract claims, including claims based on defendants’ alleged breach of the implied covenant of good faith and fair dealing.¹⁸⁷

On page 126 of their opposition brief, plaintiffs argue that these claims “do not require proof of a conspiracy or suppression by all defendants during the entire Relevant Period.” Pls. Br. at 126. Further, plaintiffs propose that claims for breach of contract and breach of the implied covenant of good faith and fair dealing are “an alternate remedy for a counterparty's own participation in LIBOR manipulation.” Id. Plaintiffs further suggest that “the record is rife with evidence showing that each defendant persistently suppressed LIBOR.” Id. at 127. This is a gross distortion of the record. In the absence of meaningful bank-by-bank evidence, there is no predicate for plaintiffs’ contract claims. However, for the sake of completion, we address plaintiffs’ claims below.

“To state a claim for breach of contract, a plaintiff must allege: (1) the existence of a contract; (2) a breach of that contract; and (3) damages resulting from the breach.” LIBOR IV, 2015 WL 6243526, at *74, slip op. at *181-82 (citing Nat'l Mkt. Share, Inc. v. Sterling Nat'l Bank, 392 F.3d 520, 525 (2d Cir. 2004)). “A covenant of good faith and fair dealing in the course of contract performance is implicit in all contracts, and thus a breach of this implied covenant constitutes a breach of the contract.” Id. (citations and quotation marks omitted). Specifically, implied in every contract is “a promise that ‘neither party shall do anything which will have the effect of destroying or injuring the right of the other party to receive the fruits of the contract.’ ” Id. (citations and quotation marks omitted). To maintain a claim for breach of the implied covenant of good faith and fair dealing, a plaintiff must make “ ‘some showing of intent to harm the other contracting party or a reckless disregard’ ” for the risk of harm. Id. at *76, slip op. at *187 (quoting LIBOR II, 962 F. Supp.2d at 634, slip op. at *60-61 (citation omitted)).

In LIBOR II, this Court defined the implied covenant entered into by the parties as an agreement that “neither party shall do anything which will have the effect of destroying or injuring the right of the other party to receive the fruits of the contract.” 962 F. Supp. 2d at 623, slip op. at *54-55. The Court subsequently held, on a motion to dismiss, that “the OTC plaintiffs [had] plausibly alleged that panel banks were ‘at least in reckless disregard of the potential harm to OTC plaintiffs’ when they suppressed LIBOR” and reached “the same conclusion with respect to any panel bank that manipulated LIBOR in a manner that harmed its own counterparties.” LIBOR IV, 2015 WL 6243526, at *76, slip op. at *187 (citation omitted).

Plaintiffs’ good faith and fair dealing claim is grounded on their claim of persistent suppression and the alleged factual support thereof, and relies heavily on this Court's upholding of the asserted good faith and fair dealing claims as against motions to dismiss. However, we are now at the summary judgment stage. In the preceding sections of this opinion, we have held that plaintiffs have failed to proffer sufficient evidence to allow a reasonable factfinder to infer persistent suppression. Given this Court's holding that plaintiffs have failed to sufficiently establish suppression, or injury, and in the absence of any additional evidence advanced by plaintiffs demonstrating defendants’ intent to harm or reckless disregard for the risk of harm to plaintiffs, plaintiffs have failed to raise a triable issue of fact with respect to defendants’ alleged breach of the implied covenant of good faith and fair dealing.

Accordingly, plaintiffs have “fail[ed] to adduce evidence from which a rational trier of fact could infer a breach of [this] implied covenant[.]” Keene Corp. v. Bogan, No. 88 Civ. 217 (MBM), 1990 WL 1864, at *14-16 (S.D.N.Y. 1990) (granting summary judgment on claim that one party breached an implied covenant of good faith where the opposing party failed to show either intent to harm the other or reckless disregard).

Separately, plaintiffs assert that any still-existing breach of express contract claims by Fannie Mae “defeat summary judgment for the same reason as the implied contract claims.” Pls. Br. at 127 n.204. However, those claims fail for the same reasons described above. In LIBOR IV, we permitted breach of express contract claims to proceed at the motion to dismiss stage where plaintiffs alleged “that some of [their] counterparties made misrepresentations when executing swaps regarding the existence of breaches of previously executed swaps.” 2015 WL 6243526, at *75, slip op. at *185. At the summary judgment stage, plaintiffs have failed to provide sufficient evidence demonstrating defendants’ alleged suppression or manipulation of LIBOR and have failed to show that defendants in fact “made misrepresentations when executing swaps[.]” Accordingly, plaintiffs have failed to show any actual breach of the parties’ contracts.¹⁸⁸

As a result, defendants’ motion for summary judgment on plaintiffs’ breach of contract and breach of implied covenant claims is granted.

4. Tortious Interference with Contract

There is a dispute between the parties as to whether any tortious interference claims remain live as to any defendant. Defs. Br. at 107; ECF No. 4583-1 at 2-3. The FDIC believes it has live claims for tortious interference with contract as to JPMorgan. Id.; see also FDIC Compl. ¶¶ 342-48; ECF No. 4583-1 at 3.¹⁸⁹ However, defendants contend that these claims were dismissed by this Court in 2019. Defs. Br. at 107 (citing ECF Nos. 2892, 2958).¹⁹⁰

Regardless of whether the FDIC's tortious interference claims are live, they are subject to dismissal. To state a claim for tortious interference with contract, a plaintiff must show: “(1) ‘the existence of a valid contract between the plaintiff and a third party’; (2) the ‘defendant's knowledge of the contract’; (3) the ‘defendant's intentional procurement of the third party's breach of the contract without justification’; (4) ‘actual breach of the contract’; and (5) ‘damages resulting therefrom.’ ” LIBOR IV, 2015 WL 6243526, at *74, slip op. at *183 (quoting Kirch v. Liberty Media Corp., 449 F.3d 388, 401-02 (2d Cir. 2006) (internal quotations omitted)).

In LIBOR IV, we wrote that, “under limited circumstances, plaintiffs may maintain claims for tortious interference.” 2015 WL 6243526, at *75, slip op. at *184. However, we also stated that “plaintiffs’ tortious interference claims add little value to this litigation[,]” noting that “plaintiffs cannot prove tortious interference against one entity without proving a breach of contract claim against another entity[.]” Id. at *75 n.108, slip op. at *184 n.108.

Defendants now assert that, because the FDIC has failed to establish the existence of any breach of contract by defendants, its surviving claim for tortious interference with contract must be dismissed, as well. Defs. Br. at 107-08. We agree. The FDIC has not sufficiently established that defendants’ actions caused an actual breach of any contract. Accordingly, summary judgment must be granted in favor of defendants. See D'Andrea v. Rafla-Demetrious, 146 F.3d 63, 66 (2d Cir. 1998) (affirming grant of summary judgment on tortious interference with contract claim where plaintiff failed to establish “actual breach” of a contract); see also Dentsply Sirona, Inc. v. Dental Brands for Less LLC, 2020 WL 1643891, at *5 (S.D.N.Y. 2020) (granting summary judgment where “the record contains no evidence that [defendant] took any steps to interfere with contracts between [plaintiff]” and third parties).

5. Unjust Enrichment

Finally, certain plaintiffs bring unjust enrichment claims. See ECF No. 4583-1; FDIC Compl. ¶¶ 279-85; OTC Compl. ¶¶ 396-99; Principal Financial Compl. ¶¶ 329-33; Principal Funds Compl. ¶¶ 326-30.

“To state a claim for unjust enrichment, a plaintiff must allege that: ‘(1) the [defendant] was enriched, (2) at [plaintiff's] expense, and (3) that it is against equity and good conscience to permit the [defendant] to retain what is sought to be recovered.’ ” LIBOR IV, 2015 WL 6243526, at *74, slip op. at *183 (quoting Mandarin Trading Ltd. v. Wildenstein, 16 N.Y.3d 173, 182 (2011) (internal quotations omitted)).

In LIBOR IV, we permitted plaintiffs to proceed with three categories of unjust enrichment claims, namely those asserting: “(1) that the defendant counterparty was enriched through its own misconduct, either as a panel bank itself or by the counterparty collaborating with a panel bank entity to manipulate LIBOR; (2) that the defendant counterparty was enriched through the misconduct of a related panel bank without any misconduct of its own[;]” and (3) “unjust enrichment claims against its own swap counterparties or obligors, but only to the extent that the plaintiff's injury was caused by the defendant's own misconduct or that of the defendant's affiliate.” 2015 WL 6243526, at *76-78, slip op. at *188-192.¹⁹¹

Each remaining category of unjust enrichment claims is premised upon the allegation that a defendant or its affiliate engaged in “misconduct” by participating in or benefiting from the suppression of LIBOR. Id. Accordingly, defendants now assert that plaintiffs “cannot prove that defendants were enriched absent proof of persistent, systematic LIBOR suppression” at plaintiffs’ expense. Defs. Br. at 108 (citation and internal quotation marks omitted). In response, plaintiffs merely reiterate their assertion that “[d]efendants persistently suppressed LIBOR.” Pls. Br. at 128. Given our determination that plaintiffs have failed to raise a genuine issue of material fact with respect to the suppression of LIBOR, supra pp. 177-244, plaintiffs have similarly failed to provide evidence demonstrating that defendants engaged in any misconduct sufficient to support a claim for unjust enrichment.¹⁹² See Grynberg v. ENI S.P.A., 503 F. App'x 42, 43-44 (2d Cir. 2012) (affirming grant of summary judgment where plaintiff failed to establish a genuine dispute of material fact as to alleged misconduct underlying unjust enrichment claim); see also Tarzy v. Dwyer, No. 18 Civ. 1456 (AT), 2021 WL 4134805, at *4 (S.D.N.Y. Sept. 10, 2021) (granting summary judgment where plaintiff has not established a genuine dispute of material fact as to defendants’ enrichment); Scott Sales, LLC v. Errico, No. 09 Civ. 7289 (WWE), 2015 WL 13879859, at *4 (S.D.N.Y. Mar. 11, 2015) (same). Accordingly, summary judgment is granted with respect to plaintiffs’ unjust enrichment claims.

VII. CLASS DECERTIFICATION

In LIBOR VII, the Court certified a class limited to OTC plaintiffs’ antitrust claims against Bank of America and JPMorgan Chase. See 299 F. Supp. 3d at 581-608, slip op. at *298-362. The OTC plaintiffs move to certify this same OTC class against defendants Credit Suisse AG, The Royal Bank of Scotland Group plc, Royal Bank of Scotland plc, and UBS AG. See ECF Nos. 4182, 4190, 4461.¹⁹³ The OTC defendants not only oppose this motion, but also seek decertification of the OTC class. See ECF Nos. 4317, 4336.

In LIBOR VII, we stated that “[o]ur decision to certify a class as to OTC plaintiffs’ antitrust claims rests on the action in its current form, including on OTC plaintiffs’ allegations of a sixteen-bank conspiracy to suppress LIBOR” and cautioned that “subsequent developments in the case [that] call into question those allegations or the other bases on which we rely” could warrant modification or decertification of the class. 299 F. Supp. 3d at 607 n.189, slip op. at *360 n.189. We further observed that OTC plaintiffs were “expected to provide ‘common evidence to show’ ” (1) “the existence of a conspiracy” and (2) that “ ‘all class members suffered some injury.’ ” Id. at 590, 592, slip op. at *320, 326 (quoting Sykes v. Mel S. Harris & Assocs., 780 F.3d 70, 82 (2d Cir. 2015) (emphasis added in LIBOR VII)).

Having found that defendants are entitled to summary judgment on all plaintiffs’ antitrust claims, supra pp. 244-246, we similarly find that OTC plaintiffs failed to provide common evidence of either conspiracy or suppression, supra pp. 29-244. Thus, we decertify the OTC class of plaintiffs. See Jin, 990 F.3d at 262 (“[D]istrict courts must ensure that a certified class satisfies Rule 23 throughout the litigation.” (citation omitted)); Wu, 2012 WL 6681701, at *5 (“A district court may—and should—decertify a class when the standards of Rule 23 have not been met.” (citations omitted)); In re Aluminum Warehousing Antitrust Litig., 336 F.R.D. 5, 63 (S.D.N.Y. 2020) (denying motion to certify a class under Rule 23(b)(3) because plaintiffs’ expert did not present “reliable common proof of classwide injury”); Laumann v. Nat'l Hockey League, 105 F. Supp. 3d 384, 398-99 (S.D.N.Y. 2015) (“Here, [the expert's] model was the common evidence -- and the model has been excluded. Therefore, no [Rule 23](b)(3) class may be certified.”); In re Pharmacy Benefit Managers Antitrust Litig., No. CV 03-4730, 2017 WL 275398, at *30 (E.D. Pa. Jan. 18, 2017) (denying certification where “there has been no actual evidence, common or otherwise, offered on the basic issue of whether a price fixing conspiracy existed”); see also Rami Abdallah Elias Rashmawi, No Injury? No Class: Proof of Injury in Federal Antitrust Class Actions post-Wal-Mart, 77 Wash. & Lee L. Rev. 1375 (2020).

VIII. CONCLUSION 194

Having found that plaintiffs failed to proffer evidence sufficient to raise a triable issue of fact as to conspiracy and suppression, all of plaintiffs’ claims fail. Accordingly, defendants’ joint motion for summary judgment is granted. Additionally, defendants’ Daubert motion as to Drs. Bernheim and Snow is granted while defendants’ Daubert motions as to Drs. Cragg and Marx are granted in part and denied in part. Finally, plaintiffs’ certification motion is denied and defendants’ decertification motion is granted. The Clerk of Court is respectfully directed to terminate all pending motions outlined in the Appendix below.

IT IS SO ORDERED.

APPENDIX

This Memorandum and Order resolves the following docket entries in the following cases:

Tabular or graphical material not displayable at this time.

FOOTNOTES

1.   Although there is a LIBOR for “each of the world's major currencies,” United States v. Connolly, 24 F.4th 821, 824 (2d Cir. 2022), “LIBOR” here refers to U.S. Dollar LIBOR, unless otherwise indicated. Cf. Laydon v. Mizuho Bank, Ltd., No. 12-cv-3419 (GBD) (S.D.N.Y.) (Yen LIBOR and the Tokyo Interbank Offer Rate); Sonterra Capital Master Fund Ltd. v. Credit Suisse Grp. AG, No. 15-cv-871 (SHS) (S.D.N.Y.) (Swiss Franc LIBOR); Sonterra Capital Master Fund, Ltd. v. Barclays Bank PLC et al., No. 15-cv-3538 (VSB) (Sterling LIBOR); Sullivan v. Barclays PLC et al., No. 13-cv-2811 (PKC) (Euribor).

2.   As this case has advanced over the last fourteen years, plaintiffs’ theories as to why defendants engaged in the alleged suppression conspiracy have changed. For example, at earlier stages of this litigation, plaintiffs alleged that the Panel Banks conspired in an effort to, among other things, “increase profits in individual financial transactions.” Gelboim, 823 F.3d at 766. After the Second Circuit rejected this profit-based theory as “strange,” id. at 783, plaintiffs then focused on proving that defendants possessed a “reputation-based motive to conspire, where increased profits followed from a positive reputation,” LIBOR VI, 2016 WL 7378980, at *5 (S.D.N.Y. Dec. 20, 2016), ECF No. 1676 at *11 (citation omitted). As we discuss below, plaintiffs’ theories of the conspiracy have evolved even further. See infra pp. 33-43. Unless otherwise noted, ECF numbers cited herein reference filings on the main MDL docket corresponding to case number 11-MD-2262.

3.   The Panel Banks during the relevant time period were: Bank of America, N.A., The Bank of Tokyo-Mitsubishi UFJ, Ltd., Barclays Bank PLC, Citibank, N.A., Coöperatieve Centrale Raiffeisen-Boerenleenbank B.A., Credit Suisse AG, Deutsche Bank AG, Bank of Scotland plc, HSBC Bank plc, JPMorgan Chase Bank, N.A., Lloyds Bank plc, The Norinchukin Bank, Royal Bank of Canada, The Royal Bank of Scotland Group plc (HBOS), Société Générale S.A., UBS AG, and WestLB AG. See ECF No. 4558 (“Defs. 56.1”) ¶ 1.We refer to “sixteen Panel Banks” throughout this opinion, even though seventeen banks made LIBOR submissions at some point during the relevant period, because only sixteen banks at a time were considered Panel Banks. The Bank of Scotland plc (HBOS), through its subsidiaries, was a Panel Bank until February 6, 2009, and Société Générale (“SocGen”) started making LIBOR submissions on February 10, 2009. See Defs. 56.1 ¶¶ 2-3; see also ECF No. 4166 (“Defs. Br.”) at xiv (defining term “Panel Banks”); ECF No. 4335 (“Pls. Br.”) at 1 (adopting defined terms listed in defendants’ briefing); ECF No. 4442 (“Defs. Reply Br.”) at 18 n.18; ECF No. 4557 (“Pls. 56.1”) at vi.

4.   Our references to the “relevant period” refer to the periods of: (i) August 2007 through August 2009 for the OTC action; and (ii) August 2007 through May 2010 for the DAPs’ actions. See Defs. Br. at xiv (defining term “Relevant Period”); Pls. Br. at 1 (adopting defined terms listed in defendants’ briefing); Defs. 56.1 at 18; Pls. 56.1 at vii; infra pp. 6-10 (defining OTC and DAPs).

5.   We note that the record contains nearly 5,000 ECF filings, and the docket sheet, if printed, would be roughly 1,400 pages long.

