NAOMI REICE BUCHWALD UNITED STATES DISTRICT JUDGE

MEMORANDUM AND ORDER

LIBOR IV TABLE OF CONTENTS

I. General Introduction ................................... 9

II. Background ........................................... 14

1. The Alleged Facts ................................... 14
2. Procedural History .................................. 19

2.1. Prior Rulings ................................... 19

2.1.1. LIBOR I ...................................... 19
2.1.2. LIBOR II ..................................... 21
2.1.3. LIBOR III .................................... 23
2.1.4. Other Orders ................................. 25

2.2. Pending Motions ................................. 27
2.3. The Individual Plaintiffs ....................... 29

2.3.1. Amabile ...................................... 29
2.3.2. BATA ......................................... 29
2.3.3. California Consolidated Plaintiffs ........... 30
2.3.4. CEMA ......................................... 32
2.3.5. Darby ........................................ 32

2.3.6. Fannie Mae ................................... 33
2.3.7. FDIC ......................................... 33
2.3.8. Freddie Mac .................................. 34
2.3.9. Houston ...................................... 35
2.3.10. Maragos (Nassau County) ..................... 36
2.3.11. NCUA ........................................ 36
2.3.12. Philadelphia ................................ 38
2.3.13. Principal ................................... 39
2.3.14. Prudential .................................. 40
2.3.15. Salix ....................................... 41
2.3.16. Schwab ...................................... 43
2.3.17. Triaxx ...................................... 45

III. Personal Jurisdiction ............................... 46

1. Preliminary Matters ................................. 49

1.1. Relation Between Jurisdictional and Merits Rulings ..................................................... 49
1.2. Plaintiffs' Burden .............................. 50
1.3. State-by-State Jurisdictional Analysis .......... 52

1.3.1. Actions Brought Outside of New York .......... 53
1.3.2. Federal Statutory Claims and Pendent Claims .. 55

2. General Personal Jurisdiction ....................... 59

2.1. General Principles .............................. 59
2.2. Application ..................................... 62

3. Specific Personal Jurisdiction ...................... 65

3.1. General Principles .............................. 65
3.2. Conspiracy ...................................... 69
3.3. LIBOR "Marketing" ............................... 71
3.4. Solicitation, Negotiation and Performance of Contracts ............................................ 75
3.5. Wrongdoing in or Directed to Forum States ....... 78
3.6. Reasonableness .................................. 81

4. Forum Selection Clauses ............................. 82

4.1. Scope of the Clauses ............................ 83
4.2. Successor Entities .............................. 86
4.3. Cases Brought Outside of New York ............... 87

5. Forfeiture .......................................... 88
6. Application ......................................... 91

6.1. Claims Against Counterparties ................... 92
6.2. Tortious Interference Claims .................... 93
6.3. Fraud and Commodities Exchange Act Claims ....... 94

7. Transfer of Venue ................................... 95

IV. Pleading ............................................. 98

1. General Pleading Standards .......................... 98
2. Pleading of Sporadic Trader-Based Manipulation ..... 100

2.1. Exchange-Based Plaintiffs (Amabile) ............ 101
2.2. Other Complaints ............................... 105

V. Conspiracy and Other Theories of Joint Liability ..... 107

1. Intra-Bank Collaboration ........................... 108

1.1. Trader-Based Manipulation ...................... 108
1.2. Persistent Suppression ......................... 110
1.3. Legal Consequences of Intra-Bank Collaboration . 111

2. Inter-Bank Collaboration ........................... 111

2.1. Trader-Based Manipulation ...................... 111
2.2. Persistent Suppression ......................... 114

3. Aiding and Abetting ................................ 121
4. Conclusions ........................................ 124

VI. Fraud and Negligent Misrepresentation ............... 125

1. The Definition of Fraud ............................ 126

1.1. Choice of Law .................................. 126
1.2. Elements of Fraud .............................. 127

2. Factual Theories ................................... 128

2.1. Misrepresentations ............................. 128
2.2. Reliance ....................................... 129

3. Legal Discussion ................................... 131

3.1. Fraud by Affirmative Misrepresentation in the Course of Offering or Trading Securities ................... 131

3.1.1. Failure to Plead Specifics .................. 131
3.1.2. Failure to Plead Misrepresentations ......... 134

3.1.2.1. LIBOR Quality ........................... 135
3.1.2.2. Good-Faith Calculations ................. 137

3.1.2.3. Compliance With Laws .................... 138

3.1.3. No-Existing-Breach Representation ........... 138

3.2. Fraud by Omission in the Course of Offering or Trading Securities .......................................... 141
3.3. Fraud by Omission After Trading ................ 144
3.3. Fraudulent Assurances About LIBOR .............. 146

3.3.1. Analysts' Statements ........................ 146
3.3.2. Official Statements ......................... 150

3.4. False LIBOR Submissions ........................ 153

3.4.1. Limiting Principles ......................... 154
3.4.2. Actual Reliance ............................. 163

3.4.2.1. Defendants' BBA Filter Argument ......... 163
3.4.2.2. Defendants' Calculation Argument ........ 165

3.5. Justifiable or Reasonable Reliance ............. 166
3.6. Damages ........................................ 169

4. Negligent Misrepresentation ........................ 170
5. Securities Fraud ................................... 172

5.1. Federal Claims ................................. 172
5.2. California Blue Sky Claims ..................... 175

6. Conclusions ........................................ 176

VII. Breach of Contract, Unjust Enrichment, and Tortious Interference ............................................ 177

1. Prior Rulings ...................................... 178

1.1. LIBOR II ....................................... 178
1.2. LIBOR III ...................................... 180
1.3. Order Dated October 8, 2014 .................... 180

2. The Elements ....................................... 181

2.1. Choice of Law .................................. 181
2.2. Contract ....................................... 181
2.3. Unjust Enrichment .............................. 183
2.4. Tortious Interference .......................... 183

3. Legal Discussion ................................... 184

3.1. Contract and Unjust Enrichment ................. 185

3.1.1. Breach of Express Contractual Terms ......... 185
3.1.2. Counterparty Requirement .................... 186

3.1.3. Wrongdoing and Scienter ..................... 187

3.1.3.1. Implied Covenant ........................ 187
3.1.3.2. Unjust Enrichment ....................... 188

3.1.3.2.1. Personal Misconduct ................. 188
3.1.3.2.2. Affiliated Misconduct ............... 189
3.1.3.2.3. Unaffiliated Misconduct ............. 191

3.1.4. Unjust Enrichment as Alternative to Contract 192
3.1.5. Free-Standing Unjust Enrichment ............. 195

3.2. Tortious Interference .......................... 197

3.2.1. Scienter.................................... 197
3.2.2. Interference with Affiliate's Contract ...... 200
3.2.3. Interference with Business Relations ........ 201

3.3. Damages ........................................ 203

4. General Application ................................ 203

4.1. Over-the-Counter Derivatives ................... 203
4.2. Exchange-Traded Derivatives .................... 204
4.3. Adjustable-Rate Bonds .......................... 204
4.4. Asset-Backed Securities ........................ 205
4.5. The BBA ........................................ 207

VIII. Antitrust ......................................... 207

1. Prior Rulings ...................................... 208

1.1. LIBOR I ........................................ 208
1.2. LIBOR II ....................................... 215

2. Plaintiffs' Federal Antitrust Claims ............... 216

2.1. The Complaints ................................. 216
2.2. Market Structure ............................... 217
2.3. Antitrust Injury ............................... 219

3. State-Law Antitrust Claims ......................... 222

IX. Consumer Protection and Unfair Business Practices ... 224

1. New York (Maragos) ................................. 224
2. California ......................................... 228

2.1. Types of Practices Actionable .................. 228
2.2. Legal Standards Applicable to NCUA ............. 230
2.3. Adequate Pleading .............................. 232

3. Conclusion ......................................... 234

X. New Jersey RICO ...................................... 234

1. Prior Rulings Regarding Federal RICO (LIBOR I) ..... 234
2. New Jersey RICO .................................... 236
3. Prudential's Complaint and the Instant Motion ...... 237
4. Extraterritoriality ................................ 240

4.1. Territorial Limits of New Jersey RICO .......... 240
4.2. Constitutionality .............................. 247
4.3. Application .................................... 250

4.3.1. New Jersey Uniform Securities Act ........... 250
4.3.2. Deceptive Business Practices ................ 252
4.3.3. Theft by Deception .......................... 253
4.3.4. Falsifying Records .......................... 254

5. Conclusion ......................................... 255

XI. Commodities Exchange Act ............................ 255

XII. Equitable Relief ................................... 257

1. Prospective Equitable Relief ....................... 258
2. Retrospective Equitable Relief ..................... 261

2.1. Constructive Trust ............................. 261
2.2. Rescission ..................................... 263

XIII. Damages ........................................... 264

1. Joint and Several Liability ........................ 264
2. Contract-Specific Liability ........................ 270

XIV. Statute of Limitations ............................. 273

1. Prior Rulings ...................................... 274

1.1. LIBOR I ........................................ 274
1.2. LIBOR III ...................................... 279

1.2.1. Period Two .................................. 279
1.2.2. Period One .................................. 280
1.2.3. Class-Action Tolling ........................ 281

1.3. Renewed Challenge to Prior Holdings ............ 281

2. Choice of Law ...................................... 283

2.1. General Principles ............................. 283
2.2. Application .................................... 284

3. Federal Extender Statutes .......................... 287

3.1. Text and History ............................... 288

3.2. Classification of Unjust Enrichment ............ 291
3.3. Successive Appointments ........................ 295
3.4. Accrual Date ................................... 298
3.5. Choice of State Law ............................ 300

4. Accrual ............................................ 301

4.1. Discovery Rule ................................. 302

4.1.1. Generally ................................... 302
4.1.2. By Jurisdiction ............................. 303

4.1.2.1. Federal ................................. 303
4.1.2.2. California .............................. 304

4.1.2.2.1. Torts ............................... 304
4.1.2.2.2. Contract and Unjust Enrichment ...... 306
4.1.2.2.3. Blue Sky ............................ 307

4.1.2.3. Connecticut ............................. 309
4.1.2.4. District of Columbia .................... 309
4.1.2.5. Iowa .................................... 310
4.1.2.6. Kansas .................................. 311
4.1.2.7. New Jersey .............................. 311
4.1.2.8. New York ................................ 312
4.1.2.9. Pennsylvania ............................ 314
4.1.2.10. Texas .................................. 315
4.1.2.11. Virginia ............................... 315

4.1.3. Application ................................. 316

4.1.3.1. Inquiry Notice .......................... 317
4.1.3.2. Effect of Inquiry Notice ................ 320

4.2. Continuing Violation Doctrines ................. 322

4.2.1. Pure CVD .................................... 323
4.2.2. Continuous Accrual .......................... 325

5. Fraudulent Concealment ............................. 326
6. Class-Action Tolling (American Pipe) ............... 328

6.1. Generally ...................................... 328
6.2. Legal Analysis ................................. 331

6.2.1. Tolling Across Jurisdictions ................ 331

6.2.1.1. California .............................. 336
6.2.1.2. Connecticut ............................. 341

6.2.1.3. Iowa .................................... 342
6.2.1.4. Kansas .................................. 343
6.2.1.5. New Jersey .............................. 344
6.2.1.6. New York ................................ 344
6.2.1.7. Ohio .................................... 346
6.2.1.8. Pennsylvania ............................ 346
6.2.1.9. Virginia ................................ 347
6.2.1.10. Conclusion ............................. 347

6.2.2. Tolling Across Claims ....................... 348
6.2.3. Tolling As to New Defendants ................ 351
6.2.4. Application to Extender Statutes ............ 353
6.2.5. Application to Statute of Repose ............ 354
6.2.6. Application to Schwab ....................... 355
6.2.7. Other Arguments ............................. 357

6.3. Predicate Cases ................................ 358

6.3.1. OTC ......................................... 358
6.3.2. Exchange-Based .............................. 364
6.3.3. Lenders ..................................... 368

7. Relation Back ...................................... 373

7.1. Defendants Added in Amended Complaints ......... 373

7.1.1. Principal ................................... 374
7.1.2. Salix ....................................... 377
7.1.3. NCUA ........................................ 378

7.2. Schwab ......................................... 380

8. Application ........................................ 381

8.1. Amabile ........................................ 383
8.2. BATA ........................................... 385
8.3. California Consolidated ........................ 387
8.4. CEMA ........................................... 390
8.5. Darby .......................................... 391

8.5.1. Darby Financial ............................. 391
8.5.2. Capital Ventures ............................ 393

8.6. Fannie Mae ..................................... 394
8.7. FDIC ........................................... 396
8.8. Freddie Mac .................................... 406

8.9. Houston ........................................ 407
8.10. Maragos ....................................... 407
8.11. NCUA .......................................... 408
8.12. Philadelphia .................................. 413
8.13. Principal ..................................... 415
8.14. Prudential .................................... 419
8.15. Salix ......................................... 420

8.15.1. Claims Assigned by FrontPoint Funds ........ 420
8.15.2. Claims Assigned by Eric Grannan ............ 422
8.15.3. Claims Assigned by Others .................. 422

8.16. Schwab ........................................ 423
8.17. Triaxx ........................................ 426

XV. Conclusion .......................................... 427

I. GENERAL INTRODUCTION

This consolidated multi-district litigation (MDL) arises from allegations that several major banks manipulated the London Interbank Offer Rate (LIBOR), a set of interest-rate benchmarks that underlie trillions of dollars of financial instruments, in order to profit in their own trading and to maintain their reputations for creditworthiness. This MDL involves U.S. Dollar LIBOR only. Cf. Laydon v. Mizuho Bank, Ltd., No. 12-cv-3419 (GBD) (S.D.N.Y.) (Yen LIBOR and the Tokyo Interbank Offer Rate).

In LIBOR I, II, and III, we evaluated claims brought by class action plaintiffs and affiliates of Charles Schwab. With limited exceptions, we found those claims wholly or substantially deficient. Perhaps attracted by treble damages, plaintiffs attempted to force injuries that were square pegs into legal theories that were round, or at best rectangular, holes. Plaintiffs failed to state racketeering claims because Congress has specifically excluded securities-related claims from the purview of RICO, the federal racketeering statute. Plaintiffs failed to state antitrust claims because their injuries were not "antitrust injuries" that flowed from the anti-competitive nature of defendants' alleged conduct. However, plaintiffs who traded exchange-based instruments successfully stated claims pursuant to the Commodities Exchange Act, and plaintiffs who traded over-the-counter instruments successfully stated contract claims against their counterparties for breaching the implied covenant of good faith and fair dealing that exists in every contract.

In this, our fourth extensive opinion, we focus on the legal sufficiency of complaints filed by plaintiffs who do not seek to represent classes or to become class members (the "Individual Plaintiffs"). This opinion is wide-ranging, addressing new antitrust theories, new assertions of common law and statutory claims, new arguments about timeliness, and new objections by some defendants to the power of United States courts to enter judgment against them.

Four years into this litigation, we have finally been presented with a viable legal theory that provides a comprehensive remedy for injuries proven to be sustained from LIBOR manipulation. According to plaintiffs' allegations, each panel bank lied to the LIBOR administrator about its own borrowing costs, knowing that entities such as plaintiffs would rely on the accuracy of that information; as was to be expected, plaintiffs then relied to their disadvantage, perhaps reasonably, on this false information. If these allegations prove true, then defendants' conduct was fraud.

Aside from our rulings on fraud, this opinion breaks little new ground regarding the legal adequacy of plaintiffs' complaints. Some of the new tortious interference claims are viable, as are some fraud claims on the theory that defendants should have revealed LIBOR manipulation before entering into contracts with plaintiffs. We adhere to and refine our prior holdings regarding pleading standards, conspiracy theories, contracts, restitution, antitrust, and the Commodities Exchange Act. The consumer protection and New Jersey RICO claims that fail are not central to any complaint.

This slip opinion devotes over 150 pages to statutes of limitations. The most that can be said in a single introductory paragraph is that, as in previous LIBOR opinions, some claims were filed on time and others not. The plaintiffs whose complaints are best preserved tend to be certain government entities that enjoy special statutes of limitations, residents of certain states (such as New York, with its six-year statute of limitations on fraud), early filers, and former members of putative classes.

We also address the contention of many defendants that plaintiffs' claims were brought in courts that lack personal jurisdiction over those defendants with respect to those claims. The essence of this argument is that it would be inconsistent with due process for courts in the states where plaintiffs brought suit to adjudicate those claims, because the claims do not arise out of defendants' contacts with those forum states. Although we reject plaintiffs' most sweeping arguments for the existence of personal jurisdiction, we uphold personal jurisdiction where a defendant's purposeful conduct is sufficiently connected to a plaintiff's claim and the forum state, and in forums where a defendant has consented to the court's jurisdiction.

The organization of this opinion is as follows. We begin with a brief overview of the allegations and the procedural history that preceded this opinion, infra at II. We then address the threshold question of whether the courts where plaintiffs filed their complaints have jurisdiction over defendants, infra at III. After reviewing the applicable pleading standards, infra at IV., we consider the sufficiency of plaintiffs' conspiracy allegations, infra at V. The next sections consider plaintiffs' substantive legal theories: fraud, infra at VI.; contract, unjust enrichment, and tortious interference, infra at VII.; antitrust, infra at VIII.; consumer protection and unfair business practices; racketeering, infra at IX.; New Jersey RICO, infra at X.; commodities manipulation, infra at XI.; and entitlement to equitable relief, infra at XII. We then briefly synthesize our thoughts regarding damages, infra at XIII. Finally, we address timeliness, infra at XIV.: when plaintiffs' claims accrued, how long the statutes of limitations run, when, if ever, the statutes of limitations were tolled, and when plaintiffs' complaints were deemed to be filed.

II. BACKGROUND

1. The Alleged Facts

The facts underlying this case have been thoroughly discussed in LIBOR I, 935 F. Supp. 2d at 677-85, and elaborated upon through LIBOR II and LIBOR III. Here, we present only a summary of the facts, with attention to the claims of specific plaintiffs. As we did in LIBOR II and LIBOR III, we supplement these facts throughout this opinion as needed.

LIBOR is a set of interest-rate benchmarks, formerly disseminated by the British Bankers' Association (the "BBA"). Although the BBA published LIBORs for ten different currencies, this case concerns only the U.S. Dollar ("USD") LIBOR, and we use "LIBOR" to refer only to USD LIBOR except where clear from context.

To calculate LIBOR, the BBA assembled a panel of sixteen major banks authorized to trade on the London money market. Aside from the BBA itself, all of the named defendants were members of the panel or affiliated with members of the panel. Each business day, the panel banks answered the following question with regard to fifteen maturities, or tenors, ranging from overnight to one year: "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]?" This question was phrased as a hypothetical; no bank was required to base its submission on actual transactions. However, each bank was supposed to respond with information about its own ability to obtain loans, and each bank was required to submit its quotes without reference to any other panelist's response.

After receiving each bank's quote, the BBA's agent determined that day's LIBOR for a given tenor by calculating the arithmetic mean of the middle two quartiles. Because the panel consisted of sixteen banks, this meant that the BBA's agent ignored the highest four and lowest four quotes and calculated the average of the middle eight. The BBA also published each bank's submission in each currency and tenor.

LIBOR was, and is, the most important benchmark for short-term interest rates. Investors and speculators commonly use LIBOR to define payments on interest rate swaps, floating-rate bonds, adjustable-rate mortgages, mortgage-backed and other asset-backed securities, auction-rate securities, and exchange-based "Eurodollar" futures and options.

The panel banks allegedly engaged in two forms of manipulation, which we have called "trader-based manipulation" and "persistent suppression."

First, many (if not all) of the panel banks employed traders to bet on interest rates, typically in instruments that referred to LIBOR. These individual traders received money, promotions, and adulation based on their personal profit and loss. To gain profits or avoid losses, therefore, a trader would sometimes ask his bank's LIBOR submitter to engage in what we call trader-based manipulation. The submitter would send a false quote in whichever currency and tenor suited the trader's book.

A typical example comes from Barclays on February 22, 2006. According to settlement documents incorporated into many of the complaints, a Barclays trader emailed a Barclays LIBOR submitter, "Hi (again) We're getting killed on our 3m resets [i.e., swap payments dependent on 3-month LIBOR], we need them to be up this week before we roll out of our positions. Consensus for 3m today is 4.78 - 4.7825, it would be amazing if we could go for 4.79...Really appreciate ur help mate [sic throughout]." Non-Prosecution Agreement Between Barclays Bank PLC and the Department of Justice, App. A ("Barclays DOJ Statement of Facts") at ¶ 14, June 26, 2012, available at http://www.justice.gov/iso/opa/resources/9312012710173426365941.pdf. The submitter replied, "Happy to help," and submitted a quote of 4.79%. This example is typical in that the trader's profit motive was manifest to the submitter, the submitter's agreement to manipulate LIBOR was clear, the trader and submitter both worked for the same bank (though perhaps for different corporate entities), the leading participants were traders and submitters well below the executive level, and the amount of manipulation was relatively small—between 0.75 and 1 basis point. This example is also not unusual in that Barclays's attempt to inflate LIBOR probably did not affect the published 3-month LIBOR (at least absent some other bank's simultaneous inflation). Only four panelists, including Barclays, submitted quotes above 4.78% and only two submitted quotes below 4.78%, so published LIBOR would have been 4.78% regardless of whether Barclays had submitted 4.78%, 4.7825%, 4.79%, or any number in between.

Second, following the "credit crunch" of 2007, each bank faced a serious risk that it would become unable to borrow money (and thus unable to survive) if lenders believed it to be a poor credit risk. Because each bank's LIBOR submission was made public, each bank feared that lenders would read a high LIBOR submission as derogatory credit information. To avoid being seen as a poor credit risk, each bank engaged in what we call persistent suppression, submitting quotes that were below their true borrowing costs.

The result of the banks' simultaneous suppression was that LIBOR was consistently well below true inter-bank borrowing rates starting in August 2007. Many plaintiffs present extensive statistical analyses to plead that LIBOR was suppressed by tens of basis points at the peak of the financial crisis and that every panel bank submitted suppressed quotes.

Persistent suppression harmed any investor or speculator who received a floating rate: one side of every swap and every futures contract, and holders of floating-rate loans.

2. Procedural History

2.1. Prior Rulings

Our prior rulings in this MDL addressed the legal sufficiency of class complaints filed on behalf of "Exchange-Based Plaintiffs" who traded exchange-traded futures and options, "OTC Plaintiffs" who traded over-the-counter instruments such as interest rate swaps, and "Bondholder Plaintiffs" who held floating-rate bonds. We also addressed an initial set of complaints filed by affiliates of Charles Schwab (the "Schwab Plaintiffs" or "Schwab").

2.1.1. LIBOR I

LIBOR I addressed the putative classes' and Schwab's complaints of persistent suppression. In so doing, we reached six key holdings, many of which are discussed in greater detail below:

• We dismissed all private antitrust claims brought under the federal Sherman and Clayton Acts and California law

because plaintiffs failed to allege an antitrust injury. 935 F. Supp. 2d at 685-95, 736.

• We sustained some of the Exchange-Based Plaintiffs' claims under the Commodity Exchange Act (CEA) because plaintiffs had adequately pleaded manipulation of Eurodollar futures contracts and because plaintiffs' claims did not involve an extraterritorial application of the CEA. (We did not, however, accept plaintiffs' theory that LIBOR was itself a "commodity" whose manipulation was actionable through the CEA.) In sustaining these claims, we permitted plaintiffs to proceed on theories of primary liability, vicarious liability, and aiding and abetting. Id. at 695-97, 713-24.

• We dismissed some of the Exchange-Based Plaintiffs' CEA claims as untimely because plaintiffs were on inquiry notice of their injury by May 29, 2008. Id. at 697-713.

• We dismissed Schwab's RICO claims because federal law does not allow a RICO plaintiff to rely on conduct that is actionable as securities fraud and because Schwab relied on an impermissible extraterritorial application of RICO. Id. at 724-34, abrogated in part, Eur. Cmty. v. RJR Nabisco, Inc., 764 F.3d 129 (2d Cir.), panel reh'g denied, 764 F.3d 149 (2d Cir. 2014), en banc reh'g

denied, 783 F.3d 123 (2d Cir. 2015), petition for cert. filed, 83 U.S.L.W. ___ (U.S. July 27, 2015) (No. 15-138).

• We dismissed the Exchange-Based Plaintiffs' unjust enrichment claims under New York law because the Exchange-Based Plaintiffs and defendants had not dealt directly with each other. 935 F. Supp. 2d at 737-38.

• Having dismissed all of the federal claims asserted by the OTC Plaintiffs and Schwab, we declined to exercise supplemental jurisdiction over their remaining state-law claims. Id. at 734-36.

2.1.2. LIBOR II

To every action, there is an opposite reaction. LIBOR I engendered responsive motions by plaintiffs and defendants, which we considered in LIBOR II. Regarding the CEA, the Exchange-Based Plaintiffs moved to certify LIBOR I for interlocutory appeal on the question of whether LIBOR was a "commodity" and moved to amend their complaint to add claims of trader-based manipulation, while three defendants moved to reconsider our holding that plaintiffs had adequately pleaded scienter. Regarding antitrust claims, plaintiffs sought leave to amend to add new allegations of antitrust injury. Finally, the OTC Plaintiffs sought to have their state-law claims heard in federal court regardless of our disposition of their federal claims, on a theory that the Class Action Fairness Act conferred diversity jurisdiction over their action. We resolved these issues as follows:

• We denied the Exchange-Based Plaintiffs' motion for interlocutory appeal. 962 F. Supp. 2d at 610-14.

• We denied without prejudice defendants' motion for reconsideration of our scienter holding, finding that several issues had not been sufficiently briefed. Id. at 614-19.

• We denied leave for the Exchange-Based Plaintiffs to add claims of trader-based manipulation because plaintiffs had failed to plead actual damages from that category of misconduct. Id. at 619-624. However, we granted the Exchange-Based Plaintiffs leave to add Société Générale as a defendant. Id. at 624.

• We denied leave for plaintiffs to add allegations addressed to antitrust injury because plaintiffs had already had an opportunity to plead their best case in a round of amended complaints and because plaintiffs' proposed amendments would not have altered the outcome of LIBOR I. Id. at 624-28.

• We acknowledged the OTC Plaintiffs' new theory of diversity jurisdiction, id. at 628, and therefore permitted them to re-plead their unjust enrichment claims and to plead new claims for breaches of the implied covenant of good faith and fair dealing. In the context of over-the-counter trading, these claims were not unduly delayed and were not futile. Id. at 628-35.

• Finally, we continued our existing stay of all complaints other than the Exchange-Based, OTC, and Bondholder class complaints. Id. at 635; cf. Mem. & Order, 2012 WL 3578149, 2012 U.S. Dist. LEXIS 120893 (S.D.N.Y. Aug. 14, 2012), ECF No. 205.

2.1.3. LIBOR III

We returned to the Exchange-Based and OTC Plaintiffs' complaints once again in LIBOR III. The Exchange-Based Plaintiffs sought reconsideration of our decision in LIBOR II that their proposed amendment failed to state trader-based injury, and attempted once again to state trader-based claims. The defendants who had moved for reconsideration in LIBOR II of LIBOR I's scienter holding renewed their motion. Defendants also renewed their motion to dismiss the Exchange-Based Plaintiffs' CEA claims that arose after plaintiffs were on inquiry notice. And defendants moved to dismiss the OTC Plaintiffs' new and revived state-law claims. We held as follows:

• As to the Exchange-Based Plaintiffs' trader-based injuries, we reaffirmed the holding of LIBOR II that plaintiffs were required to plead injury from particular instances of manipulation on particular dates. 27 F. Supp. 3d at 460-62. Even so, we found that plaintiffs' new pleading had successfully alleged injury from Barclays's and Rabobank's trader-based manipulation. Id. at 462-66.

• We agreed to reconsider our holding in LIBOR I that the Exchange-Based Plaintiffs had adequately pleaded scienter, but reached the same conclusion on a different ground, namely that defendants, regardless of their own market positions, persistently suppressed LIBOR while knowing that their conduct would affect the prices of Eurodollar futures. Id. at 466-71.

• Extending the statute of limitations holdings of LIBOR I, we held that CEA claims arising after plaintiffs were on inquiry notice but more than two years before the filing date were untimely. Id. at 471-77.

• We sustained the OTC Plaintiffs' contract and unjust enrichment claims against their counterparties, but dismissed those claims against non-counterparties. Id. at 477-84.

• Finally, we dismissed the Exchange-Based Plaintiffs' new claims against Société Générale as untimely. Although the limitations period was tolled during the pendency of a different class action that had named Société Générale as a defendant, the period of tolling was not long enough to save plaintiffs' amended complaint from a limitations bar. Id. at 484-86.

2.1.4. Other Orders

After LIBOR I, Schwab refiled its state-law claims in California state court while pursuing an appeal of its other claims. See infra at 43. Defendants removed Schwab's state-court action and state-court filings by Salix Capital US and George Maragos to federal courts pursuant to the Edge Act or the Foreign Sovereign Immunities Act. The Judicial Panel on Multidistrict Litigation transferred Schwab's and Maragos's suits to this MDL, and plaintiffs in all three cases moved to remand to state court. We denied remand, holding that (1) Edge Act jurisdiction was proper because the setting of LIBOR is an "international or financial operation," 12 U.S.C. § 632, from which plaintiffs' suits arose, and (2) the FSIA potentially provides an independent avenue to federal jurisdiction because at least one defendant in each action was majority-owned by a foreign government. See Mem. & Order, 2013 WL 6847064, 2013 U.S. Dist. LEXIS 181158 (S.D.N.Y. Dec. 30, 2013), ECF No. 520. Schwab's leading argument against removal was that the Securities Act of 1933 forbade removal of any case brought pursuant to that Act. See 15 U.S.C. § 77v(a). We noted that Schwab's Securities Act claims were barred by the Securities Act's three-year statute of repose, see § 77m, and that a statute of repose is not subject to tolling during the pendency of a related class action. Mem. & Order at 27, 2013 WL 6847064, at *10, 2013 U.S. Dist. LEXIS 181158, at *36-37 (citing Police & Fire Ret. Sys. of Detroit v. IndyMAC MBS, Inc., 721 F.3d 95, 109 (2d Cir. 2013), cert. dismissed sub nom. Pub. Emps.' Ret. Sys. of Miss. v. IndyMAC MBS, Inc., ___ U.S. ___ 135 S. Ct. 42 (2014)). Accordingly, the presence of Schwab's Securities Act claims was no bar to removal.

Last fall, we also considered and denied three requests that the Exchange-Based Plaintiffs filed in the wake of LIBOR III:

• We declined to reconsider our refusal in LIBOR III to consider plaintiffs' untimely request to add certain post-2007 allegations to their complaint. In reaching

this decision, we applied the inquiry notice holdings of LIBOR I and LIBOR III to trader-based manipulation on the grounds that news articles in Spring 2008 placed plaintiffs on notice of their injuries, regardless of the cause. Mem. & Order, 2014 WL 6488219, at *1-3, 2014 U.S. Dist. LEXIS 165126, at *10-18 (S.D.N.Y. Nov. 18, 2014), ECF No. 848.

• We denied leave to amend the complaint to add claims against affiliates of Lloyds Banking Group. One proposed claim was time-barred, and the other did not present a genuine example of an artificial submission. Id., 2014 WL 6488219, at *3-4, 2014 U.S. Dist. LEXIS 165126, at *18-19. We also denied leave to amend the complaint to add claims against Rabobank because the amendment was based on settlement documents related to Yen LIBOR rather than USD LIBOR. Id., 2014 WL 6488219, at *4, 2014 U.S. Dist. LEXIS 165126, at *19.

2.2. Pending Motions

The outstanding motions are (1) defendants' motions to dismiss the Individual Plaintiffs' claims (ECF Nos. 741, 743, 752, 974); (2) defendants' motions to dismiss putative class actions (ECF Nos. 966, 969), (3) defendants' motions to dismiss elements of the OTC Plaintiffs most recent complaint (ECF Nos. 958, 964), (4) the Exchange-Based Plaintiffs' motion for approval of their proposed process for allocating settlement proceeds (ECF No. 953), (5) the Exchange-Based Plaintiffs' motion for discovery (ECF No. 1001), and (6) the Exchange-Based Plaintiffs' motion for a pre-trial conference regarding an anticipated motion to reconsider our limitations holdings in light of BPP Illinois, LLC v. Royal Bank of Scotland Group PLC, 603 F. App'x 57 (2d Cir. 2015).

This opinion resolves almost all of defendants' motions to dismiss the Individual Plaintiffs' complaints. The remaining motions to dismiss will be decided after forthcoming oral arguments. As announced in open court on February 5, 2015, we anticipate that the Exchange-Based Plaintiffs will present us with a modified proposal for allocation and a proposed subpoena of the Chicago Mercantile Exchange that is agreeable to all parties. Finally, we believe that the present opinion may obviate the Exchange-Based Plaintiffs' request for us to separately address BPP Illinois in the context of their case. See infra at 318. Accordingly, the Exchange-Based Plaintiffs' settlement-related motions and their request for a pre-motion conference regarding BPP Illinois will be denied without prejudice in a separate order.

2.3. The Individual Plaintiffs

We next describe in brief each Individual Plaintiff: who each plaintiff is, the nature of each plaintiff's alleged injury, and the causes of action that each plaintiff asserts.

2.3.1. Amabile

Plaintiffs in Amabile v. Bank of America Corp. ("Amabile"), No. 13-cv-1700 (NRB) (S.D.N.Y.), are eighteen individuals (the "Amabile Plaintiffs") who traded LIBOR-based products such as Eurodollar futures on the Chicago Mercantile Exchange. See Am. Compl. ("Amabile Am. Compl.") ¶ 35, Amabile, ECF No. 12. Some of the Amabile Plaintiffs sue as authorized representatives of trading companies, although they state that "no customer money [was] involved." Id.

The Amabile Plaintiffs sue panel banks for violations of the Commodity Exchange Act (including theories of vicarious liability and aiding and abetting), violations of the Sherman Act, and unjust enrichment.

2.3.2. BATA

The Bay Area Toll Authority ("BATA") is the public entity responsible for administering revenue from state-owned toll bridges in the San Francisco Bay Area. See Am. Compl. ("BATA Am. Compl.") ¶ 19, Bay Area Toll Auth. v. Bank of Am. Corp. ("BATA"), No. 14-cv-1493 (N.D. Cal.), transferred to No. 14-cv-3094 (NRB) (S.D.N.Y.). BATA issued variable-rate bonds backed by its toll bridge revenue, and, in connection with these bonds, traded LIBOR-based interest rate swaps. Id. ¶¶ 19, 252-53.

BATA sues panel banks, the BBA, and some of its swap counterparties for violations of the Sherman Act, the Cartwright Act, RICO, unfair business practices (Cal. Bus. & Prof. Code § 17200), fraud (including aiding and abetting fraud), and interference with prospective economic advantage. BATA sues only the counterparty defendants for breach of contract and unjust enrichment.

2.3.3. California Consolidated Plaintiffs

Plaintiffs in the California Consolidated Cases (the "California Consolidated Plaintiffs") are public entities in California: two cities, five counties, one metropolitan planning administration, one water district, and one university system. See Direct Action Pls.' Consol. First Am. Compl. ("Cal. Consol. Compl.") ¶¶ 31-45, Oct. 8, 2014, ECF No. 684. The California Consolidated Plaintiffs traded LIBOR-based swaps and swaptions (generally in connection with municipal bond issuances) and held LIBOR-based bank notes, certificates of deposit (CDs), and corporate bonds. See Cal. Consol. Compl. ¶¶ 401-506.

The California Consolidated Plaintiffs sue panel banks and counterparty affiliates for fraud (including aiding and abetting), negligent misrepresentation, tortious interference with economic advantage, unjust enrichment, and violations of the Sherman and Cartwright Acts. They sue only the counterparty defendants for breach of contract and breach of the implied covenant of good faith and fair dealing.

2.3.4. CEMA

CEMA Joint Venture ("CEMA") is an Ohio entity that owns real estate in Akron. See Verified Compl. ¶¶ 1, 21 ("CEMA Compl."), CEMA Joint Venture v. Charter One Bank, N.A. ("CEMA"), No. 2013 CV 03 0284 (Ohio Ct. Com. Pl. Tuscarawas Cty.), removed to No. 13-cv-904 (N.D. Ohio), transferred to No. 13-cv-5511 (NRB) (S.D.N.Y.), ECF No. 1 Ex. A. CEMA traded a LIBOR-based swap. Id. ¶¶ 6, 215.

CEMA sets out four claims for relief against its two counterparties (numbered 1, 2, 3, and 5, but not 4). Although not labeled, these claims sound in antitrust, tortious interference, breach of contract, and unjust enrichment.

2.3.5. Darby

Plaintiffs in Darby Financial Products v. Barclays Bank plc ("Darby"), No. 13-cv-8799 (NRB) (S.D.N.Y.), are two related investment and trading firms: Darby Financial Products ("Darby Financial") and Capital Ventures International ("Capital Ventures"; collectively with Darby Financial, the "Darby Plaintiffs"). See Am. Compl. ("Darby Am. Compl.") ¶ 18-19, Darby, ECF No. 36. Both traded LIBOR-based swaps. Id. at ¶ 18 & Ex. A.

The Darby Plaintiffs sue their counterparties and affiliated panel banks for fraud (including aiding and abetting), tortious interference with contract, tortious interference with prospective business relations, civil conspiracy, and violations of the Sherman Act. They sue only their counterparties for unjust enrichment, breach of contract, breach of the implied covenant of good faith and fair dealing.

2.3.6. Fannie Mae

The Federal National Mortgage Association ("FNMA" or "Fannie Mae") is a federally sponsored corporation that is currently subject to a federal conservatorship. See Am. Compl. ("Fannie Mae Am. Compl.") ¶ 15, Fed. Nat'l Mortg. Ass'n v. Barclays Bank plc ("Fannie Mae"), No. 13-cv-7720 (NRB) (S.D.N.Y.), ECF No. 41; Federal Housing Finance Agency, 2014 Strategic Plan for the Conservatorships of Fannie Mae and Freddie Mac 3-4 (May 13, 2014) [hereinafter FHFA Plan], available at http://www.fhfa.gov/AboutUs/Reports/ReportDocuments/2014StrategicPlan05132014Final.pdf. Fannie Mae traded LIBOR-based swaps, held LIBOR-based mortgages and mortgage-backed securities, and issued LIBOR-based loans. See Fannie Mae Am. Compl. ¶ 38.

Fannie Mae sues its counterparties, panel banks, and the BBA for fraud, including theories of aiding and abetting and conspiracy. Fannie Mae sues only its counterparties for breach of contract, breach of the implied duty of good faith and fair dealing, and unjust enrichment.

2.3.7. FDIC

The Federal Deposit Insurance Corporation (the "FDIC") is a federally chartered corporation that insures bank deposits and acts as receiver for insured banks that become insolvent. See 12 U.S.C. § 1821(c) (2012). The FDIC sues on behalf of 38 failed banks that traded LIBOR-based swaps, issued LIBOR-based retail loans, and held LIBOR-based mortgage-backed securities. See Am. Compl. ¶¶ 9, 155-271, 429, 438 ("FDIC Am. Compl."), Fed. Deposit Ins. Corp. v. Bank of Am. Corp. ("FDIC"), No. 14-cv-1757 (NRB), ECF No. 23.

The FDIC sues the failed banks' counterparties, the panel banks, and the BBA for fraud (including aiding and abetting and civil conspiracy) and negligent misrepresentation. The FDIC sues only the panel banks and the BBA for tortious interference with contract and with prospective economic advantage (including theories of aiding and abetting and civil conspiracy), and violations of the Sherman and Donnelly Acts. The FDIC sues only the failed banks' counterparties for breach of contract, breach of the implied covenant of good faith and fair dealing, and unjust enrichment.

2.3.8. Freddie Mac

The Federal Home Loan Mortgage Corporation ("FHLMC" or "Freddie Mac") is, like Fannie Mae, a federally sponsored private corporation that is currently subject to a federal conservatorship. See [Second] Am. Compl. ("Freddie Mac Am. Compl.") ¶ 8, Fed. Home Loan Mortg. Corp. v. Bank of Am. Corp. ("Freddie Mac"), No. 13-cv-342 (E.D. Va.), transferred to No. 13-cv-3952 (NRB) (S.D.N.Y.); FHFA Plan at 3-4. Freddie Mac traded LIBOR-based swaps and held LIBOR-based mortgage-backed securities. See Freddie Mac Am. Compl. ¶ 185.

Freddie Mac sues counterparties, panel banks, and the BBA for violations of the Sherman Act, fraud, and tortious interference with contract. Freddie Mac sues only its counterparties for breach of contract.

2.3.9. Houston

The City of Houston ("Houston") traded LIBOR-based swaps and issued auction rate securities. See Direct Action Pl. City of Houston First Am. Compl. ("Houston Am. Compl.") ¶¶ 387-95, City of Houston v. Bank of Am. Corp. ("Houston"), No. 13-cv-2149 (S.D. Tex.), transferred to No. 13-cv-5616 (NRB) (S.D.N.Y.), MDL ECF No. 685.

Houston now sues its counterparties and panel banks for fraud (including aiding and abetting), negligent misrepresentation, interference with economic advantage, unjust enrichment, and violations of the Sherman Act and the Texas Free Enterprise and Antitrust Act, Tex. Bus. & Com. Code Ann. §§ 15.01-15.52 (West 2011). Houston sues only its counterparties for breach of contract and breach of the implied covenant of good faith and fair dealing.