6.   See In re LIBOR-Based Fin. Instruments Antitrust Litig. (“LIBOR VIII”), 2019 WL 1331830 (S.D.N.Y. Mar. 25, 2019), ECF No. 2837; In re LIBOR-Based Fin. Instruments Antitrust Litig. (“LIBOR VII”), 299 F. Supp. 3d 430 (S.D.N.Y. 2018), ECF No. 2452; In re LIBOR-Based Fin. Instruments Antitrust Litig. (“LIBOR VI”), 2016 WL 7378980 (S.D.N.Y. Dec. 20, 2016), ECF No. 1676, rev'd and remanded sub nom. Berkshire Bank v. Lloyds Banking Grp. Plc (“Berkshire”), No. 20-1987-CV, 2022 WL 569819 (2d Cir. Feb. 25, 2022), cert. denied, 143 S.Ct. 286 (2022), and aff'd in part, rev'd in part and remanded sub nom. Schwab Short-Term Bond Mkt. Fund v. Lloyds Banking Grp. PLC (“Schwab II”), 22 F.4th 103 (2d Cir. 2021), cert. denied, 142 S. Ct. 2852 (2022); In re LIBOR-Based Fin. Instruments Antitrust Litig. (“LIBOR V”), 2015 WL 6696407 (S.D.N.Y. Nov. 3, 2015), ECF No. 1234; In re LIBOR-Based Fin. Instruments Antitrust Litig. (“LIBOR IV”), 2015 WL 6243526 (S.D.N.Y. Oct. 20, 2015), ECF No. 1222, aff'd in part, vacated and remanded in part sub nom. Charles Schwab Corp. v. Bank of Am. Corp. (“Schwab I”), No. 16-1189-cv, 883 F.3d 68 (2d Cir. Feb. 23, 2018); In re LIBOR-Based Fin. Instruments Antitrust Litig. (“LIBOR III”), 27 F. Supp. 3d 447 (S.D.N.Y. 2014), ECF No. 568; In re LIBOR-Based Fin. Instruments Antitrust Litig. (“LIBOR II”), 962 F. Supp. 2d 606 (S.D.N.Y. 2013), ECF No. 389; In re LIBOR-Based Fin. Instruments Antitrust Litig. (“LIBOR I”), 935 F. Supp. 2d 666 (S.D.N.Y. 2013), ECF No. 286, appeals dismissed, Nos. 13-3565 (L), 13-3636 (Con), 2013 WL 9557843 (2d Cir. Oct. 30, 2013), rev'd and remanded sub nom. Gelboim v. Bank of Am. Corp., 574 U.S. 405 (2015), and vacated and remanded sub nom. Gelboim v. Bank of Am. Corp. (“Gelboim”), 823 F.3d 759 (2d Cir. 2016).

7.   In LIBOR VII, we also denied the certification motions of two other putative classes, the Exchange-based class and Lender-based class. See 299 F. Supp. 3d at 458, 471-580, slip op. at *3-4, 37-297. Following LIBOR VII, members of the Exchange-based class filed a motion for leave to appeal under Rule 23(f), which the Second Circuit denied. See In re LIBOR-Based Fin. Instruments Antitrust Litig., No. 18-728, Doc. No. 84 (2d Cir. Nov. 6, 2018). Both the Exchange-based and Lender-based classes later settled. See ECF Nos. 4110, 4197.

8.   This class defined the term “LIBOR-Based Instrument” and excluded certain individuals and entities from the class:“LIBOR-Based Instrument” means an interest rate swap or bond/floating rate note that includes any term, provision, obligation or right for the purchaser or counterparty to be paid interest by a Panel Bank (or a Panel Bank's subsidiaries or affiliates) based upon the 1 month or 3 month U.S. dollar LIBOR rate. For the avoidance of doubt, the term LIBOR-Based Instrument does not include instruments on which a Panel Bank (or a Panel Bank's subsidiaries or affiliates) does not pay interest, such as bonds/floating rate notes issued by entities other than Panel Banks (or Panel Banks’ subsidiaries or affiliates). Nor does the term include instruments that include only a term, provision, or obligation requiring the purchaser or counterparty to pay interest, such as business, home, student or car loans, or credit cards.Excluded from the class are panel banks and their employees, affiliates, parents, and subsidiaries and any judicial officers and staff presiding over this action. “Panel Bank” means Bank of America, Bank of Tokyo Mitsubishi, Barclays Bank, Citibank, Credit Suisse, Deutsche Bank, HBOS, HSBC, JPMorgan, Lloyds, Norinchukin, Rabobank, Royal Bank of Canada, Royal Bank of Scotland, Societe Generale, UBS, and West LB (n/k/a Portigon).Id. at 581, slip op. at *298.

9.   At the time that this Court issued LIBOR VII in 2018, defendants Bank of America and JPMorgan were the only defendants deemed subject to the Court's jurisdiction. 299 F. Supp. 3d at 581-82, slip op. at *298-301. The Second Circuit's 2021 decision in Schwab II held that plaintiffs had sufficiently alleged personal jurisdiction against the foreign defendants under a conspiracy theory of jurisdiction. 22 F.4th at 124-25. We refer to the defendants opposing OTC plaintiffs’ motion for class certification as OTC defendants. See ECF Nos. 4317, 4336.

10.   In August 2025, the parties submitted a Joint Memorialization which catalogued (a) all cases which have been wholly resolved by settlement or otherwise, and (b) for the cases that remain open, all claims that the parties believe to be live, not live, or in dispute as to each defendant. See ECF Nos. 4853, 4853-1.Due to the aforementioned orders and settlements, not all Panel Banks remain as defendants. The remaining defendants in the FFFP actions are: Bank of America, N.A., Bank of Scotland plc, Barclays Bank PLC; Bear Stearns Capital Markets, Inc., Coöperatieve Rabobank U.A. (f/k/a Coöperatieve Centrale Raiffeisen-Boerenleenbank B.A.), Credit Suisse AG, Credit Suisse Group AG, Credit Suisse International, Deutsche Bank AG, HBOS plc, HSBC Bank plc, HSBC Bank USA, N.A., JPMorgan Chase & Co., JPMorgan Chase Bank, N.A., J.P. Morgan Dublin plc, JPMorgan Markets Limited, Lloyds Banking Group plc, Lloyds Bank plc (f/k/a Lloyds TSB Bank plc), Merrill Lynch & Co., Merrill Lynch Capital Services, Merrill Lynch International Bank, NatWest Markets plc (f/k/a The Royal Bank of Scotland plc), Portigon AG (f/k/a WestLB AG), Royal Bank of Canada, and UBS AG. See ECF Nos. 4583, 4583-1, 4596 (dismissing remaining claims against Citibank, N.A. and Citigroup Inc.).

11.   The Court once again thanks the many talented lawyers who have engaged constructively and cordially to limit the number of discovery disputes requiring this Court's attention. For a multidistrict litigation of this magnitude and complexity, discovery has proceeded with welcomed ease and efficiency.

12.   Class plaintiffs previously acknowledged that the regulatory investigations were “based on the same conduct underlying plaintiffs’ civil claims” and “directly relate to whether the Panel Banks engaged in unlawful manipulation of LIBOR and the quantum of that manipulation.” ECF No. 1415 at 2, 4 (emphasis omitted). The DAPs have also described the regulatory productions as “undeniably relevant to [their] claims because they go to such core issues as who was involved in LIBOR manipulation, over what time period, by what means, to what extent, and the like.” ECF No. 1410 at 1. Defendants represented that the government regulators provided input into the search terms that were used for their regulatory productions, which “were constructed to identify documents related to the LIBOR-setting process and the suppression of LIBOR (if any).” ECF No. 3559 at 5.

13.   Three individual defendant-banks submitted supplemental memoranda of law, including, NatWest (f/k/a “NatWest” or “RBS”), see ECF No. 4170; JPMorgan, see ECF No. 4171; and Credit Suisse, see ECF No. 4174. Plaintiffs endeavored to rebut the individual defendant-banks’ arguments in their joint opposition to defendants’ summary judgment motion. See Pls. Br. at 114-24. The same banks also filed supplemental reply memoranda of law: NatWest, see ECF No. 4450; JPMorgan, see ECF No. 4449; and Credit Suisse, see ECF No. 4448. Individual supplemental briefs were also filed by Norinchukin Bank. ECF Nos. 4172, 4447. However, following the completion of the parties’ briefing, the Court was informed that “all remaining claims against Norinchukin in the ․ multidistrict litigation” were resolved as of April 2, 2025. ECF No. 4555.Pursuant to various sealing applications filed alongside the parties’ submissions on the Upstream Issues, which the Court entered, ECF Nos. 4195, 4293, 4346, 4347, 4463, 4464, 4540, 4541, 4559, the initial round of submissions was filed under seal, and the parties re-filed the same set of submissions on the public docket two weeks later with redactions for any sensitive material.

14.   We note that the Court received the parties’ initial joint statements and responses alongside the parties’ submission of their briefs. See ECF Nos. 4168, 4331, 4333, 4443. In addition, defendants submitted a reply to plaintiffs’ responses to defendants’ 56.1 statement, ECF No. 4444, and plaintiffs filed errata submissions, ECF Nos. 4526-1, -2. As a result of the sequence followed by the parties, there was not a single, consolidated document that reflected the parties’ positions regarding material facts, as is envisioned under 56.1(e) of the Local Rules of the United States District Courts for the Southern and Eastern Districts of New York. Accordingly, the Court wrote to the parties on March 21, 2025 requesting that “each side submit, for its respective 56.1 statement, a consolidated document reflecting the initial statement, response, reply, and errata.” ECF No. 4545. These documents were filed by the parties. See ECF Nos. 4557, 4558. Thus, for the purposes of the instant motions, the Court relies on the consolidated 56.1 statement, ECF No. 4558, and consolidated 56.1 counterstatement, ECF No. 4557.

15.   Plaintiffs initially brought a motion to exclude the expert testimony of Dr. Dennis Carlton, who was retained by JPMorgan. ECF No. 4159. However, the parties entered into a stipulation clarifying the scope of Dr. Carlton's proposed testimony, and plaintiffs withdrew their motion to exclude Dr. Carlton's testimony. ECF No. 4465.

16.   In LIBOR VII, we declined to apply Rule 403 to expert testimony offered at the class certification stage because the Court served as the trier of fact. 299 F. Supp. 3d at 469 n.7, slip op. at *32 n.7. Obviously, the same restraint is not appropriate in the context of a jury trial. See Nimely, 414 F.3d at 397 (identifying “the uniquely important role that Rule 403 has to play in a district court's scrutiny of expert testimony, given the unique weight such evidence may have in a jury's deliberations”).

17.   As we cover in a later section, infra pp. 244-269, the same principles underlying the conspiracy inquiry apply to plaintiffs’ common-law claims, though we note that some of “DAPs’ common-law claims do not require proof of a conspiracy to suppress LIBOR.” Pls. Br. at 124 n.199.

18.   We selected this formulation of the standard for its clarity, specificity, and federal uniformity. The Second Circuit has distilled this same standard into two elements: “(1) the defendants conspired to raise prices, and (2) this conspiracy caused injury to the plaintiff in the form of artificially inflated prices.” In re Publ'n Paper Antitrust Litig., 690 F.3d 51, 61 (2d Cir. 2012) (citations omitted). The alleged conspiracy does not fit perfectly onto this formulation, however, as plaintiffs allege that the Banks “colluded to depress LIBOR, and thereby increased the cost to [plaintiffs], as buyers, of various LIBOR-based financial instruments.” Gelboim, 823 F.3d at 771.

19.   We note that “once a conspiracy is shown, only slight evidence is needed to link another defendant with it.” Apex Oil Co. v. DiMauro, 822 F.2d 246, 257 (2d Cir. 1987) (citation and quotation marks omitted). “Nevertheless, [p]laintiffs must provide evidence ‘pertaining to each defendant’ to demonstrate that that defendant participated in the conspiracy.” Ross, 35 F. Supp. 3d at 438 (quoting AD/SAT, Div. of Skylight, Inc. v. Associated Press, 181 F.3d 216, 234 (2d Cir. 1999)).

20.   As the Second Circuit has explained, there is an obvious policy purpose behind this principle: “To permit an inference of conspiracy based on ambiguous evidence—that is, evidence that is equally consistent with independent conduct as with illegal conspiracy ․ would deter or penalize perfectly legitimate and procompetitive conduct.” Anderson News II, 899 F.3d at 98 (citations and quotation marks omitted).

21.   See also LIBOR VI, 2016 WL 7378980, at *3, slip op. at *6 (“[T]he object of the conspiracy that the Circuit recognized and which meets the plausibility test is the projection of financial soundness.”), id. at *5, slip op. at *11 (“The only conclusion to be drawn is that the Circuit meant ‘increased profits and the projection of financial soundness’ to describe collectively a single, reputation-based motive to conspire, where increased profits followed from a positive reputation.”); OTC Plaintiffs’ Interrogatory Responses, Resp. No. 9 at 146 (similar); DAPs’ Interrogatory Responses, Resp. No. 9 at 142 (similar).

22.   See also OTC Plaintiffs’ Interrogatory Responses, Resp. No. 9 at 147 (“[F]aced with contagious nature of banking panic, the Panel Banks had the incentive to ensure their collective interests, as opposed to individual interest.”); DAPs’ Interrogatory Responses, Resp. No. 9 at 142 (similar).

23.   It is unclear whether all plaintiffs contend that one object of the conspiracy was to disregard the LIBOR rules. As defendants point out, only DAPs included this theory in their responses to contention interrogatories. See Defs. Br. at 16-17. However, the opposition brief submitted on behalf of DAPs and OTC plaintiffs is rife with assertions to the same effect. See, e.g., Pls. Br. at 2, 75 n.110, 82, 83, 86, 93 n.145.Moreover, in their responses to contention interrogatories, DAPs assert that “Defendants also recognized that they could realize profits from lower LIBOR submissions than they otherwise would have, and that higher LIBORs could negatively impact their financial positions.” DAPs’ Interrogatory Responses, Resp. No. 9 at 150. We understand this to be a reference to defendants’ asserted motive of projecting financial strength. See LIBOR VI, 2016 WL 7378980, at *5, slip op. at *11 (“The only conclusion to be drawn is that the Circuit meant ‘increased profits and the projection of financial soundness’ to describe collectively a single, reputation-based motive to conspire, where increased profits followed from a positive reputation.”). Otherwise, it is implausible. See id. at *2, slip op. at *5 (“[T]he premise that the primary goal of the conspiracy was to increase profits by lowering the interest rate the banks had to pay when they were in the role of borrower is not plausible[.]”); Gelboim, 823 F.3d at 783 (“It seems strange that this or that bank (or any bank) would conspire to gain, as a borrower, profits that would be offset by a parity of losses it would suffer as a lender.”).

24.   See also OTC Plaintiffs’ Interrogatory Responses, Resp. No. 1 at 20 (“[T]he Panel Banks engaged in a Conspiracy to suppress LIBOR”); DAPs’ Interrogatory Responses, Resp. No. 1 at 22 (“Defendants caused LIBOR to be calculated or suppressed artificially low.”).

25.   See also OTC Plaintiffs’ Interrogatory Responses, Resp. No. 1 at 21 (“The Panel Banks’ Conspiracy was aimed at keeping all of the Panel Banks near ‘the pack’ ”); DAPs’ Interrogatory Responses, Resp. No. 1 at 21 (“Defendants acted to stay in ‘the pack.’ ”).

26.   See DAPs’ Interrogatory Responses, Resp. No. 1 at 20.

27.   See OTC Plaintiffs’ Interrogatory Responses, Resp. No. 6 at 123 (“While the Panel Banks would communicate target or consensus LIBOR submissions (directly and through Interdealer Brokers) around which the Panel Banks would make their LIBOR submissions, the Conspiracy did not involve an agreement on a specific level or magnitude of suppression.”); DAPs’ Interrogatory Responses, Resp. No. 6 at 120 (similar).

28.   Plaintiffs note that defendants did not present evidence that the panel banks were net lenders. See Pls. Br. at 87 n.136. However, defendants were not required to present evidence to this effect, as they could simply point to the absence of evidence at this critical juncture. See Brady, 863 F.2d at 210-11 (“[W]here the nonmovant will bear the ultimate burden of proof at trial on an issue, the moving party's burden under Rule 56 will be satisfied if he can point to an absence of evidence to support an essential element of the nonmoving party's claim.” (citation omitted)); Matsushita, 475 U.S. at 588 (“[t]o survive a motion for summary judgment,” a plaintiff “must present evidence ‘that tends to exclude the possibility’ that the alleged conspirators acted independently” (citation omitted)).

29.   Plaintiffs make the bizarre claim that their “experts made no such concessions and Defendants’ theory is illogical[.]” Pls. Br. at 84. At the same time, plaintiffs try to limit Dr. Cragg's “purported concession” by arguing that it “is nothing more than [his] acknowledgement that he cannot rule out the (unlikely) possibility that the Panel Banks acted contrary to their incentive to suppress LIBOR collectively.” Pls. Br. at 85 n.130.Having reviewed the record, which is quoted above, in which plaintiffs’ experts do make these concessions, the Court is reminded of a colloquy between Humpty Dumpty and Alice, found in Chapter VI of Lewis Carroll's Through the Looking Glass: “When I use a word,” Humpty Dumpty said, in rather a scornful tone, “it means just what I choose it to mean—neither more nor less.” “The question is,” said Alice, “whether you can make words mean so many different things.”Unfortunately for plaintiffs, they, unlike Humpty Dumpty, cannot choose to change the plain meaning of their experts’ testimony post hoc.

30.   This internal inconsistency raises an interesting issue for the OTC class, as “[d]irectional differences are particularly corrosive of adequacy in that they create directly conflicting incentives” and “heighten[ ] conflicts between class members’ interests.” LIBOR VII, 299 F. Supp. 3d at 539, slip op. at *198 (denying certification of proposed class concerning trader-based manipulation for failure to meet Rule 23(a)(4)’s adequacy of representation requirement). We noted this explicitly as it relates to OTC class in our class certification decision: “Unlike the trader-based class in the Exchange action, any diverging incentives within the putative OTC class (and within the proposed suppression class in the Exchange-based action) are necessarily limited by suppression’s one-directional nature.” Id. at 590, slip op. at *319 (emphasis in original).

31.   Plaintiffs note that this portion of LIBOR IV was reversed. See Pls. Br. at 84, 84 n.128. We are not relying on LIBOR IV’s precedential value for this proposition, as this portion of LIBOR IV has none, but rather noting the logic of this argument at a more advanced stage in the litigation. Cf. In re Okura & Co. (Am.), Inc., 249 B.R. 596, 611 n.10 (Bankr. S.D.N.Y. 2000) (“While it may be true that the act of vacating an opinion diminishes its value as binding precedent, it has no effect on the persuasiveness of the decision.” (citation omitted)).

32.   ECF No. 4160-5 (Cragg Sept. 11, 2024 Dep. Tr.) at 105:17-22 (“Q. You can't even tell me what the pack is, right -- how you define the ‘pack’? A. That's right. As -- as we talked about in our last meeting, I don't think a -- you know, the -- the notion of the ‘pack’ is well defined.”), 106:17-20 (“[A.] I don't have a definition specifically of what the banks meant by ‘within the pack’ ․ or being an outlier.”), 326:2-3 (“[A.] I don't have a definition of what ‘within the pack means.’ ”).

33.   To circumvent this issue, plaintiffs argue that the conspiracy did not need an enforcement mechanism. See OTC Plaintiffs’ Interrogatory Responses, Resp. No. 7 at 125 (“The Panel Bank's Conspiracy did not require a formal punishment mechanism.”); ECF No. 4160-2 (Cragg Rebuttal Report) ¶ 15 (“The Panel Banks did not need to devise an enforcement mechanism, because the agreement was self-enforcing.”); Pls. Br. at 88 (similar); but see Cragg Opening Report ¶ 125 (“Conversely, highly transparent markets can facilitate collusion, because cartel members can easily and quickly identify and respond to any deviations.”). While “[i]t is true that a cartel may exist with only soft measures of control or ineffective enforcement[,]” “the absence of evidence about enforcement does nothing to dissipate the inference of independent behavior.” Kleen Prods. LLC, 910 F.3d at 937 (citation omitted; emphasis in original).

34.   We note that, even if we found the alleged agreement to be economically sensible, “[p]laintiffs’ overall burden [would] not [be] low, ․ because the conspiracy alleged is vast and its impact is complex.” In re Foreign Exch. Benchmark Rates Antitrust Litig., No. 13 Civ. 7789 (LGS), 2022 WL 294118, at *6 (S.D.N.Y. Feb. 1, 2022).

35.   In a footnote, plaintiffs also cite statements made by the BBA's director on a Foreign Exchange & Money Markets Committee (“FXMMC”) call in August 2007 as purported direct evidence that “LIBOR submissions were false, and Defendants agreed to disregard the rules and conceal the conduct.” Pls. Br. at 75 n.110 (citing PX1005, PX1005-A). Having examined the exhibit itself, we find plaintiffs’ summary highly misleading. For instance, plaintiffs ignore the BBA director's observation that LIBOR rates are “all over the place” and another call participant's statement that “I do believe that the banks are still endeavoring to post a sensible number.” ECF No. 4310-47 (PX1005-A) at 2, 3. At most, this is “ambiguous evidence—that is, evidence that is equally consistent with independent conduct as with illegal conspiracy.” Anderson News II, 899 F.3d at 98 (citations and quotation marks omitted). Additionally, as we explain later, “the BBA was indisputably not a member of the ‘inter-bank conspiracy’ upheld in Gelboim.” In re LIBOR-Based Fin. Instruments Antitrust Litig., No. 11 MDL 2262 (NRB), 2022 WL 17082626, at *2 (S.D.N.Y. Nov. 18, 2022) (citation omitted).

36.   As we noted previously, plaintiffs initially moved to exclude the testimony of Dr. Dennis Carlton, who was retained by JPMorgan. ECF No. 4159. However, on January 24, 2025, the Court signed a Stipulation and Order submitted by the parties clarifying the scope of Dr. Carlton's opinions and terminating plaintiffs’ Daubert motion. ECF No. 4465.

37.   OTC defendants attach these reports to a declaration of their counsel Jason M. Hall, in furtherance of their motion to exclude Dr. Cragg's testimony.

38.   Dr. Cragg received his Ph.D. in Economics from Stanford University and served as an economics professor at Columbia University and University of California, Los Angeles. See Cragg Opening Report ¶ 1. He is now a Senior Partner at Keystone Strategy. See id.

39.   This ruling is consistent with the parties’ Stipulation and Order clarifying the scope of testimony offered by JPMorgan's expert, Dr. Carlton. ECF No. 4465. The Stipulation states that Dr. Carlton may testify about whether “the market structure and behavior, economic conditions, competitive incentives, conduct of JPMorgan and other market participants, and record evidence, are inconsistent with collusion and consistent with unilateral action.” Id. ¶ 2. However, Dr. Carlton may not testify as to “the ultimate conclusion that JPMorgan ․ did not enter into an agreement or conspiracy to persistently suppress USD LIBOR.” Id. ¶ 1.In any event, as we discuss later, infra pp. 162-166, Dr. Cragg's opinions on this topic do not advance plaintiffs’ efforts to defeat defendants’ summary judgment motion. See Venture Tech., Inc. v. Nat'l Fuel Gas Co., 685 F.2d 41, 47 (2d Cir. 1982) (“[O]ne who alleges that he is a victim of an antitrust conspiracy and seeks to impose the heavy sanctions of the Sherman Act upon the accused, must show more than the existence of a climate in which such a conspiracy may have been formed.”).

40.   Dr. Cragg also reviews, inter alia, various articles, economic studies, academic reports, datasets, and charts, Cragg Opening Report Appendix A, before concluding that the “economic evidence” suggests that “LIBOR was disconnected from market forces,” id. ¶ 162.

41.   This ruling is consistent with the parties’ Stipulation and Order clarifying the scope of Dr. Carlton's testimony. ECF No. 4465. The Stipulation states that Dr. Carlton is not precluded from opining “that ․ competitive incentives ․ [is] inconsistent with collusion and consistent with unilateral action.” Id. ¶ 2.As we note later, infra pp. 95-106, to the extent that Dr. Cragg's opinions on this issue are admissible, they do not assist plaintiffs in making it past the summary judgment stage. Dr. Cragg admits in his deposition testimony that each bank maintained an individual incentive to suppress its own LIBOR to portray its own financial health. See ECF No. 4160-4 (Cragg July 25, 2024 Dep. Tr.) at 107:16-20 (“Q. Did each panel bank, in your view, have a unilateral incentive to suppress LIBOR? A. They had a -- both an individual and collective interest in that.”); ECF No. 4160-5 (Cragg Sept. 11, 2024 Dep. Tr.) at 74:17-20 (“[Q.] [W]e agree that banks have a unilateral incentive to -- to portray their financial health, right? A. Yes.”). This is quintessential “[a]mbiguous evidence—evidence that is equally consistent with independent conduct as with illegal conspiracy[,]” which “is not enough” to support a claim of conspiracy. SourceOne Dental, 310 F. Supp. 3d at 357 (citation omitted).

42.   We also note that Dr. Cragg's review of the record is not “traceable to a reliable methodology.” Tourre, 950 F. Supp. 2d at 675. On this point, OTC plaintiffs argue that an expert should not “be expected to review and rely upon the entirety of such a vast discovery record.” Cragg – Pls. Br. at 10. We agree. However, that is precisely why an expert should not attempt to invade the province of the jury and “state that the defendants did or did not engage in anticompetitive conduct.” U.S. Info. Sys., 313 F. Supp. at 240.

43.   Dr. Cragg notes that the phrase “flat to LIBOR” is meant “to indicate submissions that were on average at or a fixed distance away from the LIBOR fix over a period of time.” Id. ¶ 30 n.53. Dr. Cragg insists that he relies on “extremely conservative” assumptions such that his model is intentionally biased against results that would indicate collusion. Cragg Opening Report ¶ 166. As plaintiffs explain, Dr. Cragg's model was “designed to detect only one mechanism by which the Panel banks could have stayed ‘within the pack’---periods where the a [sic] Panel Bank remained ‘flat to LIBOR,’ such as by targeting a fixed distance from the LIBOR fix, to avoid being perceived as an outlier.” Cragg – Pls. Br. at 16. Plaintiffs acknowledge that Dr. Cragg's collusion screen is not a test for parallel conduct. See Pls. Br. at 81 n.124.

44.   At his deposition, Dr. Cragg acknowledged that there existed several “potential strategies” by which the banks could collude, including conspiracies in which defendants generally agreed to ignore the LIBOR rules or agreed to make submissions that were a certain number of basis points or a percentage below its borrowing costs. See ECF No. 4160-5 (Cragg Sept. 11, 2024 Dep. Tr.) at 130:20-22, 133:20-138:14. However, Dr. Cragg admitted that “those [we]re all ․ types of strategies that [he] ha[dn't] specifically ․ tested for.” Id. at 130:20-22.

45.   Dr. Cragg's model seeks to control for various exogenous variables to avoid confounding the results. Specifically, he accounts for sixteen variables, including “LIBOR Fix, Lagged,” which accounted for the “LIBOR Fix for the previous business day close,” “Average Bank CDS, Lagged,” which accounted for the “average rate of the Credit Default Swap for Panel Banks,” “VIX, Lagged,” which controlled for the “implicit volatility of the equity market,” and several other similar factors. Cragg Opening Report ¶ 210.

46.   We are not the first court to disqualify Dr. Cragg on the basis that his expert analysis ignored key facts and data points. See The Coca-Cola Co. & Subsidiaries v. Comm'r of Internal Revenue, 155 T.C. 145, 271 (2020). While the prior case involved Dr. Cragg's analysis of various tax and accounting related items, which is, of course, quite different from the instant case, we nevertheless find it noteworthy. Id. at 272.

47.   As we address later, infra pp. 85-93, Dr. Cragg's conclusion that defendants’ LIBOR submissions were less dispersed, only after controlling for the volatility occurring as a result of the financial crisis, would not permit a jury to reasonably infer parallel conduct.

48.   Of course, Dr. Cragg may not simply repackage his excluded affirmative opinions as rebuttal opinions. “Rebuttal evidence is properly admissible when it will explain, repel, counteract or disprove the evidence of the adverse party.” Scott v. Chipotle Mexican Grill, Inc., 315 F.R.D. 33, 44 (S.D.N.Y. 2016) (citations and quotation marks omitted). And, Dr. Cragg's responses to defense experts’ critiques of his affirmative opinions are properly excluded, as well. Cf. PRCM Advisers LLC v. Two Harbors Inv. Corp., No. 20 Civ. 5649 (LAK) (BCM), 2025 WL 1276513, at *19 (S.D.N.Y. May 2, 2025) (“[I]f one expert's opinion is excluded by the court (for example, because that expert was inadequately qualified, or her methodology was unreliable), any calculations, analyses, or conclusions reached by other experts in reliance on the excluded opinion must themselves be excluded.”).

49.   FFFP plaintiffs attach these reports to a declaration of their counsel James R. Martin, in support of their opposition to defendants’ motion to exclude Dr. Marx's testimony.

50.   Unlike Dr. Cragg, Dr. Marx's reports are not cleanly divided into the above categories, as her reports contain sections with multiple, intertwined opinions. Citations to paragraphs are illustrative only, as we are “not obligated to prune away all of the problematic elements of an expert's proposed testimony to save the remaining portions, however small.” In re Pfizer Inc. Sec. Litig., 819 F.3d 642, 665 (2d Cir. 2016) (citation and quotation marks omitted). Instead, we distinguish the “unreliable portion of an opinion” so that we do not “throw the good out with the bad.” Id.

51.   Dr. Marx received her Ph.D. in Economics from Northwestern University, serves as the Robert A. Bandeen Professor of Economics at the Fuqua School of Business at Duke University, and is a partner at Bates White Economic Consulting. See Marx Opening Report ¶ 1.

52.   As we discuss later, infra pp. 162-166, Dr. Marx's opinions on this topic do not enable plaintiffs to defeat defendants’ summary judgment motion. See Venture Tech., 685 F.2d at 47 (“[O]ne who alleges that he is a victim of an antitrust conspiracy and seeks to impose the heavy sanctions of the Sherman Act upon the accused, must show more than the existence of a climate in which such a conspiracy may have been formed.”).

53.   We note that a Court in this District previously excluded Dr. Marx's opinions on this same ground. The Honorable Paul A. Crotty excluded her opinions referencing defendants’ “motivations, thought processes, and understanding” in a similar antitrust action. See Anderson News, L.L.C. v. Am. Media, Inc., No. 09 Civ. 2227 (PAC), 2015 WL 5003528, at *4 (S.D.N.Y. Aug. 20, 2015), aff'd, 899 F.3d 87 (2d Cir. 2018).

54.   Plaintiffs try to justify Dr. Marx's approach by citing the economic principle of “revealed preference,” a “standard concept” that they do not define. Marx – Pls. Br. at 16. However, even assuming that “revealed preference” is a “standard concept that has been a part of the economic landscape for decades,” id., Dr. Marx's application of this principle is circular: for a bank to “reveal” its preference that collusion is economically beneficial, one would have to conclude, ex ante, that the bank was, in fact, colluding. Otherwise, there would be nothing to reveal or observe. We reject plaintiffs’ suggestion that defendants are merely trying to “insult[ ]” and impugn Dr. Marx's reputation by raising this circularity argument, see Marx – Pls. Br. at 16, for we find the argument to be persuasive.

55.   Defendants move to exclude Dr. Marx's opinions because she did not precisely define the conspiracy. See Marx – Defs. Br. at 12-14. However, this argument is better directed to plaintiffs in the context of a summary judgment motion.

56.   To find these documents, Dr. Marx's team at Bates White “identified a set of documents that they thought were relevant to [her] opinions” and Dr. Marx also asked for specific documents as well. ECF No. 4304-24 (Marx June 25, 2024 Dep. Tr.) at 34:22-36:13.

57.   This Circuit's teaching in Anderson News II is applicable here. There, the Second Circuit stated the following about Dr. Marx's report: “[Plaintiff] again relies on Dr. Marx's report, which the District Court excluded on this point, concluding that her opinions ‘merely recite what is on the face of documents produced during discovery.’ ․ We agree with the District Court's decision to exclude those portions of Dr. Marx's opinion that merely interpret defendants’ statements.” Anderson News II, 899 F.3d at 112 (citation omitted). Similarly, in another federal case in the Northern District of California, the court allowed Dr. Marx to provide “background facts” within “the context of her econometric analysis,” but excluded Dr. Marx's “mention of strictly legal matters,” such as the “findings of government competition agencies worldwide” and refused to permit Dr. Marx “to testify about collusion, violations of antitrust law, or anticompetitive allegations undertaken by defendants” because such testimony would be “well outside Dr. Marx's domain of antitrust economics.” In re Capacitors Antitrust Litig., No. 14-3264, 2021 WL 5407452, at *5 (N.D. Cal. Nov. 18, 2021).

58.   As we address later, infra pp. 85-93, Dr. Marx's conclusion that defendants’ LIBOR submissions were less dispersed, only after controlling for the volatility occurring because of the financial crisis, would not permit a jury to reasonably infer parallel conduct.

59.   Plaintiffs’ argument that Dr. Marx's opinion on focal points is supported by Deutsche Bank submitter King's recorded statement, see Pls. Br. at 103-104 (citation omitted), is misleading and unpersuasive for the reasons we specifically discuss relating to Mr. King, infra pp. 144-147. Also, the argument further confirms that Dr. Marx's opinion on this point addresses “ ‘lay matters which a jury is capable of understanding and deciding without the expert's help.’ ” Highland Capital, 379 F. Supp. 2d at 469 (citation omitted).

60.   Defendants rely on Williamson Oil Co. v. Philip Morris USA, 346 F.3d 1287, 1323 (11th Cir. 2003), see Marx – Defs. Reply Br. at 9, which we find to be inapposite in this particular circumstance. The expert there concluded that “plaintiffs participated in an illegal price fixing conspiracy” but “defined ‘collusion’ to include conscious parallelism.” Williamson Oil, 346 F.3d at 1323. The Eleventh Circuit therefore found “no abuse of discretion in the district court's decision to exclude some of [the expert's] testimony.” Id. Here, by contrast, Dr. Marx merely opines that brokers “could facilitate exchange of information in support of collusive conduct,” Marx Opening Report ¶ 257 (emphasis added), leaving it up to the factfinder to conclude whether brokers did facilitate exchange of information.

61.   As we discuss later, infra pp. 183 n.128, the Supreme Court of the United Kingdom arrived at a similar interpretation of the LIBOR question as did the Second Circuit in Connolly. The U.K. Supreme Court unanimously held “[t]he question posed by the LIBOR ․ definition[ ] required a bank on the contributor panel of banks to submit the rate at which that bank (in the case of LIBOR), ․ could borrow money at the time of the submission. The identification of this rate required a qualitative assessment of various data sources and was a matter of subjective opinion rather than empirical fact.” R (Respondent) v Hayes (Appellant); R (Respondent) v. Palombo (Appellant) [2025] UKSC 29 (LL), ¶ 7. “It was thus in the nature of the exercise that there would on any given day be a range – which might be narrower or wider depending on market conditions – of different rates that could reasonably be selected: the figure submitted depended on the subjective judgment of the submitter.” Id. ¶ 22.