2.3.10. Maragos (Nassau County)

George Maragos appears in his official capacity as Comptroller of Nassau County, New York, which traded interest rate swaps through the Nassau County Interim Finance Authority. See Am. Compl. ("Maragos Am. Compl.") ¶¶ 8, 10 and introductory paragraph, Maragos v. Bank of Am. Corp. ("Maragos"), Index No. 12/14496 (N.Y. Sup. Ct. Nassau Cty.), removed to No. 12-cv-6294 (E.D.N.Y.), transferred to No. 13-cv-2297 (NRB) (S.D.N.Y.).

Maragos sues panel banks (one of which is a counterparty) for fraud and violations of New York consumer protection law, N.Y. Gen. Bus. Law § 349 (McKinney 2012). Maragos sues only Nassau County's counterparty for breach of contract and unjust enrichment. The Nassau County Interim Finance Authority, which traded on behalf of Nassau County, is a nominal defendant.

2.3.11. NCUA

The National Credit Union Administration (the "NCUA") is an independent federal agency that, among other things, insures deposits at credit unions and acts as conservator or liquidating agent for insured credit unions that become insolvent. See Am. Compl. ("NCUA Am. Compl.") ¶ 5, Nat'l Credit Union Admin. Bd. v. Credit Suisse Grp. AG ("NCUA"), No. 13-cv-2497 (D. Kan.), transferred to No. 13-cv-7394 (NRB) (S.D.N.Y.). The NCUA maintains its present action as liquidating agent for five failed credit unions. Four of these credit unions (U.S. Central, Western Corporate or "Wescorp," Members United Corporate or "Members United," and Southwest Corporate or "Southwest") traded LIBOR-based swaps. See id. ¶¶ 237-38, 249, 260-67, 278-84, 295. We cannot discern what the fifth credit union, Constitution Corporate ("Constitution"), held. The NCUA Amended Complaint states only that Constitution, along with the other four credit unions, "held tens of billions of dollars in investments and other assets that paid interest streams pegged to LIBOR." Id. ¶ 226.

The NCUA sues its counterparties and panel banks for violations of the Sherman Act and various state antitrust laws, unfair business practices on behalf of Wescorp only (Cal. Bus. & Prof. Code § 17200), tortious interference with contract, tortious interference with prospective economic advantage, and unjust enrichment. The NCUA also sues the swap counterparties for breach of contract.

The NCUA recently filed a separate suit directly in this Court. Nat'l Credit Union Admin. Bd. v. Credit Suisse Grp. AG ("NCUA II"), No. 15-cv-2060 (NRB) (S.D.N.Y. filed Mar. 19, 2015). The NCUA has informed us that the purpose of this substantially identical complaint is to protect against the possibility that Kansas, where NCUA was filed, lacks personal jurisdiction over some or all of the defendants. While the instant motions do not ask us to dismiss NCUA II, some of our holdings in this opinion may be binding upon the parties to NCUA II according to ordinary preclusion principles.

2.3.12. Philadelphia

Plaintiffs in City of Philadelphia v. Bank of America Corp. ("Philadelphia"), No. 13-cv-4352 (E.D. Pa.), transferred to No. 13-cv-6020 (NRB) (S.D.N.Y.), are the City of Philadelphia ("Philadelphia") and the Pennsylvania Intergovernmental Cooperation Authority ("PICA"; collectively with Philadelphia, the "Philadelphia Plaintiffs"), a special department of Pennsylvania's government that is charged with overseeing the City of Philadelphia's finances. See Second Am. Compl. ("Phila. Am. Compl.") ¶¶ 18-19, Philadelphia, ECF No. 42; 53 Pa. Stat. Ann. §§ 12720.101-.709 (West 1998). Both entities traded LIBOR-based swaps, and both entities used LIBOR calculated termination payments on LIBOR-based and non-LIBOR-based swaps. See Phila. Am. Compl. ¶¶ 285-86 and Exs.

The Philadelphia Plaintiffs sue their counterparties and affiliated panel banks for fraud (including aiding and abetting), tortious interference with contract, tortious interference with prospective business relations, civil conspiracy relating to fraud and tortious interference, and violations of the Sherman Act. The Philadelphia Plaintiffs sue only their counterparties for breach of contract, breach of the implied covenant of good faith and fair dealing, and unjust enrichment.

2.3.13. Principal

Entities associated with Principal Financial Group have filed two complaints. Principal Financial Group, Inc., Principal Financial Services, Inc., and an affiliated life insurance company (collectively, the "Principal Financial Group Plaintiffs") are plaintiffs in Principal Financial Group, Inc. v. Bank of America Corp. ("Principal Fin. Grp."), No. 13-cv-335 (S.D. Iowa), transferred to No. 13-cv-6014 (NRB) (S.D.N.Y.). Twenty-four investment funds (the "Principal Funds Plaintiffs"; collectively with the Principal Financial Group Plaintiffs, the "Principal Plaintiffs") are plaintiffs in Principal Funds, Inc. v. Bank of America Corp. ("Principal Funds"), No. 13-cv-334 (S.D. Iowa), transferred to No. 13-cv-6013 (NRB) (S.D.N.Y.). Each plaintiff traded LIBOR-based swaps and held LIBOR-based floating-rate bonds and asset-backed securities. See Am. Compl. ("Principal Fin. Grp. Am. Compl.") ¶¶ 151, 202, 216 and Ex. A, Principal Fin. Grp., ECF No. 57; Am. Compl. ("Principal Funds Am. Compl.") ¶¶ 149, 200, 214 and Ex. A, Principal Funds, ECF No. 57.

The Principal Plaintiffs sue their counterparties, panel banks, and the BBA for violations of the Sherman and Donnelly Acts and fraud (including aiding and abetting). They sue their swap counterparties and various entities associated with their bond and ABS portfolio for breach of contract (including the implied covenant of good faith and fair dealing), negligent misrepresentation, and unjust enrichment.

2.3.14. Prudential

The plaintiffs in Prudential Investment Portfolios 2 v. Bank of America Corp. ("Prudential"), No. 14-cv-3220 (D.N.J.), transferred to No. 14-cv-4189 (NRB) (S.D.N.Y.), are three entities associated with Prudential Financial, Inc. ("Prudential"). See Am. Compl. ("Prudential Am. Compl.") ¶¶ 9-14, Prudential, ECF No. 18. The lead plaintiff, Prudential Investment Portfolios 2 ("Prudential"), is an SEC-registered "open-end management investment company," while the other two plaintiffs are investment funds that Prudential controls. See id. ¶¶ 9-11. We follow the Amended Complaint's lead in treating Prudential as the sole plaintiff, suing on behalf of itself and the two funds. See id., introductory paragraph. Prudential traded LIBOR-based swaps and held LIBOR-based bonds and asset-backed securities. See id. ¶ 255 and Exs. A, B, E-G.

Prudential sues its swap counterparties, its bond obligors, and panel banks for fraud (including aiding and abetting), intentional interference with contract, intentional interference with prospective economic relations, unjust enrichment, violations of New Jersey RICO, violations of the Sherman Act, negligent misrepresentation, and civil conspiracy to commit fraud and tortious interference. Prudential sues only its swap counterparties and bond obligors for breach of contract and breach of the implied covenant of good faith and fair dealing.

2.3.15. Salix

Salix Capital US Inc. ("Salix") is the sole plaintiff in Salix Capital US Inc. v. Banc of America Securities LLC ("Salix"), Index No. 651823/2013 (N.Y. Sup. Ct. N.Y. Cty.), removed to No. 13-cv-4018 (NRB) (S.D.N.Y.). Salix was a "sub-advisor" for four investment funds (the "Frontpoint Funds"), and now brings claims as the assignee of those funds, three managers of the funds, and the funds' advisor. See Second Am. Compl. ("Salix Am. Compl.") ¶ 19, Salix, ECF No. 69.

The Frontpoint Funds traded packages of three instruments: an interest rate swap, a fixed-rate corporate bond, and a credit default swap. On the interest rate swap, the Frontpoint Funds paid a fixed rate and received a floating rate based on LIBOR. The corporate bonds paid fixed rates and were funded through repurchase agreements (or a similar mechanism) at the federal funds rate. Essentially, the Frontpoint Funds received a fixed rate in bond coupons and paid a floating rate based on the federal funds rate to the banks that financed the Frontpoint Funds' positions. The credit default swaps hedged the corporate bonds' credit risk. The net result was that the Frontpoint Funds received a floating rate based on LIBOR and paid a floating rate based on the federal funds rate. When banks' credit worsened, the Frontpoint Funds expected to make money; when banks' credit improved, the Frontpoint Funds expected to lose money. See id. ¶ 7-9. For purposes of this proceeding, we focus on the interest rate swap component, which is the only part of this transaction that involved LIBOR.

As assignee of the Frontpoint Funds, Salix sues the Funds' swap counterparties for breach of contract, breach of the implied covenant of good faith and fair dealing, and unjust enrichment. Also as assignee of the Funds, Salix sues the Funds' counterparties and affiliated panel banks for fraud, tortious interference with contract, tortious interference with prospective business advantage, civil conspiracy to commit fraud (including aiding and abetting) and tortious interference, and violations of the Sherman Act. On its own behalf and as assignee of the Funds' other managers and advisor, Salix sues the counterparties and panel banks for fraud (including aiding and abetting) and civil conspiracy to commit fraud, on the theory that the Funds' managers and advisors expected to be paid based on the Funds' profits.

2.3.16. Schwab

Plaintiffs in Charles Schwab Corp. v. Bank of Am. Corp. ("Charles Schwab"), No. CGC-13-531016 (Cal. Super. Ct. S.F. Cty.), removed to No. 13-cv-2244 (N.D. Cal.), transferred to No. 13-cv-7005 (NRB) (S.D.N.Y.), are fourteen entities (the "Schwab Plaintiffs" or "Schwab") affiliated with the Charles Schwab investment company. See Am. Compl. ("Schwab Am. Compl."), ECF No. 672. The Schwab Plaintiffs held LIBOR-based certificates of deposit (CDs), corporate debt, and mortgage-related instruments. Id. at ¶¶ 269, 277-300. Additionally (allegedly in reliance on the accuracy of LIBOR), the Schwab Plaintiffs held fixed-rate CDs and corporate debt. Id. at ¶ 270.

The Schwab Plaintiffs previously filed a separate group of cases, in which they alleged violations of state and federal antitrust law, violations of state and federal securities law, violations of federal RICO, and various other state-law claims. The Schwab Plaintiffs voluntarily withdrew their securities law claims in a failed attempt to bolster their RICO theory. We then dismissed the antitrust and RICO claims with prejudice and declined to exercise supplemental jurisdiction over the rest of the claims. Plaintiffs re-filed their state-law claims (except for the state-law antitrust claim) in California state court and appealed our antitrust and RICO holdings. The Second Circuit consolidated Schwab's appeal with that of the Bondholder Plaintiffs and dismissed both appeals on procedural grounds. See Nos. 13-3565-cv (L), -3636-cv (Con), 2013 WL 9557843, 2013 U.S. App. LEXIS 26157 (2d Cir. Oct. 30, 2013). The Bondholder Plaintiffs, but not Schwab, petitioned for certiorari and won reversal, thus reinstating the Bondholder Plaintiffs', but not Schwab's, appeal, see Gelboim v. Bank of Am. Corp., 574 U.S. ___, 135 S. Ct. 897 (2015), whereupon moved for the Second Circuit to recall the mandate of dismissal. The Second Circuit denied Schwab's motion. See Schwab Money Mkt. Fund v. Bank of Am. Corp., 2015 WL 756248, 2015 U.S. App. LEXIS 3255 (2d Cir. Feb. 10, 2015). In response Schwab petitioned for certiorari from the Circuit's February 10 order and filed a second appeal from LIBOR I. Schwab has joined in the Bondholder Plaintiffs' briefing while awaiting a ruling on defendants' motion to dismiss the Schwab Plaintiffs' second appeal as untimely. Schwab' certiorari petition remains pending.

The matter before us is a continuation of Schwab's re-filing in state court, which was removed to federal court pursuant to the Edge Act and the FSIA. In its most recent complaint, Schwab sues the sellers, issuers, and underwriters of their debt for fraud (including aiding and abetting), unfair business practices (Cal. Bus. & Prof. Code § 17200 (West 2008)), interference with prospective economic advantage, breach of the implied covenant of good faith and fair dealing, California's blue sky law (Cal. Corp. Code §§ 25400-01 (West 2006)), rescission of contract, unjust enrichment, violations of the Securities Exchange Act of 1934 and SEC Rule 10b-5 (including through a control person theory), and violations of the Securities Act of 1933.

2.3.17. Triaxx

Plaintiffs in Triaxx Prime CDO 2006-1 Ltd. v. Bank of America Corp., Index No. 654418/2013 (N.Y. Sup. Ct. N.Y. Cty.), removed to No. 14-cv-146 (NRB) (S.D.N.Y.), are, according to Triaxx's complaint "limited partnership[s] organized under the laws of the Cayman Islands with . . . principal place[s] of business in New York, New York. Am. Compl. ("Triaxx Am. Compl.") ¶¶ 14-16, Triaxx, ECF No. 14. While we do not rely on sources outside Triaxx's complaint for any conclusion in this Memorandum and Order, we think it useful to provide a more tangible description of these entities (collectively, the "Triaxx Plaintiffs" or "Triaxx"). The Triaxx Plaintiffs are three collateralized debt obligations (CDOs)—pooled investment vehicles through which investors obtained interests in residential mortgages. Investors selected interests (or "tranches") of the CDOs with different levels of risk. The CDOs paid their mortgage proceeds first to the investors in the highest and safest tranches, next to the investors in lower "mezzanine" tranches, and finally to the investors in the lowest "equity" tranche. See generally Compl. ¶¶ 18-20, SEC v. ICP Asset Mgmt., LLC, No. 10-cv-4791 (LAK) (S.D.N.Y. June 21, 2010), ECF No. 1.

The Triaxx Plaintiffs each held LIBOR-based mortgage-backed securities (MBS). See Triaxx Am. Compl. ¶ 155 and Apps. A-D.

Triaxx sues panel banks and the issuers of their MBS for fraud (including aiding and abetting) and negligent misrepresentation, and sues various entities connected with the MBS for tortious interference with economic relations.

III. PERSONAL JURISDICTION

One or more defendants in each case contest personal jurisdiction. In general, defendants acknowledge that they are bound by forum selection clauses where applicable (although there is a dispute concerning the degree to which they are so bound), and the United States defendants have refrained from challenging the existence of general personal jurisdiction in their home forums. Beyond these concessions, defendants collectively mount a substantial challenge to personal jurisdiction, particularly emphasizing two recent Supreme Court decisions on personal jurisdiction. See Walden v. Fiore, 571 U.S. ___, 134 S. Ct. 1115 (2014) (specific personal jurisdiction); Daimler AG v. Bauman, 571 U.S. ___, 134 S. Ct. 746 (2014) (general personal jurisdiction).

For their part, plaintiffs argue that we should recognize personal jurisdiction under four broad theories. First, the New York and Virginia plaintiffs argue that we should find general personal jurisdiction over certain bank defendants in those states. Second, plaintiffs present a number of arguments supporting specific personal jurisdiction against the remaining defendants in the various relevant forums. Third, plaintiffs argue that certain defendants have consented by contract to the jurisdiction of courts in New York or other states. Fourth, the Schwab Plaintiffs argue that defendants have forfeited their opportunity to challenge personal jurisdiction. For the reasons that follow, we reject plaintiffs' theories of general personal jurisdiction and forfeiture. And, while we read the relevant forum selection clauses more broadly than defendants, we reject most of plaintiffs' theories of specific personal jurisdiction.

Our rulings on specific personal jurisdiction result from the unusual factual circumstances in these cases. In an ordinary fraud case involving an out-of-forum defendant, specific personal jurisdiction might be available on the ground that the defendant aimed its fraudulent conduct at the plaintiff's home forum. In an ordinary case of commodities or securities fraud, specific personal jurisdiction might be available on the ground that the defendant aimed its manipulative conduct at a United States market. But LIBOR manipulation does not fit that mold.

Broadly speaking, the panel bank defendants are alleged to have manipulated LIBOR by persistently suppressing it and, to a lesser degree, by manipulating it on a day-to-day basis at the behest of traders. The false submissions were not intended to affect the value of plaintiffs' portfolios, and were not aimed at the United States or at any forum state. It appears likely that, at least in the context of persistent suppression, many defendants' fraudulent conduct took place outside the United States. And while the effect of LIBOR manipulation in the states in which plaintiffs sued was foreseeable, mere foreseeability does not confer personal jurisdiction.

Nor do we find personal jurisdiction on the basis of plaintiffs' allegations of conspiracy, which we reject for jurisdictional purposes as we do on the merits. Nor do we find personal jurisdiction on the basis of activities by certain panel banks and the BBA to "market" LIBOR and LIBOR-based products in the United States, because plaintiffs' claims (to the extent that they are legally viable) do not rest on those "marketing" contacts. Instead, guided by Supreme Court and Second Circuit precedents, we uphold personal jurisdiction only as to claims that arise out of or relate to the defendants' forum-related conduct.

Although such rulings may superficially stand in tension with our rulings in Part VI below upholding some of plaintiffs' fraud theories, it is elemental that the jurisdictional and merits questions are distinct. As Judge Friendly wrote, in an earlier era of learning on the topic of personal jurisdiction, "attaining the rather low floor of foreseeability necessary to support a finding of tort liability is not enough to support in personam jurisdiction." Leasco Data Processing Equip. Corp. v. Maxwell, 468 F.2d 1326, 1341 (2d Cir. 1972), abrogated on other grounds by Morrison v. Nat'l Australia Bank Ltd., 561 U.S. 247 (2010). And, because "[d]ue process limits on [a court's] adjudicative authority principally protect the liberty of the nonresident defendant—not the convenience of plaintiffs or third parties," Walden, 134 S. Ct. at 1122, it is ultimately unremarkable that we lack personal jurisdiction as to some claims that appear viable on the merits.

1. Preliminary Matters

1.1. Relation Between Jurisdictional and Merits Rulings

Throughout this discussion of personal jurisdiction, we decline to blind ourselves to our ultimate conclusions on the merits. Although personal jurisdiction is generally treated as a threshold question, a court is permitted to dismiss claims on the merits in cases "with multiple defendants—over some of whom the court indisputably has personal jurisdiction—in which all defendants collectively challenge the legal sufficiency of the plaintiff's [claims]." Chevron Corp. v. Naranjo, 667 F.3d 232, 246 n.17 (2d Cir. 2012); see, e.g., ONY, Inc. v. Cornerstone Therapeutics, Inc., 720 F.3d 490, 498 n.6 (2d Cir. 2013). Here, defendants expressly invite us to consider the merits of some claims (namely, the antitrust and RICO claims) before addressing personal jurisdiction, and defendants raise largely overlapping legal issues in their remaining merits challenges.

Thus, although we largely treat personal jurisdiction as a threshold question, we apply merits determinations where appropriate. This avoids the need to exhaustively treat the numerous personal jurisdiction issues that would arise if it were assumed that all of plaintiffs' substantive theories were meritorious.

1.2. Plaintiffs' Burden

Once a defendant has challenged personal jurisdiction, the plaintiff bears the burden of establishing it. See Penguin Grp. (USA) Inc. v. Am. Buddha, 609 F.3d 30, 34 (2d Cir. 2010). To survive a defendant's challenge to personal jurisdiction at the pleading stage, a plaintiff need only make out a prima facie showing that personal jurisdiction exists. See Dorchester Fin. Sec., Inc. v. Banco BRJ, S.A., 722 F.3d 81, 84 (2d Cir. 2013); Ball v. Metallurgie Hoboken-Overpelt, S.A., 902 F.2d 194, 197 (2d Cir. 1990). The court has "considerable procedural leeway" to "determine the motion on the basis of affidavits alone[,] . . . permit discovery in aid of the motion[,] or . . . conduct an evidentiary hearing on the merits of the motion." Dorchester Fin., 722 F.3d at 84 (quoting Marine Midland Bank, N.A. v. Miller, 664 F.2d 899, 904 (2d Cir. 1981)).

In the absence of an evidentiary hearing to resolve factual disputes, the court must "construe the pleadings and affidavits in the light most favorable to plaintiffs, resolving all doubts in their favor." Porina v. Marward Shipping Co., 521 F.3d 122, 126 (2d Cir. 2008) (citing Marine Midland, 664 F.2d at 904). However, the court is not to "draw argumentative inferences in the plaintiff's favor," Robinson v. Overseas Military Sales Corp., 21 F.3d 502, 507 (2d Cir. 1994) (internal quotation marks omitted), and is not "bound to accept as true a legal conclusion couched as a factual allegation," Jazini v. Nissan Motor Co., 148 F.3d 181, 185 (2d Cir. 1998). We therefore assume as true the factual contents of plaintiffs' pleadings and declarations.

Ordinarily, a court would assess personal jurisdiction under the standards both of the relevant jurisdictional statute—usually a state long-arm statute—and constitutional due process. See, e.g., Sonera Holding B.V. v. Çukurova Holding A.S., 750 F.3d 221, 224 (2d Cir. 2014). Here, however, defendants do not oppose personal jurisdiction on the basis of the limiting principles of any state's long-arm statute, so we may assume that the relevant long-arm statutes permit the exercise of personal jurisdiction up to the limits of due process.

1.3. State-by-State Jurisdictional Analysis

Although defendants do not rely on the limitations of state long-arm statutes, nonetheless we must analyze plaintiffs' jurisdictional allegations with respect to the states in which they brought suit. As a general matter, a federal court may exercise personal jurisdiction over a defendant only if the defendant would be "subject to the jurisdiction of a court of general jurisdiction in the state where the district court is located." Fed. R. Civ. P. 4(k)(1)(A). Thus, "a federal district court will not assert jurisdiction over a foreign corporation in an ordinary diversity case unless that would be done by the state court under constitutionally valid state legislation in the state where the court sits." Arrowsmith v. United Press Int'l, 320 F.2d 219, 222 (2d Cir. 1963) (en banc).

1.3.1. Actions Brought Outside of New York

With respect to actions brought in states other than New York, and transferred here for pretrial proceedings under the MDL statute, 28 U.S.C. § 1407 (2012), we analyze whether personal jurisdiction exists not in New York, but in the transferor court. See In re Ski Train Fire in Kaprun, Austria on Nov. 11, 2000, 343 F. Supp. 2d 208, 213 (S.D.N.Y. 2004); In re Sterling Foster & Co. Sec. Litig., 222 F. Supp. 2d 289, 300 (E.D.N.Y. 2002); see also In re "Agent Orange" Prod. Liab. Litig., 818 F.2d 145, 163 (2d Cir. 1987) ("Transfers under Section 1407 are simply not encumbered by considerations of in personam jurisdiction and venue." (quoting In re FMC Corp. Patent Litig., 422 F. Supp. 1163, 1165 (J.P.M.L. 1976))). However, we conduct this analysis according to the law not of the transferor circuit, but of the Second Circuit. See In re Methyl Tertiary Butyl Ether ("MTBE") Prod. Liab. Litig., No. 00-cv-1898 (SAS), 2005 WL 106936, at *5, 2005 U.S. Dist. LEXIS 753, at *22 (S.D.N.Y. Jan. 18, 2005); see also Menowitz v. Brown, 991 F.2d 36, 40 (2d Cir. 1993) (per curiam) ("[A] transferee court should apply its interpretations of federal law, not the constructions of federal law of the transferor circuit."). Thus, the precise question is whether a state court of general jurisdiction, in the state in which the action was filed, could constitutionally exercise personal jurisdiction under the precedents of the Supreme Court and the Second Circuit.

The Principal Plaintiffs argue that defendants' New York contacts should be assessed instead of defendants' contacts with Iowa, where they brought their suit, because they have filed an amended complaint purporting to assert venue and personal jurisdiction in New York. See Principal Fin. Grp. Am. Compl. ¶¶ 8, 13; Principal Funds Am. Compl. ¶¶ 8, 12. We disagree. A plaintiff may not unilaterally change venue by the device of amending its complaint, but instead must move for a change of venue. See, e.g., Spar, Inc. v. Info. Res., Inc., 956 F.2d 392 (2d Cir. 1992) (affirming denial of plaintiff's motion for transfer of venue pursuant to 28 U.S.C. § 1406). As discussed in section 7 below, such a motion would not be a fruitful exercise because the Supreme Court has held that a MDL transferee court may not "assign a transferred case to itself for trial." Lexecon Inc. v. Milberg Weiss Bershad Hynes & Lerach, 523 U.S. 26, 28 (1998). Thus, as with the other member cases, we must assess whether there is personal jurisdiction in the state where the action was brought.

1.3.2. Federal Statutory Claims and Pendent Claims

Some of plaintiffs' claims—although few that survive—arise under federal statutes containing provisions authorizing nationwide service of process. Courts in this Circuit commonly hold that "[w]hen the jurisdictional issue flows from a federal statutory grant that authorizes suit under federal-question jurisdiction and nationwide service of process . . . the Second Circuit has consistently held that the minimum-contacts test in such circumstances looks to contacts with the entire United States rather than with the forum state." SEC v. Straub, 921 F. Supp. 2d 244, 253 (S.D.N.Y. 2013) (quoting SEC v. Morton, No. 10-cv-1720 (LAK)(MHD), 2011 WL 1344259, at *12, 2011 U.S. Dist. LEXIS 36487, at *38 (S.D.N.Y. Mar. 31, 2011), report and recommendation adopted, 2011 WL 11768504 (S.D.N.Y. Nov. 3, 2011), and citing Chew v. Dietrich, 143 F.3d 24, 28 n.4 (2d Cir. 1998); SEC v. Unifund SAL, 910 F.2d 1028, 1033 (2d Cir. 1990), Tex. Trading & Milling Corp. v. Fed. Republic of Nigeria, 647 F.2d 300, 314 (2d Cir. 1982), overruled on other grounds by Frontera Res. Azer. Corp. v. State Oil Co. of Azer. Republic, 582 F.3d 393, 398-401 (2d Cir. 2009); Mariash v. Morrill, 496 F.2d 1138, 1143-44 (2d Cir. 1974); SEC v. Softpoint, Inc., No. 95 Civ. 2951 (GEL), 2001 WL 43611, at *5, 2001 U.S. Dist. LEXIS 286, at *15 (S.D.N.Y. Jan. 17, 2001)); see also Hallwood Realty Partners, L.P. v. Gotham Partners, L.P., 104 F. Supp. 2d 279, 283-85 (S.D.N.Y. 2000). The rationale underlying this "national contacts" approach is that "[w]hen the national sovereign is applying national law, the relevant contacts are the contacts between the defendant and the sovereign's nation." In re Oil Spill by Amoco Cadiz, 954 F.2d 1279, 1294 (7th Cir. 1992), quoted by Chew, 143 F.3d at 28 n.4. Although the Second Circuit recently stated that it had "not yet decided" whether to consider the defendants' national contacts where a "federal statute authorizes nationwide service of process," Gucci Am. v. Li, 768 F.3d 122, 142 n.21 (2d Cir. 2014), we are persuaded by the weight of authority that the national contacts approach is both sound and consistent with this Circuit's law.

Additionally, the doctrine of pendent personal jurisdiction provides that "where a federal statute authorizes nationwide service of process, and the federal and state claims derive from a common nucleus of operative fact, the district court may assert personal jurisdiction over the parties to the related state law claims even if personal jurisdiction is not otherwise available." IUE AFL-CIO Pension Fund v. Herrmann, 9 F.3d 1049, 1056 (2d Cir. 1993) (citation and internal quotation marks omitted). However, the "exercise of pendent personal jurisdiction in a particular case is within the discretion of the district court." Action Embroidery Corp. v. Atl. Embroidery, Inc., 368 F.3d 1174, 1181 (9th Cir. 2004); see Herrmann, 9 F.3d at 1059; Gen. Star Indemn. Co. v. Anheuser-Busch Cos., No. 99-7004, 199 F.3d 1322 (table), 1999 WL 1024708, at *2, 1999 U.S. App. LEXIS 29673, at *5 (2d Cir. Nov. 8, 1999) (unpublished opinion). A district court has discretion "to dismiss the pendent claims where 'considerations of juridical economy, convenience and fairness to litigants' so dictate." Oetiker v. Jurid Werke, G. m. b. H., 556 F.2d 1, 5 (D.C. Cir. 1977) (quoting United Mine Workers v. Gibbs, 383 U.S. 715, 726 (1966)).

Most of the plaintiffs in the cases to which the instant motions are addressed purported to plead claims under federal statutes, in particular the federal antitrust laws. However, while in our merits discussion, we uphold the Amabile Plaintiffs' claims under the Commodity Exchange Act to the extent that they are timely, we dismiss the remaining federal-law claims. The CEA contains a nationwide service provision, 7 U.S.C. § 25(c), pursuant to which courts assess personal jurisdiction based on national contacts. See de David v. Alaron Trading Corp., 796 F. Supp. 2d 915, 925-26 (N.D. Ill. 2010); In re Amaranth Nat. Gas Commodities Litig., 587 F. Supp. 2d 513, 526-27 (S.D.N.Y. 2008), aff'd, 730 F.3d 170 (2d Cir. 2013). Thus, as to the Amabile Plaintiffs' CEA claims, we assess defendants' contacts with the United States as a whole. Because it is apparent from the complaint in Amabile that the state-law claims for unjust enrichment derive from a common nucleus of operative fact, we also exercise pendent personal jurisdiction as to those claims.

As to the remaining cases involving federal claims, we decline to exercise pendent personal jurisdiction on the ground that the federal claims are dismissed at the outset of the litigation. See Olin Corp. v. Fisons PLC, 47 F. Supp. 2d 151, 155 (D. Mass. 1999) (declining to exercise pendent personal jurisdiction predicated on claims that had been dismissed); cf. Marcus v. AT&T Corp., 138 F.3d 46, 57 (2d Cir. 1998) ("In general, where the federal claims are dismissed before trial, [supplemental] state claims should be dismissed as well.").

2. General Personal Jurisdiction

2.1. General Principles

Where it applies, the theory of general personal jurisdiction permits suit in a given forum on "any and all claims." Goodyear Dunlop Tires Operations, S.A. v. Brown, 564 U.S. ___, ___, 131 S. Ct. 2846, 2851 (2011). In two recent decisions, Goodyear and Daimler, the Supreme Court has clarified and narrowed the bases on which a court may exercise general personal jurisdiction.

In Goodyear, the Supreme Court held that European tire manufacturers, whose products regularly arrived in North Carolina through the stream of commerce, were not subject to suit in North Carolina courts on a claim involving tires that were manufactured and sold abroad and also caused injury abroad. The Court rejected the notion that "[a] corporation's 'continuous activity of some sorts within a state' . . . 'is . . . enough to support the demand that the corporation be amenable to suits unrelated to that activity.'" Id. at 2856 (quoting Int'l Shoe Co. v. Washington, 326 U.S. 310, 318 (1945)). Instead, the Court held that general personal jurisdiction exists only where defendants' "affiliations with the State are so 'continuous and systematic' as to render them essentially at home in the forum State." 131 S. Ct. at 2851 (quoting Int'l Shoe, 326 U.S. at 317). For a corporate defendant, the "paradigm" forums are the corporation's "place of incorporation[] and principal place of business." 131 S. Ct. at 2853-54.

In Daimler, the Supreme Court held that a federal court in California lacked personal jurisdiction over a German company to adjudicate claims having no connection to California (or the United States), even though the defendant (through a United States subsidiary whose contacts were imputed to the defendant for argument's sake) did substantial business in California on a continuous basis. See 134 S. Ct. at 752; see also id. at 767 (Sotomayor, J., concurring in judgment) (noting that the subsidiary had $4.6 billion in annual California sales, accounting for 2.4% of the parent's worldwide business). The Court did not find these contacts sufficient to make the defendant "'essentially at home in the forum State,' i.e., comparable to a domestic enterprise in that State." Daimler, 134 S. Ct. at 758 n.11 (majority opinion) (quoting Goodyear, 131 S. Ct. at 2851). The Court rejected the proposition that general personal jurisdiction could be exercised in any place where a defendant's "sales are sizable." Daimler, 134 S. Ct. at 761. Explaining that "[a] corporation that operates in many places can scarcely be deemed at home in all of them," id. at 762 n.20, the Court held that only in "an exceptional case" would "a corporation's operations in a forum other than its formal place of incorporation or principal place of business . . . be so substantial and of such a nature as to render the corporation at home in that State," id. at 761 n.19.

Daimler held out Perkins v. Benguet Consolidated Mining Co., 342 U.S. 437 (1952), as such an exceptional case. In Perkins, a Philippine corporation was sued in Ohio on a claim having no relation to the defendant's Ohio activities. Id. at 438. However, at the time the corporation was sued, war in the Pacific had forced the defendant to cease its Philippine operations and to conduct its business from an interim Ohio headquarters. Id. at 447-48. The Court held that jurisdiction over the claim was not inconsistent with due process, id. at 448, because, as the Court has since explained, "Ohio was the corporation's principal, if temporary, place of business." Daimler, 134 S. Ct. at 756 (quoting Keeton v. Hustler Magazine, Inc., 465 U.S. 770, 780 n.11 (1984)).

The Second Circuit, in the wake of Daimler and Goodyear, has emphasized that "[a]side from 'an exceptional case,' . . . a corporation is at home (and thus subject to general jurisdiction, consistent with due process) only in a state that is the company's formal place of incorporation or its principal place of business." Gucci, 768 F.3d at 135 (quoting Daimler, 134 S. Ct. at 761 n.19). Thus, "even a company's 'engage[ment] in a substantial, continuous, and systematic course of business' is alone insufficient to render it at home in a forum." Sonera, 750 F.3d at 226 (quoting Daimler, 134 S. Ct. at 761).

2.2. Application

Plaintiffs argue that certain defendants are subject to general personal jurisdiction in New York and/or Virginia because of the magnitude of those defendants' continuous activities there. In New York, these defendants are Barclays Bank PLC, Credit Suisse AG, Deutsche Bank AG, Royal Bank of Canada, and Royal Bank of Scotland Group PLC, which plaintiffs dub the "Global Bank Defendants." In Virginia, the defendants are Bank of America, N.A., Barclays Bank PLC, Citibank, N.A., Credit Suisse International, Deutsche Bank AG, JPMorgan Chase Bank, N.A., RBS plc, and Société Générale, which plaintiffs dub the "Virginia Commercial Defendants."

We find no merit in plaintiffs' arguments as to the moving defendants. It is uncontested that the Global Bank Defendants are not incorporated in New York, nor is their principal place of business in New York, and that the Virginia Commercial Defendants are not incorporated in Virginia, nor is their principal place of business in Virginia. Accordingly, under Goodyear and Daimler, general personal jurisdiction may not be exercised unless these defendants present an exceptional case.

The New York plaintiffs rely principally on representations that Global Bank Defendants made to the Federal Reserve and the Federal Deposit Insurance Corporation that their New York operations are "significant to the activities of a critical operation or core business line." See Decl. of Lisa M. Kaas ("Kaas Decl."), at ¶¶ 6, 50, 54, ECF No. 892. They also rely on licenses that three Global Bank Defendants obtained from the New York State Department of Financial Services. See Kaas Decl. ¶¶ 6, 39, 53. The Virginia plaintiff, Freddie Mac, emphasizes that the Virginia Commercial Defendants have had a long-standing relationship with Freddie Mac, involving many financial transactions collectively valued in the billions or even hundreds of billions of dollars. See Pls.' PJ Mem. 29-30 & nn. 62-69 (citing numerous paragraphs of Freddie Mac's Amended Complaint and supporting declaration).

We conclude without hesitation that none of these contacts with New York and Virginia comprises an "exceptional case where [the defendant's] contacts with [those states] are so substantial as to render it 'at home' in that state." Sonera, 750 F.3d at 223. The defendant in Daimler, like the defendants here, had continuous and systematic contacts, which were both substantial in an absolute financial sense, and significant to the defendant's business. Notwithstanding the defendant's "sizable" operations, general personal jurisdiction was nonetheless inappropriate. See Daimler, 134 S. Ct. at 761. Here, even if the defendants' contacts with New York and Virginia happen to have been somewhat greater as an absolute or relative matter than were Daimler's contacts with California, the difference is not so marked as to present a case where, as in Perkins, the defendant could truly be called at home in those states. Accordingly, we reject plaintiffs' arguments that the Global Bank Defendants and Virginia Commercial Defendants are subject to general jurisdiction in New York and Virginia respectively.

3. Specific Personal Jurisdiction

3.1. General Principles

In contrast to general personal jurisdiction, specific personal jurisdiction "depends on an affiliatio[n] between the forum and the underlying controversy." Goodyear, 131 S. Ct. at 2851 (internal quotation marks omitted). Specific personal jurisdiction requires that "the defendant's suit-related conduct must [have] create[d] a substantial connection with the forum." Walden, 134 S. Ct. at 1121. A plaintiff asserting specific personal jurisdiction "must establish the court's jurisdiction with respect to each claim asserted." Sunward Elecs., Inc. v. McDonald, 362 F.3d 17, 24 (2d Cir. 2004) (emphasis in original).

The analysis of specific personal jurisdiction comprises two steps. First, the court evaluates whether the defendant has "purposefully established minimum contacts within the forum." Burger King Corp. v. Rudzewicz, 471 U.S. 462, 476 (1985). This may consist of "purposeful availment," found where "the defendant purposefully availed itself of the privilege of doing business in the forum and could foresee being haled into court there," Licci ex rel. Licci v. Lebanese Canadian Bank, SAL (Licci II), 732 F.3d 161, 170 (2d Cir. 2013) (quoting Bank Brussels Lambert v. Fiddler Gonzalez & Rodriguez, 305 F.3d 120, 127 (2d Cir. 2002)), or "purposeful direction," where "the defendant took 'intentional, and allegedly tortious, actions . . . expressly aimed' at the forum," In re Terrorist Attacks on Sept. 11, 2001, 714 F.3d 659, 674 (2d Cir. 2013) (quoting Calder v. Jones, 465 U.S. 783, 789 (1984)). Personal jurisdiction may not be exercised on the basis of a defendant's "random, fortuitous, or attenuated contacts, or the unilateral activity of another party or a third person." Burger King, 471 U.S. at 475 (citations and internal quotation marks omitted); see Walden, 134 S. Ct. at 1122. Furthermore, the mere "fact that harm in the forum is foreseeable . . . is insufficient" to establish personal jurisdiction. Terrorist Attacks, 714 F.3d at 674; see World-Wide Volkswagen Corp. v. Woodson, 444 U.S. 286, 295 (1980).

Additionally, because specific personal jurisdiction must be established with respect to "each claim asserted," Sunward Elecs., 362 F.3d at 24 (emphasis in original), the claim must arise out of or relate to the defendant's contacts with the forum. See Burger King, 471 U.S. at 472; Licci II, 732 F.3d at 170. The Supreme Court has never specified how close this nexus must be, despite a long-standing split of authority on the appropriate standard:

Some Circuits have held that in order for the defendant to be subject to the jurisdiction of a state the conduct within the state must be a proximate cause of the plaintiff's injury. . . . . Others have held that it is

sufficient if the defendant's conduct in the state is a "but for" cause of the plaintiff's injury.

Where the defendant has only limited contacts with the state it may be appropriate to say that he will be subject to suit in that state only if the plaintiff's injury was proximately caused by those contacts. . . . Where the defendant's contacts with the jurisdiction that relate to the cause of action are more substantial, however, it is not unreasonable to say that the defendant is subject to personal jurisdiction even though the acts within the state are not the proximate cause of the plaintiff's injury.

Id.

Although the Chew opinion does not expressly adopt "but for" causation as a minimum requirement of due process in the tort field, the facts of the case—and of subsequent Second Circuit cases upholding specific personal jurisdiction on the basis of limited contacts with the forum—have involved no less than a "but for" connection between the defendant's forum-directed activities and the claim. We are not aware that any federal appellate court has upheld specific personal jurisdiction on the basis of a lesser showing of relationship. To do so would "[c]onfus[e] or blend[] general and specific jurisdiction inquiries," Goodyear, 131 S. Ct. at 2851, by allowing a defendant's non-suit-related conduct to substitute, for personal jurisdiction purposes, for suit-related conduct.

At the second step of the specific personal jurisdiction inquiry, the burden shifts to the defendant to show that the exercise of jurisdiction would nonetheless be so unreasonable as to "offend 'traditional notions of fair play and substantial justice.'" Asahi, 480 U.S. 102 at 113 (quoting Int'l Shoe, 326 U.S. at 316); see Burger King, 471 U.S. at 477-78; Licci II, 732 F.3d at 170. The following factors are to be considered in analyzing reasonableness: "(1) the burden that the exercise of jurisdiction will impose on the defendant; (2) the interests of the forum state in adjudicating the case; (3) the plaintiff's interest in obtaining convenient and effective relief; (4) the interstate judicial system's interest in obtaining the most efficient resolution of the controversy; and (5) the shared interest of the states in furthering substantive social policies." Metro. Life Ins. Co. v. Robertson-Ceco Corp., 84 F.3d 560, 568 (2d Cir. 1996) (citing Asahi, 480 U.S. at 113-14).

3.2. Conspiracy

Plaintiffs' leading personal jurisdictional theory is that there existed a conspiracy among all defendants to manipulate LIBOR. We reject the applicability of the conspiracy theory of personal jurisdiction in the present context.