62.   Of course, Dr. Marx may not simply repackage her excluded affirmative opinions as rebuttal opinions. “Rebuttal evidence is properly admissible when it will explain, repel, counteract or disprove the evidence of the adverse party.” Scott, 315 F.R.D. at 44 (citations and quotation marks omitted). And, to the extent that Dr. Marx was responding to defense experts’ critiques of her affirmative opinions, those opinions are properly excluded as well. Cf. PRCM Advisers LLC, 2025 WL 1276513, at *19 (“[I]f one expert's opinion is excluded by the court (for example, because that expert was inadequately qualified, or her methodology was unreliable), any calculations, analyses, or conclusions reached by other experts in reliance on the excluded opinion must themselves be excluded.”).Defendants argue that Dr. Marx's reliance on Dr. Snow's suppression analysis is unreliable. Marx – Defs. Br. at 21-24. Given that we reject Dr. Snow's suppression analysis as unreliable, infra pp. 201-238, Dr. Marx cannot testify to this analysis either. See PRCM Advisers LLC, 2025 WL 1276513, at *19.

63.   Defendants’ suggestion that parallel conduct requires identical or uniform actions is an over-reading of LLM Bar Exam, LLC v. Barbri, Inc., 271 F. Supp. 3d 547, 579 (S.D.N.Y. 2017), aff'd, 922 F.3d 136 (2d Cir. 2019) and Anderson News II, 899 F.3d at 104. See Defs. Br. at 75-76. Defendants’ first citation, LLM Bar Exam, LLC, merely states that the plaintiff failed to allege parallel conduct where the defendants took similar actions “at different times over the span of several years.” 271 F. Supp. 3d at 579. Defendants’ second citation, Anderson News II, is similarly unhelpful. The Second Circuit noted that “[o]ne powerful form of circumstantial evidence is parallel action—proof that defendants took identical actions within a time period suggestive of prearrangement.” 899 F.3d at 104. However, the Circuit ultimately found that, while the plaintiff's complaint had survived at the motion to dismiss stage by alleging that “all of the publisher and distributor defendants ceased doing business with [it] within a span of three business days,” the evidence at the summary judgment stage established that “defendants’ conduct was not, in fact, parallel,” because “defendants’ responses were not uniform” and the “tight timeframe for those responses ․ was of [plaintiff's] own making, not the result of an unlawful agreement.” Id. at 105. We do not understand either case to have unilaterally transformed the test for parallel conduct in light of the Supreme Court's reasoning that prices “are fixed because they are agreed upon” and that “prices are fixed ․ if the range within which purchases or sales will be made is agreed upon.” Socony-Vacuum Oil Co., 310 U.S. at 222.

64.   We note that some of plaintiffs’ data visualizations do not fully capture the small yet significant variations in the banks’ LIBOR submissions. For example, plaintiffs’ line chart depicting LIBOR submissions over the relevant timeframe includes a vertical axis that runs from zero to seven percent, i.e., 700 basis points. See Pls. Br. at 76. This was, no doubt, done to capture the full range of LIBOR values over the relevant period, but this scaling masks subtle, yet important distinctions in each bank's daily submissions. Thus, by displaying LIBOR figures at a relatively high level, plaintiffs fail to capture some of the dispersion that was observable during the relevant period.

65.   Plaintiffs acknowledge that Dr. Cragg's collusion screen is “not a test for parallel conduct” and they do not rely on it to establish parallel conduct. Id. at 81 n.124.

66.   Notably, plaintiffs do not respond to defendants’ argument in a meaningful way, saying only that it is a “head[s] we win, tails they lose” argument. See Pls. Br. at 78 n.120 (quotation marks omitted). We disagree. We find that plaintiffs, not defendants, are trying to make a “head[s] we win, tails they lose” argument by contending that evidence of a “pack” conspiracy also serves as evidence of a “suppression” conspiracy, even when the two theories conflict.

67.   Plaintiffs’ argument is particularly ironic, given that, as we cover later, plaintiffs’ suppression experts only use data from outside the relevant period to construct their models and inadequately control for the potential impact the financial crisis had on the relationship between LIBOR and the Eurodollar deposit offer rate series. See infra pp. 213-229. Put another way, for the conspiracy analysis, plaintiffs argue that significant volatility within the relevant period requires significant adjustments to avoid an apples-to-oranges comparison, but for the suppression analysis, plaintiffs contend that this same significant volatility can be controlled by adjustments using out-of-sample estimates of various explanatory variables.In any event, we do not credit plaintiffs’ experts’ suppression models, see infra pp. 185-238, so we do not address defendants’ arguments that the results of plaintiffs’ econometric models fail to show parallel suppression, see Defs. Br. at 76-78; Pls. Br. at 77-78.

68.   See also ECF No. 4160-4 (Cragg July 25, 2024 Dep. Tr.) at 66:2-7 (“Q. Is it true that certain banks’ submissions were in the pack some days and outside the pack on other days? A. Yes. As I recall from my analysis of clustering and outliers, banks varied as to where they were located in the distribution.”).

69.   Both parties attempt to draw broad conclusions from narrow sample sizes. Plaintiffs assert that parallel conduct can be found in an event study which purports to show that defendants’ “submissions increased in parallel following news articles published on April 16, 2008, [which] question[ed] [d]efendants’ LIBOR submissions.” Pls. Br. at 79-80 (emphasis added); see also Defs. Reply Br. at 39-40 (similar). While this asserted evidence obviously does not support a suppression conspiracy, the presence of some divergent conduct “does not negate the allegations of other, parallel conduct.” Iowa Pub. Employees’ Ret. Sys. v. Merrill Lynch, Pierce, Fenner & Smith Inc., 340 F. Supp. 3d 285, 320 (S.D.N.Y. 2018). Similarly, defendants’ proffered evidence that the panel banks’ 3M submissions in February 2009 “shift[ed] by different magnitudes—and even different directions,” Defs. Br. at 79-80 (citations omitted), is too small of a sample size to negate “the allegations of other parallel conduct,” Iowa Pub. Employees’ Ret. Sys., 340 F. Supp. 3d at 320.

70.   Though we review this evidence within the “plus factor” framework, perhaps a more appropriate term for this evidence is a “minus factor,” as it does not lead to an inference of collusion.

71.   This element is analogous to the “object” of a conspiracy. In proving a conspiracy, plaintiffs bear the burden of establishing that the alleged co-conspirators took specific acts in furtherance of an agreed-upon goal. See Monsanto, 465 U.S. at 768 (noting evidence must “reasonably tend[ ] to prove that the [defendants] had a conscious commitment to a common scheme designed to achieve an unlawful objective”); cf. United States v. Salameh, 152 F.3d 88, 145 (2d Cir. 1998) (“It is well settled that the essential elements of the crime of conspiracy” include a conspirator's “knowing[ ] participat[ion] in the conspiracy with the specific intent to commit the offenses that were the objects of the conspiracy” and that the problematic conduct “was committed ․ in furtherance of the objectives of the conspiracy.”).

72.   As noted earlier, DAPs assert in their responses to contention interrogatories that “Defendants also recognized that they could realize profits from lower LIBOR submissions than they otherwise would have, and that higher LIBORs could negatively impact their financial positions.” DAPs’ Interrogatory Responses, Resp. No. 9 at 150. We understand this to be a reference to defendants’ asserted motive of projecting financial strength. See LIBOR VI, 2016 WL 7378980, at *5, slip op. at *11 (“The only conclusion to be drawn is that the Circuit meant ‘increased profits and the projection of financial soundness’ to describe collectively a single, reputation-based motive to conspire, where increased profits followed from a positive reputation.”). Otherwise, it is implausible. See id. at *2, slip op. at *5 (“[T]he premise that the primary goal of the conspiracy was to increase profits by lowering the interest rate the banks had to pay when they were in the role of borrower is not plausible[.]”); Gelboim, 823 F.3d at 783 (“It seems strange that this or that bank (or any bank) would conspire to gain, as a borrower, profits that would be offset by a parity of losses it would suffer as a lender.”).

73.   Plaintiffs mistakenly attribute this quote to an unidentified “HSBC submitter[ ],” Pls. Br. at 14 (citing ECF No. 4281-13 (PX61)), however, this quote is from an email by BBA's Director, John Ewan, to an HSBC submitter, among others, see ECF No. 4281-13 (PX61) at 1-2.

74.   At the motion to dismiss stage, the Second Circuit held: “The Banks argue that the ‘pack’ behavior described in the complaints is equally consistent with parallelism. Maybe; but at the motion-to-dismiss stage, appellants must only put forth sufficient factual matter to plausibly suggest an inference of conspiracy, even if the facts are susceptible to an equally likely interpretation.” Gelboim, 823 F.3d at 782 (citation omitted; emphasis in original). Now that the parties are at the motion for summary judgment stage, plaintiffs can no longer rely upon evidence that is “equally consistent with independent conduct as with illegal conspiracy.” Apple, Inc., 791 F.3d at 315 (citation omitted).

75.   We observe that “the BBA was indisputably not a member of the ‘inter-bank conspiracy’ upheld in Gelboim.” In re LIBOR-Based Fin. Instruments Antitrust Litig., No. 11 MDL 2262 (NRB), 2022 WL 17082626, at *2 (S.D.N.Y. Nov. 18, 2022) (citation omitted). “The BBA is not a financial institution whose main concern is to project financial soundness, and any act of assurance that the BBA allegedly took did not further the alleged conspiracy.” LIBOR VIII, 2019 WL 1331830, at *15, slip op. at *39. The same analysis applies to the FXMMC. “[T]he Gelboim conspiracy is limited to the 16 panel banks.” In re LIBOR-Based Fin. Instruments Antitrust Litig., No. 11 MDL 2262 (NRB), 2022 WL 17082626, at *3 (S.D.N.Y. Nov. 18, 2022).

76.   As discussed earlier, Dr. Marx insists that there is only “one answer” to the LIBOR question that would be “rule-abiding,” ECF No. 4154-5 (Marx August 26, 2024 Dep. Tr.) at 266:2-13, when the Circuit in Connolly held “there was no one true rate that [a Panel Bank] was required to submit” in response to the LIBOR question, 24 F.4th at 837.

77.   This determination is based on the record as a whole, which is voluminous, as the parties submitted over thirty boxes of evidence to this Court. If there were any pieces of the record that support plaintiffs’ arguments, it is plaintiffs’ counsel's responsibility, not the Court's, to identify them, given that “ ‘[j]udges are not like pigs, hunting for truffles buried in the record.’ ” Gonzalez v. K-Mart Corp., 585 F. Supp. 2d 501, 503 (S.D.N.Y. 2008) (quoting Albrechtsen v. Bd. of Regents of the Univ. of Wis. Sys., 309 F.3d 433, 436 (7th Cir. 2002)).

78.   The broker communications and direct interfirm communications cited in this Memorandum and Order are not exhaustive of those that appear in the parties’ voluminous papers. The Court has, however, considered all communications set forth in plaintiffs’ briefing and Rule 56.1 Statements.The cited communications are replete with spelling, grammatical, and punctuation errors. To improve the readability of the quoted communications, the term “[sic]” is not used after each error. Similarly, emphasis has been added to the cited communications unless otherwise indicated below.

79.   See also City of Moundridge v. Exxon Mobil Corp., No. CIV.A. 04-940(RWR), 2009 WL 5385975, at *9 (D.D.C. Sept. 30, 2009) (“Evidence that competitors had contact with each other or exchanged information does not establish a conspiracy.”), aff'd sub nom. City of Moundridge, KS v. Exxon Mobil Corp., 409 F. App'x 362 (D.C. Cir. 2011).

80.   “A facilitating practice is an activity that makes it easier for parties to coordinate pricing or their behavior in an anticompetitive way [and] increases the likelihood of a consequence offensive to antitrust policy.” In re Domestic Drywall Antitrust Litigation, 163 F. Supp. 3d 175, 241 n.49 (E.D. Pa. 2016) (citation and quotation marks omitted). Notably, this label “is only an invitation to further analysis, not a license for automatic condemnation.” Id. (citation and quotation marks omitted).

81.   While plaintiffs argue that “[d]etermining the purpose behind the communications by third parties within the framework of the ongoing communications by Defendants is an issue appropriate for a factfinder, not the court, on summary judgment,” Pls. Br. at 96 (citation omitted), the case they cite for this proposition, In re Domestic Airline Travel Antitrust Litig., 691 F. Supp. 3d 175 (D.D.C. 2023), stands for the unremarkable proposition that a court must deny summary judgment where “there are disputed issues of material fact,” id. at 209. We note that, in that case, the Court found that “[p]laintiffs ha[d] gone beyond ‘mere contacts, communications, and memberships’ and presented some examples illustrating that [m]oving [d]efendants may have used third-party intermediaries as conduits to facilitate conspiratorial communications.” Id.

82.   See also DAPs’ Interrogatory Responses, Resp. No. 4 at 106 (“The Interdealer Brokers acted as facilitators, conduits, and intermediaries to effectuate the Conspiracy beginning on August 1, 2007.”); OTC Plaintiffs’ Interrogatory Responses, Resp. No. 4 at 108 (“The Interdealer Brokers were co-conspirators, agents, facilitators, conduits, and/or intermediaries to effectuate the Panel Banks’ Conspiracy beginning no later than August 2007.”).

83.   See also OTC Plaintiffs’ Interrogatory Responses, Resp. No. 16 at 170 (“[OTC] Plaintiffs do not contend that all communications between Panel Banks and Interdealer Brokers were acts in furtherance of the Conspiracy.” (emphasis in original)); DAPs’ Interrogatory Responses, Resp. No. 16 at 165 (same).

84.   FFFP plaintiffs admitted that their expert, Dr. Marx, “testified that multiple interdealer brokers shared their predictions of the USD LIBOR fix with Panel Banks and other customers as a matter of practice before, during, and after the Relevant Period.” Defs. 56.1 ¶ 196 Resp. OTC Plaintiffs took no position on Dr. Marx's testimony and disputed that their expert, Dr. Cragg, “also testified that interdealer brokers interacted in the ordinary course of business with traders from Panel Banks.” Id. ¶¶ 196, 197 Resps.

85.   The decision not to depose any broker was reached despite the fact that in plaintiffs’ request to the Court, plaintiffs’ counsel represented that these brokers had “relevant personal knowledge” and refused to limit the depositions of them “to less than five hours” because it “would prejudice Plaintiffs’ ability to develop the evidence that they will introduce in opposition to summary judgment motions and at trial.” ECF No. 3822 at 3. Not so, apparently.

86.   Whether defendants use 715 days in the FFFP relevant period or 527 days in the OTC Class Period, Defs. 56.1 ¶ 13(c); Defs. Reply Br. at 22, the bottom line remains the same: the citations only reflect a small fraction of the days at issue.

87.   The Court uses an extremely simple formula to arrive at these two figures. The Court multiplies 715 (the number of days in the FFFP relevant period) by 16 (the number of Panel Banks), which totals 11440 possible communications by each bank (assuming that the banks communicated to the brokers once each day, a highly conservative assumption). Dividing 367 (the number of communications listed in Appendix A) by 11440 equals 3.2%. The Court followed the same process to calculate the appropriate figure for the OTC Class Period, which has 527 days. Additionally, the Court notes that these figures would become even more infrequent if we factor in that there were multiple individuals at each Panel Bank who communicated with multiple brokers, sometimes multiple times a day.

88.   This piece of evidence is misleadingly quoted in plaintiffs’ brief, though plaintiffs provide fuller context in their Appendix. Johnson says “I think my LIBORS are too low, but ․ they're way above anybody else's” and “I think I'm the only one being honest.” App. A (entry 20). He also says “I have managed, single-handedly, to get one month LIBOR up twenty two basis points and three months up twelve basis points in the space of a week.” Id. His desire to get the LIBOR fix “up” is plainly inconsistent with plaintiffs’ asserted suppression conspiracy.

89.   As we noted previously, this communication shows that “the banks universally felt pressure—without need for a conspiracy—to keep their ‘head[s] below the parapet.’ ” LIBOR IV, 2015 WL 6243526, at *47, slip op. at *117 (citation omitted). “The Barclays employee's metaphor is apt. The defender of a medieval castle did not shield his head behind a parapet out of an agreement with his comrades to do so, but because he wished for his own sake to avoid his enemy's arrows.” Id. at *47 n.82, slip op. at *117 n.82.

90.   Ms. Julie Doherty, who was one of UBS's LIBOR submitters and now resides in Switzerland, declined to appear voluntarily for deposition in the United States, and the Court granted UBS's application for a protective order barring plaintiffs from proceeding with a deposition in New York. ECF No. 3918. However, in the same order, the Court granted plaintiffs leave to file Letters of Request pursuant to the Hague Convention. Id. Nonetheless, plaintiffs did not obtain testimony from Ms. Doherty. See ECF No. 4169 (Declaration of Jefferson E. Bell) ¶ 7 (listing four fact witnesses from UBS, none of whom are Ms. Doherty).

91.   For clarity, the Court uses the titles of the employees rather than their names in this section.

92.   Defendants dispute this assertion, contending that plaintiffs “cited documents and testimony [which] concern only a small fraction of the more than 700 submission days during the Relevant Period.” Pls. 56.1 ¶ 164 Resp. Whether defendants use 715 days in the FFFP relevant period or 527 days in the OTC Class Period, Defs. 56.1 ¶ 13(c); Defs. Reply Br. at 22, the bottom-line remains the same: the citations only reflect a small fraction of the days at issue.Plaintiffs also rely on Dr. Marx's opinion that brokers “could facilitate exchange of information in support of collusive conduct” because “broker indications were predictive of future LIBOR fixes.” Marx Opening Report ¶ 257 (emphasis added); see also Pls. Br. at 103-04 (citing Dr. Marx's opinions). Dr. Marx's analyses contribute little. Even assuming Dr. Marx's opinion is accurate, “[t]he mere opportunity to conspire does not by itself support the inference that such an illegal combination actually occurred.” Cap. Imaging Assocs., P.C. v. Mohawk Valley Med. Assocs., Inc., 996 F.2d 537, 545 (2d Cir. 1993). Instead, plaintiffs must supply the necessary factual record demonstrating that the brokers did facilitate exchange of information in support of collusive conduct.

93.   Defendants contend that this testimony in fact undermines plaintiffs’ theory, as it shows that brokers were simply observing activity in the London interbank market, i.e. indications, as opposed to providing information about where Panel Banks were planning to submit their LIBORs. See Pls. 56.1 ¶ 174 Resp. (citing ECF No. 4527-27 (Sheehan Dep. Tr. 186:7-17) (“Q. And brokers like Mr Freese told you the rates that other panel banks were considering for their LIBOR submissions, correct? A. His expectation. Q. So brokers like Mr Freese told you their expectations of what other banks were considering for their LIBOR submissions? A. That's what it seems like.” (objections omitted))).

94.   Some of the cited record cannot even qualify as a scintilla of evidence. For example, plaintiffs cite deposition testimony from Bank of America's submitter, Ms. Louisa Sans, who testified that she “[k]new it was possible” that “information the interdealer brokers were providing [her] came from other panel banks directly.” Pls. 56.1 ¶ 175 n.253 (citing ECF No. 4304-41 (Sans Dep. Tr. 70:2-13)). This requires multiple inferences to even count as evidence in plaintiffs’ favor: (1) first, that the “information the interdealer brokers were providing” to her was not permissible market color, but rather from the other defendants’ planned submissions; (2) next, that Ms. Sans could decipher when the information being provided to her was from other defendants’ planned submissions and not permissible market color; and (3) third, that Ms. Sans was able to make this distinction with enough frequency that she could regularly make Bank of America's submissions in line with the common agreement previously struck with fifteen other banks.

95.   This discussion assumes, arguendo, that plaintiffs’ experts have accurately established that Panel Banks’ submissions were based on broker indications, and not market factors. However, defendants vigorously dispute this premise, arguing that broker indications were frequently “wrong, differed from the indications provided by other brokers, differed at different times on the same day, and/or concerned only certain tenors.” Defs. Reply Br. at 52 (citing ECF No. 4304-28 (Marx June 25, 2024 Dep. Tr.) at 218:6-9 (“Q. Did you consider the fact that different brokers offered different indications on the same day to different panel banks? A. I'm aware of that.”)). Dr. Marx admitted that the predictions of LIBOR fixes often “varied from broker to broker,” ECF No. 4175-29 (Marx Aug. 26, 2024 Dep. Tr.) at 116:7-12, 117:23-118:4; Defs. 56.1 ¶ 357, and that different Panel Banks received different predictions from different brokers, ECF No. 4175-24 (Marx June 25, 2024 Dep. Tr.) at 123:9-18; Defs. 56.1 ¶ 357, nor could she explain how Panel Banks would have known which broker prediction was to govern the conspiracy on any given day to use as a focal point. She speculated that Panel Banks could have just “bung[ed] in the number provided [ ] by the broker” but admitted she did not comprehensively test whether Panel Banks did so, ECF No. 4175-24 (Marx June 25, 2024 Dep. Tr.) at 256:2-258:13, and, tellingly, admitted that Panel Banks’ “submissions d[id] not always follow the broker indications.” And 64% of the predicted Panel Bank submissions in these communications (62 of 97) turned out to be wrong. See ECF No. 4175-25 (Asker FFFP Rebuttal Report) ¶ 64. Dr. Marx again proffered no explanation as to how Panel Banks could have used such inaccurate predictions to coordinate their submissions. Id. ¶¶ 62, 64.

96.   These are not the only words that plaintiffs latch on to. Plaintiffs wildly exaggerate the significance of a single communication referring to the panel banks as a “community” and a “syndicate.” Pls. 56.1 ¶ 171(d) (citing ECF No. 4329-22). This is risible if one actually looks at the communication: ECF No. 4329-22 at 00262 (“WSJ running an article asking how lm Libor can be 3.20ish and the 28 day TAF 3.75. Hard to see how the Libor setting community can ignore this”); id. at 00418 (“The BBA tinkering with the Libor definition I do not see as changing things, it is more a move to prevent Banks leaving the setting syndicates at a time when they are trying to get more Banks not less. Libors should be higher.”). This singular communication, discussing the potential for banks to leave the LIBOR-setting process, comes nowhere close to the nefarious meaning implied by plaintiffs.

97.   Ironically, plaintiffs supply this evidence throughout their papers. As an example, plaintiffs cite an internal email from a BTMU manager in which he states: “[A]s you know we struggle to fund the USD book at USD Libor currently. However we always maintain a Libor in line with other fixers to protect reputation of BTMU, etc.” Pls. Br. at 15 n.20 (quoting ECF No. 4318-33 (PX1137)). Notably, the manager does not say he is concerned about the reputation of the entire industry, only of his own firm.

98.   Plaintiffs’ expert, Dr. Marx, admits that Panel Banks could consider the rates at which other banks borrowed or loaned US Dollars, as well as brokers’ LIBOR indications, when making LIBOR submissions. See ECF No. 4304-29 (Marx Aug. 26, 2024 Dep. Tr.) at 98:15-21 (“Q. Was a panel bank permitted to take into account the rate at which another panel bank entered into a borrowing transaction in reasonable market size in London the same day shortly prior to 11:00 a.m.? A. I don't have any reason to think they wouldn't be able to take that into account.”); id. at 100:5-12 (“A. My understanding is they could take into account broker indications that come to them. Q. So if, for example, a panel bank regularly received an email, a broker run-through of LIBOR indications, that would be permissible for the panel bank to take into account? A. I think they could take it into account.”); see also ECF No. 4175-62 (Cragg Sept. 11, 2024 Dep. Tr.) at 247:6-12 (“Q: So – so if a bank received historical information about bids, offers and transactions from a broker but submitted a LIBOR that was consistent with its own offer rate, that would be permissible? A: Yes, I mean, so long as it's doing that independently.”).

99.   See ECF No. 4160-4 (Cragg July 25, 2024 Dep. Tr.) at 66:22-24 (“Q. Could the banks have suppressed LIBOR and been in a pack without colluding? A. Yes.”); ECF No. 4304-24 (Marx June 25, 2024 Dep. Tr.) at 89:10–90:8 (“Q. An individual panel bank decides that it wants to make a submission that is not an outlier submission. Are you with me? A. Okay. Q. That individual panel bank could do so without forming an agreement with other banks not to make outlier submissions, correct? A. I think that's correct. There was a plethora of information that would allow the banks to make submissions within the pack.”)

100.   Judge Schofield ultimately denied both parties’ motions for summary judgment based on her finding that “[v]iewing the overall picture, [p]laintiffs’ evidence tends to exclude the possibility that the [d]efendant banks acted independently, but the parties dispute the material facts of the scope of the purported conspiracy.” Id. at *11, 12. The obvious difference between our case and Judge Schofield's is that plaintiffs here have failed to provide sufficient evidence that tends to exclude the possibility that the defendant banks acted independently.

101.   As noted above, the parties often include inappropriate objections in their 56.1 statements. Here, though plaintiffs admit that defendants’ assertion is consistent with the cited deposition testimony, plaintiffs dispute the inferences that can be drawn from this assertion. Defs. 56.1 ¶ 209 Resp. These disputes “do not create a genuinely disputed fact,” Tang Cap. Partners, 757 F. Supp. 3d at 373 (collecting cases), and violate the “purpose of a [Local Rule] 56.1 response” which is “simply to advise the Court as to whether the specific fact asserted by the moving party is or is not disputed, and if it is disputed, to provide the Court with the evidence on which the non-moving party relies to dispute that particular fact. It presents no occasion for context, argument, semantic quibbles, opinions or conclusions.” Tripathy, 2024 WL 2135623, at *2; see also Kesner, 590 F. Supp. 3d at 691 (setting aside “[t]he portions of [responding party's] 56.1 Statement that contain legal argument bereft of factual matter”). The same principle applies to the other examples listed in this section where plaintiffs admit the facts but argue the law in their 56.1 responses.

102.   As noted earlier, Credit Suisse, along with three other defendant-banks, submitted supplemental memoranda of law accompanying the motion for summary judgment. See ECF No. 4174. Plaintiffs endeavored to rebut Credit Suisse's arguments in their joint opposition to defendants’ summary judgment motion. See Pls. Br. at 115-117.

103.   Credit Suisse's USD LIBOR submitter, Mr. Tony Miller, testified that he never “instruct[ed] a broker to communicate with another bank [regarding] what [Credit Suisse] had communicated to that broker about [its] intended US dollar LIBOR submissions.” Defs. 56.1 ¶ 222. Similarly, “Mr. Miller testified that it was not his practice to share where Credit Suisse was going to set its LIBOR submissions, particularly not with other banks.” Id. ¶ 217.

104.   Credit Suisse's backup submitter, Mr. White, “testified that he spoke to brokers to get a market benchmark, not to receive other Panel Banks’ LIBOR settings or to share Credit Suisse's with other banks.” Defs. 56.1 ¶ 221. Mr. Miller and backup submitter, Alan White, “explained that they would occasionally consider the ‘consensus’ view of brokers” and Mr. White testified that “[i]t might be, on occasions, in answering your questions, that we did go with the broker consensus.” Id. ¶ 216.

105.   Mr. Tony Miller testified that he never “ask[ed] a broker what another [Panel Bank] was going to be submitting.” Defs. 56.1 ¶ 222. Similarly, he “testified that he never asked what another Panel Bank was going to submit for USD LIBOR because ‘it wasn't pertinent to our way of thinking ․ [b]ecause the criteria that [Credit Suisse] used for the LIBOR submissions didn't really take notice of what other panel banks were doing.’ ” Id. ¶ 215. He “confirmed that he (i) never entered into an understanding or agreement with any other USD LIBOR Panel Bank to submit artificially low LIBOR rates at any time; and (ii) was not aware of anyone else at Credit Suisse entering into such an understanding or agreement.” Id. ¶ 218.

106.   There are material gaps in plaintiffs’ citations. In plaintiffs’ cited exhibit, PX1290, ECF No. 4323-1, Credit Suisse's submitter writes in an email: “USD libors, a combination of lack of liquidity and desire to be within some sort of consensus leaves me setting 1m 4.72 2&3m 4.93 6m 4.82 12m 4.52.” However, the exhibit does not specify the recipient of the email, despite plaintiffs’ assertion that Credit Suisse “asked those brokers for the ‘consensus.’ ” Pls. Br. at 115-116. This information cannot be found in plaintiffs’ 56.1 statement either. Indeed, Mr. Miller's “best recollection” was that “this email [was] internal to Credit Suisse.” ECF No. 4527-21 (Miller Dep. Tr.) at 137:22-138:1. As noted elsewhere, plaintiffs cannot “simply dump papers on the court and expect the court to sift through them to determine if some nugget is buried somewhere in that mountain.” Emanuel, 2022 WL 3084317, at *5 (citation and quotation marks omitted). It is the responsibility of plaintiffs, not the Court, to “present evidence sufficient to satisfy every element of the[ir] claim.” Holcomb, 521 F.3d at 137.

107.   As noted earlier, this litigation focuses solely on U.S. Dollar LIBOR. However, rates contributed pursuant to the BBA LIBOR Instruction included rates for multiple currencies, as well as rates relating to loans of various maturities and fixing dates. See, e.g., Defs. 56.1 ¶ 15.

108.   The issue of whether a plaintiff could bring a claim against the BBA was thoroughly litigated in this MDL. The FFFP plaintiffs’ complaints, but not the OTC plaintiffs’ complaint, included the BBA as a defendant. See Third Am. Compl., Fed. Nat'l Mortg. Ass'n v. Barclays Bank PLC, et al., 1:13-cv-7720-NRB (S.D.N.Y. Sept. 2, 2022), ECF No. 246 (“Fannie Mae Compl.”); Am. Compl., Fed. Deposit Ins. Corp. v. Bank of Am. Corp., et al., 1:14-cv-1757-NRB (S.D.N.Y. Oct. 6, 2014), ECF No. 23 (“FDIC Compl.”); Revised Third Am. Compl., Fed. Home Loan Mortg. Corp. v. Bank of Am. Corp., 1:13-cv-03952-NRB (S.D.N.Y. July 8, 2019), ECF No. 333 (“Freddie Mac Compl.”); Corrected Fourth Consolidated Am. Compl., Mayor & City Council of Balt., et al., v. Credit Suisse AG, 1:11-md-2262-NRB (S.D.N.Y. Oct. 12, 2022), ECF No. 3555 (“OTC Compl.”); Revised Second Am. Compl., Principal Fin. Grp., Inc. et al. v. Bank of Am. Corp., et al., 1:11-md-2262-NRB (S.D.N.Y. July 8, 2019), ECF No. 2898 (“Principal Financial Compl.”); Rev. Second Am. Compl., Principal Funds, Inc., et al. v. Bank of Am. Corp., et al., 1:11-md-2262-NRB (S.D.N.Y. July 10, 2019), ECF No. 2910 (“Principal Funds Compl.”). None of the operative complaints remaining in this litigation includes the FXMMC as a defendant. See id.As we have stated before, “the BBA was indisputably not a member of the ‘inter-bank conspiracy’ upheld in Gelboim.” In re LIBOR-Based Fin. Instruments Antitrust Litig., No. 11 MDL 2262 (NRB), 2022 WL 17082626, at *2 (S.D.N.Y. Nov. 18, 2022) (citation omitted). “The BBA is not a financial institution whose main concern is to project financial soundness, and any act of assurance that the BBA allegedly took did not further the alleged conspiracy.” LIBOR VIII, 2019 WL 1331830, at *15, slip op. at *39. The same analysis applies to the FXMMC. “[T]he Gelboim conspiracy is limited to the 16 panel banks.” In re LIBOR-Based Fin. Instruments Antitrust Litig., No. 11 MDL 2262 (NRB), 2022 WL 17082626, at *3 (S.D.N.Y. Nov. 18, 2022).In LIBOR VIII, the Court rejected an attempt by plaintiffs Freddie Mac, FDIC, Principal Financial, and Principal Funds “to cast the BBA as a member of the plausibly pled conspiracy” on the merits and on personal jurisdiction grounds. LIBOR VIII, 2019 WL 1331830, at *15, slip op. at *39. Despite the fact that no appeal from LIBOR VIII was ever filed, plaintiffs argued in 2022 that the Second Circuit's decisions in Schwab II and Berkshire reversed this Court's holding in LIBOR VIII. See In re LIBOR-Based Fin. Instruments Antitrust Litig., No. 11 MDL 2262 (NRB), 2022 WL 17082626 (S.D.N.Y. Nov. 18, 2022). We found that this argument was “without foundation and ․ analytically flawed.” Id. at *1. Plaintiffs then sought interlocutory review of LIBOR VIII, four years after this Court issued its decision, and this Court denied plaintiffs’ applications for interlocutory review. ECF No. 3634.

109.   If there were any such communications, it is plaintiffs’ counsel's responsibility, not the Court's, to identify them, given that “[j]udges are not like pigs, hunting for truffles buried in the record.” Gonzalez, 585 F. Supp. 2d at 503 (citation and quotation marks omitted).

110.   Similarly, plaintiffs assert that: “FXMMC's meetings increased in frequency during the financial crisis.” Pls. 56.1 ¶ 55 Resp.; see also Pls. Br. at 102 (similar). This unsurprising fact does not help plaintiffs, as a “mere showing of close relations or frequent meetings between the alleged conspirators ․ will not sustain a plaintiff's burden absent evidence which would permit the inference that these close ties led to an illegal agreement.” H.L. Moore, 662 F.2d at 941 (citation omitted); see also AD/SAT, 181 F.3d at 234 (“[A]lthough the nature of trade associations is such that they are frequently the object of antitrust scrutiny, every action by a trade association is not concerted action by the association's members.” (citation omitted)). Moreover, increasing the frequency of FXMMC meetings during a period of financial uncertainty or strain is equally consistent with a non-nefarious purpose and, indeed, should be expected.

111.   Plaintiffs attempt to muddy the record on this point with baseless objections. See Defs. 56.1 ¶ 28 Resp. Plaintiffs dispute that “non-Panel Bank employees were members of the FXMMC throughout the Relevant Period” without a citation to an exhibit or fact, but concede that “the Association of Corporate Treasurers attended an FXMMC meeting on March 23, 2009” and do not address defendants’ assertion that the FXMMC included meeting participants such as “the London Money Market Association, and Abbey National.” Id. Rule 56.1 “does [not] contemplate creating more or less than an admission or a denial of the truth of the allegation for the purposes of the motion.” Emanuel, 2022 WL 3084317, at *4 (emphasis added). “In order to defeat a properly supported summary judgment motion, the opposing party must proffer admissible evidence that ‘set[s] forth specific facts’ showing a genuinely disputed factual issue that is material under the applicable legal principles.” Major League Baseball Props, 542 F.3d at 310 (citation omitted). Not only are plaintiffs’ objections insufficient to create a genuine dispute of material fact, but “[a]fter reviewing the cited [evidence], the Court deems this factual assertion admitted, insofar as what is asserted ․ is supported by the evidence in the record.” Satanic Temple, Inc. v. Newsweek Mag. LLC, 774 F. Supp. 3d 688, 693 n.3 (S.D.N.Y. 2025) (citations omitted).