The underlying rationale for exercising personal jurisdiction on the basis of conspiracy is that, because co-conspirators are deemed to be each other's agents, the contacts that one co-conspirator made with a forum while acting in furtherance of the conspiracy may be attributed for jurisdictional purposes to the other co-conspirators. See, e.g., Textor v. Bd. of Regents of N. Ill. Univ., 711 F.2d 1387, 1392 (7th Cir. 1983); Chrysler Capital Corp. v. Century Power Corp., 778 F. Supp. 1260, 1266 (S.D.N.Y. 1991). However, "[t]he cases are unanimous that a bare allegation of a conspiracy between the defendant and a person within the personal jurisdiction of the court is not enough." Stauffacher v. Bennett, 969 F.2d 455, 460 (7th Cir. 1992). Nor does "the mere presence of one conspirator . . . confer personal jurisdiction over another alleged conspirator." Leasco, 468 F.2d at 1343. Instead, as plaintiffs acknowledge, courts that have recognized personal jurisdiction on the basis of conspiracy have required plaintiffs to "(1) make a prima facie factual showing of a conspiracy; (2) allege specific facts warranting the inference that the defendant was a member of the conspiracy; and (3) show that the defendant's co-conspirator committed a tortious act pursuant to the conspiracy in [the forum]." Allstate Life Ins. Co. v. Linter Grp. Ltd., 782 F. Supp. 215, 221 (S.D.N.Y. 1992) (citing Chrysler Capital, 778 F. Supp. at 1266).

In the present circumstances, this theory is simply inapplicable. In Part V below, we conclude that plaintiffs have failed to plausibly allege a conspiracy to manipulate USD LIBOR for traders' benefit or to suppress LIBOR during the financial crisis. For the same reasons, plaintiffs' assertion of the conspiracy theory of personal jurisdiction fails for want of a prima facie factual showing.

3.3. LIBOR "Marketing"

Plaintiffs argue that the exercise of jurisdiction over both panel banks and the BBA defendants is warranted because those defendants marketed LIBOR and LIBOR-based financial products within the United States. We include, under this rubric of LIBOR "marketing," plaintiffs' arguments that defendants, including the BBA, made false representations about the quality of LIBOR in order to reassure the public after the emergence of reports that LIBOR was being manipulated.

As to the panel bank defendants, plaintiffs' main factual premise is that they "made LIBOR rates available over the Internet, 'pushed' quotes for interest-rate derivative transactions to Plaintiffs directly in the forum states via electronic and instant messaging through services such as the Bloomberg Professional service (as well as phone calls), and through other licensed data vendors (some also located in forum states)." Pls.' PJ Mem. 13. Plaintiffs argue that the panel banks took these steps "for the express purpose of inducing Plaintiffs' reliance." Id.

Plaintiffs also assert that the BBA defendants marketed LIBOR at the request of BBA members including the panel banks. According to plaintiffs, the BBA registered a trademark for LIBOR with the U.S. Patent and Trademark Office and received millions of pounds from licensing that trademark. The BBA also distributed LIBOR throughout the United States, using the Wall Street Journal and proprietary computer networks such as Bloomberg. Plaintiffs also assert that the BBA, together with other defendants, made various representations (labelled by plaintiffs as the "charm offensive") intended to placate American regulators and investors and thus to induce continued reliance on LIBOR. See Pls.' PJ Mem. 5.

In support of their argument for jurisdiction based on LIBOR "marketing," plaintiffs rely on two doctrinal lines: (1) the so-called "effects" test that derives from Calder v. Jones, see Licci II, 732 F.3d at 173; Panavision Int'l, L.P. v. Toeppen, 141 F.3d 1316, 1321 (9th Cir. 1998); and (2) Internet-oriented cases, see Zippo Mfg. Co. v. Zippo Dot Com, Inc., 952 F. Supp. 1119, 1123-25 (W.D. Pa. 1997). Both of those lines of authority are, at their core, approaches to determining whether a defendant has made purposeful contacts with a forum. See, e.g., Best Van Lines, Inc. v. Walker, 490 F.3d 239, 252 (2d Cir. 2007) (concluding that although Zippo "may help frame the jurisdictional inquiry in some cases," nonetheless "traditional statutory and constitutional principles remain the touchstone of the inquiry" (internal quotation marks omitted)).

Plaintiffs' argument as to LIBOR "marketing" is flawed, principally because it seeks to use contacts that are relevant to claims that fail on the merits to support jurisdiction over separate claims that do not arise out of defendants' contacts with the relevant forum states. To the extent that panel banks or the BBA purposefully directed communications about LIBOR to plaintiffs in a given state, those contacts can in principle support personal jurisdiction over claims that those communications were fraudulent. See, e.g., Wien Air Alaska, Inc. v. Brandt, 195 F.3d 208, 212-13 (5th Cir. 1999) (fraudulent communications directed to plaintiff in plaintiff's home state constituted purposeful availment of the state). We reject such claims on the merits in Part VI below, thus making academic questions of which defendants might be subject to personal jurisdiction in connection with those claims.

Crucially, however, that a panel bank defendant engaged in LIBOR "marketing" activities which reached a given forum state does not mean that the same defendant is subject to personal jurisdiction in that state on the basis of the defendant's manipulation of LIBOR. To the extent that they are legally viable, plaintiffs' claims do not arise out of or relate to defendant's LIBOR "marketing" activities in any given forum. The plaintiffs who allege that they were defrauded by a panel bank without having any relationship with that bank would have relied on LIBOR regardless of whether any particular defendant dealt in LIBOR-based products or communicated information about LIBOR into the plaintiffs' state. It is incontrovertible that the importance of LIBOR was its universal significance, not its projection into any particular state, and plaintiffs do not plead otherwise. Plaintiffs' argument that defendants should be subject to personal jurisdiction on the basis of marketing or sales lacking a "link . . . to the allegedly tortious activity" would improperly create "de facto universal jurisdiction." Adv. Tactical Ordnance Sys., LLC v. Real Action Paintball, Inc., 751 F.3d 796, 801 (7th Cir. 2014).

3.4. Solicitation, Negotiation and Performance of Contracts

Plaintiffs' contracts with certain defendants (called "counterparties") are relevant to our jurisdictional inquiry in two ways. Here, we address plaintiffs' argument that the existence of these contractual relationships supports jurisdiction over counterparty defendants. In section 4 below, we address plaintiffs' separate argument concerning the effect of forum selection clauses in certain contracts.

For jurisdictional purposes, the most significant type of contract are the ISDA Master Agreements used in the context of swap agreements. As the name suggests, these Master Agreements are capable of governing a multitude of individual transactions. For instance, Freddie Mac alleges that it "entered into more than 150 pay-fixed swaps governed by" its master agreement with Bank of America. Freddie Mac Am. Compl. ¶ 228. In addition, a single interest rate swap may itself involve a long-term arrangement. For instance, the City of Houston alleges that, pursuant to ISDA Master Agreements, it entered into swaps with three defendants that matured over up to thirty years. See Houston Am. Compl. ¶¶ 387-392. Aside from the swap agreements, some of the contract claims involve bonds, securitized mortgages, and other asset-backed securities—contracts that were traded on open markets and did not involve any negotiation between the issuer and the ultimate holder.

The Supreme Court has cautioned that "an individual's contract with an out-of-state party" does not, by itself, "automatically establish sufficient minimum contacts in the other party's home forum." Burger King, 471 U.S. at 478. However, personal jurisdiction is likely to exist where the defendant "reach[es] out beyond one state and create[s] continuing relationships and obligations with citizens of another state." Id. at 473 (quoting Travelers Health Ass'n v. Virginia ex rel. State Corp. Comm'n, 339 U.S. 643, 647 (1950)); see, e.g., Agency Rent a Car Sys. v. Grand Rent a Car Corp., 98 F.3d 25, 32 (2d Cir. 1996). Nonetheless, "a continuing relationship is not necessary for conduct to be purposefully directed at the forum." Chloé v. Queen Bee of Beverly Hills, LLC, 616 F.3d 158, 171 (2d Cir. 2010). A court is to consider as a whole the defendants' suit-related contacts with the forum, including "prior negotiations and contemplated future consequences, along with the terms of the contract and the parties' actual course of dealing." Burger King, 471 U.S. at 479.

We conclude that, as a general matter, the swap agreements support personal jurisdiction in the plaintiffs' home forums over claims (whether pleaded in contract, unjust enrichment, or tort) concerning the contractual relationships that they embody. Although many aspects of the ISDA Agreements are standardized, they were individually negotiated by plaintiffs with the counterparty defendants (or their predecessors in interest). They do not embody one-off events as to which the plaintiffs' location is fortuitous, but rather facilitated a course of dealing with plaintiffs in their home jurisdictions over time. Accordingly, the counterparty defendants could readily have foreseen being haled into the courts of the plaintiffs' home forums for claims relating to those contracts. The exercise of personal jurisdiction as to such claims is appropriate.

In contrast, there is no basis to infer that issuers of broadly-traded securities such as bonds and MBS purposely directed those securities into plaintiffs' home forums. These securities may arrive in the hands of plaintiffs and other investors anywhere in the world by the investors' own trades—not at the direction of the issuers. Such a fortuitous, plaintiff-driven contact cannot support personal jurisdiction. See Walden, 134 S. Ct. at 1126 ("[I]t is the defendant, not the plaintiff or third parties, who must create contacts with the forum State."); Hanson v. Denckla, 357 U.S. 235, 253 (1958) ("The unilateral activity of those who claim some relationship with a nonresident defendant cannot satisfy the requirement of contact with the forum State.").

3.5. Wrongdoing in or Directed to Forum States

Plaintiffs argue that defendants' wrongdoing in forum states supports jurisdiction over defendants in those states. To the extent that plaintiffs establish a prima facie case that defendants' LIBOR manipulation took place in the relevant forum, we agree. Thus, in principle, we would uphold jurisdiction in the forum containing the office from which a defendant determined, or transmitted, a false LIBOR submission. The rationale, of course, is that the false submissions were made in the context of the defendant's purposeful availment of the privilege of doing business in the forum. See, e.g., Licci II, 732 F.3d at 170. However, assuming that (as defendants have suggested) many LIBOR submitters were in London or elsewhere abroad, it is not apparent to what degree this conclusion benefits plaintiffs.

The trader-based manipulation claims generally involve at least two actors: the panel bank's LIBOR submitter and the person who made the request of the submitter to submit a falsely low or high quote (the "requester"). As to these claims, the requester is subject to personal jurisdiction in the forum containing his office on a purposeful availment theory. In addition, personal jurisdiction is available in the requester's forum, to which the submitter aimed its conduct and intended an effect (i.e., bolstering the requester's trading position). See Licci II, 732 F.3d at 173 (holding that "exercise of personal jurisdiction may be constitutionally permissible if the defendant expressly aimed its conduct at the forum"); Terrorist Attacks, 714 F.3d at 674. Thus, for example, a panel bank that cooperated with a request made by a trader in New York would be subject to personal jurisdiction in New York in connection with a claim arising out of that request.

Plaintiffs largely refrain from arguing that defendants are subject to personal jurisdiction simply because their out-of-forum conduct had a foreseeable effect in plaintiffs' forum states. To the extent that plaintiffs make such an argument—which we discern principally in the Kansas section of their brief—we reject it. It is bedrock law that merely foreseeable effects of defendants' conduct do not support personal jurisdiction. See World-Wide Volkswagen Corp. v. Woodson, 444 U.S. at 295 (1980); Terrorist Attacks, 714 F.3d at 674. And while personal jurisdiction exists where a defendant took "intentional, and allegedly tortious, actions . . . expressly aimed at the forum," id. (internal quotation marks omitted), there is no suggestion, and it does not stand to reason, that foreign defendants aimed their manipulative conduct at the United States or any particular forum state.

Finally, to the extent that plaintiffs contend that a defendant is subject to personal jurisdiction because that defendant was subject to a government investigation, prosecution, or regulatory action or settlement in the United States or a particular state, we reject the argument, as such governmental actions are not the basis of plaintiffs' claims. Similarly, we reject in this context (as we have often rejected in other contexts) any argument that conduct related to non-USD LIBOR can support personal jurisdiction in these cases, which seek relief based on the manipulation of USD LIBOR only.

3.6. Reasonableness

Defendants argue that, to the extent that we find minimum contacts sufficient to support personal jurisdiction, we should nonetheless conclude that personal jurisdiction fails at the reasonableness step. Specifically, defendants point out that the Supreme Court has cautioned lower courts to exercise "[g]reat care and reserve . . . when extending our notions of personal jurisdiction into the international field." Asahi, 480 U.S. at 115. See also Daimler, 134 S. Ct. at 763 (commenting that the lower court's "uninhibited approach to personal jurisdiction" posed "risks to international comity"). As an initial matter, we note that this argument for the unreasonableness of exercising jurisdiction applies only to the non-United States defendants.

The burden lies with defendants to "present a compelling case" that such "considerations would render jurisdiction unreasonable." Metro. Life, 84 F.3d at 568 (quoting Burger King, 471 U.S. at 477). Unlike in Daimler, we have not recognized the existence of general personal jurisdiction over the defendants who have challenged it. Unlike in Asahi, there has been no suggestion that specific personal jurisdiction might be exercised on the basis of the merely foreseeable stream of commerce. And the approach taken by this Court to personal jurisdiction has been far from "uninhibited." The defendants against whom claims survive are sophisticated entities that, quite rightly, do not argue that they suffer any particularly compelling burden on the basis of litigating in the United States. We reject defendants' suggestion that generalized considerations of fairness, reasonableness, and international comity require us to find an absence of personal jurisdiction.

4. Forum Selection Clauses

Plaintiffs rely on the forum selection clauses in ISDA Master Agreements in support of personal jurisdiction in New York or elsewhere. To a lesser degree, defendants rely on forum selection clauses as a shield.

It is well established that "[p]arties can consent to personal jurisdiction through forum-selection clauses in contractual agreements." D.H. Blair & Co. v. Gottdiener, 462 F.3d 95, 103 (2d Cir. 2006). "Where [pre-dispute] forum-selection provisions have been obtained through 'freely negotiated' agreements and are not 'unreasonable and unjust,' their enforcement does not offend due process." Burger King, 471 U.S. at 472 n.14 (quoting The Bremen v. Zapata Off-Shore Co., 407 U.S. 1, 15 (1972), and citing Nat'l Equip. Rental, Ltd. v. Szukhent, 375 U.S. 311 (1964)). Here, no party argues that the relevant forum selection clauses are either procedurally or substantively unreasonable. Therefore, they will be enforced at the behest of either party.

4.1. Scope of the Clauses

While not disputing the enforceability of the ISDA forum-selection clauses, defendants argue that they constitute consent only to jurisdiction over claims for breach of contract and breach of the implied covenant of good faith and fair dealing, and not claims sounding in tort and unjust enrichment. We do not interpret the clauses so narrowly.

The standardized language of the ISDA Master Agreements contains a forum selection clause providing that, "[w]ith respect to any suit, action or proceedings relating to this Agreement . . . , each party irrevocably . . . submits . . . to the non-exclusive jurisdiction of the courts of the State of New York and the United States District Court located in the Borough of Manhattan in New York City." See, e.g., FDIC Am. Compl. Ex. 31 § 13(b)(i). This language applies, however, only if the parties select New York, rather than English, law to govern the Agreement. See id. There appears to be a consensus that most (but not all) of the Master Agreements involved in these cases contain a New York choice-of-law provision.

"The scope of a forum selection clause is not limited solely to claims for breach of the contract that contains it." Cfirstclass Corp. v. Silverjet PLC, 560 F. Supp. 2d 324, 329 (S.D.N.Y. 2008) (citing Roby v. Corp. of Lloyd's, 996 F.2d 1353, 1361 (2d Cir. 1993)). Determining whether claims fall within the scope of a forum selection clause requires a comparison of "the substance of those claims, shorn of their labels" to the "precise language of the clause." Phillips v. Audio Active Ltd., 494 F.3d 378, 388-89 (2d Cir. 2007). Likewise, the "applicability of a forum selection clause does not depend on [whether] the nature of the underlying action" sounds in contract or tort or derives from a statute. Couvertier v. Concourse Rehab. & Nursing, Inc., 117 A.D.3d 772, 773, 985 N.Y.S.2d 683, 684 (2d Dep't 2014); see, e.g., Bense v. Interstate Battery Sys. of Am., Inc., 683 F.2d 718 (2d Cir. 1982) (enforcing forum selection clause in federal antitrust action). Thus, courts hold that "a contractually-based forum selection clause will also encompass tort claims if the tort claims ultimately depend on the existence of a contractual relationship between the parties, or if resolution of the claims relates to interpretation of the contract, or if the tort claims involve the same operative facts as a parallel claim for breach of contract." Cfirstclass, 560 F. Supp. 2d at 329 (internal quotation marks omitted).

The forum selection clause in the ISDA Master Agreements uses the phrase "relating to this Agreement." Courts in this Circuit and elsewhere give a broad reading to the phrase "relating to" (and its sister phrases "relate to" and "related to") in forum selection clauses. See, e.g., Abbott Labs. v. Takeda Pharm. Co., 476 F.3d 421, 424 (7th Cir. 2007); Roby, 996 F.2d at 1361; Prod. Res. Grp., L.L.C. v. Martin Prof'l, A/S, 907 F. Supp. 2d 401, 414-15 (S.D.N.Y. 2012); In re Optimal U.S. Litig., 813 F. Supp. 2d 351, 367 (S.D.N.Y. 2011); DiChiara v. Ample Faith Invs. Ltd., No. 06 Civ. 3838 (DLC), 2006 WL 3431197, at *5, 2006 U.S. Dist. LEXIS 85972, at *14-15 (S.D.N.Y. Nov. 29, 2006). Here, claims brought against counterparties in their capacity as counterparties relate to the ISDA Agreements, even if they sound in unjust enrichment or fraud in the inducement, because they depend upon the existence of a contractual relationship between the parties.

We caution, however, that this conclusion does not mean that all claims against a counterparty may be brought in a contractually selected forum. The claim must relate to the particular contractual relationship. Thus, for example, we will not uphold jurisdiction over a counterparty for all fraud claims that a plaintiff might bring against that counterparty on the basis of the forum selection clause.

4.2. Successor Entities

Plaintiffs argue that forum selection clauses are applicable against successors in interest to the entities with which they contracted. Under certain circumstances, a non-signatory to a forum selection clause may be bound by, or take advantage, of such a clause. As relevant here, where the doctrine of successorship applies, the forum selection clause will bind the contracting party's successor in interest. See Aguas Lenders Recovery Grp. v. Suez, S.A., 585 F.3d 696, 701 (2d Cir. 2009); Recurrent Capital Bridge Fund I, LLC v. ISR Sys. & Sensors Corp., 875 F. Supp. 2d 297, 306 (S.D.N.Y. 2012). At this stage, we think no more needs to be said than that, where a plaintiff has made a good faith allegation that a defendant is the successor in interest to an entity whose contacts would subject it to specific personal jurisdiction, then that defendant is also subject to personal jurisdiction. As with other rulings that depend upon the facts, of course, the determination of exactly which entities stand in the necessary successor relationship is subject to further factual development.

4.3. Cases Brought Outside of New York

As explained above, for those cases that have been transferred from out of state for pretrial proceedings, our task is to determine whether the transferor court may exercise personal jurisdiction. Thus, as to the cases filed outside of New York and transferred here for pre-trial proceedings, a defendant's contractual consent to the jurisdiction of New York courts is irrelevant. It is of no use to out-of-state plaintiffs that many ISDA Master Agreements permit personal jurisdiction in New York.

The Prudential Plaintiffs, who sued in New Jersey, concede that they entered into certain ISDA Master Agreements that provide for exclusive personal jurisdiction in New York. A defendant may move to dismiss a complaint or claim where a plaintiff has brought suit in a forum excluded by a valid forum selection clause. See TradeComet.com LLC v. Google, Inc., 647 F.3d 472, 475-78 (2d Cir. 2011). Thus, we grant defendants' request to dismiss the claims falling under those clauses.

Additionally, some of the ISDA Master Agreements were customized with forum selection provisions permitting suit in other jurisdictions, such as California, Pennsylvania, or Texas. The parties do not appear to disagree that such forum selection clauses should be applied.

5. Forfeiture

The Schwab Plaintiffs argue that defendants have forfeited their opportunity to contest personal jurisdiction by not raising it as a defense in the three earlier Schwab cases that were dismissed in LIBOR I. We disagree.

A defendant waives an available defense of lack of personal jurisdiction by failing to comply with Rule 12(h) of the Federal Rules of Civil Procedure. See Fed. R. Civ. P. 12(h); China Nat'l Chartering Corp. v. Pactrans Air & Sea, Inc., 882 F. Supp. 2d 579, 588 n.52 (S.D.N.Y. 2012); Index Fund, Inc. v. Hagopian, 107 F.R.D. 95, 101 (S.D.N.Y. 1985). In addition, even a defendant that complies with the rules may forfeit or be estopped from raising the defense. See, e.g., Ins. Corp. of Ir., Ltd. v. Compagnie des Bauxites de Guinee, 456 U.S. 694, 704 (1982) (estoppel); City of New York v. Mickalis Pawn Shop, LLC, 645 F.3d 114, 134 (2d Cir. 2011) (abandonment); Hamilton v. Atlas Turner, Inc., 197 F.3d 58, 60-62 (2d Cir. 1999) (delay).

Here, there is no basis to preclude the Schwab Defendants' personal jurisdiction defense, simply because the present Schwab case is not the same as the ones that were dismissed in LIBOR I. Rule 12(h), which ordinarily precludes a Rule 12(b)(2) motion to dismiss for lack of personal jurisdiction after another Rule 12(b) motion has been brought, is inapplicable here because this is defendants' first motion directed against the present Schwab complaint. Nor have defendants been dilatory in asserting their personal jurisdiction defenses in the present context. We agree with the view of the Northern District of California that "even if actions are closely related . . . defendants do not waive their personal jurisdiction defense by raising it only in a later action, so long as the defendant is not independently seeking affirmative relief in the same court concerning the same transaction or occurrence." In re Cathode Ray Tube (CRT) Antitrust Litig., 27 F. Supp. 3d 1002, 1009 (N.D. Cal. 2014).

Our conclusion that defendants have not forfeited their opportunity to contest personal jurisdiction in the present case brought by the Schwab parties is bolstered by the Second Circuit's opinion in Gucci. There, the Second Circuit permitted the appellant, a foreign bank, to challenge personal jurisdiction for the first time on appeal on the ground that Daimler, which the Supreme Court decided during the pendency of the appeal, had abrogated "prior controlling precedent of this Circuit." Gucci, 768 F.3d at 136. The Circuit explained that "a defendant does not waive a personal jurisdiction argument . . . if the 'argument that the court lacked jurisdiction over [the] defendant would have been directly contrary to controlling precedent in this Circuit.'" Id. at 135-36 (quoting Hawknet, Ltd. v. Overseas Shipping Agencies, 590 F.3d 87, 92 (2d Cir. 2009)). We have no doubt that defendants' decision to refrain from challenging personal jurisdiction in the prior Schwab cases was informed at least in part by no-longer-valid precedents that "a foreign bank with a branch in [the forum state] was properly subject to general personal jurisdiction." Gucci, 768 F.3d at 136 (emphasis omitted). The change in the law of general personal jurisdiction means that it is not unfair to afford the Schwab defendants an opportunity to oppose jurisdiction.

We emphasize that our conclusion applies to the circumstances of the Schwab case only. Certain defendants in other cases in this MDL have argued, in motions not addressed in this opinion, that we should consider personal jurisdiction arguments not made in connection with earlier motions to dismiss. We will consider those arguments in a future decision.

6. Application

In this section, we set forth our bottom-line conclusions as to how our jurisdictional holdings apply to the claims that survive on the merits. We do not attempt to create a complaint-by-complaint, claim-by-claim, defendant-by-defendant list of conclusions due to the sheer number of allegations and the lack of clarity in many complaints as to which claims are alleged against which defendants. Instead, we direct the parties to confer and provide us with a spreadsheet containing a list of claims that, in accordance with the conclusions below, are dismissed on jurisdictional grounds. If the parties disagree as to how any ruling applies to a particular defendant in a particular case, each side may provide a brief summary of its position similar to the summaries provided in the joint spreadsheets the parties submitted following oral argument. We caution the parties that this is not an invitation to re-argue this opinion, to assert new defenses, or to present new points of law.

6.1. Claims Against Counterparties

In our merits discussion, we sustain certain state-law claims against swap counterparties and bond obligors that sound in breach of contract, unjust enrichment, and fraud in the inducement. The relevant counterparties are swap counterparties and bond obligors. As to both, we uphold personal jurisdiction where permitted by a forum selection clause, where the defendant's LIBOR submission was determined or transmitted, and where a trader requested an artificial LIBOR submission. As to swap counterparties, we also uphold jurisdiction where a plaintiff was located when it entered into the swap agreement. As to bond obligors, we also uphold jurisdiction where the bond was issued (i.e., where the bond was placed with an underwriter or agent for sale or marketing), but not necessarily where a plaintiff suffered injury because a bond, once issued, may be traded anywhere in the world without action by the issuer.

To the extent that we uphold personal jurisdiction on the basis of contracts and the forum selection clauses contained within them, we accept allegations of a successor relationship and uphold specific jurisdiction over successor entities to the same degree as jurisdiction over the entities that entered into the contracts.

Notwithstanding these rulings, we reject the exercise of personal jurisdiction where a relevant forum selection clause excludes the forum in which the action was brought.

6.2. Tortious Interference Claims

In our merits discussion, we sustain tortious interference claims connected with trader-based manipulation as pleaded against panel-bank affiliates of swap counterparties. We uphold personal jurisdiction where the LIBOR submission was determined or transmitted, or where the trader requested an artificial LIBOR submission. LIBOR submitters allegedly manipulated LIBOR in order to benefit their colleagues' swap portfolios without regard to the location of their colleagues' counterparties or the swaps' forum selection clauses, and so we do not uphold jurisdiction on the basis of a plaintiff's location or the underlying contract's forum selection clause.

We also sustain tortious interference claims against panel-bank affiliates of bond issuers. In this setting, we uphold personal jurisdiction where the bond was issued as well as where the LIBOR submission was determined or transmitted.

6.3. Fraud and Commodities Exchange Act Claims

In our merits discussion, we find that plaintiffs have stated viable claims for fraud and violations of the Commodity Exchange Act against panel banks on the basis of their allegedly false LIBOR submissions. We uphold personal jurisdiction for these claims only where the LIBOR submission was determined or transmitted. In the case of trader-based manipulation, we also uphold personal jurisdiction in the location of the person who requested the submitter to engage in manipulation. Persistent suppression was not intended to affect investments in any specific place, and so we decline to accept jurisdiction on the basis of the harm caused by defendants in a forum.

While we do not decide whether plaintiffs have stated viable fraud claims against BBA entities for publishing data that the BBA allegedly knew to be false or misleading, see infra at note 107, we do determine that courts in the United States lack personal jurisdiction over the BBA on this theory. Plaintiffs have not alleged that the BBA evaluated the accuracy of panel banks' submissions in the United States, that BBA employees in the United States made the decision to publish false data, that the BBA calculated LIBOR in the United States, or that the BBA's distribution of LIBOR in the United States was a but-for cause of plaintiffs' injuries.

7. Transfer of Venue

Plaintiffs argue, principally in a letter filed after oral argument, that if we find personal jurisdiction lacking, we should "save" their cases by transferring them to an appropriate venue under 28 U.S.C. §§ 1406(a) or 1631, rather than dismissing them. Plaintiffs argue that transfer is in the interests of justice because, if they brought new actions in appropriate jurisdictions, their claims could be deemed time-barred. See Corke v. Sameiet M. S. Song of Norway, 572 F.2d 77, 80 (2d Cir. 1978) (ordering transfer to a district where defendant was subject to personal jurisdiction, in the interest of justice, to avoid "severe[] prejudice" to plaintiff from application of statute of limitations).

Although, having written extensively on the statute of limitations issues, we are attuned to plaintiffs' concern, we reject their suggestion of transferring cases for three reasons. First, we lack power to transfer the cases under the reasoning of the Supreme Court's decision in Lexecon Inc. v. Milberg Weiss Bershad Hynes & Lerach, 523 U.S. 26 (1988). In Lexecon, the Court held that the MDL statute, which requires that a MDL member case "shall be remanded by the panel at or before the conclusion of pretrial proceedings to the district from which it was transferred unless it shall have been previously terminated," 28 U.S.C. § 1407(a), prohibits an MDL transferee court from transferring a case to itself for trial.

Although plaintiffs seek to limit the Lexecon holding to transfers under § 1404, the Second Circuit has held that "Lexecon's reasoning applies equally to transfers pursuant to any venue statute." Shah v. Pan Am. World Servs., Inc., 148 F.3d 84, 90 (2d Cir. 1998). Thus, "any . . . transfers of venue for trial under any statute must follow . . . remand [to the transferor court]." Id. at 91. Similarly, even to cure a failure of personal jurisdiction in the transferor court, Lexecon does not "leave room for the MDL transferee court to transfer MDL cases to other districts." In re Asbestos Prods. Liab. Litig. (No. VI), 965 F. Supp. 2d 612, 622 (E.D. Pa. 2013); accord Manual for Complex Litigation, Fourth (2004), § 20.132, at 223 n.666.

Second, even if we had the power to transfer cases to cure a jurisdictional problem (and even if we concluded that such a transfer would be in the interest of justice), it would be impracticable to do so in these circumstances. Because of the multiplicity of defendants joined in almost every complaint, there may be no single venue in which all defendants are subject to personal jurisdiction. Additionally, although the application of the principles described herein may result in the dismissal of some defendants, it will likely result in the dismissal of few, if any, cases. Under such circumstances, transfer would create at least as many jurisdictional problems as it would solve.

Finally, we reject plaintiffs' proposal to suggest that the Judicial Panel on Multidistrict Litigation remand any of the cases before us so that plaintiffs may move in the transferor court for a transfer to a different venue. Not only would such an extraordinarily inefficient proposal fail to address the problem that there may be no single jurisdictionally appropriate forum, but it would be fundamentally incompatible with the rationale for multidistrict litigation.

IV. PLEADING

1. General Pleading Standards

When deciding a motion to dismiss for failure to state a claim pursuant to Federal Rule of Civil Procedure 12(b)(6), the Court must accept as true all factual allegations in the complaint and draw all reasonable inferences in plaintiff's favor. Harris v. Mills, 572 F.3d 66, 71 (2d Cir. 2009); Kassner v. 2nd Ave. Delicatessen Inc., 496 F.3d 229, 237 (2d Cir. 2007). Nevertheless, a plaintiff's "[f]actual allegations must be enough to raise a right of relief above the speculative level." Bell Atl. Corp. v. Twombly, 550 U.S. 544, 555 (2007). The well-pleaded factual allegations must demonstrate "more than a sheer possibility that a defendant has acted unlawfully" in order to pass muster. Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009). If a plaintiff has "not nudged [its] claims across the line from conceivable to plausible, [the] complaint must be dismissed." Twombly, 550 U.S. at 570.

As to claims for fraud and commodities manipulation, a plaintiff must also meet heightened pleading requirements. See Fed. R. Civ. P. 9(b); LIBOR III, 27 F. Supp. 3d at 458. "In alleging fraud or mistake, a party must state with particularity the circumstances constituting fraud or mistake." Fed. R. Civ. P. 9(b). Although scienter may be pleaded "generally," id., the Second Circuit has interpreted Rule 9(b) to require plaintiff to plead circumstances sufficient to justify a "strong inference" of scienter. See LIBOR III, 27 F. Supp. 3d at 468.

To state a fraud claim, reliance must be pleaded with particularity. See Evans v. Pearson Enters., Inc., 434 F.3d 839, 852-53 (6th Cir. 2006); In re Bear Stearns Cos., Inc. Sec. Derivative & ERISA Litig., 995 F. Supp. 2d 291, 313 (S.D.N.Y. 2014). A general allegation that a plaintiff "read and relied upon" a certain representation is insufficient, see, e.g., Int'l Fund Mgmt. S.A. v. Citigroup Inc., 822 F. Supp. 2d 368, 386 (S.D.N.Y. 2011), in part because plaintiffs are "better positioned than anyone else to know what their actual reliance was," Granite Partners, L.P. v. Bear, Stearns & Co., 58 Supp. 2d 228, 259 (S.D.N.Y. 1999).

In the context of these Rule 12(b)(6) motions, we consider only the pleadings, exhibits to the pleadings, documents referred to within the pleadings, and documents subject to judicial notice. See, e.g., Faulkner v. Beer, 463 F.3d 130, 133-35 (2d Cir. 2006); Hirsch v. Arthur Andersen & Co., 72 F.3d 1058, 1088, 1092 (2d Cir. 1995). As we have implicitly done before, we take judicial notice of LIBOR-related news articles discussed in LIBOR I, not for the truth of the articles, but for the existence of the articles and their content. See Staehr v. Hartford Fin. Servs. Grp., Inc., 547 F.3d 406, 424-26 (2d Cir. 2008); In re Bear Stearns Cos., Inc. Sec., Derivative, & ERISA Litig., 763 F. Supp. 2d 423, 582 (S.D.N.Y. 2011).

2. Pleading of Sporadic Trader-Based Manipulation

To the extent that plaintiffs have attempted to allege injury from trader-based manipulation, we evaluate plaintiffs' pleadings against the standard articulated in LIBOR II and III. See LIBOR III, 27 F. Supp. 3d at 482; LIBOR II, 962 F. Supp. 2d at 621-24. A private plaintiff must plead sufficient information to show injury from a particular incident of manipulation: Which defendant bank engaged in manipulation, of which tenor of USD LIBOR, on which date, and in which direction? Further, did the manipulation affect the published rate? Finally, which instrument did the plaintiff invest in, and how did manipulation of the particular tenor affect the plaintiff's position?

2.1. Exchange-Based Plaintiffs (Amabile)

For the most part, the Amabile Plaintiffs' Exhibit A conforms to the established pleading standards. Nonetheless, we discuss their complaint to make explicit certain baseline assumptions that were implicit in LIBOR III.

First, LIBOR is published in the late morning, London time, and thus it is reasonable to infer that manipulation of a particular date's LIBOR affected trades executed during that date's business hours in Chicago.

Second, we have permitted claims to go forward even though exchange-based plaintiffs, in a chart, have omitted to mention the manipulated tenor. Where public settlement documents do not tell which tenor a bank manipulated on a particular day, it is most reasonable to infer that the bank manipulated the 3-month tenor, which (based on the pleadings) was the most common target of trader-based manipulation. Furthermore, the most common futures contracts by far are tied to 3-month LIBOR, so we assume unless otherwise specified that plaintiffs' trades were in the 3-month future. This assumption is appropriate only in the exchange-based context, because over-the-counter contracts are far more varied.

Defendants misapprehend some lines in the Amabile Plaintiffs' chart that refer to plaintiffs' trading direction as "Buyer/Seller." Defendants incorrectly argue that these lines fail to plead whether plaintiffs were net buyers or net sellers on these dates. In fact, these lines clearly allege that each Amabile Plaintiff executed a yield curve trade on these dates, buying 1-month futures and selling 3-month futures. If these Amabile Plaintiffs adequately plead and prove downward manipulation of 1-month LIBOR and upward manipulation of 3-month LIBOR, then they can recover for both instances of manipulation.

It follows that nearly all the lines in the Amabile Plaintiffs' Exhibit A meet the pleading standards. (By contrast, Exhibit B fails to allege which trades by plaintiffs were affected by defendants' manipulation.) The following chart lists the entries in Exhibit A that do not meet the pleading standards:

Based on the Amabile Plaintiffs' table in Exhibit A, the following plaintiffs have stated trader-based claims against the following defendants. In this chart, we have separated claims according to which of several periods they fell for statute of limitations purposes. (We omit plaintiffs' claims against UBS, which are properly characterized as persistent suppression claims.)

Some of the Amabile Plaintiffs (Joseph Amabile, Federighi, Gough, Krug, Lang, Olson, and Williams) failed to state trader-based claims against any defendant, apparently because each of them failed to keep adequate trading records. We will not entertain plaintiffs' suggestion to hold their pleadings open in the hopes that the Chicago Mercantile Exchange will inform them of their own trading activity. It is the duty of an investor to keep track of his own trades and the duty of a plaintiff to evaluate whether he suffered an injury before coming to court.

2.2. Other Complaints

Plaintiffs in several other cases appear to plead the requisite information, but not in a format that readily lends itself to verification. For example, in the California Consolidated Complaint, information about defendants' trader-based manipulation in USD LIBOR is interspersed with irrelevant anecdotes about Yen LIBOR, see ¶¶ 156-222 (incorporating large settlement documents as exhibits), while information about plaintiffs' holdings ranges across over 100 paragraphs, see ¶¶ 402-506. This is hardly the "short and plain statement of the claim showing that the pleader is entitled to relief" that Rule 8(a)(2) requires.

Within two weeks of this Memorandum and Order, plaintiffs shall serve upon defendants an editable Excel chart showing the date, tenor, currency, and direction of the alleged manipulation; the bank responsible for the manipulation; the effect on the manipulation on LIBOR; and the direction of plaintiff's position. Within two weeks of service, defendants shall mark any lines on plaintiffs' chart that they believe fail the pleading standards, and shall file the completed document on ECF.

We envision this procedure as a mechanical exercise, not an opportunity to insert new facts into plaintiffs' already bulky amended pleadings or to reargue pleading standards. It would be particularly unfair to allow new facts to be pleaded when we have declined to allow some of the Amabile Plaintiffs to supplement their complaints with new discoveries about their own trading. Accordingly, plaintiffs' list should not include any information beyond their existing pleadings, and plaintiffs should cite the locations in their existing complaints where each piece of information is already pleaded. Having addressed pleading standards several times, we believe that there should be few, if any, disagreements as to which claims conform.

V. CONSPIRACY AND OTHER THEORIES OF JOINT LIABILITY

Assertions of conspiracy and collaboration course through plaintiffs' allegations and arguments. Plaintiff rely on theories of joint action to support antitrust claims, RICO claims, contract claims against non-panelist counterparties, joint and several liability, and personal jurisdiction over certain defendants. Close examination of these allegations reveals:

• Many plaintiffs have failed to distinguish between trader-based manipulation and persistent suppression. Especially puzzling are the complaints that dwell at length on the most scandalous instances of trader-based manipulation, but appear to allege injury only from persistent suppression.

• Despite wide-ranging investigations of LIBOR since at least 2011 by the Securities Exchange Commission, the Commodities Futures Trading Commission, the Department of Justice, the New York State Attorney General, and numerous foreign regulators, and despite public settlements and plea agreements involving Barclays, Citi, Deutsche Bank, JPMorgan, Rabobank, RBS, Société Générale, UBS, and brokers, there has been no exposure of a broad conspiracy among traders at different banks to fix USD LIBOR, or of any conspiracy to persistently suppress LIBOR during the financial crisis.

• To the contrary, the number of examples of traders agreeing to manipulate USD LIBOR is actually quite small when repeated claims and claims involving other currencies are excluded from consideration.

• Many complaints contain sufficient allegations that multiple entities within a single banking conglomerate collaborated to engage in trader-based manipulation and possibly persistent suppression.

To evaluate and explain the sufficiency or lack thereof of plaintiffs' claims, we consider each mode of LIBOR manipulation separately. We begin with the intra-bank context and then turn to the inter-bank context, dividing each discussion into trader-based manipulation and persistent suppression subsections.

1. Intra-Bank Collaboration

1.1. Trader-Based Manipulation

In the intra-bank context, trader-based manipulation refers to the scenario of a derivatives trader asking a LIBOR submitter to alter the bank's LIBOR quote to suit the trader's book. This scenario, if proven, would result in liability for the employer of both the trader and the submitter if each worked for different entities.

The pleadings are sufficient to support such a claim against Barclays. As part of a settlement agreement with Barclays, the Commodity Futures Trading Commission found that both Barclays Bank PLC and Barclays Capital Inc. attempted to manipulate LIBOR and made false LIBOR submissions. See Order Instituting Proceedings ("Barclays CFTC Order"), Barclays PLC, No. 12-25 (C.F.T.C. June 27, 2012). Likewise, as part of a settlement agreement between the Department of Justice and Barclays Bank PLC, Barclays Bank PLC admitted that Barclays Capital Inc. manipulated LIBOR. See Non-Prosecution Agreement Between Barclays Bank PLC and the Department of Justice, App. A ("Barclays DOJ Statement of Facts"), June 26, 2012, available at http://www.justice.gov/iso/opa/resources/9312012710173426365941.pdf. While neither document specifies each entity's contribution to LIBOR manipulation, it is nonetheless plausible at this stage that both Barclays Bank PLC and Barclays Capital Inc. assisted the manipulation in some way. Cf. Carpenters Pension Trust Fund of St. Louis v. Barclays PLC, 56 F. Supp. 3d 549, 559 (S.D.N.Y. 2014) (citing Department of Justice settlement to sustain complaint's allegation that Barclays Capital Inc. submitted false LIBOR quotes).

The same reasoning holds as to Lloyds Banking Group plc (formerly Lloyds TSB Group plc) and Lloyds Bank plc (formerly Lloyds TSB Bank plc), both of which admitted misconduct in a CFTC Order. See Order Instituting Proceedings ("Lloyds CFTC Order"), Lloyds Banking Group plc, No. 14-18 (C.F.T.C. July 28, 2014). Thus, it is plausible that Lloyds Banking Group plc contributed to manipulation committed by Lloyds TSB Bank plc, and to manipulation committed by HBOS plc and affiliates after January 19, 2009, when Lloyds TSB Group plc purchased HBOS plc.