112.   An innocuous, yet typical, example of plaintiffs’ imprecision follows: while plaintiffs assert that “On May 19, 2008, the BBA held a meeting” that “[r]epresentatives of Bank of America, Barclays, Deutsche Bank, HBOS, Lloyds, and [Natwest] attended,” Pls. 56.1 ¶ 91, their cited documents only reference the banks who were invited to the meeting, not who attended, see id. ¶ 91 n.144.Plaintiffs also make arguments based on out-of-context quotations from exhibits. For instance, plaintiffs argue that “the BBA was complicit, at one point seeking to destroy an email that referred to its ‘colluding members’ after the orchestrated effort to float LIBORs up in response to media scrutiny in April 2008.” Pls. Br. at 106 (citing Pls. 56.1 ¶ 221). However, we must look to the underlying context. In the underlying document, ECF No. 4310-54 (PX1017), Ewan, the BBA's LIBOR director forwards a Bloomberg article which states that the “cost of borrowing in dollars for three months rose the most since August after the British Bankers’ Association threatened yesterday to ban members that deliberately understate their borrowing costs,” id. at 2. Ewan's supervisor, Mr. Merriman, replies from his BlackBerry device, saying “Not an altogether negative story. Think again: we have done something positive to shake up the market, no matter what our – colluding – members think.” Id. Mr. Ewan replies “I think you might want to ask Richard or Dave to retract and completely delete the email below․ I am paranoid about this, I know, but as we've seen lately, reputations can be tarnished without evidence at the moment.” Id. A plain reading of this document does not unambiguously demonstrate that the BBA was “complicit” in any conspiracy, and an errant reference to “colluding members” is not a smoking gun. If this exchange has any actual significance, it is evidence that the BBA was not part of any alleged conspiracy.Moreover, plaintiffs’ proffered evidence, as whole, tends to establish that the BBA, FXMMC, and Steering Group did not operate to facilitate a conspiracy. For example, while plaintiffs assert that the BBA and FXMMC were part of the conspiracy, see OTC Plaintiffs’ Interrogatory Responses, Resp. No. 1 at 21; DAPs’ Interrogatory Responses, Resp. No. 2 at 66, plaintiffs cite Mr. Ewan's witness statement that “[b]y 2009, the concerns surrounding LIBOR were such that the FXMMC sought to further tighten the monitoring regime,” Pls. 56.1 ¶ 235 n.482 (quoting ECF No. 4310-55 (PX1019 (Statement of Witness [John Ewan] ¶ 43))).

113.   We separately address plaintiffs’ arguments concerning defendants’ alleged concealment efforts. See infra pp. 166-169.

114.   See also Tera Grp., Inc. v. Citigroup, Inc., No. 24-135-CV, 2024 WL 4501967, at *4 (2d Cir. Oct. 16, 2024) (holding at the pleading stage that the “mere opportunity to conspire does not by itself support the inference that such an illegal combination actually occurred” (citation and quotation marks omitted)); In re London Silver Fixing, Ltd., Antitrust Litig., 213 F. Supp. 3d 530, 560 (S.D.N.Y. 2016) (“[E]ven at the pleading stage, the structure of the Silver Fixing does not constitute a ‘plus factor’ in support of Plaintiffs’ claims.” (citation omitted)); but see Mosaic Health, Inc. v. Sanofi-Aventis U.S., LLC, 147 F.4th 183, 188, 199 (2d Cir. 2025) (reversing dismissal of complaint based, in part, on “the low pleading threshold for surviving a motion to dismiss” and “the inference that the [d]efendants’ sharing of lobbying services and joint participation on the PhRMA board suggests that the [d]efendants had ample opportunity to conspire”).

115.   Given the repeated holdings in the Second Circuit rejecting market structure's susceptibility to collusion as a plus factor, plaintiffs’ reliance on a Seventh Circuit case which may be understood as holding to the contrary does not advance plaintiffs’ argument.

116.   Plaintiffs also rely on the purported testimony of BBA LIBOR Director Ewan who stated that he was often put in a “difficult position in terms of enforcing adherence to the LIBOR definition.” Pls. Br. at 105 (citing Pls. 56.1 ¶ 154, § V.A). In another instance of plaintiffs’ imprecision, neither of the two cites in plaintiffs’ 56.1 statement supports plaintiffs’ assertion. The cited 56.1 statement paragraph does not relate to Mr. Ewan's testimony about a potential conflict of interest. See Pls. 56.1 ¶ 154 n.227. Likewise, the cited section in the 56.1 statement, Pls. 56.1 § V.A, also does not have this testimony. Regardless, even assuming that Mr. Ewan did testify that he was put in a “difficult position in terms of enforcing adherence to the LIBOR definition,” Pls. Br. at 105 (citations omitted), this fact is not material for the reasons listed above.

117.   The Wheatley Review, The Wheatley Review of LIBOR: final report (Sept. 2012) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/191762/wheatley_review_libor_finalreport_280912. pdf (the “Wheatley Report”). This report, conducted in 2012 by the director of England's Financial Services Authority, found that the LIBOR structure presented a conflict of interest because “ ‘the BBA acts as the lobby organisation for the same submitting banks that they nominally oversee.’ ” Pls. Br. at 47-48, 104-105 (citing the Wheatley Report).

118.   Deutsche Bank's submitters “admitted to including ‘CYA’ (‘Cover Your Ass’) language in emails discussing Deutsche Bank's LIBOR submissions to make it look as if they were setting them appropriately.” Pls. 56.1 ¶ 223. This admission is afforded little weight because, as noted already, “regulatory and law enforcement agencies concluded that Deutsche Bank traders engaged in opportunistic, up-and-down manipulation and attempted manipulation intended to benefit trading positions.” Defs. 56.1 ¶ 581. Thus, this “CYA” activity did not concern plaintiffs’ asserted sixteen bank, multi-year conspiracy. See LIBOR VI, 2016 WL 7378980, at *5 & n.8, slip op. at *11 & n.8 (“Profit-motivated trader-based manipulation, which was sporadic and would result in both the inflation and deflation of LIBOR submissions, has nothing to do with the persistent suppression conspiracy that is at issue in the antitrust claims.” (citations omitted)).

119.   As discussed previously, one simply cannot ascribe to the BBA any of the shifting objects of the purported sixteen-bank conspiracy. See LIBOR VIII, 2019 WL 1331830, at *15, slip op. at *39.

120.   Plaintiffs ignore complaints of banks about other banks, which would be contrary to the presumed conspiracy. See supra pp. 118-126; see, e.g., App. A (entry 179) (A Deutsche Bank employee told a Tullett Prebon broker, “You have to talk to these idiots at [Natwest] and fucking Lloyd's ․ If the cash is 290 bid, you should not but a 275 fucking fixing.”).Moreover, Dr. Hubbard's “calculations establish that Defendants’ daily LIBOR submissions overwhelmingly were within or above the range of each Panel Bank's corresponding borrowing rates on submission days with associated borrowing transactions.” See Defs. Br. at 20 (citation omitted).

121.   Defendants assert that this is true for an eighth panel bank, MUFG, Defs. Br. at 25, but their cited evidence does not support this conclusion.

122.   Plaintiffs make a series of legal arguments, masquerading as 56.1 objections to this paragraph, which are not appropriate to consider. Defendants’ 56.1 statement asserts: “No finding or other statements published by the DOJ, CFTC, FCA, or State AGs indicate that Barclays conspired with other USD LIBOR Panel Banks to persistently suppress USD LIBOR.” Defs. 56.1 ¶ 287. Plaintiffs point out that the agencies found that Barclays’ submitters, in certain instances, submitted rates that were in line with other contributing banks. Id. However, these findings are equally consistent with the possibility of independent action, as Barclays could have made submissions in line with other banks without collusion. See supra pp. 95-111.

123.   As noted supra pp. 95-106, these findings also tend to undermine plaintiffs’ asserted common motives to conspire. For instance, “as part of its resolution with the DOJ, Barclays admitted ․ that ‘on some occasions’ between August 2007 and January 2009 Barclays unilaterally made USD LIBOR submissions that were ‘lower than Barclays otherwise would have submitted’ because Barclays ‘sought to avoid inaccurate negative media attention about Barclays’ financial health as a result of its high LIBOR submission relative to other banks.” Defs. 56.1 ¶ 284 (emphasis added).Additionally, government investigations “found that Soc Gen suppressed LIBOR but the time period covered by the investigation is nearly all outside the Relevant Period in this case.” Pls. Br. at 110 n.171.

124.   Plaintiffs’ reliance on these investigations must be viewed in light of the Court's order providing that all documents the Panel Banks provided to regulators be reproduced to plaintiffs. See ECF Nos. 1441, 1461. As a result, the Panel Banks reproduced to plaintiffs “over 3.4 million documents, spanning 15.9 million pages and 88,000 audio files, from over 500 custodians,” generally covering the LIBOR setting process in the “period of August 2007 to May 2010” that they had previously produced to regulators. Defs. 56.1 ¶ 5. Furthermore, plaintiffs had the opportunity to serve additional document requests, resulting in the production of an additional 400,000 documents, take the depositions of 59 current or former Panel Bank employees and three former BBA employees, and serve on defendants “more than 700 interrogatories (inclusive of subparts) and more than 300 requests for admission of facts.” Defs. 56.1 ¶¶ 5, 6, 8. Consequently, our decision today rests upon the totality of the underlying evidence.

125.   As noted earlier, though we review this evidence within the “plus factor” framework, perhaps a more appropriate term for this evidence is a “minus factor,” as it does not lead to an inference of collusion.

126.   At times in their submissions, the parties describe the second Upstream Issue of “the alleged suppression of LIBOR” as whether plaintiffs have proven damages for their claims. See, e.g., Defs. Br. at 102; Pls. Br. at 55-56. We do not address damages, which is a downstream issue, in the current opinion. See ECF No. 3687 (Scheduling Order identifying “the alleged suppression of LIBOR and/or any conspiracy to suppress LIBOR” as “Upstream Issues”).The parties agree that the same principles underlying the injury-in-fact inquiry applies to plaintiffs’ common-law claims. Compare Defs. Br. at 102 (“The Court should grant summary judgment in favor of Defendants on Plaintiffs’ surviving claims, all of which fail in the absence of persistent suppression or a conspiracy.”) with Pls. Br. at 56 n.88 (“Conceptually, the same principles apply to DAPs’ common law claims because they similarly allege persistent LIBOR suppression.”).We note that antitrust injury poses a legal question unique to antitrust claims: “whether any such injury is ‘injury of the type the antitrust laws were intended to prevent and that flows from that which makes defendants’ acts unlawful.’ ” LIBOR VII, 299 F. Supp. 3d at 592, slip op. at *324 (quoting Cordes & Co., 502 F.3d at 106). The Second Circuit has already answered this question in the affirmative as to certain of plaintiffs’ antitrust claims. See Gelboim, 823 F.3d at 772-75. This issue is not disputed by the parties today.

127.   To be more specific, plaintiffs allege they were injured because they “entered into a variety of financial transactions at interest rates that reference LIBOR” and “beginning in 2007, the Banks engaged in a horizontal price-fixing conspiracy, with each submission reporting an artificially low cost of borrowing in order to drive LIBOR down.” Gelboim, 823 F.3d at 765, 766. LIBOR-based financial instruments included: “(1) asset swaps, in which the owner of a bond pegged to a fixed rate pays that fixed rate to a bank or investor while receiving in return a floating rate based on LIBOR; (2) collateralized debt obligations, which are structured asset-backed securities with multiple tranches, the most senior of which pay out at a spread above LIBOR; and (3) forward rate agreements, in which one party receives a fixed interest rate on a principal amount while the counterparty receives interest at the fluctuating LIBOR on the same principal amount at a designated endpoint.” Id. at 765; see also OTC Compl. ¶¶ 40-41 (similar).

128.   Subsequent to the briefing of the pending motions, the Supreme Court of the United Kingdom arrived at a similar conclusion as the Second Circuit in Connolly. On July 23, 2025, the U.K. Supreme Court unanimously reversed the fraud-related convictions of a LIBOR trader, Tom Hayes, and a trader of Euro Inter-Bank Offered Rate (“EURIBOR”), Carlo Palombo. See R (Respondent) v Hayes (Appellant); R (Respondent) v. Palombo (Appellant) [2025] UKSC 29 (LL), ¶¶ 1, 162, 234, 235. The Court held “[t]he question posed by the LIBOR ․ definition[ ] required a bank on the contributor panel of banks to submit the rate at which that bank (in the case of LIBOR), ․ could borrow money at the time of the submission. The identification of this rate required a qualitative assessment of various data sources and was a matter of subjective opinion rather than empirical fact.” Id. ¶ 7. “It was thus in the nature of the exercise that there would on any given day be a range – which might be narrower or wider depending on market conditions – of different rates that could reasonably be selected: the figure submitted depended on the subjective judgment of the submitter.” Id. ¶ 22 (emphasis added).In arriving at this conclusion, the Supreme Court of the United Kingdom explicitly considered and approvingly cited the Second Circuit's Connolly decision. See id. ¶¶ 27, 38, 39, 75, 79, 80, 81, 154. However, the U.K. Supreme Court noted that Connolly did not fully accord with English law because “in English law a submission can be false and fraudulent even though the rate at which the bank could have borrowed funds at the relevant time is a matter of opinion and not verifiable fact,” id. ¶ 80, or stated otherwise, “[i]n submitting a rate for use in setting LIBOR, the person doing so represented that the figure submitted was his or her opinion of the relevant borrowing rate[;] If it was not, the representation was false,” id. ¶ 89. Still, the Court held that the trial court erred by removing Mr. Hayes’ “defence from the jury's consideration” that he “had intended only that, where there was a range of rates which the submitter regarded as equally valid estimates of the rate at which the bank would have borrowed funds, a figure at the high (or low) end of the range should be submitted.” Id. ¶¶ 129, 130.

129.   See, e.g., OTC Compl. ¶ 5 (“By acting together and in concert to knowingly understate their true borrowing costs, the panel banks caused LIBOR to be calculated or suppressed artificially low, reaping hundreds of millions, if not billions, of dollars in ill-gotten gains.” (emphases added)).

130.   We note the Supreme Court of the United Kingdom, which explicitly considered and approvingly cited the Second Circuit's Connolly decision, also uses the term “range”: “It was thus in the nature of the exercise that there would on any given day be a range – which might be narrower or wider depending on market conditions – of different rates that could reasonably be selected: the figure submitted depended on the subjective judgment of the submitter.” R (Respondent) v Hayes (Appellant); R (Respondent) v. Palombo (Appellant) [2025] UKSC 29 (LL), ¶ 22 (emphasis added); see also id. 27, 38, 39, 75, 79, 80, 81, 154 (citing Connolly).

131.   OTC plaintiffs relied upon testimony from Dr. Douglas Bernheim to demonstrate predominance and numerosity at the class certification stage. At that stage, OTC Defendants did not bring a Daubert motion against Dr. Bernheim. See LIBOR VII, 299 F. Supp. 3d at 583, 586, slip op. at *301-302, 309.

132.   Defendants submit five reports from Dr. Hubbard: (1) an initial report, dated April 19, 2024 (Decl. of Patrick W. Blakemore, Ex. 16, October 4, 2024, ECF No. 4153); (2) a report rebutting OTC expert Dr. Bernheim, dated June 18, 2024 (Decl. of Patrick W. Blakemore, Ex. 17, October 4, 2024, ECF No. 4153); (3) a report rebutting FFFP expert Dr. Snow, dated June 18, 2024 (Decl. of Patrick W. Blakemore, Ex. 13, October 4, 2024, ECF No. 4153); (4) a reply report responding to OTC expert Dr. Bernheim, dated August 16, 2024 (Decl. of Patrick W. Blakemore, Ex. 18, October 4, 2024, ECF No. 4153); and (5) a reply report responding to FFFP expert Dr. Snow, dated August 16, 2024 (Decl. of Patrick W. Blakemore, Ex. 8, October 4, 2024, ECF No. 4153). We refer to these as the Hubbard Initial Report, Hubbard OTC Rebuttal Report, Hubbard FFFP Rebuttal Report, Hubbard OTC Reply Report, and Hubbard FFFP Reply Report, respectively.Dr. Hubbard provided testimony at the class certification stage on behalf of a defendant in the Exchange-based action. LIBOR VII, 299 F. Supp. 3d at 473, slip op. at *40-41. Though his expert testimony was challenged, we noted then that “Dr. Hubbard is qualified to offer the challenged opinions” and denied the Exchange plaintiffs’ motion to exclude his opinions. LIBOR VII, 299 F. Supp. 3d at 516, 609, slip op. at *143, 365. Indeed, Dr. Hubbard serves as Dean Emeritus and Russell L. Carson Professor of Economics and Finance at the Columbia Graduate School of Business and has served on the Federal Reserve Bank of New York's Panel of Economic Advisors. Dr. Hubbard holds a Ph.D. and an A.M. in Economics from Harvard University, and has written extensively on various topics in finance and economics.

133.   Dr. B. Douglas Bernheim, Ph.D., is presently the Edward Ames Edmons Professor of Economics at Stanford University, a Senior Fellow of the Stanford Institute of Economic Policy Research, and a Research Associate at the National Bureau of Economic Research. See Bernheim Initial Report App. A. He has previously held positions at Princeton University and Northwestern University. See id.

134.   “According to the U.S. Federal Reserve, ‘Eurodollars are bank deposit liabilities denominated in U.S. dollars but not subject to U.S. banking regulations. For the most part, banks offering Eurodollar deposits are located outside the United States.’ The Eurodollar market originally developed after World War II in Europe; although dollar deposits are now transacted in all major global financial centers, they are still known as Eurodollars.” Bernheim Initial Report ¶ 66 (citations omitted).As will be discussed further below, the brokerage firm “ICAP publishes two daily Eurodollar series, a ‘high’ and a ‘low’ series, representing their brokers’ best estimates of the range within which most Eurodollar transactions by upper tier banks would take place.” Id. ¶ 73. Tullet Prebon, another broker, also reports “two daily Eurodollar series” and Dr. Bernheim uses the Tullet Prebon ask rate, which he contends “is very similar” to ICAP High “for both the 1M and the 3M tenors.” Id. ¶ 75. Bloomberg operates a trading platform and reports two daily Eurodollar series, the best ask for Eurodollar transactions and the best bid. Id. ¶¶ 76, 79. According to Dr. Bernheim, the Bloomberg ask series is “more volatile than either of the offer rates from ICAP and Tullett Prebon” because “Bloomberg does not filter posted quotes or make judgments based on other market information” and “[c]onsequently, the series represent[s] unfiltered market data.” Id. ¶ 80.