Other settlement documents do not give rise to similar inferences of inter-entity cooperation. Admissions by both RBS Securities Japan Ltd. and UBS Securities Japan Co., Ltd., to manipulating Yen LIBOR are not relevant to this MDL, which deals only with USD LIBOR. Not surprisingly, then, neither entity has been named as a defendant here.

1.2. Persistent Suppression

As with trader-based manipulation, we accept as plausible, based on public settlement documents, that Barclays Bank PLC and Barclays Capital Inc. collaborated in persistently suppressing LIBOR during the financial crisis, as did Lloyds Banking Group plc and its affiliates. Again, the public settlement documents do not make it clear what role (if any) each entity played in persistent suppression, and, as noted earlier, at least one opinion in this Court has sustained an allegation that the head of Barclays's money market desk was an employee of Barclays Capital Inc. rather than Barclays Bank PLC. See Carpenters Pension, 56 F. Supp. 3d at 556.

Because the internal division of labor between Barclays and Lloyds entities is peculiarly within the knowledge of those institutions, we do not hold plaintiffs to an especially high standard of particularity with respect to pleading the activities of each entity.

1.3. Legal Consequences of Intra-Bank Collaboration

Before turning to inter-bank collaboration, we explicitly state a point that we believe plaintiffs accept. Collusion within a bank will not support a claim pursuant to section 1 of the Sherman Act. See Copperweld Corp. v. Independence Tube Corp., 467 U.S. 752, 771 (1984). Nor would such collusion support a federal RICO claim, because a federal RICO enterprise must be distinct from a set of corporate affiliates and their employees. See Black Radio Network, Inc. v. NYNEX Corp., 44 F. Supp. 2d 565, 581 (S.D.N.Y. 1999).

2. Inter-Bank Collaboration

We now turn to allegations of inter-bank collaboration.

2.1. Trader-Based Manipulation

The Barclays settlement documents reveal a few examples of trader-based inter-bank collaboration in USD LIBOR, which in at least two cases would have affected the published rates:

• On October 26, 2006, a Barclays submitter decreased Barclays's 3-month submission from 5.38% to 5.375% to assist a former Barclays employee at another institution. See Barclays DOJ Statement of Facts ¶ 26; Financial Services Authority Final Notice to Barclays Bank plc ("Barclays FSA

Final Notice") ¶ 83, June 27, 2012; Amabile Am. Compl. ¶¶ 263, 272.

• On February 28, 2007, a Barclays trader asked another bank's trader to increase his 3-month submission above 5.345%. See Barclays FSA Final Notice ¶ 91; Amabile Am. Compl. ¶ 264. If the other bank's trader complied, then this affected LIBOR.

• On March 29, 2007, a Barclays submitter decreased Barclays's 3-month submission from 5.35% to 5.345% to assist a trader at another institution. See Barclays DOJ Statement of Facts ¶ 27; Amabile Am. Compl. ¶ 265.

If the other bank in each of these instances can be identified through discovery, then both Barclays and the other bank will be jointly liable for both banks' manipulation of LIBOR on these dates. And, of course, disclosure of additional examples of collusion involving defendants against whom claims survive may serve as the predicate for additional claims.

Significantly, apart from these sporadic examples, the public settlement agreements do not reveal any broad pattern of collusion in USD LIBOR, in sharp contrast to the far more colorful admissions regarding Yen LIBOR. It is well-documented that a broker who called himself "Lord LIBOR" received kickbacks from a UBS affiliate in order to secure favorable Yen LIBOR submissions from several panel banks. See generally Order Instituting Proceedings, ICAP Europe Ltd., No. 13-38 (C.F.T.C. Sept. 25, 2013). When plaintiffs highlight an RBS trader's statement that LIBOR is a "cartel now in London" (and nearly every plaintiff's complaint highlights this quote as exhibit A supporting allegations of collusion), plaintiffs routinely forget to mention that a "Senior Yen Trader" wrote this line in reference to Yen LIBOR. See Order Instituting Proceedings 14-15, The Royal Bank of Scotland plc, No. 13-14 (C.F.T.C. Feb. 6, 2013).

We continue to reject the impermissible inference that defendants' reprehensible behavior in one product (or even many products: Yen LIBOR, TIBOR, Swiss Franc LIBOR, EURIBOR, foreign exchange, precious metals, mortgages, auction-rate securities, foreign tax shelters, and so on) suffices to overcome deficiencies in the pleading of actionable bad behavior in USD LIBOR. Cf. Fed. R. Evid. 404(b)(1).

Indeed, some of the plaintiffs' own complaints show that USD LIBOR behaved in a different manner from LIBORs in other currencies. Specifically, one of the key pieces of evidence demonstrating persistent suppression in USD LIBOR is that some banks' implied credit spreads in U.S. dollars became markedly different from the implied credit spreads in other currencies during the financial crises, indicating that suppression largely affected USD LIBOR alone. As suppression was confined to USD LIBOR, there is no logical barrier to concluding that a concerted conspiracy involving traders of Yen interest rates was confined to Yen LIBOR.

2.2. Persistent Suppression

None of plaintiffs' complaints raises a plausible inference that banks conspired to suppress LIBOR. Instead, the pleaded facts support a conclusion that every panel bank had the same incentive to suppress LIBOR. See, e.g., BATA Compl. ¶ 48; NCUA Compl. ¶ 59; Principal Funds Compl. ¶ 49. Every bank worried that it would be viewed as a credit risk if it were to report an honest account of its borrowing costs. This common incentive created a vicious cycle: the more each bank suppressed LIBOR, the greater the incentive for every other bank to do the same.

Plaintiffs' allegations offered to support a persistent suppression conspiracy may be categorized as follows:

• Parallel suppression. Many of the complaints feature a chart showing all of the panel banks' quotes gradually increasing from mid-September to mid-October 2008, and then gradually falling through the end of October. E.g., FDIC Am. Compl. Fig. 6.

• Circumstantial and direct evidence that banks attempted to stay in or near the bottom of the "pack." Some complaints show statistical evidence that banks' quotes were artificially "bunched," see, e.g., BATA Am. Compl. ¶¶ 100-08, and several settlement documents report that

banks explicitly adopted a "pack" strategy. See, e.g., Barclays FSA Final Notice ¶ 116, June 27, 2012; Barclays DOJ Statement of Facts ¶¶ 40, 43; Non-Prosecution Agreement Between UBS AG and the Department of Justice, App. A ("UBS DOJ Statement of Facts") ¶¶ 105, 129, Dec. 18, 2012, available at http://www.justice.gov/iso/opa/resources/6942012121911725320624.pdf; Lloyds CFTC Order 14-15.

• Banks' awareness of other banks' suppression. Barclays CFTC Order 22; Barclays FSA Final Notice ¶ 117; UBS DOJ Statement of Facts ¶¶ 101, 117.

• The BBA's institution of a non-public committee to oversee LIBOR, the BBA's practice of informing panel banks whose submissions were out of line with others, and major banks' opposition to the BBA's plans to abandon control of LIBOR. See, e.g., FDIC Am. Compl. ¶¶ 67-68, 386; Principal Fin. Grp. Am. Compl. ¶ 154.

• The failure of a broker-dealer named ICAP (incidentally, a key participant in Yen LIBOR manipulation) to create an alternative to LIBOR in May 2008. See Gavin Finch & Ben Livesey, ICAP's Libor Alternative Lacks 'Concrete Timetable', Bloomberg (May 14, 2008), http://bloomberg.com/apps/news?pid=newsarchive&sid=asgvKFvA0iio.

• The return of "competition" to LIBOR once Barclays revealed its misconduct in 2012.

We address these sets of allegations seriatim.

First, parallel conduct need not imply a conspiracy, and certainly not where each supposed conspirator independently had the same motive (namely, to protect its own reputation for creditworthiness) to engage independently in the same misconduct.

Moreover, it is hardly surprising that different banks' LIBOR submissions generally rose and fell together. Each submission by a panel bank is the sum of several factors, including the risk-free rate, the creditworthiness of banks in general, the creditworthiness of the panel bank relative to other banks, some error based on sincere uncertainty about the panel bank's own borrowing costs, and, presumably, manipulation. At least two of these factors—the risk-free rate and general credit spreads—are identical for every panel bank. These factors fluctuated widely in September and October 2008 and represent the most plausible explanation for why the submissions generally moved together. See Federal Reserve Bank of St. Louis, "3-Month Treasury Bill: Secondary Market Rate," FRED, http://research.stlouisfed.org/fred2/series/DTB3; Federal Reserve Bank of St. Louis, "3-Month Treasury Constant Maturity Minus Federal Funds Rate," FRED, http://research.stlouisfed.org/fred2/series/T3MFF. Plaintiffs' complaints offer little reason to infer that manipulation was behind the roughly parallel changes to the banks' LIBOR submissions, or that collusion was the natural explanation for parallel changes in the manipulation component.

Second, we do not infer collusion from the fact that many banks knew that other banks were suppressing LIBOR and tried to stay in or at the bottom of "the pack." It was to be expected that a panel bank would know whether its co-panelists could actually obtain loans at the rates they quoted to the LIBOR panel. After all, panel banks were chosen to be panel banks because, as leaders in the inter-bank lending market, they normally had information about inter-bank lending.

Furthermore, the banks universally felt pressure—without need for a conspiracy—to keep their "head[s] below the parapet," Barclays DOJ Statement of Facts ¶¶ 40, 43. 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. In short, no collusion was necessary; all that was required to stay in the pack was the very awareness of money-market conditions on which the LIBOR system has always relied.

In fact, in this regard, the most that can be said is that, on a few occasions, Barclays employees were apparently aware of other banks' forthcoming bids. Specifically, on November 28, 2007, a Barclays submitter stated that he would set Barclays's 2- and 3-month quotes at 5.13% and 5.12% even though "brokers tell me that [RBS] is going to set at 5.15 for both." Barclays CFTC Order at 20; Barclays FSA Final Notice ¶ 117. The next day, credit conditions had worsened substantially, and every bank significantly increased its submissions, particularly at the 1-month tenor. A Barclays submitter (perhaps the same one) planned to submit a 1-month quote of 5.50%, but was overruled because no other bank was expected to submit a quote higher than 5.30%. Barclays FSA Final Notice ¶ 118.

These scraps of advance information about other panelists' bids that Barclays possessed during a single two-day window hardly imply a conspiracy. Indeed, the November 28 email goes on to say, "Not really sure why contributors are keeping [LIBOR quotes] so low. . . . [R]eality seems to be setting in for a few libor contributors who are belatedly moving libors up in line with where money is really trading." Id. at ¶ 117. These statements belie plaintiffs' notion of a bank that, through membership in a conspiracy, knew exactly why quotes were so low and preferred for reality not to set in. Likewise cutting against plaintiffs' argument is the fact, discussed in the same settlement documents, that Barclays's employees advised regulators that other banks' submissions were artificially low, and that Barclays could "submit honest rates without standing out from other members . . . if other banks submitted honest rates." Barclays DOJ Statement of Facts ¶ 42.

Some plaintiffs also point to a telephone transcript dated October 24, 2008, in which a Barclays sales staffer states to a representative of the Federal Reserve Bank of New York that "three-month libor is going to come in at 3.53[%]," "a touch lower than yesterday's." Telephone Tr., available at http://www.newyorkfed.org/newsevents/news/markets/2012/libor/October_24_2008_transcript.pdf. In fact, practically every verifiable statement in this transcript is mistaken. Three-month LIBOR was 3.51625% on October 24, 2008, and 3.50625% the next business day. The salesman even got Barclays's own LIBOR quote wrong. Barclays's 3-month LIBOR quote was not 4%, but rather 3.95% on October 24, 2008, and 3.9% the next day.

It is evident that this salesman knew as little as anyone else about other banks' LIBOR-setting strategies. When asked "why are these [quotes] being set [low]," he replies, "Yeah, I'd love to know. I really would love to know." Regarding WestLB's submitter, he remarks, "I don't know where he gets his libor indications from. I can't imagine anyone would want to lend him any money." The salesman's bemusement, combined with his modestly inaccurate estimates of forthcoming LIBOR, bear out our conclusion that non-conspiratorial (but parallel) reputation-driven suppression is the only logical explanation.

Third, there is nothing inherently wrong or suspicious about banks forming a committee within the BBA to administer LIBOR. It is not uncommon for an industry association to publish market data for its members' convenience (the historical purpose of LIBOR), and not unreasonable for a trade organization to keep legitimate meetings among its members private.

Fourth, there is no allegation that any panel bank threatened to retaliate against ICAP when ICAP announced inchoate plans to develop an alternative interest-rate benchmark. The complaints that develop this theory suggest that the panel banks had the market power to do so, but do not state with any degree of precision which banks allegedly connived to push ICAP out of the benchmark market or by what means. Such innuendos are no substitute for well-pleaded allegations.

Fifth, some complaints point to various post-2012 changes in the benchmark marketplace as evidence that competition re-emerged after Barclays announced its wrongdoing. These changes are far more consistent with regulatory pressure than competition, and therefore do not lead to a conclusion that competition was suppressed during the financial crisis. Oddly, at least some of the "new" alternatives that plaintiffs mention (repo rates, for instance) were available throughout the financial crisis.

We conclude by noting that plaintiffs' conspiracy theories suffer by comparison to the antitrust complaint that the Supreme Court rejected in Bell Atlantic Corp. v. Twombly, 550 U.S. 544 (2007). In that case, the complaint attempted to allege a conspiracy based on parallel conduct, communications through seven trade organizations, and conduct inconsistent with the defendant telephone companies' separate profit motives. See Cons. Am. Class Action Compl. ¶¶ 37-46, Twombly v. Bell Atl. Corp., No. 02-cv-10220 (GEL) (S.D.N.Y. Apr. 14, 2003), ECF No. 24. As in Twombly, the Individual Plaintiffs allege parallel conduct. As in Twombly, the Individual Plaintiffs non-specifically assert that defendants communicated through a trade organization. But, unlike in Twombly, each panel bank's decision to suppress LIBOR was consistent with its own independent motive to preserve its reputation for creditworthiness.

3. Aiding and Abetting

In LIBOR I, we accepted three theories of aiding and abetting liability under section 22(a) of the Commodities Exchange Act, 7 U.S.C. § 25 (2012). Our analysis there was not specific to the CEA's treatment of aiding and abetting, and we accordingly follow the approach of LIBOR I in the context of common law torts. While we do not retreat from our earlier ruling, we do clarify one aspect.

One theory we accepted was that Barclays could be liable on an aiding and abetting theory for its admitted trader-based conspiracy. See supra at 108; LIBOR I, 935 F. Supp. 2d at 723. Two other theories of aiding and abetting did not depend on the existence of a conspiracy, although they may lead to the same result—joint liability of all panel banks for each other's persistent suppression—that a conspiracy theory would achieve.

First, we accepted the Exchange-Based Plaintiffs' theory that "each bank, by allegedly submitting artificial LIBOR quotes, furthered other banks' manipulation" by making it more difficult to detect the other banks' manipulation. Id. at 723. This theory passes the pleading stage because it is plausible that each panel bank knew (1) that each other bank was suppressing LIBOR, and (2) that its own suppression would help each other bank suppress LIBOR.

With the benefit of further briefing and subsequent experience distinguishing between persistent suppression and sporadic trader-based manipulation, we now clarify that this last theory applies only to persistent suppression—not to sporadic trader-based manipulation. If only one bank or a few banks had engaged in persistent suppression at the magnitude alleged, then those banks' submissions would have been glaring. Because each persistent suppressor needed all (or at least significant number) of the other banks to suppress LIBOR simultaneously in order to avoid immediate detection, each persistent suppressor plausibly bears vicarious liability for harm caused by the others' persistent suppression. Not so for trader-based suppression, which all evidence indicates was conducted at a much lower magnitude. A trader could (and at least some did) adjust his own bank's quote by a few basis points without relying on any other bank to manipulate LIBOR at the same time or in the same direction.

We also clarify that the aiding and abetting theory applies only to CEA claims and to "false data fraud" claims (defined infra at 125). No pleaded facts indicate that banks were aware of each other's contracts, and so we do not recognize aiding and abetting in relation to breach of contract or tortious interference.

Second, we accepted the Exchange-Based Plaintiffs' theory that multiple panel banks' quotes jointly influenced the final LIBOR fix. This theory is essentially a theory of joint and several liability, which we discuss further in our discussion of damages. See infra at 266.

4. Conclusions

For these reasons, we accept the "aid to concealment" theory of vicarious liability for persistent suppression, but reject the same theory for trader-based manipulation. The "joint causation" theory survives for persistent suppression and for trader-based manipulation as well, to the extent that any two banks engaged in trader-based manipulation on the same dates. Indeed, the "joint causation" theory could also make a persistent suppressor liable for another bank's trader-based suppression and vice versa, provided that both occurred on the same day and in the same direction.

As for conspiracy, we sustain only the specific and limited theories that Barclays traders communicated with other banks' traders to engage in sporadic trader-based manipulation, and that entities within Barclays and Lloyds/HBOS collaborated to manipulate LIBOR. Plaintiffs' remaining allegations do not support the pleading of a broad-based conspiracy to manipulate USD LIBOR for traders' benefit or to suppress LIBOR during the financial crisis.

VI. FRAUD AND NEGLIGENT MISREPRESENTATION

Plaintiffs' fraud claims fall into two broad categories. The first category ("LIBOR quality fraud") encompasses claims that defendants made false statements or omissions about LIBOR, causing plaintiffs to invest in LIBOR-based instruments. This category also includes claims that counterparties made false statements or omissions about LIBOR after plaintiffs had traded securities, as well as claims that panel banks and the BBA made false statements or omissions about LIBOR after news articles criticized LIBOR. The second category ("false data fraud") encompasses claims that defendants made false LIBOR submissions, causing plaintiffs to pay or receive flawed amounts related to their LIBOR-based investments.

Most of the LIBOR quality claims are inadequately pleaded. Plaintiffs either fail to identify a false statement or fail to allege reliance with sufficient particularity. However, some plaintiffs adequately allege that their counterparties were aware of LIBOR manipulation and fraudulently or negligently omitted to alert plaintiffs before entering into LIBOR-based agreements. Furthermore, Fannie Mae adequately pleads a fraudulent misrepresentation based on its ISDA agreements.

The false data claims generally survive against panel banks. Except with respect to some of the Salix and Schwab Plaintiffs' claims, plaintiffs were within the class of persons whom defendants expected to rely on LIBOR quotes. We accept the Individual Plaintiffs' allegations of reliance, notwithstanding that plaintiffs utilized their agents or counterparties to calculate payments based on a combination of the LIBOR quotes.

1. The Definition of Fraud

1.1. Choice of Law

The parties generally agree on the choice of substantive law, at least at this stage, according to the conflicts rules of each transferor state. For the most part, New York law applies, but the parties agree that California, New Jersey, Pennsylvania, Texas, or Virginia law applies to some cases and plaintiffs.

One plaintiff in the Darby case, Capital Ventures, is a Cayman Islands entity, and now argues that Caymanian law should govern its claims. In its initial briefing, Capital Ventures did not mention Caymanian law, and only raised this issue after we asked the parties at oral argument to file a joint spreadsheet showing their positions on choice of substantive law. Defendants contend that we should follow New York substantive law because plaintiff failed to raise Caymanian law previously. Contrary to Rule 44.1 of the Federal Rules of Civil Procedure, neither party has submitted evidence of Caymanian law. "Since neither party has suggested that [foreign] law differs from New York law in any relevant respect, we have not embarked on an independent investigation of the matter." Bartsch v. Metro-Goldwyn-Mayer, Inc., 391 F.2d 150, 155 n.3 (2d Cir. 1968). Instead, we will assume that the common law in the Cayman Islands is essentially the same as that of New York.

1.2. Elements of Fraud

To state a claim for fraud, a plaintiff must plead the following:

False statement— The defendant made a misrepresentation (or an omission, if the defendant has a duty to disclose information).

Scienter— 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." Restatement (Second) of Torts § 531. When the defendant makes a misrepresentation to a person other than the plaintiff, the defendant will still be liable if he "intend[ed]" or had "information that g[a]ve[] him special reason to expect" that the information would be relayed to plaintiff. § 533 & cmt. d. The defendant may be liable to a class of plaintiffs even though the

defendant had no particular victim in mind, § 533 cmt. g, and had no interest in the plaintiff's transactions, § 533 cmt. e.

Reliance — The plaintiff actually relied on the misrepresentation.

Reasonable reliance — The plaintiff's reliance was reasonable or justifiable.

Proximate causation — The plaintiff's reliance must have proximately caused the plaintiff injury.

2. Factual Theories

The factual basis of plaintiffs' false data claims is straightforward: defendants submitted false data to the BBA. The LIBOR quality claims are more varied, thus necessitating an articulation of the several factual theories that plaintiffs pursue.

2.1. Misrepresentations

First, plaintiffs offer a set of related theories that their counterparties committed fraud. Some plaintiffs claim that their counterparties made false statements about LIBOR in the course of offering or trading LIBOR-based securities. Others, that their counterparties omitted to mention LIBOR manipulation in the course of offering or trading LIBOR-based securities. And yet others, that their counterparties failed to reveal LIBOR manipulation once the counterparties learned of LIBOR manipulation.

Second, some plaintiffs claim that panel banks and the BBA committed fraud by rendering false assurances about the quality of LIBOR after critical news articles were published in 2008. Plaintiffs also claim that, throughout the periods of LIBOR manipulation, the panel banks and the BBA had a duty to reveal LIBOR manipulation to the public at large.

2.2. Reliance

Next, we must classify plaintiffs' theories of reliance and harm. One set of theories is that plaintiffs relied on representations in making their investment decisions: decisions to trade (or issue) LIBOR-based securities; decisions to trade non-LIBOR-based securities instead of LIBOR-based securities; decisions to close out positions in LIBOR-based securities (thus incurring a LIBOR-based termination fee); and decisions to continue holding LIBOR-based securities. These investment decisions all turned on information about the quality of LIBOR as a reflection of economic realities in the underlying inter-bank money market. Any "investment decision" theory of harm must therefore involve a "LIBOR quality" misrepresentation or omission upon which the plaintiff relied. In this regard, a plaintiff may not allege reliance on daily LIBOR quotes, because daily LIBOR quotes, whether honest or not, do not state anything about the quality of LIBOR as a benchmark.

Plaintiffs also allege that they relied on LIBOR in calculating payments related to swaps, bonds, and mortgages. Typically, these payments were routine payments that were required to be based on LIBOR pursuant to some contract—a swap agreement, bond, mortgage, etc. In this circumstance, it is fair to say that, in at least some sense, plaintiffs relied upon LIBOR and hence upon the quotes that LIBOR incorporated. It would be incorrect, however, to assert that plaintiffs relied on information about the quality of LIBOR. By the time of calculation, plaintiffs had already committed, for richer or for poorer, to pay and receive payments with reference to LIBOR.

Another variation involves the situation in which a plaintiff used LIBOR to calculate a payment in the absence of a contractual obligation to do so. In particular, several plaintiffs allege incurring LIBOR-based termination fees by closing out swaps early. It is at least conceivable that the parties to these swaps agreed to base the termination fee upon an estimate of future LIBOR or upon the counterparty's cost of funds, and then used current LIBOR information to estimate the basis of the termination fee. In this case, plaintiffs may have relied on the panel banks' LIBOR submissions, and also may have relied on false statements or omissions about LIBOR, because the parties in this scenario could have used information other than LIBOR to arrive at a fair termination fee.

With this framework of misrepresentations and reliance in mind, our next step is to explore the plausibility of each theory. The key questions are whether plaintiffs muster enough facts to plead the existence of an actual misrepresentation and actual reliance with particularity and whether there are any legal reasons not to accept plaintiffs' theories of fraud.

3. Legal Discussion

3.1. Fraud by Affirmative Misrepresentation in the Course of Offering or Trading Securities

3.1.1. Failure to Plead Specifics

Most plaintiffs vaguely allege that a counterparty defendant made a misrepresentation regarding LIBOR when entering into a contract. The following are examples:

Defendants made materially false or misleading statements . . . by misleadingly representing, in materials disseminated to Plaintiff during the Relevant Period in connection with Plaintiff's purchase of LIBOR-based financial instruments, that the rates of return assigned to those financial instruments were tied or indexed to, or otherwise derived from, a USD LIBOR that reflected the LIBOR panel banks' true costs of borrowing . . . .

The Defendants, and each of them, knowingly and intentionally, or with reckless disregard for the truth, made, authorized, and caused to be made material representations concerning their borrowing costs and the proper level of LIBOR that were false and misleading, including but not limited to making affirmative misrepresentations directly to

Plaintiffs as well as making affirmative misrepresentations to third parties such as the [BBA], that they knew, intended and expected to be relied upon by Plaintiffs. . . .

Defendants made these misrepresentations . . . of material fact while entering directly into transactions with Plaintiffs which involved LIBOR in the determination of either the value of the transaction or financial instruments or the amount that would be paid to the Plaintiffs.

Defendants made, authorized, and caused false statements or omissions to be made to Plaintiffs to induce Plaintiffs to enter into the swaps.

Defendants misrepresented the basis of payments Plaintiffs would receive under the swaps . . . .

The Contracting Defendants made these misrepresentations . . . during negotiations for each pay-fixed swap . . . in order to induce the Contracting Closed Banks to enter into these transactions.

As the examples just cited demonstrate, these plaintiffs' pleadings are wholly inconsistent with Rule 9(b)'s demand for particularity. While the precise requirements of this rule vary with the facts of each case, it is normally incumbent upon the plaintiff to specify the statements that were false or misleading, to state when and where those statements were made, and to identify those responsible. Lundy v. Catholic Health Sys., 711 F.3d 106, 119 (2d Cir. 2013). A generalized "defendants lied" allegation fails "to give defendants notice of the particular misconduct which is alleged to constitute the fraud charged so that they can defend against the charge and not just deny that they have done anything wrong." Semegan v. Weidner, 780 F.2d 727, 731 (9th Cir. 1985).

Finally, some plaintiffs allege a false statement that payments would be based on the "published definition of LIBOR." E.g., Principal Fin. Grp. Am. Compl. ¶¶ 228-29. This is slightly more specific than the allegations discussed above, but still falls short of pleading with particularity, because it fails to indicate where and when each defendant made such a statement.

Rule 9(b) also protects defendants "against spurious charges of immoral and fraudulent behavior." Durham v. Bus. Mgmt. Assocs., 847 F.2d 1505, 1511 (11th Cir. 1998). In the context of this case, this concern requires us to distinguish carefully between the branches of a large banking institution. Whether torts or crimes have been committed by the entities that submitted false LIBOR quotes for profit and reputation, it remains entirely possible that some of the banks' customer-facing entities were innocent of misconduct when they marketed swaps and bonds to plaintiffs.

This is not a case where the information absent from the complaint (assuming its existence) is in defendants' sole possession, such that plaintiffs should be excused from pleading specifics. Cf. Freitas v. Wells Fargo Home Mortg., Inc., 703 F.3d 436, 440 (8th Cir. 2013); In re Rockefeller Ctr. Props., Inc. Sec. Litig., 311 F.3d 198, 216 (3d Cir. 2003). If a plaintiff truly relied on some representation, then that he should know with some precision what that representation was and how he relied on it. Nor is this a case in which plaintiffs provide a fair selection of representative examples. Cf. United States ex rel. Tiesinga v. Dianon Sys., Inc., 231 F.R.D. 122, 123-24 (D. Conn. 2005); Commercial Prop. Invs., Inc. v. Quality Inns Int'l, Inc., 61 F.3d 639, 646 n.6 (8th Cir. 1995).

Rule 9(b) commands particularity instead of generality, concrete examples instead of abstraction. The plaintiffs whose complaints are cited above have not approached the threshold for pleading fraud against their counterparties, and so their claims fail.

3.1.2. Failure to Plead Misrepresentations

Some plaintiffs have in fact pleaded misrepresentations with enough specificity to approach or surpass the Rule 9(b) bar, but each such claim turns out not to involve a genuine misrepresentation.

3.1.2.1. LIBOR Quality

Some plaintiffs allege that defendants' offering materials made specific misrepresentations regarding the quality of LIBOR. E.g., Prudential Am. Compl. ¶ 372(a), (f); id. Exs. C-D (listing examples of statements); Triaxx Am. Compl. ¶ 155(E) (same). Typical examples are a private placement memorandum for a security with CUSIP 83050XAN2 (Prudential allegation against Bank of America Corp.), and a document related to the mortgage-backed security "BOAA 2005-3" (Triaxx Am. Compl. ¶ 155(E)(i)). First, the Prudential example:

Interest rate: One-month LIBOR, reset on a monthly basis on each interest reset date, plus the applicable spread. The interest rate for the initial interest reset period will be one-month LIBOR, to be determined two London business days prior to the original issue date, minus the applicable spread for such period of .01%. Interest on the Notes will be computed on the basis of the actual number of days elapsed over a 360-day year.

One-month LIBOR will be determined by the calculation agent, initially Deutsche Bank Trust Company Americas, as of the applicable interest determination date in accordance with the following provisions:

(i) LIBOR will be determined on the basis of the offered rates for deposits in U.S. dollars having a one-month maturity, commencing on the

second London business day immediately following such interest determination date, which appears on Moneyline Telerate Page 3750 (as defined below) as of approximately 11:00 a.m., London time, on such interest determination date. "Moneyline Telerate Page 3750" means the display designated on page "3750" on that service, any successor service or such other service or services as may be nominated by the British Bankers' Association for the purpose of displaying London interbank offered rates for U.S. dollar deposits).

Triaxx's example is considerably shorter: "The Floating Rate and Inverse Floating Rate Certificates will bear interest at their respective pass-through rates, which are each based on LIBOR determined by the Trustee as described below."

These allegations pass the Rule 9(b) bar, in that they put defendants on notice of the facts that plaintiffs believe to have been misrepresentations. Nevertheless, even a specific pleading must plead a statement that was plausibly false.

Fairly read, these passages simply prescribed, in varying degrees of detail, the recipes for each security's calculation agent to calculate payments. The agent was to access a particular data source at a particular time, record the LIBOR reported by that data source for a particular currency and tenor, and plug the reported LIBOR into a prescribed calculation. Nothing about these passages guaranteed, or really even suggested, that LIBOR would be an accurate, precise, or unbiased source of inter-bank interest rates. The instructions are equally consistent with LIBOR being accurate as with LIBOR being somewhat inaccurate but advantageous for other reasons (convenience, cost, familiarity, network effects, continuity). The contracting parties could have devised some other approach, such as requiring the calculation agent to verify LIBOR's accuracy against evidence of true interest rates, but chose not to do so. The parties instead preferred to use LIBOR as is.

3.1.2.2. Good-Faith Calculations

Some plaintiffs allege that defendants promised to make calculations in good faith and in a commercially reasonable manner. E.g., Principal Fin. Grp. ¶ 228 ("Each Swap Counterparty Defendant also represented that, as calculation agent, it would 'make each calculation [under the Contracts] in good faith and in a commercially reasonable manner."); Darby Am. Compl. ¶ 442(c) (similar).

The cited passages do not suggest that the calculation agents were involved in creating LIBOR. The only reasonable reading is that these passages refer to the task that the contract commands the calculation agent to perform: to incorporate LIBOR (as published) into various contract-specific calculations.

Furthermore, these contractual provisions define performance under the contract, and non-performance is not actionable as fraud. See, e.g., MTA v. Triumph Advertising Prods., Inc., 116 A.D.2d 526, 527, 497 N.Y.S.2d 673, 675 (1st Dep't 1986) ("[A] cause of action for fraud does not arise when the only alleged fraud relates to a breach of contract.").

3.1.2.3. Compliance With Laws

The Darby Plaintiffs allege that their counterparties promised to comply with applicable laws "if failure so to comply would materially impair its obligations under the swaps." Darby Am. Compl. ¶ 442(d). However, this theory fails because plaintiffs fail to identify a statute or regulation that defendants violated, or a specific obligation that was impaired by defendants' alleged violations. Also, this passage too appears to define performance of the contract, and thus a fraud claim based on a breach of this clause would be duplicative of a contract claim.

3.1.3. No-Existing-Breach Representation

One plaintiff, Fannie Mae, has sufficiently alleged fraud against some counterparties in a specific context. Specifically, each time that a counterparty traded a swap pursuant to an ISDA agreement, Fannie Mae's counterparty thereby renewed a promise that the counterparty was not in breach of any previous swap agreement. According to Fannie Mae, this promise was false whenever a counterparty was in breach of a previous swap's implied covenant of good faith. Cf. infra at 203.

We focus on the Barclays ISDA agreement (Fannie Mae Am. Compl. Ex. 17) that Fannie Mae discusses at length in its complaint. In its ISDA agreement, Barclays promises that certain representations are true "on each date on which a Transaction [i.e., a new swap] is entered into," § 3. Among these representations is a promise that no "Potential Event of Default" has occurred, § 3(b). "Potential Events of Default" include breaches of the ISDA agreement, breaches of any credit support documents associated with the ISDA agreement, and previous misrepresentations. §§ 5(a)(ii), (iii)(1), (iv), 14.

The logic of Fannie Mae's argument proceeds as follows. A breach of good faith on one swap was a breach of contract, and thus a Potential Event of Default. When Barclays traded a subsequent swap with Fannie Mae, Barclays represented that no Potential Event of Default, and thus no breach, existed. Therefore, whenever Barclays traded a subsequent swap with plaintiff and failed to reveal its ongoing bad faith on the previous swap, Barclays made a misrepresentation.

This logic is sound. The potential weakness in Fannie Mae's argument is that the definition of "Potential Event of Default" could plausibly be read to include or to exclude a breach of the implied covenant of good faith and fair dealing. The text refers to a "[f]ailure by the party to comply with or perform any agreement or obligation . . . to be complied with or performed by the party in accordance with this Agreement." The words "agreement or obligation" could be read narrowly to mean only explicit obligations found within the four corners of an agreement, or the same words could be read more broadly to include promises that the law imports into an agreement. Because it would be premature to resolve this ambiguity, Fannie Mae's claim survives this stage.

Assuming that this is a genuine misrepresentation, the requirement to allege actual and reasonable reliance is easily met. It is hard to imagine a reasonable person signing a new deal (in swaps or anything else) with a counterparty who has just acknowledged breaching an earlier deal in bad faith.

Moreover, this fraud claim is distinct from any contract claim. The fraud claim on a later swap, although predicated on a breach of an earlier swap, is distinct from a contract claim on an earlier swap because the fraud claim springs from a false representation in the formation of the later swap and seeks damages only on the later swap. The fraud claim is also distinct from a contract claim on the later swap, because the misrepresentation in the formation of the later swap related to a then-existing fact (a breach of the earlier swap), not to the performance of the second swap.

However, Fannie Mae's claim fails insofar as Fannie Mae relies on a representation regarding the absence of litigation. Initially, it is questionable whether LIBOR litigation was pending or threatened at the time that defendants traded swaps with plaintiff. More fundamentally, however, no aspect of this LIBOR litigation "is likely to affect the legality, validity or enforceability against [defendants] of" standard swap agreements and credit annexes. § 3(c).

We conclude that Fannie Mae, but no other plaintiff, has adequately stated a claim for fraud by affirmative misrepresentation in the inducement of a contract.

3.2. Fraud by Omission in the Course of Offering or Trading Securities

Almost every plaintiff submits that their counterparties committed fraud by failing to inform plaintiffs of LIBOR manipulation.

Defendants correctly point out that defendants were not fiduciaries of plaintiffs, that parties to arm's-length transactions are not obligated to disclose information to each other, and that defendants (with the one exception above) made no affirmative misrepresentations that demanded correction.

However, a contracting party bears a duty to disclose facts when it knows that the other party is mistaken as to facts that are basic to the transaction and when the customs of the trade or other objective circumstances would reasonably demand disclosure. Cf. LiMandri v. Judkins, 52 Cal. App. 4th 326, 336, 60 Cal. Rptr. 2d 539, 543 (4th Dist. 1997) (holding that nondisclosure may constitute fraud "when the defendant had exclusive knowledge of material facts not known to the plaintiff"); Restatement (Second) of Torts § 551(2). While not every material fact is "basic to the transaction," the interest rate on a loan or interest rate swap is fundamental. Just as a seller must disclose to a buyer a termite-filled house or a diseased herd of cattle, a counterparty to an interest rate swap has a duty to disclose what he knows of the distortion of an interest rate.

We cannot, at this stage, say that the relevant information was equally within the grasp of plaintiffs and defendants. Although many of the plaintiffs were sophisticated and possessed some information about LIBOR, it is plausible that at least some counterparties had much superior information to plaintiffs, either because defendants themselves were manipulating LIBOR or (in the case of entities that were not on the LIBOR panel) because defendants were major players in the inter-bank lending market with access to non-public data regarding real market transactions.

Express disclaimers do not bar a fraudulent omission claim on a "special facts" or "superior knowledge" theory. See, e.g., Banque Arabe et Internationale d'Investissement v. Md. Nat'l Bank, 57 F.3d 146, 155 (2d Cir. 1995); TIAA Global Invs., LLC v. One Astoria Square LLC, 127 A.D.3d 75, 87, 7 N.Y.S.3d 1 (1st Dep't 2015); Tahini Invs., Ltd. v. Bobrowsky, 99 A.D.2d 489, 490, 470 N.Y.S.2d 431, 433 (2d Dep't 1984) (citing Danann Realty Corp. v. Harris, 5 N.Y.2d 317, 322, 157 N.E.2d 597, 599-600 (1959)).

Most plaintiffs' attempts to plead omissions on a "special facts" or "superior knowledge" theory are sufficient. Although fraud by omission must be alleged with particularity, there is no need for plaintiffs to cite specific terms of a contract, because the point of an omission is that information was missing from the contract and from negotiations. It is sufficient for a plaintiff to name a particular contract (or set of contracts) and a particular counterparty (or set of counterparties) that failed to divulge information about the quality of LIBOR.

Even now that LIBOR manipulation has been widely reported, whether defendants possessed superior knowledge would remain within the particular knowledge of those defendants. Consequently, plaintiffs need not plead scienter with great precision. It is sufficient for plaintiffs to state plausibly that defendants were either themselves manipulating LIBOR or that defendants were large banking institutions with access to non- public data about real inter-bank transactions. This much may be fairly imputed as to transactions that defendants executed (1) after a particular counterparty-defendant itself began to manipulate LIBOR, or (2) from August 2007 onward, when LIBOR suppression became so widespread that any major bank plausibly knew of manipulation.

3.3. Fraud by Omission After Trading

Many plaintiffs allege that defendants omitted to disclose manipulation at some time after entering into contracts. See Cal. Cons. Compl. ¶ 565; Houston Am. Compl. ¶ 454; Philadelphia Am. Compl. ¶¶ 435(d), (g), 438; Darby Am. Compl. ¶¶ 441, 442(e), 445; Prudential Am. Compl. ¶¶ 372(d), 375; Salix Am. Compl. ¶¶ 476(d), 479; Triaxx Am. Compl. ¶ 155(C)-(D).

This formulation of a fraud claim adds little, if anything, to plaintiffs' arsenal. If a counterparty knew of LIBOR manipulation when it formed a contract with plaintiff, then this approach adds nothing to the claim that the counterparty committed fraud at the time of formation. If, however, a counterparty learned about manipulation after the inception of a contract, then there was no duty to inform plaintiffs later, because new information would not alter plaintiffs' commitment to pay and receive money based on published LIBOR.

The only exceptional case occurs when a plaintiff agreed to base a swap termination fee on an estimate of future LIBOR. Unless a pre-existing swap agreement defines the parties' termination payments, the parties must form a new agreement as to the amount payable, possibly using LIBOR to estimate future LIBOR or the swap dealer's cost of funds. Plaintiffs in Philadelphia (Am. Compl. Ex. A) and Salix (Am. Compl. Ex. A) allege this scenario by listing specific termination dates, payments, counterparties and swaps, and thus allege a fraudulent omission at each termination.

Some plaintiffs argue that non-counterparty defendants owed a duty to disclose LIBOR manipulation to the public. While open confession may be good for the soul, it is not a legal duty. Whether in the realm of defective manufacturing, environmental harm, or financial misconduct, a wrong committed against the public does not become fraud simply because the wrongdoer keeps silent. See, e.g., Keck v. Dryvit Sys., Inc., 830 So. 2d 1 (Ala. 2002); Connick v. Suzuki Motor Co., Ltd., 174 Ill. 2d 482, 499-501, 675 N.E.2d 584, 593 (1996); Taggart v. Ford Motor Credit Co., 462 N.W.2d 493, 501 (S.D. 1990).

3.3. Fraudulent Assurances About LIBOR

Plaintiffs allege that banks and the BBA made false public statements about the quality of LIBOR. These allegations are potentially stronger than plaintiffs' "public omission" claims, because a person does have a general duty to tell the truth when he knows that a specific class of people will rely on his statements. These allegations fall into two categories. Some statements are simply opinions of analysts who happened to work for a panel bank or a panel bank's affiliate. These are not actionable because plaintiffs make insufficient allegations of scienter and reasonable reliance. Other statements speak with authority regarding particular banks' submissions or about the mechanics of LIBOR. These would be actionable, except that plaintiffs fail to explain their reliance with any degree of particularity. Thus, as we discuss below, these claims are dismissed.