135.   Dr. Bernheim estimates the “spread between the LIBOR fix or individual submissions and the Eurodollar deposit rate” in two ways. Bernheim Initial Report ¶¶ 93, 117. First, in his “bottom-up” approach, Dr. Bernheim predicts a but-for LIBOR submission for each tenor and panel bank and then calculates the trimmed mean from the LIBOR formula by “mimic[king] the LIBOR-determination process.” Id. ¶ 117. Second, in his “top-down” approach, he applies his model to the LIBOR fix. Id. Both approaches yield similar predictions. Id.

136.   In his initial report, Dr. Bernheim opines that these “simple comparisons between LIBOR and Eurodollar deposit offer rates,” which he refers to as “unadjusted comparisons,” “provide reasonable approximations of LIBOR suppression.” Bernheim Initial Report § VI.C, ¶ 81. However, in his reply report, Dr. Bernheim reverses himself: “To be clear, I am not suggesting that the differences between [London interbank deposit] offer rates and the ICAP High or any other benchmark Eurodollar deposit offer rate are negligible, or that it is appropriate to reach conclusions about suppression based on unadjusted comparisons between LIBOR and such benchmarks.” Bernheim Reply Report ¶ 40 (emphasis in original).

137.   Dr. Bernheim finds similar results using the top-down estimate. Using the top-down estimate, Dr. Bernheim calculates an average suppression of 20 basis points for the 1M LIBOR fix and 19 basis points for the 3M LIBOR fix. See Bernheim Initial Report ¶ 118.However, in his reply report, Dr. Bernheim also estimates a more conservative specification that uses the midpoint of ICAP High and ICAP Low as a benchmark for LIBOR, which generates suppression of 11 to 12 basis points for the one-month tenor and 7 to 9 basis points for the three-month tenor. Bernheim Reply Report ¶ 367 & Fig. 76.

138.   Specifically, to create this new metric, Dr. Bernheim “uses a regression model explaining the observed spread between non-Panel Prime Banks’ borrowing costs on Eurodollar transactions brokered through ICAP and the ICAP Eurodollar offer rate during the Class Period.” Bernheim Rebuttal Report ¶ 98.

139.   Dr. Karl N. Snow is an Adjunct Professor of Finance at the University of Maryland and holds a Ph.D. in economics from the University of Chicago with an emphasis on finance and econometrics. See Snow Initial Report App. A.

140.   For purposes of the analysis below, our references to the “relevant period” include the OTC Class Period, which is between August 2007 and August 2009, which Dr. Bernheim examined, and the Conduct Period, August 2007 and May 2010, which Dr. Snow tested. Bernheim Initial Report ¶¶ 14, 24; Snow Initial Report ¶ 14.

141.   As explained above, our references to the “relevant period” include the OTC Class Period, which is between August 2007 and August 2009, which Dr. Bernheim examined, and the Conduct Period, August 2007 and May 2010, which Dr. Snow tested. Bernheim Initial Report ¶¶ 14, 24; Snow Initial Report ¶ 14.

142.   “Somewhat confusingly, the ‘before-and-after’ approach” of analyzing antitrust impact, which is separate from the yardstick approach, “is sometimes also referred to as the ‘benchmark’ approach.” Keurig Green Mountain Single-Serve Coffee, 2025 WL 354671, at *10 n.6 (citations omitted). For purposes of our analysis here, we review plaintiffs’ yardstick approach, as plaintiffs admit they did not use a “before or after” approach. See Bernheim/Snow – Pls. Br. at 15-16. “Courts in this Circuit have repeatedly stated that a ‘yardstick’ methodology is an accepted method to measure antitrust impact and damages.” Iowa Pub. Employees’ Ret. Sys. v. Merrill Lynch, Pierce, Fenner & Smith Inc., No. 17 Civ. 6221 (KPF) (SLC), 2024 WL 5004632, at *14 (S.D.N.Y. Dec. 6, 2024) (collecting cases).

144.   Plaintiffs admitted that ICAP served interrogatory responses to this effect. See Defs. 56.1 ¶ 636(d) Resp.

145.   See Snow Initial Report ¶ 264 (“[I]t is possible that riskier banks were a part of the market information used to formulate its Eurodollar rates.”); id. ¶ 263 n.253 (noting that the term “prime bank[s]” includes Panel Banks, but also includes “any bank rated A1/P1 or better that was either A) a top 50 bank as measured by total assets in 2008, [or] B) a bank that had a large amount of Eurodollar transactions in the data produced by ICAP”); Bernheim Reply Report ¶ 267 (explaining “the average CDS spread of Prime Banks may not fully reflect the credit risk of the set of banks associated with the measures of Eurodollar offer rates I consider”); Bernheim Initial Report ¶ 69 (detailing differences between LIBOR and Eurodollar deposit offer rates); see also Defs. 56.1 § IV.B.3 (similar).

146.   See Bernheim Initial Report ¶¶ 131-38; Snow Initial Report ¶¶ 158, 160; Snow Reply Report ¶ 180; ECF No. 4153-7 (Snow June 12, 2024 Dep. Tr.) at 176:5-7, 185:5-8 (Dr. Snow testifying that he uses ICAP High in his baseline model).

147.   See ECF No. 4153-29 (ICAP Cap. Mkts. LLC's Second Resps.) at 2 (“ICAP interpreted the range that it quoted as a high and a low for the expected range for trades done by upper tier banks. Bloomberg's methodology converted that to a bid and an offer, which was no longer a relevant framework for the market.”); see also id. (“The high and low in ICAP's range were both essentially offered rates, but they were indicative offered rates designed to give a sense of where the bulk of activity would take place.”); Defs. 56.1 Resps. ¶ 636(d) (plaintiffs did not dispute ICAP's description of ICAP High and ICAP Low).

148.   See also id. (Dr. Bernheim's “estimates of average suppression are 96 percent lower on average for the one-month tenor and 131 percent lower on average for the three-month tenor” at the individual Panel Bank level.); Hubbard FFFP Rebuttal Report ¶ 83 (“The corresponding percentage changes in Dr. Snow's estimates of alleged LIBOR fix suppression are 88, 69, 100, and 103 percent, respectively, for the one-month, three-month, six-month and 12-month tenors, when using the ICAP Low rate.”).

149.   The footnote in Dr. Bernheim's report quotes the following working paper as stating “the H.15 Eurodollar deposit rate, an offered broker quote reported in the Federal Reserve's H.15 report based on data from ICAP.” Bernheim Reply Report ¶ 70 n.74 (quoting Dennis Kuo, David Skeie, and James Vickery, “A Comparison of LIBOR to Other Measures of Bank Borrowing Costs” (working paper, Federal Reserve Bank of New York, 2012)). This statement does not contradict ICAP's interrogatory responses. In fact, a version of this same working paper published in 2018 undermines Dr. Bernheim's assertion that ICAP High is an appropriate comparator to LIBOR: “ ‘One of the yardsticks that we and other analysts have used to evaluate LIBOR is the Eurodollar deposit rates[.] ․ They reflect the upper end of an indicative run that ICAP furnishes to data-vendors that is intended to capture the bulk of the rates paid by A1/P1 banks at any given time․ The LIBOR fixings have at times appeared to be too low, but picking the upper limit of rates being paid by active participants almost by definition produces a rate that is too high.’ ” Dennis Kuo, David Skeie and James Vickery, “A Comparison of LIBOR to Other Measures of Bank Borrowing Costs” (working paper, Federal Reserve Bank of New York, 2018) at 27 n.36 (quoting “Fed Policy” Research Commentary, Wrightson ICAP, May 2, 2008)).

150.   In his reply report, Dr. Bernheim asserts, without evidence, that “offer rates received by Panel Banks during the Class Period would likely have fallen between the ICAP High and the Low, and closer to the High than to the Low,” and then estimates a more conservative specification that uses the midpoint of ICAP High and ICAP Low as a benchmark for LIBOR, which generates suppression of 11 to 12 basis points for the one-month tenor and 7 to 9 basis points for the three-month tenor. Bernheim Reply Report ¶¶ 366, 367 & Fig. 76. This internally inconsistent result points to the unreliability of Dr. Bernheim's model. See, e.g., Louis Vuitton Malletier v. Dooney & Bourke, Inc., 525 F. Supp. 2d 558, 569 (S.D.N.Y. 2007) (finding “unexplained inconsistency between the results” produced by two iterations of the same methodology to be a basis for exclusion). Moreover, “there is simply too great an analytical gap between the data and the opinion proffered” for this estimate to be considered admissible. Joiner, 522 U.S. at 146.

151.   As explained above, our references to the “relevant period” include the OTC Class Period, which is between August 2007 and August 2009, which Dr. Bernheim examined, and the Conduct Period, August 2007 and May 2010, which Dr. Snow tested. Bernheim Initial Report ¶¶ 14, 24; Snow Initial Report ¶ 14.

152.   Indeed, Dr. Bernheim cites to academic research noting that the divergence occurred following the financial crisis. See Bernheim Initial Report ¶ 68 (“Bonaldi (2024) mentions that ‘Libor followed the EDR [Eurodollar deposits rate] closely since its inception until late 2007, mostly lying a few basis points above it․The Libor-EDR spread becomes negative and relatively large in magnitude around August 2007, when BNP Paribas announced that it had suspended redemption in two subprime mortgage funds. It then drops drastically with Lehman's collapse reaching its lowest historical value (approx. -200 bp). As the crisis evolves, this spread narrows but remains negative and large in magnitude compared to its pre-crisis levels.’ Similarly, an article published by the International Monetary Fund mentions that LIBOR has traditionally tracked similar rates, but ‘[t]he glaring exception was the period immediately after the September 2008 failure of the New York investment banking firm Lehman Brothers, which triggered the global financial crisis. The three-month U.S. LIBOR diverged from two publicly available similar short-term rates—the ICAP NYFR and the three-month rate on Eurodollar deposits (․)’ ”).As noted elsewhere, other academic research points out why it is difficult to use Eurodollar deposit rate series to evaluate but-for LIBOR during this time period. “ ‘One of the yardsticks that we and other analysts have used to evaluate LIBOR is the Eurodollar deposit rates[.] ․ They reflect the upper end of an indicative run that ICAP furnishes to data-vendors that is intended to capture the bulk of the rates paid by A1/P1 banks at any given time․ The LIBOR fixings have at times appeared to be too low, but picking the upper limit of rates being paid by active participants almost by definition produces a rate that is too high.’ ” Dennis Kuo, David Skeie and James Vickery, “A Comparison of LIBOR to Other Measures of Bank Borrowing Costs” (working paper, Federal Reserve Bank of New York, 2018) at 27 n.36 (quoting “Fed Policy” Research Commentary, Wrightson ICAP, May 2, 2008)).

153.   To be more precise, Drs. Bernheim and Snow attribute to suppression the entire difference between actual LIBOR and but-for LIBOR, which they estimate based on data of the LIBOR-Eurodollar deposit rate spread and a collection of variables outside the relevant time period. See supra pp. 188-193, 198-200; Hubbard OTC Rebuttal Report ¶ 108; Hubbard FFFP Rebuttal Report ¶ 107; see also Bernheim/Snow – Pls. Br. at 3-4, 6-8, 22-25 (similar). As discussed further below, infra pp. 223-228, even though the experts identified several relevant variables that might explain the difference between LIBOR and Eurodollar deposit offer rates during the relevant period, supra pp. 188-193, 198-200, the experts’ suppression estimates differed very little from the spread of LIBOR and Eurodollar deposit offer rates. See Snow Initial Report Figs. 95-98; Bernheim Initial Report Fig. 38; Hubbard FFFP Reply Report Exs. 6A-6D.

154.   See Snow Initial Report ¶ 264 (“[I]t is possible that riskier banks were a part of the market information used to formulate its Eurodollar rates.”); id. ¶ 263 n.253 (noting that that the term “prime bank[s]” includes Panel Banks, but also includes “any bank rated A1/P1 or better that was either A) a top 50 bank as measured by total assets in 2008, [or] B) a bank that had a large amount of Eurodollar transactions in the data produced by ICAP”); Bernheim Initial Report ¶ 69 (detailing differences between LIBOR and Eurodollar deposit offer rates).

155.   As for Dr. Bernheim's claim that Dr. Hubbard's analysis ignores OTC plaintiffs’ allegations that the panel banks concealed their suppression, Bernheim Reply Report ¶ 169, OTC plaintiffs do not provide a basis in law for why an expert in an antitrust case can presume the concealment of liability in addition to liability itself. See Bernheim/Snow – Pls. Br. at 23. This assumption also runs counter to the representations of plaintiffs’ counsel concerning plaintiffs’ allegations. See, e.g., Pls. Br. at 57 (“Plaintiffs here are not required to prove falsity to prevail on their antitrust claims.”). Either falsity is, or is not, integral to plaintiffs’ claims. Plaintiffs cannot pick-and-choose whatever favorable theory will advance them to the next stage of the litigation. In addition, as we note above, plaintiffs’ purported evidence of concealment is weak and does not establish a “unity of purpose with respect to fixing” LIBOR. Aiyer, 470 F. Supp. 3d at 405; supra pp. 166-169.

156.   Dr. Hubbard identified six other interest rate series that tracked ICAP High closely outside of the relevant period but diverged dramatically during the relevant period. See Hubbard FFFP Rebuttal Report ¶¶ 95-102, Exs. 21A-26C; Hubbard OTC Rebuttal Report ¶¶ 94-101, Exs. 22A-28B. Plaintiffs do not allege that any of these other rate series were suppressed. Id. Plaintiffs’ experts try to explain away this result by doubling down on their mischaracterization of the LIBOR rate, arguing that LIBOR is an offer rate and these other rates are materially different in that they are transaction rates or bid rates. See Bernheim Reply Report ¶ 312; Snow Reply Report ¶ 316. But again, plaintiffs’ experts too narrowly define LIBOR, such that they ignore the “and accepting” language in the LIBOR question. Supra pp. 204-208.

157.   We note that both experts believe that any factors, other than suppression, which caused the LIBOR-Eurodollar deposit offer rates spread to diverge were “minor.” Bernheim Reply Report ¶¶ 165, 165 n.151, 310; see also Snow Reply Report ¶ 264 (similar). Dr. Snow so concludes because “[m]odeling the spread between two interest rates removes factors that tend to impact both interest rates at the same time.” Snow Reply Report ¶ 264. However, even though Dr. Bernheim concedes that it is not “appropriate to reach conclusions about suppression based on unadjusted comparisons between LIBOR and such benchmarks,” Bernheim Reply Report ¶ 40, and the experts identified several variables that might explain the difference between LIBOR and Eurodollar deposit offer rates during the relevant period, supra pp. 188-193, 198-200, the experts’ suppression estimates differed very little from the spread of LIBOR and Eurodollar deposit offer rates, see Snow Initial Report Figs. 95-98; Bernheim Initial Report Fig. 38; Hubbard FFFP Reply Report Exs. 6A-6D; Hubbard OTC Rebuttal Report ¶ 9(b).

158.   As explained above, our references to the “relevant period” include the OTC Class Period, which is between August 2007 and August 2009, which Dr. Bernheim examined, and the Conduct Period, August 2007 and May 2010, which Dr. Snow tested. Bernheim Initial Report ¶¶ 14, 24; Snow Initial Report ¶ 14.

159.   See Hubbard FFFP Rebuttal Report ¶¶ 11(b), 114-118; Hubbard OTC Rebuttal Report ¶¶ 11(b), 113-117; Bernheim Initial Report ¶ 106 (“[T]he financial crisis ․ was in many ways an exceptional event.”), id. ¶ 53 (detailing liquidity and credit concerns during the financial crisis); Snow Initial Report ¶ 53 (similar); see also LIBOR VII, 299 F. Supp. 3d at 553, slip op. at *231 (“By construction, a regression does not incorporate time-series data from outside the benchmark period in order to estimate the relationship between the variables in question.”).

160.   The exception is the Relative CDS spreads, Hubbard FFFP Rebuttal Report ¶ 116 n.237, which is further discussed below.

161.   See Bernheim Initial Report ¶ 105; Snow Initial Report ¶ 234; Hubbard OTC Rebuttal ¶ 150; Hubbard FFFP Rebuttal ¶ 148; see also Bernheim Reply Report ¶ 267 (“I include the cross-sectional standard deviation of CDS spreads across the full set of Prime Banks as a control in my regressions because the average CDS spread of Prime Banks may not fully reflect the credit risk of the set of banks associated with the measures of Eurodollar offer rates I consider.”).

162.   Similarly, even though the mean of Relative CDS spreads in the benchmark period does not differ significantly over all tenors from that of the relevant period in Dr. Snow's report, Hubbard FFFP Rebuttal Report ¶ 116 n.237, excluding Relative CDS spreads from the models of Dr. Bernheim and Dr. Snow would yield largely identical results, see Hubbard OTC Rebuttal ¶ 164; Hubbard FFFP Rebuttal ¶ 161, indicating that their models are improperly specified in any event. See In re Air Cargo Shipping Servs. Antitrust Litig., No. 06 MDL 1175 (JG) (VVP), 2014 WL 7882100, at *23-24 (E.D.N.Y. Oct. 15, 2014) (excluding defense expert who claimed to use “all the data” but effectively discarded 99% of data in regression analysis by multiplying data by improper variable, as the analysis was “highly misleading,” “ha[d] no probative value, and carrie[d] the potential to confuse or mislead the jury”), report and recommendation adopted, No. 06 MDL 1775 (JG) (VVP), 2015 WL 5093503 (E.D.N.Y. July 10, 2015).Moreover, defendants’ expert is not the only one who opined on the Relative CDS spread. Dr. Cragg, OTC plaintiffs’ conspiracy expert, found that “the relationship between LIBOR submissions and the CDS spreads broke down during the Class Period,” though, of course, he opines that this “is consistent with the conclusion in my Opening Report that LIBOR submissions were the result of collusion by Panel Banks.” Cragg Rebuttal Report ¶ 103. Fundamental disagreements between a party's own experts cannot “be considered a mere ‘battle of the experts’ ” and warrant exclusion. LIBOR VII, 299 F. Supp. 3d at 481-482, slip op. at *60-61 (citing Deutsch v. Novartis Pharm. Corp., 768 F.Supp.2d 420, 469 (E.D.N.Y. 2011) (identifying inconsistencies between two of the same party's experts as a basis for exclusion)).