3.3.1. Analysts' Statements

Plaintiffs point to several now-questionable statements about LIBOR from analysts who were affiliated with panel banks, including the following:

• Jeffrey Rosenberg, head of credit strategy for Banc of America Securities, attributed low LIBOR to technical

effects involving "dispersion of risks across banks." Schwab Am. Compl. ¶ 248; Philadelphia Am. Compl. ¶ 344; Fannie Mae Am. Compl. ¶ 94.

• A Bank of America "Global Rates Strategy Report" concluded that LIBOR was not being manipulated because it didn't move in the banks' favor on particular dates when swap agreements with Freddie Mac and Fannie Mae were commonly reset. Fannie Mae Am. Compl. ¶ 94.

• Patrick Duthie of Bank of America stated that the "consensus seems to [be] that LIBOR problems are primarily due to bank balance sheet constraints (and balance sheet unpredictability)." Fannie Mae Am. Compl. ¶ 94.

• Peter Ristine of Citibank wrote in an email: "With the interbank market freezing up almost completely at times, it's not inconceivable that banks don't know their 'true' cost of funds interbank." Fannie Mae Am. Compl. ¶ 97.

• Kal Vadasz of Citibank wrote: "[M]uch of what makes it into the press does not stand up to scrutiny. As I said in an earlier piece, Libor isn't perfect but it works." Fannie Mae Am. Compl. ¶ 97.

• Dominic Konstam of Credit Suisse attributed low LIBOR to easy money being available to U.S. banks from depositors and the Federal Reserve. Schwab Am. Compl. ¶¶ 247, 249; Philadelphia Am. Compl. ¶¶ 344-45; Principal Fin. Grp. ¶ 141.

• Sean Keane of Credit Suisse commented that "LIBORs are remaining sticky at higher levels than they should be." Fannie Mae Am. Compl. ¶ 96.

• Mustafa Chowdhury of Deutsche Bank wrote: "[T]here is little evidence that rate manipulation, if it exists, has been appreciably affecting the LIBOR fixing. . . ." Collaboration among a large number of the survey banks to submit non-market-based quotes is also highly unlikely." Fannie Mae Am. Compl. ¶ 97.

• Terry Belton of JPMorgan wrote that a Wall Street Journal article was "deeply flawed" and that LIBOR was not "biased too high or too low." Fannie Mae Am. Compl. ¶ 98.

• A head of global fixed-income strategy for JPMorgan criticized the Wall Street Journal for using too high a risk-free rate in its analysis of LIBOR. Philadelphia Am. Compl. ¶ 355.

• A JPMorgan report said that the deviation between reported LIBOR and "implied USD Libor from GBP/USD and EUR/USD forwards" could be explained by heterogeneity of market participants." Fannie Mae Am. Compl. ¶ 94.

• Lauren Cantor of UBS reported conflicting media reports about whether LIBOR was artificially high or artificially low. Fannie Mae Am. Compl. ¶ 96.

It is well-established that "opinions are not actionable in securities fraud merely because they are misguided, imprudent or overly optimistic," In re Salomon Analyst Level 3 Litig., 350 F. Supp. 2d 477, 492 (S.D.N.Y. 2004) (Lynch, J.), and we discern no reason to evaluate a similar claim of common law fraud any differently. See also Omnicare, Inc. v. Laborers Dist. Council Constr. Indus. Pension Fund, 575 U.S. ___, ___, 135 S. Ct. 1318, 1328-31 (2015) (holding that a stated opinion is actionable as securities fraud only if the opinion is insincere or if the opinion falsely implies a non-existent basis for forming the opinion). Some of these speakers managed greater intelligence, clarity, and persuasive force than others, but, in context, none purported to do anything more than to state a sincere opinion based on publicly available information. Furthermore, there are no allegations that analysts were falsely held out to be independent of their employers' other branches. Cf. Lapin v. Goldman Sachs Grp., Inc., 506 F. Supp. 2d 221 (S.D.N.Y. 2006) (sustaining allegation that defendant falsely held out its research analysts as independent). Thus, these allegations fail the scienter prong.

Moreover, in this circumstance, plaintiffs' allegations also fail the reasonable reliance prong, as it would have been patently unreasonable for traders to treat an analyst's views as a guarantee that his bank's cash desk was submitting honest LIBOR bids.

3.3.2. Official Statements

More troublesome are official press statements by the banks and the BBA denying the existence of LIBOR manipulation, even while it now appears that some banks were manipulating LIBOR and were aware of severe discrepancies between other banks' LIBOR bids and true inter-bank lending rates. We list some examples:

• In response to a press inquiry, Citibank stated that its LIBOR quotes reflected its perception of the market. Schwab Am. Compl. ¶ 252; Philadelphia Am. Compl. ¶ 355; Principal Fin. Grp. Am. Compl. ¶ 144.

• Colin Withers, the managing director and head of short-term products for Citibank, stated, "We need to let the dust settle, markets stabilize and then have a review. But the measures we are using are historic—up to 30 to 40 years old." Schwab Am. Compl. ¶ 251; Philadelphia Am. Compl. ¶ 347; Principal Fin. Grp. ¶ 143. (The context of this statement is not entirely clear, but Mr.

Withers's title suggests that he was in a position to know about Citibank's alleged manipulation.)

• HBOS responded to a press inquiry by stating that its LIBOR quotes reflected reality. Schwab Am. Compl. ¶ 252.

• JPMorgan responded to a press inquiry by attributing "rate volatility" to "reluctance among banks to lend to each other amid the credit crunch," and denied "any bias in the fixing process." Schwab Am. Compl. ¶ 250; Philadelphia Am. Compl. ¶¶ 346-47; Principal Fin. Grp. Am. Compl. ¶ 142.

• UBS stated, "[W]e don't even know if contributing banks are mis-pricing Libor in the first place." Fannie Mae Am. Compl. ¶ 94.

• The BBA publicly stated that it would expel any panel bank that deliberately submitted false quotes, that it intended to conduct an "intensive" review of its own processes, and that it had confirmed the honesty and accuracy of the panel banks' quotes. Fannie Mae Am. Compl. ¶¶ 108-09, 133; Freddie Mac Am. Compl. ¶¶ 155, 291-97; Principal Fin. Grp. Am. Compl. ¶ 145; Schwab Am. Compl. ¶ 247.

• The BBA also stated (not just during the period of persistent suppression) that LIBOR reflected the average

interest rate at which certain banks could borrow funds. Principal Fin. Grp. Am. Compl. ¶ 34.

These statements might well be actionable if any plaintiff sufficiently alleged actual reliance. Actual reliance would mean that a plaintiff actually learned of these statements, took these statements seriously (which is not the only way to treat a bank's self-serving publicity), and incorporated these statements into its decision-making.

However, the plaintiffs before us have not alleged these facts with any degree of particularity. Their complaints convey no precision about when and how plaintiffs learned of these alleged misrepresentations, how plaintiffs assessed the credibility of these statements, or how plaintiffs balanced these statements against other information about LIBOR. Plaintiffs simply recite that they relied and that they did so reasonably. See, e.g., BATA Am. Compl. ¶ 304 ("Plaintiff reasonably relied on those false representations of material fact in deciding whether to do business with Panel Bank Defendants."); Fannie Mae Am. Compl. ¶ 140 ("Fannie Mae reasonably and justifiably relied on Defendants' false representations and misleading omissions by entering into and performing under huge volumes of Libor-indexed transactions."); Freddie Mac Am. Compl. ¶ 296 ("Freddie Mac reasonably relied on these false representations in deciding whether to enter into transactions tied to USD bbaLIBOR™ and whether to continue holding [LIBOR-based instruments]."); Philadelphia Am. Compl. ¶ 442 ("Plaintiffs and the Libor Third Parties reasonably and justifiably relied on Defendants' false representations and misleading omissions."); Principal Fin. Grp. Am. Compl. ¶ 233 ("Plaintiffs reasonably relied on each Defendant's material misrepresentations and omissions in entering financial transactions tied to USD Libor."); Triaxx Am. Compl. ¶ 161 ("Triaxx reasonably and justifiably relied on Defendants' false representations and misleading omissions.").

3.4. False LIBOR Submissions

We now turn to the second category of fraud claims, the allegations that banks submitted false LIBOR quotes to the BBA. The complaints generally posit that these false submissions entailed a daily representation that submissions were consistent with the BBA's definition of LIBOR. See BATA Am. Compl. ¶ 295(i); Schwab Am. Compl. ¶¶ 307(i), 310; Cal. Consol. Compl. ¶ 562; Houston Am. Compl. ¶ 451; Philadelphia Am. Compl. ¶¶ 434, 435(b); Darby Am. Compl. ¶¶ 440, 442(a); Prudential Am. Compl. ¶¶ 371, 372(b); Salix Am. Compl. ¶¶ 475, 476(b); FDIC Am. Compl. ¶ 288; Freddie Mac Am. Compl. ¶ 282; Fannie Mae Am. Compl. ¶ 132; Maragos ¶ 80; Principal Fin. Grp. Am. Compl. ¶ 255; Triaxx ¶ 155(A). Several complaints explicitly allege that the false submissions caused LIBOR to misrepresent banks' borrowing rates, a position that is implicit in all of the complaints. See Philadelphia Am. Compl. ¶ 435(c); Darby Am. Compl. ¶ 442(a); Prudential Am. Compl. ¶ 372(c); Salix Am. Compl. ¶ 476(c); Triaxx Am. Compl. ¶ 155(B).

Plaintiffs in this catalog of cases sufficiently allege persistent, intentional suppression of LIBOR quotes, leading to suppression of LIBOR in a large number of instances. In the context of persistent suppression, "LIBOR quotes" are sufficiently identifiable to pass muster under Rule 9(b). Artificially low quotes constituted false statements, because manipulated quotes are literally false or insincere responses to questions about banks' ability to borrow money at volume.

We also conclude that plaintiffs adequately plead indirect reliance. Accordingly, subject to the limiting principles discussed below, plaintiffs state claims for fraud against the panel banks.

3.4.1. Limiting Principles

No bank is too big to sue and no fraud too big to remedy. Nevertheless, defendants cannot be liable for every ill effect that befell every person as a result of LIBOR manipulation. Defining a limiting principle raises two closely related questions: (1) whether plaintiffs were in the "class of persons" whom defendants expected would rely on LIBOR, Restatement (Second) of Torts § 534, and (2) which injuries were proximately caused by defendants' misrepresentations.

Defendants argue that all plaintiffs were outside the scope of expected reliance, but especially press the point as to some of the more unusual claims, such as Salix's claim for lost hedge-fund management fees. We conclude that plaintiffs who traded well-known LIBOR-based securities, such as interest rate swaps, mortgage-backed securities, and floating-rate bonds, were within the class of persons who would be expected to rely on the soundness of LIBOR. Several lines of authority guide our formulation of the appropriate limiting principle.

First, at the threshold, we note that a third party to a transaction (i.e., a panel bank) can commit fraud, see, e.g., Restatement (Second) of Torts § 533, and can even do so gratuitously. The Restatement gives the example of a bank that issues a false bond prospectus to investors. Although the bank's financial interest in the bond may be evidence of fraud, such a financial interest is not an element of the claim. Id. § 533 cmt. e. Provided that scienter is adequately pleaded, it is commonplace for accountants to be liable to investors who rely on false audits. See, e.g., In re AOL Time Warner, Inc. Sec. & "ERISA" Litig., 381 F. Supp. 2d 192, 239-40 (S.D.N.Y. 2004); Jordan v. Madison Leasing Co., 596 F. Supp. 707, 710 (S.D.N.Y. 1984); State St. Trust Co. v. Ernst, 278 N.Y. 104, 15 N.E.2d 416 (1938); cf. Citizens Nat'l Bank of Wisner v. Kennedy & Coe, 232 Neb. 477, 480, 441 N.W.2d 180, 182 (1989) (third-party accountant not liable except in cases of fraud); Ultramares Corp. v. Touche, 255 N.Y. 170, 189, 174 N.E. 441 (1931) (similar).

Second, we examine Rio Grande Royalty Co. v. Energy Transfer Partners, L.P., 786 F. Supp. 2d 1202 (S.D. Tex. 2009), the most factually similar case available. In Rio Grande, the defendant had suppressed a natural gas price index (the Houston Shipping Channel or HSC Index) by accurately reporting the prices and volumes of manipulated trades to the index's publisher. Although defendant allegedly commented in reference to the scheme, "these producers they're going to hate it," the court nevertheless held that defendants had no "reason to expect" plaintiffs to rely on the HSC Index.

As the Rio Grande court described it, natural gas producers customarily rely on the HSC Index to price their sales. 786 F. Supp. 2d at 1208-09. We agree with Rio Grande that a well-known market custom, by itself, is insufficient to support third-party liability. As the court noted, this conclusion follows from Ernst & Young, L.L.P. v. Pacific Mutual Life Insurance Co., 51 S.W.3d 573, 580 (Tex. 2001), a case against an auditor in which the Texas Supreme Court declined to extend the "expected reliance class" to purchasers of bonds issued by a different company than the company that the defendant had audited. Even though such reliance might have been reasonable and consistent with market custom under the peculiar circumstances of the case, the audit itself was not directed to investors of any company other than the company that was the subject of the audit. The Supreme Court expressly reserved the question of whether the same auditor could have been liable to investors in the company that the auditor had actually audited. See id. at 582.

Fourth, it is insufficient for a plaintiff to allege relying on a statement that was not intended for him. For example, in City of New York v. Smokes-Spirits.com, Inc., the municipal plaintiff alleged reliance on false tax reports that defendants filed with New York State. Because the federal statute, 15 U.S.C. § 376 (2012), required defendants to make such filings for the benefit of the State's tax authorities alone, the City could not make out a fraud claim, even though the City's reliance had been perfectly reasonable. 541 F.3d 425, 454 (2d Cir. 2008), rev'd on other grounds sub nom. Hemi Grp., LLC v. City of New York, 559 U.S. 1 (2010).

Finally, it is insufficient for a plaintiff to allege a distant injury on account of some other person's reliance. In Lollo, a construction company falsely told its contractors that it had paid certain contributions to workers' union funds. In reliance on these representations, the contractors did not withhold union contributions, leaving the unions short of revenue. The union funds sued the construction company. The construction company was not liable to the union funds because (according to the District Court's fact-finding at a bench trial) the union funds were not deceived by defendant's false statements. Only the contractors had been deceived, to the detriment of the funds. Cement & Concrete Workers Dist. Council Welfare Fund, Pension Fund, Legal Servs. Fund & Annuity Fund v. Lollo, 148 F.3d 194, 196 (2d Cir. 1998).

Having addressed the foregoing lines of authority, which help guide our formulation of the appropriate limiting principle, we return to the central inquiry. Consistent with the Restatement and the case law, we ask which class of persons were the targets of a representation. For example, a business is liable for fraud when it falsely represents its economic condition to a credit agency in the expectation that lenders will rely on the agency's opinion. See Eaton, Cole & Burnham Co. v. Avery, 83 N.Y. 31, 31 (1880); cf. Issen v. GSC Enters., Inc., 538 F. Supp. 745, 751 n.10 (N.D. Ill. 1982) (stating that the duty to supplement misleading information "runs to those members of the investing public who rely or can be expected to rely on" incomplete misrepresentations); Vikse v. Flaby, 316 N.W.2d 276, 284 (Minn. 1982) (defendant liable for misrepresentations that defendant "knew and intended" a third party to pass on to potential investors). Likewise, a forger of stock certificates is liable in fraud to all subsequent holders. See Bruff v. Mali, 36 N.Y. 200 (1867).

We agree entirely with the conclusion in Pasternack v. Laboratory Corp. of America, No. 10-cv-4426 (PGG), 2014 WL 4832299, at *17-18, 2014 U.S. Dist. LEXIS 137671, at *45-51 (S.D.N.Y. Sept. 29, 2014), appeal docketed, No. 14-4101-cv (2d Cir. argued June 2, 2015), that Eaton and Bruff are consistent with Smokes- Spirits.com and Lollo. In both Eaton and Bruff, but not in Lollo or Pasternack, the plaintiff relied on the defendant's false representation. And in both Eaton and Bruff, but not in Smokes-Spirits.com or Pasternack, the plaintiff was among the intended recipients of the defendant's false representation.

Within the framework of this litigation, we must ask what the intended purposes of LIBOR submissions were. It would be facetious to pretend that nobody (save perhaps the BBA) was expected to use LIBOR. By the mid-2000s, the network effects described by some plaintiffs (see, e.g., FDIC Am. Compl. ¶¶ 50, 56) had turned LIBOR into the benchmark of choice for interest rate swaps, floating-rate loans, and adjustable mortgages. The principal use of LIBOR, and thus the principal use of LIBOR quotes, was to calculate payments on these kinds of LIBOR-based financial instruments. It follows that almost every plaintiff is within the class of persons to whom defendants' LIBOR quotes were directed.

Certain exceptions illuminate the standard. One exception is the claim in Salix on behalf of hedge fund managers that they were unable to charge high management fees to their funds because the funds lost money on LIBOR-related transactions that the managers selected. Whether viewed from the "expected reliance" or "proximate cause" perspective, this is too far removed from LIBOR manipulation to support recovery. It was the funds themselves who suffered direct, foreseeable injury through their LIBOR-based instruments. That, by private contract, some of this loss passed through to the funds' managers means only that the managers suffered indirectly. Even if this harm was in some sense foreseeable (for we accept that major banks know that hedge fund managers take a substantial percentage of their funds' profits), we do not view hedge fund managers as being within the class of persons towards whom LIBOR was directed. Plaintiffs' theory is especially troublesome because it would create a double recovery, an intrinsic danger of ignoring the proximate causation element: the hedge funds themselves would recover for the full amount of their losses and the managers would recover for their management fees.

It is a close question whether plaintiffs who used LIBOR-based pricing to decide whether to invest in LIBOR-based instruments were within the "class of persons" who were expected to rely on LIBOR. Although it was foreseeable that LIBOR quotes would affect the pricing of LIBOR-based instruments, this reliance may take innumerably different forms depending on an investor's goals and the available alternatives. We need not resolve our doubts because this theory (unlike plaintiffs' other theories) is essentially a "fraud on the market" theory that efficient market forces embedded defendants' false information in otherwise reliable prices. Cf. Basic Inc. v. Levinson, 485 U.S. 224, 246- 47 (1988) (adopting fraud on the market for federal securities fraud). As defendants point out, the common law does not generally recognize a "fraud on the market" theory of reliance. See Ex parte Household Retail Servs. Inc., 744 So. 2d 871, 880 n.2 (Ala. 1999); Mirkin v. Wasserman, 5 Cal. 4th 1082, 1100-08, 858 P.2d 568, 579-84 (1993); Kaufman v. i-Stat Corp., 165 N.J. 94, 113-118, 754 A.2d 1188, 1198-1201 (2000); Recent Case, Kaufman v. i-Stat Corp., 754 A.2d 1188 (N.J. 2000), 114 Harv. L. Rev. 2550, 2550 (June 2001) ("[A]s yet, no state appellate court has incorporated [fraud on the market] into the common law."); cf. Aspinall v. Philip Morris Cos., 442 Mass. 381, 399 n.23, 813 N.E.2d 476, 490 n.23 (2004) (distinguishing plaintiffs' theory from fraud on the market and declining to adopt fraud on the market doctrine); but cf. Minpeco, S.A. v. Hunt, 718 F. Supp. 168, 175 (S.D.N.Y. 1989) (predicting that New York would accept fraud on the market); Hurley v. FDIC, 719 F. Supp. 27, 34 n.4 (D. Mass. 1989) (same, predicting Massachusetts law). In light of this consistent precedent, we will not permit a fraud claim on the theory that plaintiffs relied on LIBOR-based pricing to decide whether to invest in LIBOR-based products.

3.4.2. Actual Reliance

Plaintiffs, together with their counterparties and issuers, incorporated LIBOR into contracts to define payments. This was reliance. When LIBOR was lower, the pay-fixed side of a swap willingly paid more money or accepted less money; when LIBOR was higher, the pay-floating side did the same.

3.4.2.1. Defendants' BBA Filter Argument

Defendants argue that plaintiffs did not "rely" on LIBOR quotes because each bank's submission was filtered through the BBA's calculation process. We find defendants' approach overly formalistic, because each panel bank's LIBOR submission affected plaintiffs' calculations of their cash flows in a direct and measurable way. Although no single defendant's manipulation makes it liable (absent vicarious liability) for the full extent of LIBOR suppression, each defendant is liable for whatever change it helped cause.

On this point, defendants tend to overstate the cases they cite. In the unusual case of SIPC v. BDO Seidman L.L.P., 95 N.Y.2d 702, 709-11, 746 N.E.2d 1042, 1047-48 (2001), the defendant auditor communicated a false report about a company to the National Association of Securities Dealers (NASD). NASD was required to divulge negative information about this company to plaintiff. Having received no such negative information, NASD failed to disclose any such information, and thus plaintiff inferred that defendant had issued a clean audit. The Court of Appeals held that plaintiff's inference was misguided. As a general matter, NASD exercised significant case-by-case discretion in deciding which information to communicate to plaintiff, and so NASD's silence did not logically imply a direct misrepresentation from defendant.

BDO Seidman actually supports plaintiffs, because it appears that SIPC would prevailed if the inference of a clean audit had been clearer. As a "general and unremarkable principle," the Court wrote, "liability for fraud can be imposed through communication by a third party." 746 N.E.2d at 1047, 95 N.Y.2d at 710. See also Gawara v. U.S. Brass Co., 63 Cal. App. 4th 1341, 1357-58, 74 Cal. Rptr. 2d 663, 672 (4th Dist. 1998) (accepting reliance through indirect communications as a legal theory, but finding that evidence in the cases did not establish actual reliance).

Municipality of Bremanger v. Citigroup Global Markets, Inc., No. 09-cv-7058 (VM), 2013 WL 1294615, at *13-18, 2013 U.S. Dist. LEXIS 49603, at *37-56 (S.D.N.Y. Mar. 28, 2013), another case cited by defendants, simply stands for the proposition that a defendant is not liable for making a statement that, as stated, was neither false nor misleading. In contrast, the allegation here is that the panel banks' LIBOR submissions were false as transmitted to the BBA.

Finally, defendants rely on an unexceptional stray quote in In re Enron Corp. Securities, Derivative & ERISA Litigation, 56 pages away from the holding of the case, which has nothing to do with the element of actual reliance. Compare 762 F. Supp. 2d 942, 967 (S.D. Tex. 2010) ("Plaintiff must show that it actually and justifiably relied on the defendant's representation."), with id. at 1022 (dismissing misrepresentation claims against auditor because plaintiffs were not especially likely to rely on representations, and dismissing omission claims because auditor had no duty to plaintiffs).

Having reviewed these cases, we conclude that plaintiffs relied on false LIBOR quotes despite the BBA's mechanical process for combining quotes into a single number. To the extent that a bank's LIBOR quote caused LIBOR to be different than it should have been, the bank can be liable in fraud for the damages that it actually caused.

3.4.2.2. Defendants' Calculation Argument

Defendants next argue that plaintiffs did not rely on LIBOR because some other person typically performed the LIBOR-based calculations that plaintiffs' contracts required. For over-the-counter contracts, a schedule to the counterparties' master ISDA agreement typically names the swap dealer as the "Calculation Agent." For asset-backed securities, a "loan servicer" calculates the (LIBOR-based) amount to bill each borrower.

This argument fails because the calculation agent or loan servicer is a true agent, who undertakes certain duties on behalf of both counterparties (in the case of a swap) or on behalf of the trust and its beneficiaries (in the case of asset-backed securities). An agent's reliance is equivalent to a principal's reliance, so the calculation agent's reliance is sufficient to assert fraud claims. See, e.g., Wyeth, Inc. v. Weeks, 159 So. 3d 649, 693 n.26 (Ala. 2014); Thrifty-Tel, Inc. v. Bezenek, 46 Cal. App. 4th 1559, 1568, 54 Cal. Rptr. 2d 468, 474 (4th Dist. 1996). Furthermore, some agreements provide that both counterparties will independently calculate payments and will follow a prescribed procedure for disputing the calculation agent's assessments. In such a case, the counterparties rely on LIBOR to perform their calculations just as much as the calculation agent does.

3.5. Justifiable or Reasonable Reliance

As to both false data fraud and LIBOR quality fraud, defendants argue that reliance on published LIBOR was unreasonable because of warning signs that LIBOR was a corrupted or corruptible benchmark. This argument is based largely on our previous holdings that widely published news articles placed plaintiffs on inquiry notice of their claims in Spring 2008.

At the outset, whether a plaintiff took reasonable advantage of access to information is an "always nettlesome," "fact-intensive" question. Schlaifer Nance & Co. v. Estate of Warhol, 119 F.3d 91, 98 (2d Cir. 1997). Reasonableness is normally "not to be decided on a motion to dismiss," Barron Partners, LP v. Lab123, Inc., No. 07-cv-11135 (JSR), 2008 WL 2902187 at n.3, 2008 U.S. Dist. LEXIS 56899 at n.3 (S.D.N.Y. July 25, 2008), because a plaintiff must only plead facts sufficient to "render the claim" of reasonableness "plausible." Thomas H. Lee Equity Fund V, L.P. v. Grant Thornton LLP, 586 F. Supp. 2d 119, 135 (S.D.N.Y. 2008).

Justifiable reliance is different from inquiry notice. Inquiry notice depends on whether a plaintiff was obligated to investigate signs of fraud. Justifiable reliance depends on whether a plaintiff was obligated to cease all trading in LIBOR-based instruments.

Here, the pleadings show that news articles in Spring 2008 revealed a meaningful probability that LIBOR had been, and continued to be, manipulated. Other reports delivered conflicting information. We have previously stated, and strongly adhere to, a rule that a prospective plaintiff faced with conflicting information bears a duty to investigate the possibility of injury. Nevertheless, it was plausibly reasonable for plaintiffs to continue using LIBOR with trepidation in the face of conflicting information about the benchmark's accuracy. Plaintiffs may have faced real difficulty or expense in achieving their financial objectives (for example, hedging interest-rate risk, speculating on short-term interest rates, or receiving credit-risky income without duration risk) if they had refused to deal in LIBOR-based securities. Whether these decisions were reasonable depends on each particular plaintiff's reasons for investing in LIBOR-based instruments, the alternatives available to each particular plaintiff, each particular plaintiff's ability to investigate the possibility of LIBOR manipulation, and how much credence a reasonable investor would have lent to news articles criticizing LIBOR.

Finally, we note that justifiability is measured at the time that plaintiff committed to rely on LIBOR. This is especially relevant to the false data claims, as plaintiffs may have relied on LIBOR to calculate particular payments years after committing to do so. If the commitment at the time of executing a swap agreement was reasonable, then the reliance that necessarily followed, even years later, is actionable.

3.6. Damages

Defendants advance three arguments directed to whether particular plaintiffs have adequately pleaded damages.

First, defendants argue that certain plaintiffs (those in BATA, Fannie Mae, FDIC, Houston, and Triaxx) fail to plead damages, asserting only that their "LIBOR-based financial products" suffered losses. Given the complexity and broad scope of plaintiffs' alleged losses, the complaints in each case give adequate notice of plaintiffs' positions in interest rate swaps and other LIBOR-based products. See BATA Am. Compl. ¶ 252-53; Fannie Mae Am. Compl. ¶¶ 38-39, Exs. 25-28; FDIC Am. Compl. Exs. 22-73; Houston Am. Compl. ¶¶ 387-95; Triaxx Am. Compl. ¶ 134, App. C. Defendants may seek additional information in discovery.

Second, Houston's allegation that it was harmed (through its auction-rate securities) by an isolated incident of upward manipulation is not inconsistent with its allegation that it was also harmed (through its interest rate swaps) by persistent suppression. Houston simply must plead isolated manipulation in the manner we have prescribed. See supra at 100.

Third, defendants submitted that punitive damages are not available in certain states that defendants consider relevant to the FDIC. Although we are inclined to agree with defendants, defendants now appear to accept (for purposes of this motion) that New York law applies. See Joint Conflicts Spreadsheet at 13. Furthermore, the parties' briefs did not examine whether the availability of punitive damages is a conduct-regulating or loss-allocating rule according to New York's conflicts principles, or how New York's loss-allocation doctrine should apply. Accordingly, the FDIC's claims for punitive damages survive this motion.

4. Negligent Misrepresentation

Plaintiffs' negligent misrepresentation claims are directed to defendants' false statements and omissions about LIBOR, not to false LIBOR submissions. These negligent misrepresentation claims largely fail for the same reasons that plaintiffs' claims for fraudulent misrepresentations fail: redundancy with contract claims, lack of a pleaded misrepresentation, lack of reliance, and absence of duty. However, plaintiffs adequately state claims for fraudulent omissions in the course of offering or transacting LIBOR-based securities, see supra at 141, and so we examine whether certain plaintiffs (those in the California Consolidated Cases, FDIC, Houston, the Principal Cases, Prudential, and Triaxx) may pursue an alternative theory of negligent omission.

The general rule is that negligent misrepresentation is actionable as to omissions, so long as a defendant had a special duty to communicate information. See Gibbs v. Ernst, 538 Pa. 193, 213-15, 647 A.2d 882, 892-93 (1994) (negligent failure to disclose child's history of physical and sexual abuse before placing child for adoption); Pearson v. Simmonds Precision Prods., Inc., 160 Vt. 168, 170-72, 624 A.2d 1134, 1135-37 (1993) (employer's negligent failure to disclose imminent layoffs to new employee); Restatement (Second) of Torts § 551 cmt. d; but see Ross v. Kirner, 162 Wash. 2d 493, 499-500, 172 P.3d 701, 704-05 (2007) ("An omission alone cannot constitute negligent misrepresentation.").

As to California in particular, defendants cite cases stating that a claim for negligent misrepresentation may not be founded on an omission. See Wilson v. Household Fin. Corp., No. Civ. S-12-1413 (KJM), 2013 WL 1310589, at *4, 2013 U.S. Dist. LEXIS 46126, at *12 (E.D. Cal. Mar. 28, 2013) (citing Yanase v. Auto. Club of So. Cal., 212 Cal. App. 3d 468, 472-73, 260 Cal. Rptr. 513, 516 (4th Dist. 1989)). However, neither case dealt with a situation in which the defendant bore an affirmative duty to convey information to the plaintiff. The publisher of a tourist guide, for example, has no duty to convey information about neighborhood safety as was alleged in Yanase. If anything, California law is especially favorable to plaintiffs. In California, if a defendant has a duty to reveal information, then the duty "swallows up plaintiff's . . . cause of action for negligence." Wells v. John Hancock Mut. Life Ins. Co., 85 Cal. App. 3d 66, 73, 149 Cal. Rptr. 171, 176 (2d Dist. 1978). California lacks a "negligent omission" rule only because the negligent failure to convey information in the face of a duty to do so constitutes a fraudulent omission.

Given our holding that defendant counterparties had a duty to communicate what if anything they knew of LIBOR manipulation before transacting, claims based on a failure to communicate that knowledge may proceed on a negligence theory as well as a theory of deceit.

5. Securities Fraud

5.1. Federal Claims

Schwab pursues federal claims pursuant to section 10(b) of the Securities Exchange Act and SEC Rule 10b-5 on what appear to be three factual theories:

1. Schwab bought LIBOR-based bonds during the persistent suppression period at an artificial price.

2. Schwab bought fixed-rate bonds during the persistent suppression period in preference to artificially-priced floating-rate bonds.

3. Schwab received artificial interest payments during the persistent suppression period.

Schwab's first theory is illogical. If LIBOR was (as Schwab pleads) persistently suppressed when Schwab bought LIBOR-based bonds, then the bond's expected future interest payments would also have been suppressed (or, at the very least, not inflated). Because a bond's price is equal to the present value of its expected future interest and principal payments, the bond's purchase price would also necessarily have been suppressed, so that Schwab may reap a windfall now that suppression has ended. In other words, as a purchaser of bonds during the suppression period, Schwab was benefited rather than injured by any persistent suppression. This observation is not "fact-finding," as Schwab would have it, but is the kind of "common economic experience," Bell Atl. Corp. v. Twombly, 550 U.S. 544, 565 (2007), that may appropriately bear on a motion to dismiss. We need not defer to pleading that suppression of a bond's coupon payments caused the bond's price to be inflated any more than we must defer to a pleading that falsely optimistic information about a company harmed sellers of the company's stock.

Schwab's second theory fares no better. Even if artificially cheap pricing of floating-rate bonds could have induced Schwab to buy fixed-rate bonds, federal securities law allows a private right of action only "in connection with the purchase or sale" of a security. Exchange Act § 10(b), 15 U.S.C. § 78j(b); SEC Rule 10b-5. Schwab's claim is essentially that it declined to purchase manipulated securities. Such a claim is not viable. See Blue Chip Stamps v. Manor Drug Stores, 421 U.S. 723, 754 (1975) (declining to recognize a private right of action in favor of an offeree who allegedly declined to purchase stock in reliance on an overly pessimistic prospectus).

Schwab's third theory is at least factually plausible, as it is the same factual theory that we sustain in the context of common law fraud claims. However, plaintiff has cited no case (and we have found none) treating the receipt of a cash dividend or interest payment as a "purchase or sale" of securities. Cf. Int'l Controls Corp. v. Vesco, 490 F.2d 1334, 1346 (2d Cir. 1974) (holding that the spin-off of corporate stock by way of an in-kind dividend constituted a "sale"); In re Adelphia Commc'ns Corp. Sec. & Deriv. Litig., 398 F. Supp. 2d 244, 260 (S.D.N.Y. 2005) (holding that the distribution of stock as a dividend did not constitute a "sale"). The Supreme Court has repeatedly cautioned against extending Rule 10b-5's implied private right of action without congressional guidance, and we will not do so in this context. See Stoneridge Inv. Partners, LLC v. Scientific-Atlanta, 552 U.S. 148, 158 (2008) (declining to extend private right of action to aiders and abettors); Blue Chip Stamps.

Our dismissal of Schwab's Exchange Act claims is not inconsistent with our earlier dismissal of Schwab's RICO claims. As we explained in LIBOR I, "the dispositive inquiry" as to whether a RICO claim is barred "is whether the alleged predicate acts could form the basis for a securities fraud suit by the SEC." 935 F. Supp. 2d at 726 (emphasis added). Because "defendants' misrepresentations to the BBA would likely be grounds for a securities fraud claim to the SEC," id. at 729, Schwab's RICO claim was barred regardless of whether Schwab itself possessed a private right of action.

5.2. California Blue Sky Claims

Schwab's claims pursuant to California's blue sky law is essentially a claim that broker-dealer defendants committed fraud in the sale or issuance of bonds. Schwab has failed to plead that any entity both sold bonds to Schwab and possessed private information about LIBOR manipulation. (In fact, Schwab does not name any brokerage house as a defendant or plead facts indicating that any named defendant was a control person of a particular brokerage. Instead, Schwab simply lists several broker-dealers that were affiliated with some panelist defendants and concludes that the panelist defendants "controlled or otherwise directed or materially participated in" the broker-dealers' operations. See Schwab Am. Compl. ¶¶ 274-75.) Accordingly, Schwab's state-law securities claims are dismissed.

6. Conclusions

The Individual Plaintiffs adequately allege common law fraud in three ways. First, plaintiffs allege that their 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. Second, Fannie Mae alleges that its counterparties breached swap agreements and then falsely represented that no such breach existed each time they entered into subsequent swap agreements. Third, and most importantly, plaintiffs allege that LIBOR panel banks committed fraud by submitted false quotes to the BBA.

VII. BREACH OF CONTRACT , UNJUST ENRICHMENT, AND TORTIOUS

INTERFERENCE

Most Individual Plaintiffs attempt to state claims for breach of contract (including the implied covenant of good faith and fair dealing), unjust enrichment, and tortious interference. Because these claims are to some degree related, we treat them together.

We adhere to our prior rulings in sustaining most claims against counterparties for unjust enrichment and breach of the implied covenant of good faith and fair dealing (hereinafter the "implied covenant"), and in dismissing claims for breaches of express contractual terms. In our prior opinions, we recognized implied covenant claims and unjust enrichment claims as alternative remedies for a counterparty's own participation in LIBOR manipulation. Here, we extend this holding to permit unjust enrichment as a remedy for damages caused by a counterparty's affiliate.

Some tortious interference claims against non-counterparty panel banks survive. To prevail, each plaintiff must ultimately demonstrate (1) that the plaintiff's counterparty breached the implied covenant with the non-counterparty defendant's assistance or encouragement, and (2) that the non-counterparty defendant either knew of the plaintiff's particular contract or else manipulated LIBOR for the specific purpose of affecting a portfolio of contracts belonging to plaintiff's counterparty.

1. Prior Rulings

1.1. LIBOR II

In LIBOR II we recognized LIBOR manipulation as a breach of the implied covenant that inheres in every LIBOR-based swap contract, writing:

In entering into these contracts, plaintiffs allege, they expected LIBOR to be set according to its definition, such that it reflected the average interest rate being charged in the London interbank lending market. Such an expectation would have been integral to the "bet" that is one purpose of entering into a swap. . . . By allegedly manipulating LIBOR downward, . . . defendants depressed the consideration plaintiffs received pursuant to their contracts and undermined the contractual bargain whereby plaintiffs agreed to pay a certain fixed rate in exchange for receiving a rate that reflected prevailing interest rates. In other words, plaintiffs have alleged that defendants "injured their right to receive the fruits of the contract."

Although we accepted plaintiffs' implied-covenant theory, we rejected plaintiffs' theory of express breach, because contracts referring to "LIBOR" did not "require[] the benchmark used to calculate the floating amount to be based on the LIBOR definition." Id. at 631 n.31. We noted that, "given that LIBOR is, by definition, an average of eight banks' submissions to the BBA, no one bank could possibly guarantee that a particular LIBOR fix was determined in a manner that wholly complied with the BBA's rules." Id.

We also permitted plaintiffs to plead unjust enrichment in the alternative to their contractual claims, because "the contracts did not 'clearly cover' the subject matter now at issue, namely whether defendants were permitted to manipulate LIBOR itself and thereby depress the amount they were required to pay plaintiffs." Id. at 630 (quoting Clark-Fitzpatrick, Inc. v. Long Island R.R. Co., 70 N.Y.2d 382, 389, 516 N.E.2d 190, 193 (1987)) (internal modifications omitted). In so holding, we agreed with plaintiffs that defendants allegedly "reaped a 'net gain' at the expense of plaintiffs' 'net loss.'" Id. "[P]laintiffs entered into swap contracts directly with defendants, and the allegation is that defendants benefited at plaintiffs' expense by paying plaintiffs less on those contracts." Id. at 631.

While these holdings pertained to swaps, the logic of LIBOR II applies equally to other instruments.

1.2. LIBOR III

LIBOR II permitted the OTC Plaintiffs to amend their complaint to state implied covenant and unjust enrichment claims, having merely found that those claims were not futile. In LIBOR III, we addressed defendants' motion to dismiss those same claims.

In LIBOR III, we made explicit the "counterparty requirement" for contract and unjust enrichment actions, holding that plaintiffs may not state contract or unjust enrichment claims against defendants that were not their own counterparties. 27 F. Supp. 3d at 478-79. Further, we held that conclusory allegations of a non-counterparty's financial benefit were insufficient to support an unjust enrichment claim, and that conspiracy pleading was insufficient to overcome the lack of a relationship with a particular defendant. Id. at 479.

As to counterparty defendants, we reiterated our holdings of LIBOR II, concluding that implied covenant and unjust enrichment claims were adequately pleaded.

1.3. Order Dated October 8, 2014

Following LIBOR III, defendant Credit Suisse Group AG brought to our attention that LIBOR III did not distinguish between it and Credit Suisse International, and asked us to clarify that Credit Suisse Group AG, a non-counterparty, was dismissed from the OTC class action. See Kurtzberg Letter, Aug. 13, 2014, ECF No. 590. In a two-paragraph order, we agreed and dismissed Credit Suisse Group AG. See Order, Oct. 8, 2014, ECF No. 682. Although we noted that the OTC plaintiffs' claims against Credit Suisse International survived LIBOR III, this meant only that the claims against Credit Suisse International survived the arguments that we had addressed in LIBOR III. The Individual Plaintiffs misread our Order as foreclosing defendants from arguing that only a panelist entity can be liable for a breach of the implied covenant. In fact, as we conclude below, a non-panelist counterparty cannot be liable for breaching the implied covenant unless the counterparty itself participated in manipulating LIBOR.

2. The Elements

2.1. Choice of Law

As with the fraud claims, the parties generally agree on the choice of substantive law. New York law governs most claims, especially the contract claims. Also, for the reasons stated above, we apply New York law rather than un-briefed Caymanian law to Capital Ventures' claims. See supra at 126.

2.2. Contract

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." Nat'l Mkt. Share, Inc. v. Sterling Nat'l Bank, 392 F.3d 520, 525 (2d Cir. 2004) (applying New York law).