163.   A “structural break” test evaluates the stability of the estimated effect of the explanatory variables used by Dr. Snow and Dr. Bernheim. See Bernheim/Snow – Defs. Br. at 20 n.65 (citations omitted).

164.   In his reply, Dr. Snow states that his baseline model produces reliable results because it “(1) fits the data well during the period when LIBOR was not alleged to have been suppressed and (2) generates suppression results during the Conduct Period that are consistent with documentary evidence not used in the estimation of the model.” Snow Reply Report ¶ 284.This reply does not establish the model's reliability. Of course Dr. Snow's model fits the data well during the period when LIBOR was not alleged to have been suppressed-- his model uses this data to calculate the but-for LIBOR fix. The problem, as articulated above, is that Dr. Snow's model fails to produce reliable results within the period when LIBOR was allegedly suppressed. Moreover, the documentary evidence of a handful of dates cannot prove systematic suppression throughout the relevant period. Infra pp. 229-231.

165.   As noted supra pp. 118-126, evaluating the percentage of days these communications cover understates the infrequency of these communications. Plaintiffs’ cited set of communications is exceedingly infrequent, considering the number of days in the relevant period, that sixteen Panel Banks were allegedly involved in this purported conspiracy, and that multiple employees at each Panel Bank was communicating about LIBOR, sometimes multiple times a day.

166.   While plaintiffs claim that “Dr. Snow calculated but-for LIBOR based on actual borrowing data,” Bernheim/Snow – Pls. Br. at 30, according to defendants’ calculations, Dr. Snow only incorporates transaction data on 27%, 16%, 7%, and less than 2% of days for the 1M, 3M, 6M, and 12M tenors respectively, Bernheim/Snow – Defs. Br. at 7 (citations omitted). Because Dr. Snow's alternative model primarily relies upon his ICAP High-based “but for” submissions for the majority of days, it is additionally unreliable for the same reasons why his econometric model is unreliable. See supra pp. 202-229.

167.   Dr. Bernheim's starting premise regarding suitability is reasonable and well-supported: “Transactional borrowing costs are not offer rates and are on average lower than offer rates.” Bernheim Reply Report ¶ 368.

168.   Specifically, to create imputed borrowing costs, Dr. Bernheim “uses a regression model explaining the observed spread between non-Panel Prime Banks’ borrowing costs on Eurodollar transactions brokered through ICAP and the ICAP Eurodollar offer rate during the Class Period.” Bernheim Rebuttal Report ¶ 98.

169.   Faced with the many deficiencies outlined above, plaintiffs misdescribe Dr. Bernheim's methodology. Plaintiffs assert that “Dr. Bernheim's second method for modeling LIBOR suppression uses interbank borrowing costs on consummated London interbank deposit transactions.” Bernheim/Snow – Pls. Br. at 4 (citations omitted). However, as articulated above, Dr. Bernheim finds that “[o]n most days during the Class Period, most Panel Banks did not have [London interbank deposit] transactions of reasonable size.” Bernheim Rebuttal Report ¶ 30. Therefore, in all four of estimates under this model, Dr. Bernheim uses some form of constructed “estimate[ ] of LIBOR suppression,” whether that is imputed borrowing costs or predictions from his baseline model based on ICAP High, on most days without any borrowing transactions. Bernheim Rebuttal Report ¶¶ 30, 95. Moreover, in three of four estimates, Dr. Bernheim adjusts the transactional costs when he does use this data set. Supra pp. 193-198.

170.   Following the submission of the parties’ briefing related to Upstream Issues, both DAPs and OTC plaintiffs filed letters moving to strike certain arguments from defendants concerning inconsistent results produced by the models of Drs. Snow and Bernheim. ECF Nos. 4523, 4530. This Court does not rely upon the contested portions of defendants’ arguments and plaintiffs’ letter motions to strike are moot. See Snyder v. Perry, No. 14 Civ. 2090 (CBA) (RER), 2015 WL 1262591, at *13 (E.D.N.Y. Mar. 18, 2015) (collecting cases).

171.   This conclusion is buttressed, not contradicted, by defendants’ string-cite of exhibits in two footnotes of their memorandum of law in support of their motion for summary judgment. See Defs. Br. at 47 n.21 (stating “Contemporaneous BBA guidance confirms that the ‘rate at which each contributor submits must be formed from that bank's perception of its costs of funds in the interbank market.’ Defs. 56.1 ¶ 22 ․ Terms of Reference for LIBOR Contributor Banks, BBA, July 15, 2009, RABO_METZLER_0078539 at -539) (LIBOR submissions should reflect ‘a fair and accurate reflection of that bank's opinion of its cost of funds’))․ (Mar. 21, 2024 Dep. of John Ewan (BBA) Tr.) 174:6-24) (‘Q. Now, the LIBOR definition calls for the rate at which a bank could borrow funds, if it were to do so, by asking and then accepting an offer, correct? A. Yes. Q. What do you understand the inclusion of the word, ‘accepting’ in the definition to mean? A. Saying that the terms under which it was being offered cash were acceptable to it. That it wished to do so. Q. And that is because the banks might receive offers they would not accept, correct? A. Yes.’ ”))); Defs. Br. at 47 n.22 (citing Defs. 56.1 ¶ 91 (Declaration of Alan C. Turner (“JPMorgan Decl.”), Ex. 1 (Dec. 8, 2023 Dep. of Amanda Adams Dep. Tr. (JPMorgan) 44:21-24)) (“Q. And the LIBOR definition asked for the rate at which banks could borrow funds by asking for and accepting interbank offers, correct? A. That is correct.”), ¶ 49 (Declaration of Andrew A. Ruffino (“Citi Decl.”), Ex. 10 (Mar. 26, 2024 Dep. of Andrew Thursfield Tr. (Citi) 44:13-17 (“It was a rate at which judgment was required to formulate to come up with where you believed you would be able to borrow, but also where you would be willing to accept the rate because it was asking and then accepting.”), 161:19-21))), ¶ 163 (Rabobank Decl., Ex. 3 (Feb 28, 2024 Deposition of Anthony Conti Tr. (Rabobank) 191:2-4)) (“I would kind of calculate the range where I perceived an offer would be offered that I would find acceptable.”), ¶ 522 (Declaration of Jefferson E. Bell (“UBS Decl.”), Ex. 5 (Apr. 4, 2024 Dep. of Etienne Perret Tr. (UBS) (“E. Perret Dep. Tr.”) 78:23-79:4)) (“․ where we would have made the expert judgment of where we were likely to be offered and accept funding in the London market”)).

172.   See also OTC Compl. ¶ 5 (“By acting together and in concert to knowingly understate their true borrowing costs, the panel banks caused LIBOR to be calculated or suppressed artificially low, reaping hundreds of millions, if not billions, of dollars in ill-gotten gains.”); FDIC Compl. ¶ 428 (“In a competitive environment, Defendants could not have sustained their conspiracy and USD bbaLIBOR™ would have been calculated according to its published methodology and therefore would have been consistently higher throughout the conspiracy period”); Principal Funds Compl. ¶ 234 (“Defendants’ collusion caused Plaintiffs to pay more for or receive less from their USD Libor tied investments than they would have had competitive market forces in the London interbank lending market (and not collusion) determined USD Libor.”); Principal Financial Compl. ¶ 237 (same); Fannie Mae Compl. ¶ 163 (similar); Freddie Mac Compl. ¶¶ 3, 7, 303 (similar).

173.   As we noted previously, many documents cited by plaintiffs undermine their assertions of collusive suppression. See, e.g., supra pp. 118-126.

174.   All remaining plaintiffs assert antitrust claims. However, not all claims remain live. See ECF No. 4583-1. The OTC plaintiffs assert a live antitrust claim, see ECF No. 4583-1; OTC Compl. ¶¶ 377-82. The majority of FFFP plaintiffs have live antitrust claims. See ECF No. 4583-1; FDIC Compl. ¶¶ 375-433; Freddie Mac Compl. ¶¶ 284-347; Principal Financial Compl. ¶¶ 230-78; Principal Funds Compl. ¶¶ 227-75. Fannie Mae's antitrust claim is not live. See ECF No. 4583-1 at 1; Fannie Mae Compl. ¶¶ 161-68.

175.   See also OTC Compl. ¶ 5 (“By acting together and in concert to knowingly understate their true borrowing costs, the panel banks caused LIBOR to be calculated or suppressed artificially low, reaping hundreds of millions, if not billions, of dollars in ill-gotten gains.”); FDIC Compl. ¶ 428 (“In a competitive environment, Defendants could not have sustained their conspiracy and USD bbaLIBOR™ would have been calculated according to its published methodology and therefore would have been consistently higher throughout the conspiracy period”); Principal Funds Compl. ¶ 234 (“Defendants’ collusion caused Plaintiffs to pay more for or receive less from their USD Libor tied investments than they would have had competitive market forces in the London interbank lending market (and not collusion) determined USD Libor.”); Principal Financial Compl. ¶ 237 (same); Fannie Mae Compl. ¶ 163 (similar); Freddie Mac Compl. ¶¶ 3, 7, 303 (similar).

176.   We note that, throughout this section of their briefing, plaintiffs rely heavily on the Court's ruling in LIBOR IV, which was issued at a far earlier stage in this litigation. See Pls. Br. at 124-128. At the current summary judgment stage, which follows extensive discovery and legal developments, LIBOR IV is of limited precedent.

177.   Apparently recognizing the importance of Connolly in this MDL, plaintiffs have gone to great lengths to dismiss its significance. We note that in addressing their antitrust claims, plaintiffs assert that Connolly is irrelevant because it is a fraud case. See Pls. Br. at 56-58. In addressing their fraud claims, plaintiffs find new and equally unconvincing ways to dismiss Connolly. See Pls. Br. at 68-73. Plainly stated, plaintiffs’ attempts to distinguish Connolly are unavailing. It is immaterial that Connolly did not involve plaintiffs’ specific fraud theories or that the government had a different standard of proof to convict the Deutsche Bank traders. See id. Rather, what is important about Connolly here is that the Second Circuit determined what would and would not qualify as a false response to the LIBOR question. 24 F.4th at 835-840.

178.   We note that, in accordance with Connolly’s holding, Dr. Hubbard, defendants’ expert, submitted “calculations [which] establish that Defendants’ daily LIBOR submissions overwhelmingly were within or above the range of each Panel Bank's corresponding borrowing rates on submission days with associated borrowing transactions.” Defs. Br. at 20 (citation omitted). Though plaintiffs challenged the reliability of Dr. Hubbard's model, see, e.g., Pls. Br. at 67-68, plaintiffs did not move to exclude Dr. Hubbard's opinions. Moreover, adoption of Dr. Hubbard's report is unnecessary, as “the moving party's burden under Rule 56 will be satisfied if he can point to an absence of evidence to support an essential element of the nonmoving party's claim.” Brady, 863 F.2d at 210-11 (citing Celotex, 477 U.S. at 323-26).

179.   We note that the parties dispute whether Freddie Mac's common law fraud remains live. See ECF No. 4853-1 at 4.

180.   The parties do not address whether New York or Washington, D.C. law applies to Fannie Mae's fraudulent inducement claims. In determining which substantive law applies to a party's claim, “[f]ederal courts sitting in diversity look to the choice-of-law rules of the forum state.” Int'l Bus. Mach. Corp. v. Liberty Mut. Ins. Co., 363 F.3d 137, 143 (2d Cir. 2004) (citing Curley v. AMR Corp., 153 F.3d 5, 12 (2d Cir. 1998)). However, “[i]n the absence of substantive difference [between the laws of competing jurisdictions], a New York court will dispense with choice of law analysis; and if New York law is among the relevant choices, New York courts are free to apply it.” Id. (citations omitted). Here, there is no substantive difference between the elements of a fraudulent inducement claim under New York and D.C. law. Compare Axginc Corp., 759 Fed. App'x at 30 (listing elements of fraudulent inducement claim under New York law) with Africare, Inc. v. Xerox Complete Document Sols. Maryland, LLC, 436 F. Supp. 3d 17, 38 (D.D.C. 2020) (listing elements of fraudulent inducement claim under D.C. law). Accordingly, we apply New York law to evaluate Fannie Mae's fraudulent inducement claim.

181.   As plaintiffs note, Pls. Br. at 125 n.201, defendants do not separately address Fannie Mae's fraudulent inducement claim. This is immaterial.

182.   Neither party addresses the choice of law to be applied to these claims. However, given that both parties cite to New York law in their briefing, the Court will apply New York law to this issue.

183.   Defendants also assert that, in order to prevail on their aiding and abetting claims, plaintiffs “must prove that each Panel Bank (i) was a persistent suppressor and (ii) knew ‘that its own suppression would help each other bank suppress LIBOR.’ ” Defs. Br. at 105 (quoting LIBOR IV, 2015 WL 6243526, at *49, slip op. at *123). However, this selective quotation relates to only one theory of aiding and abetting liability outlined in LIBOR IV. See 2015 WL 6243526, at *49, slip op. at *123.

184.   Again, neither party addresses the choice of law to be applied to these claims. However, given that both parties cite to New York law in their briefing, the Court will apply New York law to this issue.

185.   Plaintiffs assert that their conspiracy to commit fraud claims remain viable because they “have come forward with substantial evidence of suppression.” Pls. Br. at 125. However, as defendants note, Defs. Br. at 106, and as explored in more detail supra pp. 177-244, plaintiffs have not presented evidence sufficient to raise a genuine issue of material fact with respect to suppression.Plaintiffs inexplicably argue that, if the Court concludes that the DAPs have not proffered sufficient evidence of their fraud claims against “certain [d]efendants[,]” their conspiracy to commit fraud claims “can proceed against those [d]efendants so long as [the] DAPs proffer sufficient evidence to support their fraud claim against other [d]efendants.” Pls. Br. at 125 n.202. In light of this Court's determination that plaintiffs have failed to establish the necessary underlying fraud claim against all defendants, we need not reach this argument.

186.   The parties dispute whether Principal Funds’ negligent misrepresentation claim remains live. See ECF No. 4853-1 at 9.

187.   See ECF No. 4583-1; Fannie Mae Compl. ¶¶ 115-21 (breach of contract), 122-25 (breach of implied covenant); FDIC Compl. ¶¶ 272-78 (breach of implied covenant); Freddie Mac Compl. ¶¶ 348-54, 355-61, 362-68, 376-82, 390-96, 397-403 (breach of contract claims); OTC Compl. ¶¶ 383-95 (breach of implied covenant); Principal Financial Compl. ¶¶ 287-98, 299-301 (breach of contract and implied covenant); Principal Funds. Compl. ¶¶ 284-95, 296-98 (breach of contract and implied covenant).

188.   The same analysis applies to the extent Freddie Mac still has live claims for breach of express contract, as shown in the joint memorialization chart provided by the parties. See ECF No. 4853-1.

189.   In its complaint, the FDIC brings a series of tortious interference claims against the Panel Bank defendants and the BBA, namely: tortious interference with contract, ¶¶ 480-86; aiding and abetting tortious interference with contract, ¶¶ 487-91; civil conspiracy to commit tortious interference with contract, ¶¶ 492-96; tortious interference with prospective economic advantage, ¶¶ 497-502; aiding and abetting tortious interference with prospective economic advantage, ¶¶ 503-07; and civil conspiracy to commit tortious interference with prospective economic advantage, ¶¶ 508-512.

190.   In 2019, this Court endorsed two letters submitted by the parties in which they memorialized their understanding of the status of live and dismissed claims in this action. In these documents, ECF Nos. 2892, 2958, the parties represented that all the FDIC's tortious interference claims had been dismissed. Id. However, plaintiffs assert that “an open question still exists as to whether the FDIC-R's tortious interference claims were revived by appellate decisions subsequent to the 2019 stipulation[,]” Pls. Br. at 126 n.203, citing a list compiled by the parties in 2022 identifying the survival of the FDIC-R's tortious interference claims as one of many “Disputed Issues” relating to this litigation, ECF No. 3501-2 at 3-4.

191.   We noted that this last category includes “[a defendant's] own or its affiliate's assistance of an unrelated panel bank's manipulation.” LIBOR IV, 2015 WL 6243526, at *78, slip op. at *192.

192.   While plaintiffs also assert that defendants “benefitted reputationally and, for at least some banks, financially from a lower LIBOR and projecting financial strength,” Pls. Br. at 128, such allegations are similarly insufficient in light of this Court's holding that plaintiffs have not shown sufficient evidence that defendants, in fact, suppressed (i.e., lowered) LIBOR.

193.   As discussed earlier, at the time that this Court issued LIBOR VII in 2018, defendants Bank of America and JPMorgan were the only two defendants deemed subject to the Court's jurisdiction. 299 F. Supp. 3d at 581-82, slip op. at *298-301. The Second Circuit's 2021 decision in Schwab II held that plaintiffs had sufficiently alleged personal jurisdiction against the foreign defendants under a conspiracy theory of jurisdiction. 22 F.4th at 124-25.Plaintiffs define the proposed OTC Class as “All persons or entities residing in the United States that purchased, directly from a Panel Bank (or a Panel Bank's subsidiaries or affiliates), a LIBOR–Based Instrument that paid interest indexed to a U.S. dollar LIBOR rate set any time during the period August 2007 through August 2009 (“Class Period”) regardless of when the LIBOR–Based Instrument was purchased.” ECF No. 4190 at 1. This is the same OTC Class that this Court certified against JPMorgan and Bank of America in LIBOR VII. Id.

NAOMI REICE BUCHWALD UNITED STATES DISTRICT JUDGE