"A covenant of good faith and fair dealing in the course of contract performance" is "[i]mplicit in all contracts," and thus a breach of this implied covenant constitutes a breach of the contract. Dalton v. Educ. Testing Serv., 87 N.Y.2d 384, 389, 663 N.E.2d 289, 291 (1995) (citing Van Valkenburgh, Nooger & Neville, Inc. v. Hayden Publ'g Co., 30 N.Y.2d 34, 45, 281 N.E.2d 142, 144 (1972)); accord Restatement (Second) of Contracts § 205. 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." Dalton, 87 N.Y.2d at 389, 663 N.E.2d at 296 (quoting Kirke La Shelle Co. v. Armstrong Co., 263 N.Y. 79, 87, 188 N.E. 163 (1933)) (internal quotation marks omitted). In particular, when a counterparty plays a discretionary role in setting a contractual price, then the counterparty must exercise its discretion in good faith. See Cal. Lettuce Growers v. Union Sugar Co., 45 Cal. 2d 474, 483-84, 289 P.2d 785, 791 (1955) (sugar processor not permitted to alter its accounting system to reduce calculations of payments to grower of sugar beets). That said, the implied covenant arises "only in connection with the rights or obligations set forth in the terms of the contract," and "cannot create duties that negate explicit rights under a contract." Paul v. Bank of Am. Corp., No. 09-cv-1932 (ENV)(JMA), 2011 WL 684083, at *6, 2011 U.S. Dist. LEXIS 15569, at *17 (E.D.N.Y. Feb. 16, 2011); LJL 33rd St. Assocs., LLC v. Pitcairn Props. Inc., 725 F.3d 184, 195 (2d Cir. 2013).

2.3. Unjust Enrichment

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." Mandarin Trading Ltd. v. Wildenstein, 16 N.Y.3d 173, 182, 944 N.E.2d 1104, 1110 (2011) (internal quotations omitted); see also Restatement (Third) of Restitution and Unjust Enrichment § 1.

2.4. Tortious Interference

To state a claim for tortious interference with a contract, a plaintiff must allege: "(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.'" Kirch v. Liberty Media Corp., 449 F.3d 388, 401-02 (2d Cir. 2006) (quoting Lama Holding Co. v. Smith Barney Inc., 88 N.Y.2d 413, 424, 668 N.E.2d 1370, 1375 (1996)). To state a claim for tortious interference with business relations, prospective economic advantage, or the like, a plaintiff must allege: "(1) business relations with a third party; (2) the defendant's interference with those business relations; (3) [that] the defendant acted with the sole purpose of harming the plaintiff or used dishonest, unfair, or improper means; and (4) injury to the business relationship." Nadel v. Play-By-Play Toys & Novelties, Inc., 208 F.3d 268, 382 (2d Cir. 2000).

3. Legal Discussion

We now turn to the parties' disputes concerning these contract-related claims. Most of the arguments regarding contract and unjust enrichment are retreads of arguments raised in previous motions, and do not persuade us to alter our considered earlier holdings. As for tortious interference, defendants principally argue that they lacked the requisite contract-specific scienter, and that a corporation cannot as a matter of law tortiously interfere with its affiliate's contracts. We conclude that, under limited circumstances, plaintiffs may maintain claims for tortious interference.

3.1. Contract and Unjust Enrichment

3.1.1. Breach of Express Contractual Terms

With one exception, plaintiffs' "express breach" claims fail for the reasons stated in LIBOR II.

Plaintiffs chiefly attempt to distinguish LIBOR II by citing contracts that expressly require defendants to calculate payments in good faith. This argument fails because those contractual provisions deal with the mechanical calculation of payments based on published LIBOR, not with the calculation of LIBOR itself.

Some contracts (most notably ISDA agreements) treat misrepresentations as breaches. As discussed supra at 138, Fannie Mae adequately alleges that some of its counterparties made misrepresentations when executing swaps regarding the existence of breaches of previously executed swaps. Insofar as these misrepresentations were also warranties, Fannie Mae may proceed on a misrepresentation theory or a breach of warranty theory. See Merrill Lynch & Co. v. Allegheny Energy, Inc., 500 F.3d 171, 184 (2d Cir. 2007) (finding "that the alleged misrepresentations would represent, if proven, a breach of the contractual warranties as well does not alter the result" that the misrepresentation is collateral to contract performance and therefore actionable.).

3.1.2. Counterparty Requirement

We adhere to our conclusion in LIBOR III that contract and unjust enrichment claims may only be alleged against a counterparty. See LIBOR III, 27 F. Supp. 3d at 478-79; C.J.S. Contracts § 824. The converse is not, however, true. That is, the bare fact that a defendant transacted with a plaintiff does not open the door to liability.

In particular, plaintiffs' allegations fail against entities that were merely involved in the sale of LIBOR-related securities—underwriters, brokers, dealers, agents, managers, bookrunners, and so on. Most of these vendors agreed to do nothing more than to sell or offer securities at a bargained-for price. They had no duty under contract law to advise sophisticated investors of LIBOR-related risks, no duty to deal at any particular price, and no ongoing duties of good faith after concluding a sale on bargained-for terms. Even when a defendant had some ongoing relationship with a plaintiff (for instance, a role in processing an MBS's mortgage payments), it is far from clear that the defendant can be liable for breaching the implied covenant, because LIBOR manipulation was too distant from the defendant's duties to the plaintiff. Furthermore, these vendors were not enriched by LIBOR manipulation. Suppression benefited obligors who paid less in home-loan interest, bond interest, or swap payments.

The Amabile Plaintiffs' exchange-based claims against panel banks also fail. In the context of trading on an exchange, the panel banks did not contract directly with the Amabile Plaintiffs, and so the panel banks' unjust gains, if any, cannot be identified as corresponding to the Amabile Plaintiffs' losses.

3.1.3. Wrongdoing and Scienter

3.1.3.1. Implied Covenant

To maintain an implied covenant claim, a plaintiff must make "some showing of intent to harm the other contracting party or a reckless disregard" of the risk of harm. LIBOR II, 962 F. Supp. 2d at 634 (quoting Paul v. Bank of Am. Corp., No. 09-cv-1932 (ENV) (JMA), 2011 WL 684083, at *6, 2011 U.S. Dist. LEXIS 15569, at *18 (E.D.N.Y. Feb. 16, 2011)). We previously held that the OTC plaintiffs plausibly alleged that panel banks were "at least in reckless disregard of the potential harm to OTC plaintiffs" when they suppressed LIBOR, LIBOR III, 27 F. Supp. 3d at 483, and we reach the same conclusion with respect to any panel bank that manipulated LIBOR in a manner that harmed its own counterparties.

Contrary to defendants' argument, an implied covenant claim is not absolutely barred when plaintiff's counterparty was a distinct entity from its affiliated panel bank. Such a claim can be asserted when plaintiffs can allege that the counterparty entity participated somehow in the panel bank's illicit manipulation. As discussed supra at 108, such a claim is adequately pleaded with respect to subsidiaries of Barclays and Lloyds.

3.1.3.2. Unjust Enrichment

A claim for unjust enrichment requires "circumstances . . . such that equity and good conscience require defendants to make restitution," CBS Broad. Inc. v. Jones, 460 F. Supp. 2d 500, 505 (S.D.N.Y. 2006), but not necessarily a showing of wrongdoing or bad faith by the specific recipient of unjust enrichment. See, e.g., Simonds v. Simonds, 45 N.Y.2d 233, 242, 380 N.E.2d 189, 194 (1978) (second wife liable for restitution when husband promised to name first wife as beneficiary of life insurance policy but named second wife instead).

In the LIBOR context, there are three types of claims to analyze: (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; (3) that the defendant counterparty was enriched through the misconduct of unrelated entities.

3.1.3.2.1. Personal Misconduct

In LIBOR II and III, we accepted restitution claims against entities that themselves manipulated LIBOR. When an entity itself participated in manipulating LIBOR, restitution is co-extensive with or alternative to liability for breaching the implied covenant of good faith.

3.1.3.2.2. Affiliated Misconduct

In LIBOR II and III, we did not consider whether an innocent affiliate of a panel bank may be liable in restitution. The precise question, which we now address, is whether good conscience requires one corporate entity to make restitution when it is benefits from the misdeeds of its corporate affiliate or shareholder.

Restitution is due (subject to equitable defenses) when one entity receives benefits through "some related third party's acts against the plaintiff, when it would be 'unconscionable for the receiving party to retain'" the benefits. Cty. of El Paso v. Jones, No. EP-09-cv-119 (KC), 2009 WL 4730345, at *15, 2009 U.S. Dist. LEXIS 113141, at *35 (W.D. Tex. Dec. 4, 2009) (quoting Mowbray v. Avery, 76 S.W.3d 663, 679 (Tex. App. Corpus Christi 2002)) ; see also Teachers Ins. & Annuity Ass'n of Am. v. CRIIMI MAE Servs. L.P., No. 06-cv-392 (LAK) (KNF), 2007 WL 7753901, at *9-10, 2007 U.S. Dist. LEXIS 28279, at *26-28 (S.D.N.Y. Mar. 20, 2007) (corporate entity liable for restitution when enriched by wrongdoing of parent), adopted in relevant part, 2007 WL 7569162, at *1, 2007 U.S. Dist. LEXIS 101862, at *2 (S.D.N.Y. Sept. 7, 2007), aff'd, 481 F. App'x 686 (2d Cir. 2012); cf. Montanez v. HSBC Mortg. Corp. (USA), 876 F. Supp. 2d 504, 517 (E.D. Pa. 2012) (dismissing unjust enrichment claim against wrongdoer's affiliate when mortgagor failed to plead that affiliate received inflated mortgage payments); Bank of Am. Corp. v. Gibbons, 173 Md. App. 261, 274-75, 918 A.2d 565, 573 (2007) (requiring innocent defendant to restore benefits derived from spouse's tort); Crossgates Realty, Inc. v. Moore, 279 Pa. Super. 247, 252, 420 A.2d 1125, 1128 (1980) (same). Extension of a restitution remedy in this context is just and necessary. Otherwise, a corporate tortfeasor could evade restitution simply by ensuring that a separate entity receives the profits of the tort.

Contrary to defendants' argument, applying the law of restitution to a corporate affiliate is not "veil piercing," and corporate formalities do not dictate the result. The doctrine of "veil piercing" requires one entity to pay for liability that the substantive law imposes upon a separate entity. Here, the substantive law of restitution makes a counterparty entity liable for its own enrichment even though the enrichment was caused by a different entity's conduct.

This theory is also not one of "group pleading," because the relevant fact—receiving and unjustly retaining an unjust enrichment—pertains to a specific counterparty entity that is liable for restitution.

Finally, the presence of a contract does not preempt this type of liability. Just as the contracts do not "clearly cover" whether a counterparty is "permitted to manipulate LIBOR itself," the contracts also do not "clearly cover" whether a counterparty may retain the benefits of its affiliates' manipulation.

3.1.3.2.3. Unaffiliated Misconduct

It is not clear whether plaintiffs contend that a counterparty may be liable for its gains on account of an unaffiliated panel bank's manipulation. Regardless of what position plaintiffs may wish to take on this point, we now hold that no unjust enrichment claim lies against a counterparty based on the misconduct of an entity that was unrelated to both parties.

In LIBOR II, we held that unjust enrichment was available as a remedy for a counterparty's misconduct, as the swap contracts did not "clearly cover" a counterparty's own misconduct. However, absent some misconduct as between the parties or their affiliates (as we have just held), the parties are bound by their decision to use LIBOR to define their mutual obligations, despite whatever flaws might have been inherent in LIBOR. One party's enrichment in this circumstance is not unconscionable, but simply fortuitous and well within the range of risks that both parties to a contract assumed.

If accepted, plaintiffs' theory would be disturbingly broad, as this theory would require restitution from counterparties who had nothing to do with manipulating LIBOR. For example, Goldman Sachs, a non-LIBOR bank, traded in interest rate swaps and possibly benefited from LIBOR manipulation. But it is not unconscionable for Goldman to keep its profits, because Goldman simply received the benefits of its bargains.

Accordingly, we conclude that a plaintiff may maintain 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. Potentially, this may include its own or its affiliate's assistance of an unrelated panel bank's manipulation. See supra at 121.

3.1.4. Unjust Enrichment as Alternative to Contract

To some extent, defendants revive their argument that unjust enrichment is barred when a contract covers the transaction. While accepting that LIBOR II and LIBOR III control as to most states, defendants assert that they have a strong argument under the laws of Nevada, Pennsylvania, and Puerto Rico.

We have no need to rule as to Nevada or Puerto Rico law. Although defendants initially argued that the law of these jurisdictions applies to some of the claims in FDIC, they have withdrawn that argument for purposes of this motion. See Joint Conflicts Spreadsheet.

As to Pennsylvania, defendants rely on Wilson Area School District v. Skepton, 586 Pa. 513, 520, 895 A.2d 1250, 1254 (2006). In that case, the plaintiff school district awarded a plumbing contract to the defendant. Under the terms of the contract, the plumber was responsible for paying certain permitting fees, and did so, but later recouped most of the fees by challenging the validity of the local government's fee structure. The school district sued the plumber for restitution on the theory that the plumber's retention of the recouped fees was an unwarranted windfall.

Three out of six justices subscribed to the following passage, upon which defendants now rely:

[I]t has long been held in this Commonwealth that the doctrine of unjust enrichment is inapplicable when the relationship between parties is founded upon a written agreement or express contract, regardless of how harsh the provisions of such contracts may seem in the light of subsequent happenings. While it does not appear that this Court has expounded upon this rule of law, it has been recognized that this bright-line rule not only has a distinguished common-law pedigree, but it also derives a great deal of justification from bedrock principles of contract law.

Two justices wrote that they "would apply the bright-line rule that a written contract precludes recovery based on unjust enrichment as a general, but not an inviolable, one." 586 Pa. at 520, 895 A.2d at 1256 (Saylor, J., concurring). The application of the rule, these justices explained, can depend on how the parties intended to allocate risks, and whether the plaintiff was aware of a particular risk. The Wilson Area School District lost because it had failed to submit evidence on these points.

One justice dissented (586 Pa. at 525, 895 A.2d at 1257 (Newman, J., dissenting)), and incorporated his previous opinion, Skepton v. Borough of Wilson, 562 Pa. 344, 353-56, 755 A.2d 1267, 1272-74 (2000) (Newman, J., dissenting), stating that the fees should have been refunded in the first place to the school district rather than to the plumber. The remaining justice did not participate in Wilson Area School District, but had previously joined Justice Newman's dissent in Borough of Wilson.

As is evident from this recapitulation, the sweeping bright-line rule that defendants rely upon failed to command a majority, and is not binding. Even if it were binding, the plurality relies on cases holding that a contract will not be implied in law when an express contract exists in fact. To that extent, the bright-line rule is correct. But, to the extent that a restitution claim derives from a tort, none of the plurality's cases suggest that the mere presence of a contract will bar a claim for unjust enrichment.

To illustrate, suppose A agrees to provide certain services to B in exchange for B devising A certain property. Both parties perform their contract. A then murders B to advance his inheritance. Even without a statutory "slayer rule" barring inheritance, A is liable to B's contingent beneficiary for unjust enrichment arising from A's tort. See generally Restatement (Third) of Restitution and Unjust Enrichment § 45 & cmt. b. That a contract governed some aspects of A and B's relationship would surely not bar such an unjust enrichment claim.

The controlling aspect of Wilson Area School District is the concurrence's statement that, when a contract governs the relationship between two parties, a plaintiff must bring forward evidence of how the parties intended to allocate a windfall. As applied to LIBOR, it would be premature to grant the motion to dismiss, because it is unclear that the parties intended for a counterparty to reap the benefits of its own (or an affiliate's) manipulation of LIBOR.

3.1.5. Free-Standing Unjust Enrichment

Defendants argue that New Jersey does not recognize a stand-alone action for unjust enrichment. To the contrary, we find New Jersey law to be typical of American jurisprudence. "To establish unjust enrichment, a plaintiff must show both that defendant received a benefit and that retention of that benefit without payment would be unjust." VRG Corp. v. GKN Realty Corp., 135 N.J. 539, 554, 641 A.2d 519, 526 (1994). Apart from the scenario of a contract implied in law for gratuitous services, restitution provides an alternative remedy to contract and tort. See Palmeri v. LG Elecs. USA, Inc., Civ. No. 07-5706 (JAG), 2008 WL 2945985, at *7, 2008 U.S. Dist. LEXIS 55391, at *17 (D.N.J. July 30, 2008). Although restitution "is limited for the most part to its use as a justification for other torts," Castro v. NYT Television, 370 N.J. Super. 282, 299, 851 A.2d 88, 98 (App. Div. 2004), and may "often turn[] out to be superfluous in light of other causes of action," Avram v. Samsung Elecs. Am., Inc., Civ. Nos. 2:11-6973 (KM), 2:12-976 (KM), 2013 WL 3654090, at *21, 2013 U.S. Dist. LEXIS 97341, at *67 (D.N.J. July 11, 2013), those considerations alone do not render restitution worthy of dismissal.

Defendants rely on Swift v. Pandey for the proposition that "New Jersey does not recognize unjust enrichment as an independent tort cause of action." Civ. No. 13-649 (JLL), 2013 WL 6022093, at *6, 2013 U.S. Dist. LEXIS 161590, at *19 (D.N.J. Nov. 13, 2013) (citing Castro and Steamfitters Local Union No. 420 Welfare Fund v. Philip Morris, Inc., 171 F.3d 912, 936 (3d Cir. 1999)). We think that Swift overstates both of the cases it relies on for this sweeping statement. Castro recognized that unjust enrichment has limited utility as a tort remedy, but did not purport to hold that unjust enrichment is unavailable. Steamfitters also recognized that unjust enrichment has become less valuable in many tort cases (because some of the "imperfections in the tort remedies . . . have now been removed"), but did not hold that an unjust enrichment must derive from an underlying tort claim.

In light of this analysis, we find no reason to dismiss unjust enrichment claims under New Jersey law.

3.2. Tortious Interference

3.2.1. Scienter

The two sides differ regarding the scienter element of tortious interference. Defendants maintain that a defendant must have known about a specific contract or business relationship and must have specifically intended to interfere with that contract or business relationship. To the contrary, plaintiffs argue that it is sufficient if the defendant knows that a plaintiff engages in a general type of transaction, and that no specific intent to interfere is required.

The most illuminating case cited by either side is Lightning Lube, Inc. v. Witco Corp., 4 F.3d 1153, 1169-72 (3d Cir. 1993) (applying New Jersey law). In that case, plaintiff was a quick-lube franchisor that bought oil-dispensing equipment from defendant on credit for installation at its franchise locations. In the midst of a commercial dispute between plaintiff and defendant, defendant caused plaintiff's franchisees to doubt that plaintiff owned the equipment, and threatened to confiscate the equipment from franchise locations unless the franchisees terminated their contracts with plaintiff. Defendant's scheme was successful. Several franchisees, fearing loss of their equipment, terminated their franchise agreements.

The court held that plaintiff could state a tortious interference claim by pleading either (1) knowledge of a specific contract and substantial certainty (cf. Restatement (Second) of Torts § 766 cmt. j) that defendant's conduct would interfere with that contract, or (2) specific intent to interfere with a category of contracts. These are a reasonable pair of limiting principles, as they require a substantial nexus between the contract (or business relation) and the wrongfulness of defendant's conduct.

In each case cited by defendants, no liability was found because of the plaintiff's inability to establish the existence of a breach of contract, the defendant's knowledge of specific contracts, or the defendant's specific intent to cause a breach. For example, in Boehner v. Heise, no liability attached to the defendants' actions in delaying plaintiffs' importation of ginseng because, as the court found at summary judgment, the defendants did not "kn[o]w of any specific contract that Plaintiffs had with their customers" and did not "direct[] any action at those customers in order to secure a breach of their contracts with Plaintiffs." 734 F. Supp. 2d 389, 404 (S.D.N.Y. 2010). In Mercury Skyline Yacht Charters v. Dave Matthews Band, Inc., defendants were not liable for interfering with a charter boat's contracts with passengers when defendants dumped human waste from a bridge onto the passengers, because defendants lacked knowledge of plaintiff's specific contracts and did not specifically intend to strike plaintiffs' passengers. No. 05-cv-1698, 2005 WL 3159680, at *7-10, 2005 U.S. Dist. LEXIS 29663, at *21-30 (N.D. Ill. Nov. 22, 2005). Finally, in Trump Taj Mahal Assocs. v. Costruzioni Aeronautiche Giovanni Agusta, S.p.A., defendants were not liable for tortious interference when their defective helicopter killed plaintiff's employees, even though defendants had marketed their helicopter for executive use, because defendants did not know of plaintiff's specific employment contracts and did not specifically intend to kill plaintiff's executive staff. 761 F. Supp. 1143 (D.N.J. 1991), aff'd, 958 F.2d 365 (Table) (3d Cir. 1992). As the Lightning Lube court explained, "[i]n those cases which require knowledge of the specific contract and prohibit third parties from recovering for tortious interference, the defendant did not intend to harm the suing parties." 4 F.3d at 1170; see also G.K.A. Beverage Corp. v. Honickman, 55 F.3d 762, 768 (2d Cir. 1995) (dismissing tortious interference claim when defendant's motive was to monopolize a soda bottling market, without regard to the effect on plaintiffs' contracts with a smaller bottler).

The Lightning Lube rule is likewise consistent with plaintiffs' point that a person can commit tortious interference without knowing the details of a contract, including the contract's parties. When a defendant has specific intent to interfere with a contract or relationship, there is no need for the plaintiff to plead defendant's knowledge of the contract's details.

Accordingly, we will follow the rule of Lightning Lube and will permit claims to proceed when a defendant (1) knew of a specific contract and knew to a substantial certainty that its conduct would induce a breach, or (2) specifically intended to induce a breach of a category of contracts. The leading example of the second prong in the context of LIBOR occurs when plaintiffs allege that the employees of two entities agreed that one would submit a false LIBOR quote to improve the other's trading profits. In this situation, it is plausible that the panel bank's employee specifically intended to assist the counterparty's employee in breaching the counterparty's implied covenant of good faith.

3.2.2. Interference with Affiliate's Contract

It is axiomatic that an entity cannot tortiously interfere with its own contractual obligations. Defendants ask us to extend this rule and to hold that one corporate entity cannot tortiously interfere with its affiliate's contractual obligations.

Some cases hold that a corporate shareholder is entitled to induce his corporation to breach a contract or to break a business relationship. Certainly, a shareholder may persuade a company's officers to breach a contract for the company's benefit. See, e.g., Deauville Corp. v. Federated Dep't Stores, Inc., 756 F.2d 1183, 1196 (5th Cir. 1985); Felsen v. Sol Cafe Mfg. Corp., 24 N.Y. 682, 687, 249 N.E.2d 459, 461 (1969).

But it is not the rule that "a sole shareholder is privileged to employ any means, no matter how improper, to induce a breach of contract involving its own company." Boulevard Assocs. v. Sovereign Hotels, Inc., 72 F.3d 1029, 1037 (2d Cir. 1995) (applying Connecticut law, and citing Connecticut and Louisiana cases). The more complete rule is that a "person with a financial interest [in the obligor] may safely interfere with a contract if his purpose is to protect his own interests and if he does not employ improper means." Record Club of Am., Inc. v. United Artists Records, Inc., 611 F. Supp. 211, 217 (S.D.N.Y. 1985) (citing Felsen); see also Woods v. Fox Broad. Sub., Inc., 129 Cal. App. 4th 344, 352, 28 Cal. Rptr. 3d 463, 470 (2d Dist. 2005). A shareholder (or other corporate affiliate) who employs improper means does not merit a "financial interest" defense because financial interest is essentially a justification, qualified by the duty not to commit an independent crime or tort. See Boulevard Assocs., 72 F.3d at 1037. Furthermore, we have found no case suggesting that a shareholder is justified in encouraging a breach of the implied covenant of good faith.

We have previously held that, as pleaded in some complaints, LIBOR manipulation was an independent tort—a fraud and a violation of the Commodities Exchange Act. Accordingly, no defendant is immune from tortious interference on the grounds that it interfered with its own affiliate's or subsidiary's contract.

3.2.3. Interference with Business Relations

Some plaintiffs press claims of tortious interference with business relations, prospective business advantage, prospective contracts, and so forth. However stated, the gist of these torts is that a plaintiff loses a customer through a defendant's wrongful conduct. See, e.g., Hanney Corp. v. GMI, Inc., 140 F.3d 194, 205 (2d Cir. 1998) ("[Plaintiff] lost a valuable business relationship when S & S Japan signed the exclusive distribution agreement with [defendant]."); Korea Supply Co. v. Lockheed Martin Corp., 29 Cal. 4th 1134, 131 Cal. Rptr. 2d 29 (2003) (military contract awarded to defendant as result of corrupt influence despite plaintiff's superior bid); Restatement (Second) of Torts § 766B. Such is not the case here. No plaintiff lost a customer or contractor as a result of defendants' manipulation of LIBOR. The only possible injury was that plaintiffs' contracts were worth less than they otherwise would have been. If any tortious claim lies, it is a traditional claim for tortious interference with contract. Plaintiffs' assertion of these other torts appears to be an attempt around the general rule that "deliberate interference with plaintiff's contractual rights that causes damage" is not "punishable as tortious interference" if the contract was not breached. NBT Bancorp v. Fleet/Norstar Fin. Grp., 87 N.Y.2d 614, 620, 664 N.E.2d 492, 495 (1996).

Because plaintiffs' complaints simply do not present actual claims for interference with business relations, we need not examine the heightened scienter requirements that courts have sometimes imposed. Compare NBT Bancorp, 87 N.Y.2d at 621, 664 N.E.2d at 496 ("[P]laintiff must show more culpable conduct [for the prospective tort].") (citing Guard-Life Corp. v. Parker Hardware Mfg. Corp., 50 N.Y.2d 183, 193-94, 406 N.E.2d 445, 451 (1980)), with Korea Supply Co. v. Lockheed Martin Corp., 29 Cal. 4th 1134, 1157, 131 Cal. Rptr. 2d 29, 48 (2003) ("[W]e find the intent requirement to be the same.").

3.3. Damages

The parties have not dwelt on questions of damages. We pause simply to note that a defendant is liable only for the portion of damages that the defendant caused or assisted in causing. In many cases, this may mean that a plaintiff may utilize the contract-related causes of action discussed in this section only to recover for the portion of the plaintiff's damages ascribable to its own counterparty's manipulation.

4. General Application

4.1. Over-the-Counter Derivatives

With respect to OTC claims, the relevant contract is between a plaintiff and its counterparty, typically a swaps dealer.

The counterparty is liable for a breach of the implied covenant if it was a panel bank or otherwise participated in manipulation. At this point, we do not dismiss such claims against counterparties, because it is generally plausible that banks' swap desks participated in associated panel banks' manipulation. The counterparty is also liable for unjust enrichment if the counterparty itself or an affiliated panel bank manipulated LIBOR.

An affiliated panel bank is potentially liable for tortious interference with contract if the counterparty committed a breach of the implied covenant. Thus, tortious interference claims against counterparties' panelist affiliates survive at this stage.

An unaffiliated panel bank is not liable for tortious interference, because the complaints give no reason to believe that any panel bank specifically targeted its conduct towards another bank's swap portfolio, or that any panel bank knew of any plaintiff's particular swaps with other banks.

4.2. Exchange-Traded Derivatives

In this category, the Amabile Plaintiffs attempt to state unjust enrichment claims. These claims fail because plaintiffs' futures and options contracts were not contracts with particular defendant counterparties.

4.3. Adjustable-Rate Bonds

A bond is a contract between plaintiff and plaintiff's obligor, typically a corporate issuer. A bond underwriter or seller also has knowledge of the bond's existence and terms, but is not a counterparty to the bond itself.

The obligor is liable for a breach of the implied covenant if the obligor was a panel bank or otherwise participated in manipulation, and the obligor is liable for unjust enrichment if the obligor itself or an affiliated panel bank manipulated LIBOR.

An obligor's affiliated panel bank is conceivably liable for tortious interference. There is, however, no suggestion in any of the pleadings that any panelist collaborated with an affiliate with specific intent to suppress bond payments, so a tortious interference claim can succeed only on the theory that a defendant panel bank knew of an affiliate's specific bond issuance. It is plausible that corporate affiliates are aware of each other's financing arrangements, so tortious interference claims may proceed against panel banks as to bonds issued by corporate affiliates. Of course, plaintiffs must ultimately prove that the issuing entity breached the implied covenant by assisting in the panel entity's LIBOR manipulation.

As discussed above, underwriters, vendors, and similar entities are not liable for breach of contract or unjust enrichment. Furthermore, although such entities were aware of particular contracts, there is no suggestion that they assisted obligors in suppressing coupon payments. Accordingly, claims against such entities fail.

4.4. Asset-Backed Securities

When an investor holds an asset-backed security, the investor actually holds a certificate as evidence that the investor is entitled to certain disbursements as beneficiary of a trust. The trust's res is a collection of assets that "back" the trust, such as a collection of homeowners' notes and mortgages. The trust has legal personality and acts through its trustee, who (at least following a default) is a fiduciary for the investors collectively. Cf. Trust Indenture Act § 314(c), 15 U.S.C. § 77mmm(c) (2012) ("The indenture trustee shall exercise in case of default . . . such of the rights and powers vested in it by such indenture, and [] use the same degree of care and skill . . . as a prudent man would exercise or use under the circumstances."). To administer this arrangement, the trust enters into contracts with other persons, including (1) borrowers, who are obligated to pay money on the debts that the trust holds; (2) a servicer, an agent of the trust who collects money from the borrowers; (3) a payment agent, an agent of the trust who disburses money to beneficiaries; (4) an issuer, who conveys assets to the trust at its inception, and who may continue to owe duties to the trust (such as a duty to repurchase defective loans); and sometimes (5) a guarantor, who guarantees the borrowers' payments. Other entities, such as underwriters and dealers, are involved in the marketing of the asset-backed security.

No action for breach of the implied covenant lies against the various agents and issuers, because LIBOR manipulation did not intersect with their limited duties. No action for unjust enrichment lies against defendants either, because it was the underlying borrowers who innocently profited from manipulation. And, because plaintiffs have not identified any breach of contract, plaintiffs may not maintain an action for tortious interference.

An additional challenge for plaintiffs—one that the parties did not address in their briefing—is that the trust (rather than plaintiffs as beneficiaries) may be the proper party to maintain claims against the trust's counterparties. We need not address this point, because plaintiffs' claims fail for other reasons.

4.5. The BBA

No complaint plausibly alleges that the BBA was unjustly enriched by plaintiffs, knew of plaintiffs' specific contracts, or specifically intended to induce a breach of any plaintiffs' contract. Accordingly, all unjust enrichment and tortious interference claims against the BBA fail.

VIII. ANTITRUST

Nearly every plaintiff asserts private antitrust claims. Most do so only for preservation, as we have previously dismissed antitrust claims on grounds that apply broadly to all LIBOR plaintiffs. A few plaintiffs attempt to distinguish our previous holdings. Although we reject plaintiffs' new arguments, we emphasize that our antitrust holdings should not be understood as an endorsement of defendants' alleged misconduct, which, if proven, the Commodities Exchange Act and the common law of fraud condemn. We simply hold that federal and state antitrust laws are the wrong vehicle for plaintiffs to obtain the damages they seek.

1. Prior Rulings

1.1. LIBOR I

We first ruled on federal antitrust claims in LIBOR I, 935 F. Supp. 2d at 686-95. There, we dismissed plaintiffs' antitrust claims for the reason that the plaintiffs' alleged injuries were not "antitrust injuries" "attributable to an anti-competitive aspect of the practice under scrutiny," Atl. Richfield Co. v. USA Petrol. Co. (ARCO), 495 U.S. 328, 334 (1990).

As we explained, it is not sufficient for plaintiffs to plead that defendants committed an antitrust violation (even a per se violation), or even that defendants committed an antitrust violation that harmed plaintiffs. LIBOR I, 935 F. Supp. 2d at 686 (citing ARCO, 495 U.S. at 339 n.8, 344). Rather, a plaintiff must allege that his injury resulted from the anti-competitive nature of the defendants' anti-competitive conduct. LIBOR I, 935 F. Supp. 2d at 686 (citing Nichols v. Mahoney, 608 F. Supp. 2d 526, 543-44 (S.D.N.Y. 2009)). Relying on the leading Supreme Court cases, ARCO and Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477 (1977), we then concluded that a plaintiff does not suffer an antitrust injury when the plaintiff's injury was consistent with free competition. See LIBOR I, 935 F. Supp. 2d at 689-92.

A recent case in this Court, In re Foreign Exchange Benchmarks Antitrust Litig., disagreed with our legal analysis while acknowledging that the two cases are factually distinguishable. See Nos. 13-cv-7789, 13-cv-7953, 14-cv-1364 (LGS), ___ F. Supp. 3d ___, ___, 2015 WL 363894, at *10-13, 2015 U.S. Dist. LEXIS 9826, at *38-46 (S.D.N.Y. Jan. 28, 2015), appeal on other grounds docketed sub nom. Larsen v. Barclays Bank plc, No. 15-574-cv (2d Cir. filed Feb. 26, 2015). As discussed below, we believe that the criticisms of Foreign Exchange are not well-founded, and we therefore stand by LIBOR I.

First, Foreign Exchange criticized LIBOR I for relying on summary judgment and post-trial cases in deciding a motion to dismiss at the pleading stage. It is true that, when the court must decide a question of factual sufficiency at the pleading stage, a court should avoid relying on summary judgment and post-trial opinions. See Swierkiewicz v. Sorema N.A., 534 U.S. 506 (2002) (holding that, at the pleading stage of a discrimination case, the plaintiff need not allege facts that would be sufficient to shift the evidentiary burden to the defendant at summary judgment or trial); Anderson News, L.L.C. v. Am. Media, Inc., 680 F.3d 162, 184 (2d Cir. 2012) (holding that facts alleged by an antitrust plaintiff need not rule out the possibility of parallel action at the pleading stage). But when the court must decide a purely legal question of how a civil cause of action is defined, opinions resolving motions on the pleadings, motions for summary judgment, and post-trial appeals are fungible. This is because the only difference among opinions from cases in different procedural postures is the set of facts—plausibly pleaded, genuinely disputed, or proven—that a court must match against the offense's legal definition: the offense's definition remains constant.

In LIBOR I, we adhered to this principle, relying on ARCO and Brunswick only to support our understanding of what constitutes a legally actionable "antitrust injury," and then applying this legal definition to the pleaded facts. This is entirely consistent with numerous cases that have relied on those same two cases at the pleading stage, and with Brunswick's own reliance on pleading cases. See, e.g., Associated Gen. Contractors of Cal., Inc. v. Cal. State Council of Carpenters, 459 U.S. 519, 540 (1983) (citing Brunswick in holding that plaintiff had failed to plead an antitrust injury); Blue Shield of Va. v. McCready, 457 U.S. 465, 479 (1982) (quoting Brunswick in holding that plaintiff had succeeded in pleading an antitrust injury); Gatt Commc'ns, Inc. v. PMC Assocs., L.L.C., 711 F.3d 68, 76-78 (2d Cir. 2013) (citing Brunswick and ARCO in holding that plaintiff had failed to plead an antitrust injury); Port Dock & Stone Corp. v. Oldcastle Ne., Inc., 507 F.3d 117, 122 (2d Cir. 2007) (same); see also Brunswick, 429 U.S. at 488 n.13 (citing as analogous cases Peterson v. Borden Co., 50 F.2d 644 (7th Cir. 1931) (affirming judgment sustaining defendant's demurrer), and Kirihara v. Bendix Corp., 306 F. Supp. 72 (D. Haw. 1969) (dismissing for failure to state cause of action)).

Second, Foreign Exchange disagreed with our analysis that the plaintiffs in Brunswick and ARCO lost because they "could have suffered the same harm under normal circumstances of free competition." LIBOR I, 935 F. Supp. 2d at 689. While we suppose that one could reduce these two cases—or any case—to their particular facts, we continue to believe that a district court has a duty to synthesize holdings into a coherent doctrine, and that our synthesis is correct.

In Brunswick, the plaintiff, a bowling alley operator, sought damages by alleging that a manufacturer of bowling equipment had violated antitrust laws by purchasing another chain of bowling alleys. The Court assumed that the manufacturer's purchase violated antitrust laws and that the purchase caused economic losses to the plaintiff, because the competing chain of bowling alleys would otherwise have gone out of business. The Court reasoned that the plaintiff "would have suffered the identical 'loss' . . . had the acquired centers" been saved by means other than an antitrust violation, such as "obtain[ing] refinancing or be[ing] purchased by 'shallow pocket' parents." 429 U.S. at 487. This was sufficient for the Court to decide Brunswick; the Court's additional point, that a private suit on Brunswick's facts was "inimical to the purposes of [antitrust] laws," id. at 488, was unnecessary to the disposition of the case.

ARCO supports this reading. The private plaintiff there was a gasoline marketer who complained of injury from a vertical price-capping agreement involving other gasoline marketers. Even though the plaintiff advanced record evidence to show that the plaintiff suffered economic harm as a result of this clear antitrust violation, the Court still found that the plaintiff's injury was not an "antitrust injury." The Court held that such a harm "cannot be viewed as an 'anticompetitive' consequence of the claimed violation," 495 U.S. at 337, because "cutting prices in order to increase business often is the very essence of competition," id. at 338 (quoting Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 594 (1986)). Notably, there was no suggestion that the low prices in ARCO were themselves the "essence of competition." Indeed, those low prices were the direct consequence of a per se antitrust violation. The ARCO plaintiff lost because its injury was of the type associated in typical circumstances with competition rather than collusion. Thus, ARCO supports our view that a court should consider typical alternative causes to decide whether an injury is an "antitrust injury."

This is not to say that antitrust injury exists only when the anticompetitive activity is the sole possible cause of a plaintiff's injury. The antitrust injury analysis is not akin to but-for causation, but rather to proximate causation. See Blue Shield of Va., 457 U.S. at 476-78 & n.13 (citing, among other cases, Palsgraf v. Long Island R. Co., 248 N.Y. 399, 162 N.E. 99 (1928)). Hypothetical alternative causes are relevant to this analysis to the extent they can assist in deciding whether the plaintiff's injury flowed from anticompetitive conduct or more directly from intervening causes, including other tortious aspects of defendant's conduct.

With this understanding of antitrust doctrine, LIBOR I then rejected the plaintiffs' two theories of how defendants engaged in anti-competitive conduct that caused antitrust injury to the plaintiffs.

First, focusing on LIBOR itself, we noted that "the process of setting LIBOR was never intended to be competitive," such that defendants' submission of false information "subverted [a] cooperative process." LIBOR I, 935 F. Supp. 2d at 688. Thus, we concluded that the plaintiffs' injuries were not antitrust injuries because they "resulted from defendants' misrepresentation, not from harm to competition." Id. Second, focusing on financial products, we held that the plaintiffs had failed to allege any restraint on competition in the market for LIBOR-based financial instruments. Whatever collusion may have been alleged, it affected the non-competitive process of setting LIBOR, rather than the intensely competitive process of dealing in loans and interest-rate derivatives on the open market. See id. at 688-89.

We confirmed this analysis by observing that, "[a]s . . . in Brunswick and ARCO, the harm alleged here could have resulted from normal competitive conduct." Id. at 690. The plaintiffs' injury was consistent with "normal commercial incentives facing defendants," such as bolstering defendants' reputations and paying lower floating interest rates.

1.2. LIBOR II

In LIBOR II, we denied motions to amend the complaints of several plaintiffs whose antitrust claims had been dismissed in LIBOR I. See 962 F. Supp. 2d at 624-28. We denied those motions first on the grounds that the plaintiffs, each represented by experienced counsel, had already had an ample opportunity to offer their strongest possible complaints and had unduly delayed in seeking to bolster their allegations of antitrust injury.

Independently (and with more relevance to the present motions), we determined that the plaintiffs' proposed amendments failed to plead antitrust injury. Relying again on Brunswick and ARCO, we wrote:

Plaintiffs' allegations include new ways of packaging previously known facts, such as arguing that the LIBOR-setting rules themselves give rise to competition, and new theories for how defendants compete, such as that they compete over their creditworthiness, that they compete to offer customers the best interest rate benchmark on financial instruments, or that they compete by 'keeping other banks honest' and reporting any improper conduct by them. However, regardless of the creativity they display, none of plaintiffs' allegations make plausible that there was an arena in which competition occurred, that defendants' conduct harmed such competition, and that plaintiffs suffered injury as a result. . . . [Plaintiffs] have not plausibly alleged that each defendant failed to act in its independent individual self-interest. In other words, even if we grant . . . that they suffered harm as a result of defendants' conduct [] they have not plausibly alleged . . . that the process of competition was

harmed because defendants failed to compete with each other or otherwise interacted in a manner outside the bounds of legitimate competition.

2. Plaintiffs' Federal Antitrust Claims

2.1. The Complaints

Notwithstanding our dismissal of antitrust claims in LIBOR I, plaintiffs in four cases (the "Antitrust Plaintiffs") assert that they have alleged new facts and new legal theories to support federal antitrust claims under the Sherman and Clayton Acts. See Leveridge Letter, Feb. 12, 2015, ECF No. 1018 (declining to request partial judgment on the Antitrust Plaintiffs' federal antitrust claims); Joint Mem. of Law in Opp. to Defs.' Mot. to Dismiss Direct Action Antitrust Claims Based on Prior Rulings ("Pl. Antitrust Mem."), Dec. 8, 2014, ECF No. 883.

The complaints in eighteen other cases included federal antitrust claims for preservation. After the present motion was filed, these plaintiffs requested that partial judgment be entered on these claims under Rule 54(b) of the Federal Rules of Civil Procedure, enabling these plaintiffs to appear in a pending appeal from LIBOR I. We granted that request, the Clerk entered judgment, and these plaintiffs' appeals are pending before the Second Circuit. See Partial J., Feb. 23, 2015, ECF No. 1053. Therefore, the remainder of this discussion addresses only the Antitrust Plaintiffs that have chosen to pursue their antitrust claims in this Court.

2.2. Market Structure

We begin by summarizing the Antitrust Plaintiffs' theory of how defendants violated antitrust laws and how plaintiffs suffered antitrust injuries.

In the Antitrust Plaintiffs' view, two markets relate to LIBOR, each of which is competitive in its own way (or would be competitive in the absence of collusion). First, there is an upstream market in "interest-rate benchmarks," in which LIBOR (trademarked by the BBA as "bbaLIBOR") dominated competition at the times relevant to this case. Second, there is a multi-brand downstream market in "benchmark-based products," such as adjustable-rate loans and interest-rate derivatives. These products incorporate interest-rate benchmarks such as LIBOR to determine the stream of payments between contracting parties. To make an analogy to physical things, interest-rate benchmarks are components in the manufacture of benchmark-based products.

According to the Antitrust Plaintiffs, LIBOR came to dominate the upstream market in indices for two cumulative reasons. First, defendants collectively dominated a substantial portion of the downstream market for benchmark-based products, and thus were in a position to prevent competition from emerging in the upstream market for benchmarks. Second, network effects in the upstream market tend to entrench the position of a product that is favored by a substantial portion of benchmark consumers. The artificial monopoly in the upstream market allowed defendants to degrade LIBOR's quality with impunity. In other words, plaintiffs posit that defendants would have submitted misleading LIBOR bids to no avail if there had been competition in the benchmark market, because benchmark users could have switched to a competing benchmark.

Plaintiffs allege that they each traded in benchmark-based products, and thus place themselves as consumers in the upstream benchmark market. They allege that they were effectively forced to use the LIBOR monopoly in their downstream products and then lost money from the fraud that LIBOR's corruption permitted. Without the combination of LIBOR's market dominance and defendants' suppression of LIBOR, plaintiffs' benchmark-based products would have instead incorporated a reliable benchmark (LIBOR or otherwise).

Plaintiffs principally contend that (1) this way of viewing benchmarks as a relevant upstream market and benchmark-based products as a multi-brand downstream market is the correct framework for considering their allegations of antitrust injury, and (2) a number of new facts support their central allegation that defendants colluded to extinguish competition in the benchmark market.

Eight years after questions of LIBOR manipulation first arose, four years after the first suit, two years after our first examination of conspiracy pleading, and with the benefit of over a dozen consent decrees involving LIBOR and similar benchmarks, plaintiffs have not pleaded the existence of a conspiracy to persistently suppress LIBOR, and have found only sporadic agreements to engage in trader-based manipulation. We do not address plaintiffs' conspiracy theory further in this section, and instead turn directly to the legal issue of antitrust injury.

2.3. Antitrust Injury

Plaintiffs' theory fails even if we assume that: (1) their analysis of the relevant markets is correct; (2) there was a conspiracy to make LIBOR more susceptible to manipulation; and (3) such a conspiracy violated the Sherman Act. Regardless of whether a conspiracy to degrade a dominant benchmark is a per se violation or subject to the "Rule of Reason," the same conclusion follows: plaintiffs' injuries are not of the sort that stem from anti-competitive conduct.

The essence of an antitrust violation is a conspiracy to alter the structure of a market. A conspiracy to commit fraud may resemble an antitrust conspiracy, in that it causes the public to pay higher prices or to receive worse goods, but a conspiracy to commit fraud is not thereby transmuted into a violation of the Sherman Act. It was, and remains, our conclusion that plaintiffs' injuries in this MDL followed immediately from false LIBOR quotes. A conspiracy (if any existed) to submit these false LIBOR quotes was entirely fortuitous, as each bank separately had powerful incentives to manipulate LIBOR. And a conspiracy (again, if any existed) to make LIBOR more susceptible to manipulation was two steps removed from plaintiffs' injuries. It was the manipulation rather than LIBOR's susceptibility to manipulation that proximately caused plaintiffs' injuries. According to either conspiracy theory, plaintiffs' injuries are not "attributable to an anti-competitive aspect of the practice under scrutiny." ARCO, 495 U.S. at 334.

To form an analogy to a more tangible conspiracy, suppose that car manufacturers conspire to sell cars without side airbags. As with the alleged conspiracy to issue a degraded product into the "benchmark market," this hypothetical conspiracy seeks to degrade output in the automobile market. The immediate anticompetitive consequences of such a conspiracy are (1) that any consumer who would have bought a car with side airbags at the fair market price instead receives a car without side airbags, and (2) that airbag manufacturers lose the profits they would have reaped from manufacturing side-airbags. We assume that this antitrust violation would be actionable by the consumers (suing for the reduced value of their new cars), the airbag manufacturers (suing for lost profits), or government agencies.

Extending this hypothetical, posit a driver who is grievously injured in a car accident because, as a result of the manufacturers' conspiracy, his car lacked a side airbag. Does the Clayton Act offer federal jurisdiction over and treble damages for this driver's personal injury? We have assumed that the conspiracy violates federal law and that the conspiracy is a but-for cause of the injury. Nevertheless, this is plainly not an antitrust injury, because physical injury is the kind of injury that commonly and directly follows from a defective product design, regardless of whether an antitrust conspiracy somehow enabled the design defect in a particular case.

Similarly, in the case of LIBOR, it does not ultimately matter whether plaintiffs plead or prove that an anticompetitive conspiracy enabled the panel banks to manipulate LIBOR. Plaintiffs' injuries are the proximate consequence of defendants' allegedly untruthful representations, and plaintiffs may therefore recover for torts that arise from any such misrepresentation.

3. State-Law Antitrust Claims

The Antitrust Plaintiffs in three cases assert antitrust claims pursuant to New York's Donnelly Act, but the parties appear to agree that these claims rise or fall with the federal claims. We concur, because both New York law and federal law require a plaintiff to plead an "antitrust injury." See Gatt Commc'ns, 711 F.3d at 81-82; Continental Guest Servs. Corp. v. Int'l Bus Servs., Inc., 92 A.D.3d 570, 573, 939 N.Y.S.2d 30, 35 (1st Dep't 2012) ("An antitrust injury is an injury 'attributable to an anti-competitive aspect of the practice under scrutiny.'" (quoting ARCO, 495 U.S. at 334)).

Plaintiffs in thirteen of the cases in which we granted partial judgment also raised antitrust claims under state law. We did not order partial judgment to be entered on these claims because the pending Second Circuit appeal is addressed solely to claims arising under federal law. Plaintiffs in these thirteen cases have not attempted to distinguish state law from federal law for at least one obvious reason: each relevant state's antitrust doctrine requires plaintiffs to plead an element analogous to the "antitrust injury" requirement of federal law. See LIBOR I, 935 F. Supp. 2d at 735-36 (citing Flagship Theatres of Palm Desert, LLC v. Century Theatres, Inc., 198 Cal. App. 4th 1366, 1380, 131 Cal. Rptr. 3d 519, 529-30 (2d Dist. 2011), and other cases) (California law); 740 Ill. Comp. Stat. 10/11 (2014); O'Regan v. Arbitration Forums, Inc., 121 F.3d 1060, 1066 (7th Cir. 1997) (citing Illinois ex rel. Burris v. Panhandle E. Pipe Line Co., 935 F.2d 1469, 1479-80 (7th Cir. 1991)) (Illinois law); Holzrichter v. Cty. of Cook, 231 Ill. App. 3d 256, 266, 595 N.E.2d 1237, 1243 (1st Dist. 1992) (citing Brunswick, 429 U.S. at 488); Orr v. Beamon, 77 F. Supp. 2d 1208, 1211 (D. Kan. 1999); O'Brien v. Leegin Creative Leather Prods., Inc., 294 Kan. 318, 334-35, 277 P.3d 1062, 1075 (2012) ("The concept of antitrust injury . . . equates to the Kansas concept of causation, or the requirement that a plaintiff's theory of damages correspond to an economic effect that the statute or case law rule invoked as the basis for liability aims to prevent."); McPeters v. LexisNexis, 11 F. Supp. 3d 789, 796-97 (S.D. Tex. 2014) (citing cases of the Court of Appeals of Texas). Therefore, we now dismiss all state-law antitrust claims.

IX. CONSUMER PROTECTION AND UNFAIR BUSINESS PRACTICES

Plaintiffs in three cases, Maragos, Charles Schwab, and NCUA, have each attempted and failed to state claims pursuant to consumer protection and unfair competition statutes of New York and California. We reject the legal theories of Maragos (New York law) and Schwab (California law) completely. Although the NCUA presents a viable legal theory on behalf of Wescorp Credit Union, the NCUA fails to adequately plead that Wescorp was harmed by the manipulation that the NCUA alleges. 1. New York (Maragos)

Article 22-A of New York's General Business Law, entitled "Consumer Protection from Deceptive Acts and Practices," prohibits "[d]eceptive acts or practices in the conduct of any business, trade or commerce or in the furnishing of any service in [New York]," N.Y. Gen. Bus. Law § 349(a) (McKinney 2012) (the "GBL"), and provides a private right of action for actual or liquidated damages, § 349(h).

Here, Nassau County (suing through plaintiff Maragos) presents a theory that false LIBOR submissions "constituted a dissemination of false information to the public," in violation of the General Business Law. Maragos Am. Compl. ¶ 95. As the beneficiary of interest rate swaps that its finance authority traded with several counterparties, the County seeks relief pursuant to section 349.

"[A]s a threshold matter, plaintiffs claiming the benefit of section 349—whether individuals or entities . . . —must charge conduct of the defendant that is consumer-oriented." Oswego Laborers' Local 214 Pension Fund v. Marine Midland Bank, N.A., 85 N.Y.2d 20, 25, 647 N.E.2d 741, 744 (1995). The test is not whether a particular plaintiff is an individual consumer, for in Oswego itself, a union pension fund was allowed to state a section 349 claim involving a savings account at a bank. Rather, the test is whether the "acts or practices have a broader impact on consumers at large," id., where a "consumer" is "one 'who purchases goods and services for personal, family or household use.'" Exxonmobil Inter-Am., Inc. v. Advanced Info. Eng'g Servs., Inc., 328 F. Supp. 2d 443, 447 (S.D.N.Y. 2004) (quoting Sheth v. N.Y. Life Ins. Co., 273 A.D.2d 72, 709 N.Y.S.2d 74 (1st Dep't 2000)). Business-to-business transactions and private contract disputes unique to the parties do not typically give rise to section 349 claims. See Exxonmobil, 328 F. Supp. 2d at 448; Oswego, 85 N.Y.2d at 25, 647 N.E.2d at 744.

Consistent with these principles, "consumer-oriented" transactions that support GBL claims have involved a savings account at a bank branch (Oswego), homeowners' insurance (Riordan v. Nationwide Mut. Fire Ins. Co., 977 F.2d 47, 51-53 (2d Cir. 1992)); a public stamp auction frequented by novices and the elderly (New York v. Feldman, 210 F. Supp. 2d 294, 301-02 (S.D.N.Y. 2002)); gasoline (In re Methyl Tertiary Butyl Ether ("MTBE") Prods. Liab. Litig., 175 F. Supp. 2d 593 (S.D.N.Y. 2001)); an individual health benefit plan (Wurtz v. Rawlings Co., No. 12-cv-1182 (JFB), 2014 WL 4961422, 2014 U.S. Dist. LEXIS 141896 (E.D.N.Y. Oct. 3, 2014)); a residential home loan, together with a home appliance warranty and maintenance plan (Delgado v. Ocwen Loan Servicing, LLC, No. 13-cv-4427 (NGG), 2014 WL 4773991, 2014 U.S. Dist. LEXIS 135758 (E.D.N.Y. Sept. 23, 2014)); and a home loan modification (Harte v. Ocwen Fin. Corp., No. 13-cv-5410 (MKB), 2014 WL 4677120, 2014 U.S. Dist. LEXIS 132611 (E.D.N.Y. Sept. 19, 2014)).

Non-consumer-oriented transactions that do not support GBL claims have involved "employee dishonesty" insurance (N.Y. Univ. v. Continental Ins. Co., 87 N.Y.2d 308, 662 N.E.2d 763 (1995)); general contracting (Teller v. Bill Hayes, Ltd., 213 A.D.2d 141, 630 N.Y.S.2d 769 (2d Dep't 1995)); a complex truck-tracking computer system (Exxonmobil); advertising in the Yellow Pages (Cruz v. Nynex Info. Res., 263 A.D.2d 285, 703 N.Y.S.2d 103 (1st Dep't 2000)); and a commercial office lease (Circle Click Media LLC v. Regus Mgmt. Grp. LLC, No. 12-4000 (SC), 2013 WL 57861, at *11-13, 2013 U.S. Dist. LEXIS 1604, at *32-36 (N.D. Cal. Jan. 3, 2013)).

Most relevant here, two courts have held that section 349 does not regulate interest rate swaps. See Regions Bank v. SoFHA Real Estate, Inc., No. 2:09-cv-57, 2010 WL 5488471, at *3, 2010 U.S. Dist. LEXIS 138614, at *8-10 (E.D. Tenn. Dec. 3, 2010), aff'd without objection, 2011 WL 13386, 2011 U.S. Dist. LEXIS 890 (E.D. Tenn. Jan. 4, 2011); Merrill Lynch Capital Mkts. AG v. Controladora Comercial Mexicana SAB de C.V., No. 603214/08, 2010 WL 5827550, at *14, 2010 N.Y. Misc. LEXIS 6743, at *29-33 (Sup. Ct. N.Y. Cty. Mar. 16, 2010).

We concur without hesitation. As an instrument of high finance, an interest rate swap is hardly a product that individuals purchase for "personal, family or household use." Swap traders in general are sophisticated financiers, bond issuers, investors, and speculators, who deal with each other at arm's length, can obtain independent financial and legal advice, and often negotiate custom agreements. They are not the sort of unwary "consumers" that the New York Legislature and courts have chosen to protect through Article 22-A of the General Business Law. See Teller v. Bill Hayes, Ltd., 213 A.D.2d 141, 148, 630 N.Y.S.2d 769, 774 (2d Dep't 1995) ("The statute was intended to empower consumers; to even the playing field in their disputes with better funded and superiorly situated fraudulent businesses. It was not intended to supplant an action to recover damages for breach of contract between parties to an arm's length contract."). Following both Regions Bank and Merrill Lynch Capital Markets, we dismiss Nassau County's consumer protection claim.

2. California

California's Unfair Competition Law (UCL) bans "any unlawful, unfair or fraudulent business act or practice." Cal. Bus. & Prof. Code § 17200 (West 2008). Its private right of action for injunctive relief, enacted at section 17204, has been held to provide a right to restitution. See Kasky v. Nike, Inc., 27 Cal. 4th 939, 950, 45 P.3d 243, 249 (2002).

The NCUA (on behalf of one California credit union, Wescorp) presents a theory that defendants breached the UCL by violating the Cartwright Act, California's antitrust statute. The Schwab plaintiffs base their UCL claims on (1) violations of federal and state securities laws, and (2) misrepresentations to purchasers of LIBOR-based instruments.

2.1. Types of Practices Actionable

California, unlike New York, does not require that an improper practice be directed at ordinary individual consumers. The California statute does not mention consumers, but California cases ask whether the practice affects customers and the public generally or a single, specially situated party. The breach of an individual, sui generis contract does not generate a UCL claim. See, e.g., Dollar Tree Stores Inc. v. Toyama Partners LLC, 875 F. Supp. 2d 1058, 1083 (N.D. Cal. 2012); In re Webkins Antitrust Litig., 695 F. Supp. 2d 987, 998-99 (N.D. Cal. 2010). However, some contracts, even between sophisticated parties, can form the basis for a UCL claim, so long as the defendant perpetrates a practice against a number of customers. See Int'l Union of Operating Eng'rs v. Bank of N.Y. Mellon Corp., No. C. 11-3620 (WHA), 2012 WL 476526, 2012 U.S. Dist. LEXIS 18281 (N.D. Cal. Feb. 14, 2012) (permitting UCL claims involving hidden foreign-exchange fees charged to a pension fund when trading foreign securities).

In the securities context, fraud in a particular securities transaction is not a valid UCL predicate. See, e.g., Betz v. Trainer Wortham & Co., 829 F. Supp. 2d 860, 866 (N.D. Cal. 2011) (dismissing UCL claims for mismanagement of an investment account); Deitrich v. Bauer, 76 F. Supp. 2d 312, 351 (S.D.N.Y. 1999) (dismissing UCL claims relating to a penny-stock manipulation); Bowen v. Ziasun Techs., Inc., 116 Cal. App. 4th 777, 11 Cal. Rptr. 3d 522 (4th Dist. 2004) (comparing UCL to Federal Trade Commission Act). Only the general business practices of a company in the securities business may be actionable through the UCL, because general business practices affect the investing public generally. See, e.g., Strigliabotti v. Franklin Res., Inc., No. C. 04-883 (SI), 2005 WL 645529, 2005 U.S. Dist. LEXIS 9625 (N.D. Cal. Mar. 7, 2005) (permitting UCL claims relating to overcharges of an investment management fee); Overstock.com , Inc. v. Gradient Analytics, Inc., 151 Cal. App. 4th 688, 61 Cal. Rptr. 3d 29 (1st Dist. 2007) (permitting a business's UCL claims that its stock price was diminished by a securities analyst's defamatory reports).

Applying this standard, the Schwab Plaintiffs' claims fail. These claims relate to misrepresentations in the sales of securities (especially adjustable-rate bonds). Courts have consistently held that an investor may not state a UCL claim for straightforward securities fraud, and we adhere to their conclusion.

In contrast to Schwab's claims, the NCUA's claim is based on antitrust violations, and defendants do not argue that any "securities exception" bars the NCUA's UCL claims. Assuming for argument that an antitrust conspiracy existed, the harm would have been felt by many participants in the financial markets.

2.2. Legal Standards Applicable to NCUA

We assuming without deciding that the NCUA may predicate a UCL claim upon an antitrust violation, even though it lacks standing to assert a private claim for the remedies that antitrust law makes available directly. See Korea Supply Co. v. Lockheed Martin Corp., 29 Cal. 4th 1134, 1143-44, 63 P.3d 937, 943 (2003) (permitting a private plaintiff to state a UCL claim predicated on a violation of the Foreign Corrupt Practices Act); Wash. Mut. Bank, FA v. Super. Ct., 75 Cal. App. 4th 773, 783, 89 Cal. Rptr. 2d 560, 568 (2d Dist. 1999) (permitting a private plaintiff to state a UCL claim predicated on portions of a federal regulation that did not directly provide a private right of action); but see Feitelson v. Google, Inc., Case No. 14-cv-02007-BLF, ___ F. Supp. 3d ___, ___, 2015 WL 740906, at *11, 2015 U.S. Dist. LEXIS 20778, at *35-36 (N.D. Cal. Feb. 20, 2015); Universal Grading Serv. v. eBay, Inc., No. c-09-2755 (RMW), 2012 WL 70644, at *10, 2012 U.S. Dist. LEXIS, at *29-30 (N.D. Cal. 2012).

We also assume, without deciding, that the New York choice-of-law provision in Wescorp's contract with Barclays does not foreclose a UCL claim. Cf. Wescorp-Barclays ISDA Agreement, Schedule § 4(h), ECF No. 662-5.

2.3. Adequate Pleading

To state a UCL claim, the NCUA must adequately plead that each defendant participated in an antitrust violation, that each defendant's antitrust violation caused Wescorp to be injured, and that each defendant obtained something of value from Wescorp, to which the defendant was not entitled. This last requirement flows from the fact that the UCL permits a plaintiff to recover only restitution, not general civil damages. See Korea Supply, 29 Cal. 4th at 1144, 63 P.3d at 943; Kraus v. Trinity Mgmt. Servs., Inc., 23 Cal. 4th 116, 126-27, 999 P.2d 718, 725 (2000), abrogated on other grounds, Prop. 64 (adopted at general election Nov. 2, 2004).

As discussed supra at 111, the NCUA's complaint lacks substantial allegations of inter-bank conspiracy. For the most part, the NCUA alleges intra-bank misconduct, inter-bank misconduct in non-USD LIBOR, and parallel misconduct.

The one exception is that the NCUA incorporates an FSA Final Notice, stating that "[b]etween February 2006 and October 2007, Barclays Derivatives Traders made at least 63 requests to external traders with the aim that those traders would pass on the requests for EURIBOR and US dollar LIBOR submissions to their banks' submitters." Barclays FSA Final Notice ¶ 89 (mis-cited as ¶ 71(i) at NCUA Am. Compl. ¶ 87). Specifically, on October 26, 2006, Barclays collaborated with another panel bank to depress 3-month USD LIBOR on October 26, 2006, and to inflate 3-month USD Libor on February 28, 2007. See Barclays FSA Final Notice ¶¶ 83, 91.

Even assuming that this is sufficient to plead the existence of antitrust violations by Barclays and its unidentified accomplices on those dates, the NCUA's UCL claim against Barclays fails for failure to plead any injury from a specific instance of trader-based manipulation. Every swap that Wescorp traded with Barclays depended on 1-month LIBOR rather than the 3-month LIBOR that Barclays allegedly manipulated on those dates. See NCUA Am. Compl. App. C at 2. The NCUA adequately alleges pay-fixed 3-month LIBOR swaps with Bank of America on October 26, 2006, but the NCUA may not collect restitution from Barclays for Wescorp's overpayments to Bank of America. Furthermore, although the NCUA adequately alleges that many of Wescorp's pay-fixed swaps were open on October 26, 2006, the NCUA fails to allege that any swap payment was calculated based on the value of LIBOR on that date.

3. Conclusion

Each consumer claim fails for a different reason. Maragos fails to allege a consumer-oriented harm, Schwab fails to allege unfair dealing beyond a particular securities transaction, and the NCUA, suing on behalf of Wescorp Credit Union, fails to allege that Wescorp was injured by conspiratorial conduct. The GBL and UCL claims are therefore dismissed.

X. NEW JERSEY RICO

1. Prior Rulings Regarding Federal RICO (LIBOR I)

In LIBOR I, defendants moved to dismiss the Schwab Plaintiffs' federal RICO claims. Schwab alleged that defendants formed an association-in-fact whose purpose was to profit by setting LIBOR at an unnaturally low rate. To advance this common enterprise, Schwab alleged that defendants committed predicate acts of wire fraud, mail fraud, and bank fraud, each involving false LIBOR submissions to the BBA.

We dismissed Schwab's RICO claims for two independent reasons. First, Schwab's RICO claims were barred by the PSLRA. Under the PSLRA's "RICO Amendment," no RICO claim may rely on a predicate act that "would have been actionable as fraud in the purchase or sale of securities." 18 U.S.C. § 1964(c). Because defendants' statements to Schwab and to the BBA were allegedly intended to defraud purchasers of securities, Schwab's RICO claims were barred under this provision. See LIBOR I, 935 F. Supp. 2d at 727-31. Second, we held that Schwab relied on an impermissibly extraterritorial application of RICO. See id. at 731-34. Our analysis, which focused on the foreign location of the alleged RICO enterprise, is no longer good law now that the Second Circuit has held that RICO "applies extraterritorially if, and only if, liability or guilt could attach to extraterritorial conduct under the relevant RICO predicate." Eur. Cmty. v. RJR Nabisco, Inc., 764 F.3d 129, 136 (2d Cir.), panel reh'g denied, 764 F.3d 149 (2d Cir. 2014), en banc reh'g denied, 783 F.3d 123 (2d Cir. 2015), petition for cert. filed, 83 U.S.L.W. ___ (U.S. July 27, 2015) (No. 15-138). We have no reason to conduct a federal extraterritoriality analysis under the Circuit's new approach because no federal RICO claim is disputed in this motion.

2. New Jersey RICO

Prudential asserts a claim under New Jersey's RICO statute, N.J. Stat. Ann. §§ 2C:41-1 to -6.2 (West 2005) ("New Jersey RICO").

New Jersey's RICO statute provides:

It shall be unlawful for any person employed by or associated with any enterprise engaged in or activities of which affect trade or commerce to conduct or participate, directly or indirectly, in the conduct of the enterprise's affairs through a pattern of racketeering activity or collection of unlawful debt.

New Jersey courts have recognized that, "because . . . New Jersey RICO is modeled upon its federal counterpart, it is appropriate to accept guidance from the federal RICO cases." State v. Cagno, 211 N.J. 488, 508, 49 A.3d 388, 399-400 (2012) (citing federal cases regarding RICO statute of limitations); see also State v. Sparano, 249 N.J. Super. 411, 423, 592 A.2d 608, 614 (App. Div. 1991) (commenting that New Jersey RICO "essentially incorporated federal case law"). Nevertheless, New Jersey RICO is "broader than federal RICO in a number of respects." Horowitz v. Marlton Oncology P.C., 116 F. Supp. 2d 551, 554 n.1 (D.N.J. 1999). For example, some courts have held that only an association with an "ascertainable structure" can be a federal RICO enterprise, but New Jersey has rejected such a requirement. See State v. Ball, 141 N.J. 142, 155-63, 661 A.2d 251, 257-61 (1995). As another example, incidents of racketeering activity must exhibit "continuity" to constitute a federal "pattern of racketeering activity," but need only be "related" to constitute a "pattern" under New Jersey law. See id., 141 N.J. at 163-69, 661 A.2d at 261-65.

3. Prudential's Complaint and the Instant Motion

Like its federal model, New Jersey RICO prohibits persons associated with a so-called "RICO enterprise" from conducting the enterprise's affairs through "RICO predicates"—multiple violations of certain laws associated with organized crime. See Ball, 141 N.J. at 155, 661 A.2d at 257.

Here, Prudential alleges that "[d]efendants' collective association, including as members of the BBA's USD Libor panel, constitutes the racketeering enterprise." Prudential Am. Compl. ¶ 460. The purpose of this enterprise was to "obtain[] pecuniary gain . . . in connection with suppressing Libor and in transacting with such investors as [Prudential]." ¶ 463. The enterprise engaged in "trade or commerce . . . in connection with the sale and purchase of securities in . . . New Jersey." ¶ 464.

The alleged New Jersey RICO predicates consist in violations of the New Jersey Uniform Securities Act (N.J. Stat. Ann. § 49:3-52 (West 2001)), deceptive business practices (§ 2C:21-7(i) (West 2005)), theft by deception (§ 2C:20-4), and falsifying records (§ 2C:21-4(a)). Prudential Am. Compl. ¶¶ 469-99. Each of these statutes can provide the basis for pleading a New Jersey RICO predicate. See § 2C:41-1(a)(1)(n) to (a)(1)(p). To plead these predicates, Prudential alleges two sets of facts. First, that certain defendants, identified in exhibits to Prudential's complaint, made false statements in the course of offering and selling securities to Prudential. Second, that certain defendants (presumably the panel banks) made false statements to the BBA. See generally Prudential Am. Compl. ¶¶ 470-99.

Defendants have moved to dismiss Prudential's racketeering claim on two grounds. First, defendants argue that New Jersey RICO does not apply extraterritorially on the pleaded facts. Second, defendants argue that extraterritorial application would unconstitutionally transgress upon the federal power to regulate interstate and foreign commerce. We hold that some (but not all) of Prudential's allegations refer to conduct within the scope of New Jersey RICO, but that Prudential has failed to plead any of the legally viable predicates.

Furthermore, we are skeptical that Prudential's complaint would survive a motion directed to the "enterprise" element. Plaintiffs have been unable to proffer information to make it plausible that banks worked together to manipulate USD LIBOR. See supra at 111. Prudential's complaint is replete with allegations of intra-bank communications and allegations having to do with interest-rate benchmarks other than USD LIBOR, but offers very little to suggest a joint enterprise to defraud counterparties by manipulating USD LIBOR. See, e.g., Prudential Am. Compl. ¶¶ 131-36 (describing RBS collusion to manipulated Yen LIBOR); ¶¶ 148-152 (describing fines imposed by the European Commission for conspiracies related to Yen LIBOR and Euribor). As Prudential itself points out, defendants "were motivated by the direct desire to line their own pockets by way of their own exposure to interest-rate risk," and "were also motivated to understate their borrowing costs to avoid negative publicity." ¶¶ 154, 159.

4. Extraterritoriality

4.1. Territorial Limits of New Jersey RICO

New Jersey courts presume that New Jersey laws should not be applied extraterritorially. See Archut v. Ross Univ. Sch. of Vet. Medicine, Civ. No. 10-1681 (MLC), 2012 WL 5867148, at *12, 2012 U.S. Dist. LEXIS 164960, at *34-36 (D.N.J. Nov. 19, 2012) (declining to apply New Jersey's Law Against Discrimination against a Kittitian veterinary school); cf. Buccilli v. Timby, Brown & Timby, 283 N.J. Super. 6, 10, 660 A.2d 1261, 1263 (App. Div. 1995) ("New Jersey law regulates conduct in New Jersey, not outside the state."). We agree that some territorial rule must constrain the application of New Jersey RICO.

Mainly in the context of federal RICO, courts have developed two approaches to limiting the territorial scope of a RICO statute. One approach focuses on the scope and location of the RICO predicates committed by the particular defendant. See Eur. Cmty., 764 F.3d at 136; United States v. Chao Fan Xu, 706 F.3d 965, 978 (9th Cir. 2013) ("[I]t is highly unlikely that Congress was unconcerned with the actions of foreign enterprises where those actions violated the laws of this country while the defendants were in this country.") (as amended on denial of reh'g); Hourani v. Mirtchev, 943 F. Supp. 2d 159, 165 (D.D.C. 2013) (internal quotation marks omitted) (declining "to assess the extraterritorially of Plaintiffs' RICO claim by examining the location of the enterprise"), aff'd on other grounds, Nos. 13-7088, -7089, ___ F.3d ___, 2015 WL 4590324, 2015 U.S. App. LEXIS 13342 (D.C. Cir. July 31, 2015); Chevron Corp. v. Donziger, 871 F. Supp. 2d 229, 244-45 (S.D.N.Y. 2012) (applying RICO when some predicate acts were committed in the United States); CGC Holding Co. v. Hutchens, 824 F. Supp. 2d 1193, 1209-10 (D. Colo. 2011) (same); United States v. Philip Morris USA, Inc., 783 F. Supp. 2d 23, 29 (D.D.C. 2011) ("Whether or not a criminal enterprise committed a predicate act with extraterritorial scope . . . there is no evidence that Congress intended to criminalize foreign racketeering activities under RICO.").

The other approach focuses on the location and acts of the RICO enterprise. See LIBOR I, 935 F. Supp. 2d at 732-34, abrogated by Eur. Cmty.; Mitsui O.S.K. Lines, Ltd. v. Seamaster Logistics, Inc., 871 F. Supp. 2d 933, 939 (N.D. Cal. 2012) ("The relevant question is simply whether the enterprise is extraterritorial or not."); In re Le-Nature's, Inc., No. 9-mc-162, 2011 WL 2112533, at *3, 2011 U.S. Dist. LEXIS 56682, at *18, (W.D. Pa. May 26, 2011) (finding no bar "if a complaint avers a domestic enterprise"); Cedeño v. Intech Grp., Inc., 733 F. Supp. 2d 471, 474 (S.D.N.Y. 2010) ("RICO evidences no concern with foreign enterprises."), aff'd, 457 F. App'x 35 (2d Cir. 2012), and abrogated by Eur. Cmty.; State v. Casilla, 362 N.J. Super. 554, 563-66, 829 A.2d 1095, 1100-01 (App. Div. 2003) (reversing New Jersey RICO conviction because jury made no finding as to whether defendant's enterprise affected New Jersey commerce).

While clearly some limiting principle on New Jersey RICO is necessary, none of the leading federal tests fits easily with New Jersey's statute and policy.

The European Community rule, that RICO can be applied to extraterritorial conduct when the statute defining the predicate itself applies outside the United States, would not effectively limit New Jersey RICO, because New Jersey expressly defines many ordinary criminal statutes of other jurisdictions to be New Jersey RICO predicates. See N.J. Stat. Ann. § 2C:41-1(a)(1). Thus, the European Community rule, if literally applied to the New Jersey RICO statute, would fashion practically any serious crime committed anywhere in the world into a New Jersey RICO violation.

The "location of the enterprise" test of LIBOR I and other cases is likewise inappropriate for New Jersey RICO. It is implausible to think that the legislature in New Jersey was unconcerned with the in-state activities of out-of-state enterprises. Indeed, New Jersey courts routinely enter RICO judgments against members of criminal enterprises that are based outside New Jersey, so long as the defendant or the enterprise has a sufficient connection to New Jersey. See, e.g., State v. Taccetta, 301 N.J. Super. 227, 234, 693 A.2d 1229, 1333 (App. Div. 1997) (affirming New Jersey RICO convictions of members of "the New York-based Lucchese family and the Philadelphia-based Bruno-Scarfo family").

One intermediate state-court case, Casilla, coped with the territoriality problem by inventing an element without significant textual basis. Casilla held that New Jersey RICO applies only when the RICO enterprise affects trade or commerce in New Jersey. See 362 N.J. Super. at 563-66, 829 A.2d at 1100-01. In dicta, Casilla even suggested that a New Jersey RICO enterprise must affect purely intrastate trade or commerce. Id., 362 N.J. Super. at 564-65, 829 A.2d at 1101 (distinguishing "trade or commerce in New Jersey" from "interstate trade or commerce"). This test is both too restrictive and too permissive. It is too restrictive because there is no reason to believe that the New Jersey Legislature wished to absolve an out-of-state defendant of RICO liability for racketeering activity perpetrated in New Jersey, even if the main operations of the defendant's enterprise affect out-of-state commerce. The Casilla rule is also too permissive. If an out-of-state enterprise affects New Jersey's internal commerce in some way, then the Casilla test would allow New Jersey RICO to punish practically any crime committed by members of the enterprise anywhere in the world, regardless of whether any connection exists between the crimes and the enterprise's New Jersey operations.

No subsequent case has relied on Casilla to bar a New Jersey RICO claim, although some have paid it lip service. Two federal decisions have held that an enterprise affects trade or commerce in New Jersey when an individual defendant commits predicate acts in New Jersey involving sales of securities. See Prudential Ins. Co. of Am. v. Credit Suisse Sec. (USA) LLC, Civ. No. 12-7242 (KSH), 2013 WL 5467093, at *21, 2103 U.S. Dist. LEXIS 142191, at *62-64 (D.N.J. Sept. 30, 2013); Prudential Ins. Co. of Am. v. Goldman, Sachs & Co., Civ. No. 12-6590 (SDW), 2013 WL 1431680, at *10-11, 2013 U.S. Dist. LEXIS 50788, at *33-34 (D.N.J. Apr. 9, 2013). The court in Prudential v. Credit Suisse specifically rejected defendants' (and, apparently, Casilla's) view that New Jersey RICO is unavailable when the enterprise affects only interstate commerce. A third federal decision favorably cited Casilla in the course of dismissing a New Jersey RICO count, but ultimately relied on the fact that "the alleged pattern of racketeering activity did not occur in New Jersey commerce"—not on whether the defendants' enterprise affected New Jersey commerce. Trans USA Prods., Inc. v. Howard Berger Co., Civ. No. 07-5924 (JAP), 2008 WL 3154753, at *5, 2008 U.S. Dist. LEXIS 61069, at *14-16 (D.N.J. Aug. 4, 2008).

We prefer to express forthrightly our belief that the New Jersey Supreme Court would apply a different rule than that in Casilla. Fortunately, we have a statutory beacon to guide our course.

The New Jersey criminal code (of which New Jersey RICO is a part) holds a person accountable for a crime (1) when "the conduct which is an element of the offense" occurs in New Jersey, (2) when "the result which is . . . an element" occurs in New Jersey, or (3) when "[t]he offense is based on a [New Jersey] statute . . . which expressly prohibits conduct outside [New Jersey], when the conduct bears a reasonable relation to a legitimate interest of [New Jersey] and the actor knows or should know that his conduct is likely to affect that interest." N.J. Stat. Ann. § 2C:1-3(a)(1), (6). For example, the absence of a territorial limitation in New Jersey's homicide statute, see N.J. Stat. Ann. § 2C:11-2(a), does not render all the world's murderers accountable to people of New Jersey, because section 2C:1-3 limits the homicide statute's territorial scope. A murder is properly prosecuted in New Jersey only if the victim's death or the cause of the victim's death occurred in New Jersey. See State v. Denofa, 187 N.J. 24, 45, 898 A.2d 523, 536 (2006).

The general rule of section 2C:1-3 calls for us to examine each of New Jersey RICO's elements in turn. New Jersey RICO applies if any conduct or result element occurs in New Jersey, or if New Jersey RICO protects a legitimate New Jersey interest by expressly prohibiting extraterritorial conduct.

The first element is employment by or association with a RICO enterprise. Employment or association is properly an attendant circumstance rather than a conduct or result element, and so cannot be a basis for territoriality. In any case, Prudential's complaint does not suggest that any defendant associated itself with a RICO enterprise in New Jersey, so the association element does not supply a basis for applying New Jersey RICO. Cf. LIBOR I, 935 F. Supp. 2d at 733-34 (finding the LIBOR-setting enterprise to be located in London, England).

The second element is the commission of multiple related predicates (a "pattern of racketeering activity") in the conduct of an enterprise's affairs. Thus, the second element effectively incorporates the elements of the predicate offense or offenses that the defendant is alleged to have committed. It follows, then, that provision (1) of subsection 2C:1-3(a) permits a New Jersey RICO suit if a conduct element or a result element of any predicate offense occurs in New Jersey.

The second element of the RICO statute also clearly prohibits conduct outside New Jersey, as non-New Jersey offenses can constitute New Jersey RICO predicates. Cf. Eur. Cmty., 764 F.3d at 136 (noting that some federal RICO predicates can only be violated outside the United States, and concluding that Congress intended federal RICO to have some extraterritorial effect). In this context, New Jersey has a "legitimate interest," § 2C1-3(a)(6), in applying its RICO statute to out-of-state conduct that advances a substantial New Jersey-focused aim of a RICO enterprise.

4.2. Constitutionality

Defendants argue that it would be unconstitutional to apply New Jersey RICO to extraterritorial conduct. So long as New Jersey RICO is confined to the bounds we have set for it, we disagree. As applied here, New Jersey RICO simply creates heightened civil and criminal penalties for violations of New Jersey's fraud, theft, and securities laws. To the extent that New Jersey can enforce these predicate offenses constitutionally, there is no reason why New Jersey cannot also apply RICO as well.

Courts have always recognized that states have a legitimate interest in regulating securities transactions that occur, in relevant part, within the state. See, e.g., A.S. Goldmen & Co. v. N.J. Bur. of Sec., 163 F.3d 780, 789 (3d Cir. 1999) (upholding provision of New Jersey Uniform Securities Act that requires securities "offered or sold" in New Jersey to be registered or exempt from registration); Houston v. Seward & Kissel, LLP, No. 07-cv-6305 (HB), 2008 WL 818745, at *5-6, 2008 U.S. Dist. LEXIS 23914, at *18-23 (S.D.N.Y. Mar. 27, 2008) (upholding application of Oregon securities law to transaction between a New York business and an Idaho resident within Oregon); cf. Hall v. Geiger-Jones Co., 242 U.S. 539 (1917) (upholding Ohio law regulating dispositions of securities within Ohio). Likewise, state laws barring theft (such as New Jersey's "theft by deception" statute) "do not violate the Dormant Commerce Clause" even if the stolen property flows through interstate commerce. Flo & Eddie, Inc. v. Sirius XM Radio, Inc., No. 13-cv-5784 (CM), 2014 WL 7178134, at *5, 2014 U.S. Dist. LEXIS 174907, at *15 (S.D.N.Y. Dec. 12, 2014).

Defendants' cases on this subject are off point. In Grand River Enterprises Six Nations, Ltd. v. Pryor, 425 F.3d 158 (2d Cir. 2005), a number of states had banded together to force cigarette manufacturers to pay a certain amount of money depending on the manufacturers' nationwide market share. The aggregate effect of this regulation was thus to increase cigarette prices nationwide, and this nationwide interference rendered the states' agreement unconstitutional. See id. at 171-73. New Jersey's regulations, as we construe them, are distinguishable because they only affect transactions with a significant relation to New Jersey.

In American Libraries Ass'n v. Pataki, 969 F. Supp. 160 (S.D.N.Y. 1997), New York had criminalized the use of a computer to communicate certain sexual material to minors. Id. at 169. As interpreted by the district court, this statute "d[id] not import any restriction that the criminal communication must take place entirely within the State of New York." Id. Thus, the statute impermissibly "export[ed] [New York's] domestic policies," as a California artist wishing to transmit nude art to Oregon could not reliably avoid sending it unwittingly to a New York minor over the Internet. Id. at 174. Here, there is nothing unwitting when a major financial company sending offering memoranda to another financial company in New Jersey, so New Jersey is well within its authority to regulate misrepresentations within those offering memoranda.

Finally, in Backpage.com , LLC v. Hoffman, No. 13-cv-3952, 2013 WL 4502097, at *11-12, 2013 U.S. Dist. LEXIS 119811, at *31-33 (D.N.J. Aug. 20, 2013), the court temporarily restrained enforcement of a New Jersey statute that purported to criminalize "any advertisement for a commercial sex act, which is to take place in [New Jersey] and which includes the depiction of a minor," N.J. Stat. Ann. § 2C:13-10(b)(1) (eff. July 1, 2013). As construed by the district court, this law regulated advertisements outside of New Jersey. Backpage.com, 2013 WL 4502097, at *11, 2013 U.S. Dist. LEXIS 119811, at *32. Thus, like the New York law in American Libraries Ass' n, and unlike the narrowly applied securities regulation in this case, New Jersey's advertisement law impermissibly regulated activities in other states.

4.3. Application

Prudential's complaint alleges New Jersey RICO predicates under four New Jersey statutes: N.J. Stat. Ann. § 49:3-52(b) (the New Jersey Uniform Securities Act); § 2C:21-7(i) (deceptive business practices); § 2C:20-4 (theft by deception); and § 2C:21-4(a) (falsifying records). We consider in turn to determine whether Prudential may rely on each alleged predicate to state a New Jersey RICO claim and (if so) whether Prudential has adequately pleaded each predicate.

4.3.1. New Jersey Uniform Securities Act

The New Jersey Uniform Securities Act makes it "unlawful for any person, in connection with the offer, sale, or purchase of any security, directly or indirectly . . . (b) To make any untrue statement of a material fact." N.J. Stat. Ann. § 49:3-52. Prudential may therefore rely on an "offer, sale, or purchase" in New Jersey or an untrue statement in New Jersey to establish a Uniform Securities Act predicate.

Insofar as Prudential's NJUSA predicates are based on misrepresentations within offering materials that counterparties delivered to Prudential in New Jersey, Prudential's RICO claim survives the territoriality analysis because Prudential's Amended Complaint adequately alleges receiving offering materials in New Jersey, see ¶ 16, making it plausible that defendants intentionally directed those offering materials into New Jersey.

However, the "offering materials" theory fails because there is no plausible connection between individual securities offerings and the alleged RICO enterprise. The purpose of "[d]efendants' collective association, including as members of the BBA's USD Libor panel," was to publish LIBOR. The panel banks did not, by virtue of their membership on the LIBOR panel, have a stake in any particular security offered by any other panel bank. Because the New Jersey RICO statute reaches only those predicate offenses that are committed "in the conduct of the enterprise's affairs," the "offering materials" theory fails.

Insofar as Prudential's NJUSA predicates are based on fraudulent LIBOR submissions (none of which occurred in New Jersey), Prudential must rely on the "offer, sale, or purchase" element to establish a New Jersey nexus. No offer of securities took place in New Jersey, because an offer takes place where the offeror is. A.S. Goldmen & Co. v. N.J. Bur. of Sec., 163 F.3d 780, 787 (3d Cir. 1999). It is theoretically possible, however, that the sale and purchase of Prudential's securities occurred in New Jersey, because a sale takes place where "irrevocable liability [i]s incurred or title was transferred." Absolute Activist Value Master Fund Ltd. v. Ficeto, 677 F.3d 60, 68 (2d Cir. 2012).

Even so, Prudential's allegation that defendants "offered and sold" securities to Prudential in New Jersey, Prudential Am. Compl. ¶¶ 16, 475, is too conclusory to survive a motion to dismiss. It may well be, as the Amended Complaint states, that Prudential maintains its principal place of business in New Jersey, that Prudential made its own investment decisions in New Jersey, that Prudential's decisions to rely on representations took place in New Jersey, that Prudential communicated from within New Jersey with defendants, and that Prudential received LIBOR quotes and offering materials in New Jersey. ¶¶ 9, 16. None of this demonstrates, however, that Prudential irrevocably incurred liability for its LIBOR-based transactions in New Jersey.

Accordingly, Prudential's RICO claim fails insofar as it relies on a NJUSA predicate.

4.3.2. Deceptive Business Practices

Section 2C:21-7 of the New Jersey Statutes provides that a person "commits an offense if in the course of business he . . . (i) Makes a false or misleading written statement for the purpose of promoting the sale of securities, or omits information required by law to be disclosed in written documents relating to securities." This provision contains a single "conduct" element: the making or omission of certain information. Thus, Prudential may legally establish a predicate by pleading that a defendant made a false or misleading written statement in New Jersey.

However, as discussed supra at 135, Prudential's complaint fails to allege the making of a false or misleading statement, and none of the properly alleged omissions are omissions of "information required by law to be disclosed in written documents." Accordingly, Prudential's RICO claim fails insofar as it relies on a predicate of deceptive business practices.

4.3.3. Theft by Deception

Section 2C:20-4 provides that a person "is guilty of theft by deception if he purposely obtains property of another by deception." 2C:20-4. "A person deceives if he purposely: a. Creates or reinforces a false impression . . . ." Id. This provision contains a conduct element (creating or reinforcing a false impression) and a result element (obtaining property of another). Thus, Prudential may legally establish a predicate by pleading that a defendant performed acts in New Jersey that created or reinforced a false impression, or by pleading that defendants obtained property in New Jersey.

The complaint adequately pleads that Prudential's counterparties sent offering materials into New Jersey, which would establish the New Jersey nexus. Nevertheless, Prudential fails to plead that any defendant sent offering materials to Prudential with "specific intent to cheat or defraud," Selective Ins. Co. v. McAllister, 327 N.J. Super. 168, 176, 742 A.2d 1007, 1011 (App. Div. 2000). At most, Prudential's pleadings tend to state general intent—that is, knowledge of LIBOR manipulation, knowledge that offering materials failed to inform Prudential of LIBOR manipulation, awareness that Prudential would trade in reliance on the soundness of LIBOR, and indifference to the Prudential's losses. While this level of intent may support a Uniform Securities Act or civil fraud claim, it is not sufficient to support a theory of theft by deception. Accordingly, Prudential's RICO claim fails insofar as it relies on a predicate of theft by deception.

4.3.4. Falsifying Records

Subsection 2C:21-4(a) provides that a person "commits a crime . . . if he falsifies . . . any writing or record, or utters any writing or record knowing that it contains a false statement or information, with purpose to deceive or injure anyone or to conceal any wrongdoing." This provision contains a single conduct element: the falsification of a record. Thus, Prudential may legally establish a predicate by pleading that a defendant falsified a record in New Jersey.

To establish this predicate, Prudential relies entirely on false LIBOR submissions. Since these false LIBOR submissions did not occur in New Jersey, Prudential may not rely on falsification of records as a New Jersey RICO predicate.

5. Conclusion

Defendants' motion is granted as to all of Prudential's New Jersey RICO claims, as each predicate lacks either the requisite connection to New Jersey territory or the requisite connection to the "conduct of the [LIBOR panel's] affairs," N.J. Stat. Ann. § 2C:41-2(c). In closing, we also repeat our skepticism that the members of the LIBOR panel (or any other collection of defendants) constituted an "enterprise" within the meaning of New Jersey RICO. In light of the sustained claims available to Prudential, we see no reason to transmogrify ordinary fraud, breaches of contract, and violations of securities law into racketeering.

XI. COMMODITIES EXCHANGE ACT

The Amabile plaintiffs maintain an individual suit separate from the Exchange-Based Plaintiffs' class action. Nevertheless, none of their allegations can be distinguished from the Exchange-Based Plaintiffs whose complaints we considered in LIBOR I, II, and III. We adhere to those earlier holdings, which we summarize now.

• "Plaintiffs' claims involve manipulation of the price of domestically traded futures contracts," and so are not impermissibly extraterritorial. LIBOR I, 935 F. Supp. 2d at 697.

• Allegations such as those in the Amabile Plaintiffs' complaint suffice to plead that defendants inflated Eurodollar prices by persistently suppressing LIBOR. LIBOR I, 935 F. Supp. 2d at 718-19; LIBOR III, 27 F. Supp. at 460-61. Plaintiffs need not allege specific dates on which they traded Eurodollar derivatives or specific tenors because LIBOR was allegedly artificially suppressed throughout the suppression period across a broad range of tenors. Thus, all of the Amabile Plaintiffs state CEA claims of persistent suppression, although they must ultimately prove that they bought futures at inflated prices and sold at un-inflated or less inflated prices.

• Likewise, the trader-based allegations of some Amabile Plaintiffs, see supra at 101, suffice to plead that defendants altered Eurodollar prices by manipulating LIBOR, although plaintiffs again must ultimately prove that they settled their position or engaged in an

offsetting trade at a price that was not manipulated. See LIBOR III, 27 F. Supp. 2d at 466.

• Plaintiffs' allegations suffice to plead vicarious liability for manipulation committed by defendants' agents. See LIBOR I, 935 F. Supp. 2d at 721-22.

• Plaintiffs' allegations suffice to plead aiding and abetting of persistent suppression, on the theory that each defendant, by itself suppressing LIBOR, knowingly made it easier for each other defendant to suppress LIBOR. See id. at 722-23. As to trader-based manipulation, plaintiffs' allegations suffice to plead aiding and abetting on the part of Barclays. See id. at 723.

Accordingly, we conclude that each Amabile Plaintiff states CEA claims against each panel-bank defendant, although only some state claims on a trader-based factual theory. See supra at 101.

XII. EQUITABLE RELIEF

Most of the individual complaints recite a boilerplate demand for unspecified injunctive, equitable, and declaratory relief. Defendants submit that plaintiffs are not entitled to such relief because money damages can make plaintiffs whole and because plaintiffs lack standing to seek prospective injunctions. We agree with defendants that prospective relief is unwarranted and retrospective relief largely unnecessary.

1. Prospective Equitable Relief

To have standing to seek an injunction, "a plaintiff must show that he is under threat of suffering 'injury in fact' that is concrete and particularized; the threat must be actual and imminent, not conjectural and hypothetical." Summers v. Earth Island Institute, 555 U.S. 488, 493 (2009). This concrete and particularized injury must be imminent, and thus capable of redress, "at that moment" when the plaintiff filed the complaint. Cty. of Riverside v. McLaughlin, 500 U.S. 44, 51 (1991). Accordingly, if the "objectionable practice ceased altogether before the plaintiff filed his complaint," then the plaintiff may not seek an injunction. Id. (distinguishing from City of Los Angeles v. Lyons, 461 U.S. 95 (1983)).

For the period of the financial crisis, approximately 2007 to 2010, plaintiffs support their persistent suppression claims with ample well-pleaded facts: cross-currency analyses, clustering analyses, and persuasive graphical comparisons to alternative benchmarks. For a similar period, plaintiffs support their trader-based claims with defendants' specific public admissions. In contrast, plaintiffs only supply barebone conclusions that LIBOR manipulation endures. E.g., Pl. Fraud Mem. 49 (citing, with added emphasis, one complaint's allegation that manipulation continued "until at least May of 2010").

Plaintiffs' best argument is that defendants, at least those who remain on the LIBOR panel, retain motives to manipulate LIBOR. It may well be that traders still have a motive to manipulate LIBOR for the benefit of their books, but motive is not enough to make manipulation "actual and imminent" instead of "conjectural and hypothetical." None of the settlement documents or other sources on which plaintiffs rely state that traders continue to manipulate LIBOR for their own profit. Furthermore, the reputational motive for persistent suppression has completely dissipated. The "credit crunch" has eased, and no complaint lodges an allegation that any panel bank is concerned about signaling its inability to borrow.

Significantly, LIBOR is a fundamentally different benchmark now than it was in 2008. Previously, the banks' own trade organization administered LIBOR; now, a division of the Intercontinental Exchange (ICE), the world's largest network of financial exchanges, administers it. Previously, a closed committee of banks implemented LIBOR for the BBA; now, the ICE Benchmark Oversight Committee publishes its minutes publicly, it has adopted a formal Conflict of Interest Policy, and it includes several central banks as formal observers. Previously, no agency regulated LIBOR; now, the United Kingdom's Financial Conduct Authority (FCA) regulates it. It is simply not enough for plaintiffs to say that the new LIBOR might be manipulated as the old one was. Plaintiffs must come forward with well-pleaded facts suggesting that LIBOR manipulation still occurs.

Injunctive relief is especially inappropriate now that the FCA, a foreign regulator, has assumed oversight of LIBOR. See Chad Bray, British Regulator to Oversee 7 Additional Financial Benchmarks, N.Y. Times Dealbook (Dec. 22, 2014), available at http://dealbook.nytimes.com/2014/12/22/british-regulator-to-oversee-7-additional-financial-benchmarks/; Financial Conduct Authority, Business Plan 2014/15 ("We will deliver on our commitment to establish a robust framework of supervision for LIBOR—covering both the submitting banks and the new administrator."), available at http://www.fca.org.uk/static/documents/corporate/business-plan-2014-2015-interactive.pdf. We agree with our colleague's comment in the context of a dispute over control of a foreign company that courts "must use the most drastic remedies with great caution, and must consider issues of international comity." E.ON AG v. Acciona, S.A., 468 F. Supp. 2d 559, 584-85 (S.D.N.Y. 2007). Absent concrete facts showing a real need for the United States' judicial intervention, we will not interfere with the United Kingdom's regulation.

2. Retrospective Equitable Relief

2.1. Constructive Trust

Some plaintiffs seek the imposition of a constructive trust as a remedy for unjust enrichment. To the extent that the Individual Plaintiffs state claims for unjust enrichment, see supra at 185, this is permissible, as a constructive trust is simply a mechanism by which restitution can be made effective. Although it is "[m]ost frequently . . . the existence of a confidential relationship which triggers the equitable considerations leading to the imposition of a constructive trust," Sharp v. Kosmalski, 40 N.Y.2d 119, 121, 351 N.E.2d 721, 723 (1976), the "constructive trust doctrine is not rigidly limited," Simonds v. Simonds, 45 N.Y.2d 233, 241, 380 N.E.2d 189, 194 (1978). "[I]ts applicability is limited only by the inventiveness of men who find new ways to enrich themselves unjustly by grasping what should not belong to them," Latham v. Father Divine, 299 N.Y. 22, 27, 85 N.E.2d 168, 170 (1949). Whether a counterparty unjustly retained money that should have been paid or received money that should not have, equity may impose a constructive trust.

When a defendant inequitably retains property, the plaintiff may seek money damages and constructive trust as alternative remedies, even (in many states) when money damages constitute a sufficient remedy. See Wilson v. Scruggs, 371 F. Supp. 2d 837, 842 (S.D. Miss. 2005) ("In such cases [where a defendant enriches himself by investing funds and retaining a profit], although a legal remedy may adequately redress the plaintiff's loss, the equitable remedy nevertheless remains available to prevent the defendant's unjust enrichment."); Heckmann v. Ahmanson, 168 Cal. App. 3d 119, 134, 214 Cal. Rptr. 177, 187 (2d Dist. 1985) (citing Bacon v. Grosse, 165 Cal. 481, 492, 132 P. 1027, 1032 (1913)) ("A constructive trust is the usual theory upon which a plaintiff recovers wrongfully acquired assets. Only where the constructive trustee has dissipated the fund that would constitute the res of the constructive trust is it proper to award a judgment for money damages."). However, in New York (and perhaps some other states), a constructive trust may not be imposed when money damages are adequate. See Anwar v. Greenwich Ltd., 728 F. Supp. 2d 372, 419-20 (S.D.N.Y. 2010) (explaining that a constructive trust is a remedy to be imposed if money damages are inadequate); Pons v. People's Rep. of China, 666 F. Supp. 2d 406, 415 (S.D.N.Y. 2009) ("The adequacy of monetary damages . . . forecloses [a constructive trust remedy], because 'a constructive trust should not be imposed unless it is demonstrated that a legal remedy is inadequate.'" (quoting Bertoni v. Catucci, 117 A.D.2d 892, 895, 498 N.Y.S.2d 902, 905 (3d Dep't 1986))).

Accordingly, at this stage, we dismiss demands for a constructive trust only where the substantive law of New York applies.

2.2. Rescission

In general, rescission is a remedy for fraud in the inducement of a contract, but only when legal remedies are inadequate. Provided that plaintiffs can prove the extent of LIBOR manipulation, plaintiffs can gain full satisfaction through readily calculable money damages. By seeking rescission, plaintiffs in effect ask to unwind the portion of their investment losses that derived from falling interest rates, unrelated to any manipulation. Where plaintiffs' damages are wholly monetary and simple to measure, see infra at 264, there is no reason to permit an equitable remedy that transgresses beyond the injury to be remedied.

Finally, we mention that a plaintiff must be diligent in pursuing rescission. See Ballow Brasted O'Brien & Rusin P.C. v. Logan, 435 F.3d 235, 239 (2d Cir. 2006). We do not address this issue, which is rarely decided at the pleadings stage, see IOP Cast Iron Holdings, LLC v. J.H. Whitney Capital Partners, No. 14-cv-816 (NRB), ___ F. Supp. 3d ___, ___, 2015 WL 1005961, at *19, 2015 U.S. Dist. LEXIS 29070, at *44-46 (S.D.N.Y. Mar. 3, 2015), but express some doubt that plaintiffs demanded rescission quickly enough to obtain it now.

XIII. DAMAGES

Although not strictly necessary to dispose of the present motions, issues of damages are ever-present whether explicitly or implicitly. For the benefit of all parties going forward, we share our current thinking regarding the framework for evaluating damages.

We proceed from the fundamental principle that a defendant is liable only for the portion of a plaintiff's injuries that are fairly attributable to that defendant and for the enrichment that the defendant unjustly receives. Only rarely may a plaintiff's recovery exceed his injury. Statutory or punitive damages may serve to punish especially reprehensible conduct or to deter misconduct that is difficult to detect. To date, we have not expressed an opinion as to whether punitive damages would be available at all in this MDL and we do not do so now.

1. Joint and Several Liability

Under a variety of circumstances, multiple defendants may be liable for a single injury and a plaintiff may proceed to recover against any or all of them. This election is not an exception to the fundamental principle of single recovery, but an elaboration upon it. When an injury is not clearly attributable to a particular wrongdoer or when an injury is not easily susceptible to division amongst wrongdoers, it is fair to hold all of the wrongdoers jointly liable to the plaintiff without allowing for a total recovery greater than the loss

In the context of LIBOR manipulation, joint and several liability may arise in at least three ways.

First, to the limited extent that plaintiffs have plausibly alleged and can ultimately prove a conspiracy, each defendant may be jointly liable for harm caused by any conspirator, so long as the defendant was aware of the scope of the conspiracy. See Kashi v. Gratsos, 790 F.2d 1050, 1055-56 (2d Cir. 1986); Fidelity Funding of Cal., Inc. v. Reinhold, 79 F. Supp. 2d 110, 124 (E.D.N.Y. 1997).

Second, to the extent that plaintiffs can prove their theory that each panel bank's persistent suppression substantially assisted each other panel bank's persistent suppression, each panel bank may be jointly liable for harm caused by any other panel bank. See Diduck v. Kaszycki & Sons Contractors, Inc., 974 F.2d 270, 284 (2d Cir. 1992) (applying federal law); In re Parmalat Sec. Litig., 377 F. Supp. 2d 390, 413-14 (S.D.N.Y. 2005) (applying Illinois law); Restatement (Second) of Torts § 876(b).

Third, for tort and CEA claims arising out of the panel banks' false submissions, two banks may be jointly liable if their false submissions jointly influenced the final LIBOR fix. As a concrete example, suppose that three panel banks faced true borrowing costs of 2% and thirteen faced true borrowing costs of 3% on a particular day. Because the LIBOR calculation discards the lowest four quotes, none of the thirteen high-interest panelists could unilaterally suppress LIBOR by altering its bid. However, any two of the high-interest panelists could suppress LIBOR by simultaneously submitting artificially low quotes. If two of the high-interest panelists in fact submitted artificially low quotes, then the two would be jointly and severally liable despite the absence of a conspiracy. When the independent acts of two tortfeasors combine to cause injury, both are liable for the entire injury though neither could have caused injury alone. See, e.g., Restatement (Second) of Torts § 433A(2), Illustration 12 (bystander injured by automobile accident that either of two negligent drivers could have avoided); see also Project Hope v. M/V IBN SINA, 250 F.3d 67, 76 (2d Cir. 2001) (adopting as federal common law the Restatement's principle that "[i]f the tortious conduct of each of two or more persons is a legal cause of harm that cannot be apportioned, each is subject to liability for the entire harm, irrespective of whether their conduct is concurring or consecutive"); § 870 cmt. l (commenting that the causation rules for negligence generally apply with equal force to intentional torts).

Conversely, two tortfeasors are jointly and severally liable when each one alone could have caused the injury that the two in fact caused together. See, e.g., Restatement (Second) of Torts § 433A(2), Illustrations 14-15 (harm caused by fire after two defendants spilled oil into a stream); Kingston v. Chi. & N.W. Ry., 191 Wis. 610, 211 N.W. 913 (1927) (defendant liable for causing one of two fires, each of which was sufficient to destroy plaintiff's property). In the LIBOR setting, suppose that five panel banks faced true borrowing costs of 2% and eleven faced true borrowing costs of 3%. Here, any one of the five low-interest panelists could unilaterally inflate LIBOR by inflating its quote, but two of the low-interest panelists could not inflate LIBOR any more than one alone could. So if two of the low-interest panelists independently inflated their quotes by the same amount, then both are jointly and severally liable for the resulting inflation of the published rate.

In contrast to these examples, consider an example of divisible injury and separate liability. Suppose that four panel banks faced borrowing costs of 2%, eight faced borrowing costs of 2.5%, and four faced borrowing costs of 3%. Suppose further that several of the mid-interest panelists suppressed their quotes by 0.25%. Each bank's suppression would have caused LIBOR to fall by 0.03125%, regardless of whether the other mid-interest banks had also suppressed their quotes. Therefore, each one is separately responsible for suppressing LIBOR by 0.03125%.

Two final examples demonstrate the limits of joint and several liability in the unique context of LIBOR. In the penultimate example, suppose that eight panel banks faced true borrowing costs of 2% and eight faced true borrowing costs of 3%, and suppose that all eight of the high-interest banks suppressed their quotes by 0.50%. The combined effect of this manipulation was to suppress LIBOR by 0.25%. Critically, though, regardless of what combination of other high-interest banks had simultaneously suppressed LIBOR, no one of the high-interest banks' suppressed quotes could have reduced LIBOR by more than 0.0625% (one-fourth of the total suppression), no two of the high-interest banks' suppressed quotes could have reduced LIBOR by more than 0.125% (two-fourths or one-half of the total suppression), no three of the high-interest banks' suppressed quotes could have reduced LIBOR by more than 0.1875% (three-fourths of the total suppression), and no four of the high-interest banks' suppressed quotes could have reduced LIBOR by more than 0.25% (the full amount of suppression). Therefore, the eight banks' liability is joint and several only to a limited extent. An injured plaintiff may not collect more than one-fourth of his damages from any one of the eight culpable panel banks, more than one-half of his damages from any two, more than three-fourths of his damages from any three, or more than the whole amount of his damages from any four or more.

Finally, suppose that each of the sixteen panel banks suppressed its quote from 3% to 2%. The combined effect of this manipulation was, obviously, to reduce LIBOR by 1%. Here, any one bank's suppressed quote would have been the but-for cause of a 0.125% reduction in LIBOR (one-eighth of the actual total suppression) if between four and eleven of the other banks had simultaneously submitted suppressed quotes. Likewise, depending on the number of banks that had simultaneously suppressed their quotes, any two banks' suppressed quotes could have affected LIBOR by up to 0.25% (two-eighths of the actual total suppression), any three banks' suppressed quotes could have affected LIBOR by 0.375% (three-eighths of the actual total suppression), and so forth. Therefore, an injured plaintiff may not collect more than one-eighth of his damages from any one panel bank, more than two-eighths of his damages from any two, more than three-eighths of his damages from any three, and so forth.

When joint and several liability is available, defendants may hope to recover from each other for contribution. We are inclined to agree with the First Circuit that Texas Industries, Inc. v. Radcliff Materials, Inc., 451 U.S. 630 (1981), forecloses defendants who are in pari delicto from seeking contribution for liability under the Commodity Exchange Act. See Fleming v. Lind-Waldock & Co., 922 F.2d 20, 27-28 (1st Cir. 1990). Contribution may be available to some extent for common law claims, although we recognize that state law may limit that possibility.

2. Contract-Specific Liability

The potential liability for plaintiffs' contract-specific claims (breach of contract, unjust enrichment, and tortious interference) is more circumscribed than the potential liability for "false data" claims of fraud and CEA violations. A plaintiff may lodge its false data claims against all of the panel banks without regard to contractual privity and may thereby recover for the full extent of the plaintiff's injury even without proving that the panel banks conspired or assisted each other. However, a plaintiff may recover on its contract-specific claims only for the portion of the plaintiff's injury that may be ascribed to the plaintiff's counterparty or its affiliates. Theoretically, if the plaintiff can prove a conspiracy or aiding and abetting theory, then the plaintiff's recovery may extend to the entire amount of its injury. If, however, the plaintiff cannot plead or prove such a theory, then the plaintiff may recover only for the portion of its injury caused by its counterparty's own manipulation or that of the counterparty's affiliate. Furthermore, a plaintiff may not recover twice for the same panel bank's manipulation through an unjust enrichment or contract claim against a counterparty entity and a fraud or tortious interference claim against an affiliated panel bank.

Damages for fraud in the inducement of a contract depend on the nature of the promise. For example, the affirmative misrepresentation alleged by Fannie Mae, see supra at 138, promised at the time of swap execution that Fannie Mae's counterparty was not then in breach of an earlier swap agreement. This promise is specific to the honest business relations of Fannie Mae's own counterparty, and so Fannie Mae can rely on this claim only to recover the portion of its injury that can be ascribed to its counterparty. By contrast, fraudulent omissions alleged by several plaintiffs, see supra at 141, relate to the quality of LIBOR as a whole. If a plaintiff's counterparty fraudulently omitted to inform the plaintiff of LIBOR's susceptibility to manipulation, then the plaintiff may potentially recover from the counterparty for the full amount of damages suffered through LIBOR manipulation in relation to the contract.

We repeat our conclusion, see supra at 261, that rescission is an inappropriate remedy for fraud in the inducement. Given the recent climate of low interest rates, rescission of a pay-fixed swap would allow a plaintiff to harvest an inequitable windfall.

Another potential difference between tort and contract remedies bears mentioning. As discussed above, we recognize a form of joint and several liability when the conjunction of several panel banks' manipulation altered LIBOR. However, it is not clear that the law of contract permits joint and several liability. Compare Cal. & Haw. Sugar Co. v. Sun Ship, Inc., 794 F.2d 1433 (9th Cir. 1986) (applying Pennsylvania law and permitting plaintiff to recover liquidated damages from two counterparties whose separate breaches concurrently caused injury), amended and reh'g denied, 811 F.2d 1264 (9th Cir. 1987), with Bd. of Educ. of Hudson City Sch. Dist. v. Sargent, Webster, Crenshaw & Folley, 71 N.Y.2d 21, 28-30, 517 N.E.2d 1360, 1364-65 (1987) (declining to recognize contribution claim arising from breaches of contract). Therefore, it is possible that "causation" in the context of contract claims is not properly measured according to the principles of joint and several liability described supra at 266-269, but by examining the effect that the breaching counterparty alone had upon published LIBOR. Following this approach, a counterparty whose quote was in the upper quartile could not be liable for any damages relating to suppression, even though such liability is possible in tort, see supra at note 140.

XIV. STATUTE OF LIMITATIONS

News of LIBOR manipulation first garnered widespread public attention in April 2008, when the Wall Street Journal published an article suggesting that "bankers and traders are expressing concerns that the London interbank offered rate, known as Libor, is becoming unreliable." Carrick Mollenkamp, Bankers Cast Doubt on Key Rate amid Crisis, Wall St. J., Apr. 16, 2008. By late May, several more public articles, some quoting inside sources and others analyzing public market data, concluded that many banks were probably underreporting their borrowing costs. By 2010, the heyday of the alleged manipulation had ended, presumably as major banks no longer found it necessary to bolster their reputations or found it safe to bolster their trading profits by manipulating LIBOR.

Nevertheless, no private plaintiff filed a LIBOR-related suit until government reports had been published in early 2011, and many of the Individual Plaintiffs still waited years after that to sue either the panel banks or their own counterparties. Because of this delay, timeliness issues have featured prominently in this MDL. This opinion is no different.

This section is organized as follows: After reviewing and reaffirming our prior rulings, we turn to choice of law questions that arise in the present motion, including how federal "extender statutes" properly apply to the FDIC's and NCUA's claims. We then address when an action accrues—discovery rules and continuing violation doctrines—followed by tolling doctrines—fraudulent concealment and American Pipe tolling. After examining some complaint-specific issues of relation back, we apply our legal holdings to each action.

1. Prior Rulings

1.1. LIBOR I

In LIBOR I, we considered defendants' challenges to the Exchange-Based Plaintiffs' claims pursuant to the Commodities Exchange Act. See 7 U.S.C. § 25(c) (2012) (providing for a limitations period of "two years after the date the cause of action arises"); LIBOR I, 935 F. Supp. 2d at 697-713. Although few of the individual plaintiffs state CEA claims, many of the principles in LIBOR I apply to the individual plaintiffs' other causes of action.

First, we determined that a plaintiff has a duty to inquire into circumstances that suggest a CEA violation, and that the limitations period runs from the moment of "inquiry notice" against a plaintiffs who fails to inquire. We found the language of the CEA statute of limitations to be a close match to that of RICO, which the Second Circuit recently interpreted in Koch v. Christie's International PLC, 699 F.3d 141 (2d Cir. 2012). Both statutes' limitations periods begin when "the cause of action arises" or "accrues." See 7 U.S.C. § 25(c), 28 U.S.C. § 1658(a). Courts infer a limited "discovery rule" for both statutes. "[W]hen the circumstances would suggest to an investor of ordinary intelligence the probability that she has been defrauded," the investor has a duty to inquire. LIBOR I, 935 F. Supp. 2d at 699 (quoting Koch, 699 F.3d at 151). If the investor makes some inquiry, then knowledge is imputed when a reasonably diligent investor would have discovered the fraud. See id. If, however, the investor makes no inquiry, then knowledge is imputed from the date when the duty arose. See id.

Inquiry notice can arise long before a plaintiff possesses enough facts to adequately plead every element of a claim, and therefore the statute of limitations can begin to run before a plaintiff possesses such facts. This distinguishes the CEA from the Securities Exchange Act, whose statute of limitations runs upon "the discovery of the facts constituting the violation." 28 U.S.C. § 1658(b)(1); see Merck & Co. v. Reynolds, 559 U.S. 633 (2010).

Next, we examined the record of public articles and reports regarding LIBOR that were published between April and May 2008. These sources reported that:

(1) since August 2007, LIBOR had diverged from benchmarks with which it should have been correlated, (2) independent experts had confirmed this comparative methodology and concluded that LIBOR was too low, (3) the BBA had accelerated its review of the LIBOR submissions process and publicly declared that a bank submitting false rates would be disqualified from the LIBOR panel, (4) LIBOR quotes jumped abnormally on the day following the BBA's announcement, and (5) market actors had begun to shift away from LIBOR-based instruments toward instruments based on alternative benchmarks because of their distrust of recent LIBOR fixes.

Next, we addressed plaintiffs' argument that the banks' (and the BBA's) consistent denials of LIBOR manipulation negated their inquiry notice. Reassurances serve to dissipate a plaintiff's duty of inquiry "only if an investor of ordinary intelligence would reasonably rely on the statements to allay the investor's concern." Id. at 705 (quoting LC Capital Partners, LP v. Frontier Ins. Grp., Inc., 318 F.3d 148, 155 (2d Cir. 2003)). We held that a person of ordinary intelligence would have understood the banks' and the BBA's strong incentives to deny manipulating LIBOR. Plaintiffs should have regarded such assurances cautiously, if at all, and should not have forgone a reasonable investigation of public data on the basis of such self-serving statements. Id. at 705.

Finally, we rejected plaintiffs' argument that the statute of limitations should be tolled because of defendants' fraudulent concealment of their conduct. This argument failed for two reasons. First, plaintiffs did not "remain[] unaware of [a] violation during the limitations period" because plaintiffs were on sufficient notice by May 29, 2008. Id. at 710 (quoting In re Natural Gas Commodity Litig., 337 F. Supp. 2d 498, 513 (S.D.N.Y. 2004)). Second, and perhaps more fundamentally, defendants' LIBOR manipulation was not "inherently self-concealing" as plaintiffs proposed because each individual bank's LIBOR submission was published every day. Id. at 710-11.

To apply these legal conclusions to the facts of LIBOR, we divided the Exchange-Based Plaintiffs' injuries into three time periods. "Period One" ran from the beginning of the financial crisis in August 2007 (when banks first had a major reputational incentive to underreport their borrowing costs) to May 29, 2008, when plaintiffs were placed on inquiry notice. "Period Two" ran from May 29, 2008, to April 15, 2009 (two years before the Exchange-Based Plaintiffs filed their complaint). "Period Three" ran from April 15, 2009, until the end of the Exchange-Based Plaintiffs' class period.

We continue to use this terminology in a slightly more general way. "Period One" starts on August 9, 2007. "Period Two" starts on May 29, 2008, the date by which news articles had established the strong possibility of LIBOR manipulation. Depending on the context, "Period Three" refers to claims that are timely without application of a discovery rule. We also denominate the period before August 9, 2007 as "Period Zero."

In our conclusion, we held that Period One claims were untimely because of inquiry notice in May 2008. Period Three claims (which defendants had not challenged on limitations grounds) were timely regardless of inquiry notice, because the complaint was filed within two years of those claims. We declined to dismiss Period Two claims based on the information available to us because plaintiffs who purchased LIBOR-based instruments after the articles of April and May 2008 "may not have had reason to follow LIBOR-related news." Id. at 92.

1.2. LIBOR III

1.2.1. Period Two

We returned to the Period Two claims in LIBOR III, dividing the Period Two plaintiffs into two groups: Period Two plaintiffs who had traded Eurodollar futures during Period One and plaintiffs who had first traded during Period Two.

Plaintiffs in the first group were, as we held in LIBOR I, on inquiry notice at the end of Period One, and we considered it "nonsensical to assume that the[ir] minds . . . were wiped clean" before they traded again during Period Two. 27 F. Supp. 3d at 473. Thus, plaintiffs remained on inquiry notice as to their Period Two trading unless their duty to inquire somehow dissipated.

To test for dissipation, we examined the significance of the disclosed problems, the recurrent nature of the problems, and the weight of the reassurances. Each of these factors weighed against holding that the inquiry notice had dissipated. Plaintiffs themselves pleaded (and it is beyond cavil) that LIBOR manipulation was extremely significant, articles published during Period Two suggested that the problem was recurrent, and the banks' own reassurances "rang hollow in light of the BBA's continued failure to implement meaningful changes." Id. at 475.

For many of the same reasons, we held that plaintiffs who had first traded during Period Two were on inquiry notice as well. "When . . . information about probable artificiality during Period 1 is combined with the chorus of articles discussing the BBA's general inaction during Period 2, the logical conclusion for an ordinary Period 2 purchaser was that LIBOR remaining subject to manipulation." Id. at 476.

To avoid inquiry notice during Period Two, plaintiffs proffered three articles, none of which dissuaded us from these conclusions. Two articles acknowledged that market participants widely suspected that LIBOR was artificial. One other article bore the title "Recent Concerns Regarding LIBOR's Credibility." Set against a "drumbeat of suggestions that LIBOR was artificial," these articles "would have been insufficient to change the view of an ordinary investor that LIBOR was probably being set at artificial levels." Id. at 476-77.

We therefore concluded that Period Two CEA claims were time-barred.

1.2.2. Period One

We also briefly reconsidered our holdings regarding Period One claims, even though the Exchange-Based Plaintiffs had not filed a motion for reconsideration of LIBOR I. Plaintiffs' only significant new argument was that the news articles spoke only of LIBOR and not of Eurodollar futures, so that futures traders were not placed on inquiry notice. LIBOR III, 27 F. Supp. 3d at 472-73. We rejected this argument because Eurodollar futures are explicitly tied to LIBOR, a fact that formed the "centerpiece" of plaintiffs' merits argument.

1.2.3. Class-Action Tolling

The exchange-based plaintiffs in Metzler first named Société Générale as a defendant in their second amended complaint, filed in advance of LIBOR III. Société Générale moved to dismiss for untimeliness, whereupon plaintiffs argued that their claims against Société Générale had been tolled during the pendency of Laydon v. Credit Suisse Group AG, a separate exchange-based class action that had named Société Générale as a defendant.

We agreed with plaintiffs that the pendency of Laydon tolled their (federal) claims. See LIBOR III, 27 F. Supp. 3d at 485-86 (citing Am. Pipe & Constr. Co. v. Utah, 414 U.S. 538 (1974)). However, the limitations period was tolled only for those plaintiffs who were members of the Laydon class, and only during the period from Laydon's filing date until the date when a consolidated complaint in Metzler supplanted the Laydon complaint. This period of time, approximately one year, was not long enough to make plaintiffs' complaint timely. Id. at 486.

1.3. Renewed Challenge to Prior Holdings

The Amabile Plaintiffs, who attempt to state CEA claims, argue that we should set aside our conclusion that the articles published by May 29, 2008, "would [have] suggest[ed] to an investor of ordinary intelligence the probability that she has been defrauded." LIBOR I, 935 F. Supp. 2d at 698 (quoting Koch, 699 F.3d at 151). Plaintiffs rely on a collection of declarations from traders and other market participants who now attest that their personal failures to respond to LIBOR news were reasonable.

The Amabile Plaintiffs' declarations offer a variety of bad reasons to stop investigating a potential injury, many of which we have considered in our previous opinions: faith in the Chicago Mercantile Exchange; an assumption that the banks were adversely exposed to a decrease in LIBOR because of traditional lending; a belief that no bank would submit false information to the BBA because doing so would risk a run on the bank; a belief that collusion between banks was "not fathomable," the "irate" denials of a BBA representative in discussions with the CME, promises by the BBA that it would scrutinize outliers, the presence of LIBOR banks on a BBA oversight committee, and the similarity of LIBOR to an alternative bank-reported index.

We will not beat this dead horse any further. The law demands that an investor conduct a reasonable inquiry of suspicious circumstances, and does not indulge wishful thinking. We adhere to our prior holdings and hold that the limitations period began to run on the Amabile Plaintiffs' CEA claims on May 29, 2008, at the latest.

2. Choice of Law

2.1. General Principles

A federal court sitting in diversity normally applies the choice-of-law rules of the state in which the court sits, including the principles that select a statute of limitations to apply to common-law claims. See Klaxon v. Stentor Elec. Mfg. Co., 313 U.S. 487 (1941). However, in the MDL cases that were transferred here for pre-trial purposes, we stand in the shoes of each transferor court, and therefore apply the choice-of-law rules of the state where each separate case was filed. See, e.g., In re Mirena IUD Prods. Liab. Litig., 29 F. Supp. 3d 345, 350 (S.D.N.Y. 2014).

2.2. Application

The parties largely agree on which statute of limitations applies to each case. By and large, the limitations law of each case's forum state (California, Kansas, New Jersey, New York, Ohio, Pennsylvania, Texas, and Virginia) applies. We mention only the exceptions.

The most significant exceptions apply to certain plaintiffs who filed in Kansas or New York. Courts in Kansas and New York generally apply their own state's statute of limitations. However, when a cause of action accrues outside the forum state in favor of an out-of-state plaintiff, then the plaintiff's claim must be timely under both the forum state's limitations law and the law of the state where the action accrued. See Kan. Stat. Ann. § 60-516 (West 2008); N.Y. C.P.L.R. 202 (Consol. 2008). All relevant parties agree that these rules affect the plaintiffs in Darby (New York and Pennsylvania law for plaintiff Darby, and New York and Cayman Islands law for plaintiff Capital Ventures), FDIC (New York law and the law of each failed bank's home state), NCUA (Kansas law and the law of each failed credit union's home state), and both Principal Cases (New York and Iowa law).

The parties dispute whether New York's borrowing rule applies to certain claims brought by Fannie Mae, whose principal place of business is in the District of Columbia. While there is no doubt that Fannie Mae is a non-resident of New York, the parties dispute whether Fannie Mae's claims accrued in New York.

"When an alleged injury is purely economic, the place of injury usually is where the plaintiff resides and sustains the economic impact of the loss." Global Fin. Corp. v. Triarc Corp., 93 N.Y.2d 525, 529, 715 N.E.2d 482, 485 (1999). In some circumstances, however, a plaintiff may overcome this presumption by showing that his injury was particularly connected to a state other than his residence. For example, when an out-of-state tourist loses money playing three-card monte in Times Square, we have no doubt that the fraud accrues in New York.

Courts have sometimes held that a fraud claim accrues where a plaintiff suffers injury to a discrete financial base that he has established away from home. See Lang v. Paine, Webber, Jackson & Curtis, Inc., 582 F. Supp. 1421, 1426 (S.D.N.Y. 1984); cf. In re Countrywide Fin. Corp. Mortg.-backed Sec. Litig., 834 F. Supp. 2d 949, 959 n.10 (C.D. Cal. 2012) (observing that the Court of Appeals cited Lang favorably in Global Financial). The usual rule, however, that an economic injury does not accrue in New York simply because a plaintiff holds money in a New York bank account or conducts some of its financial operations in New York. See, e.g., Deutsche Zentral Genossenschaftsbank AG v. HSBC N. Am. Holdings, Inc., No. 12-cv-4025 (AT), 2013 WL 6667601, at *5-7, 2013 U.S. Dist. LEXIS 178462, at *13-20 (S.D.N.Y. Dec. 17, 2013); Robb Evans & Assocs. LLC v. Sun Am. Life Ins., No. 10-cv-5999 (GBD), 2013 WL 123727, at *1, 2013 U.S. Dist. LEXIS 4683, at *4-5 (S.D.N.Y. Jan. 8, 2013); Countrywide, 834 F. Supp. 2d at 959-60. Because many institutions have some financial connection to New York, an overbroad "financial base" exception would swallow the rule that economic injury attaches at the place of residence.

Fannie Mae relies on the fact that many of its swap agreements required Fannie Mae to post collateral in New York. See Fannie Mae Am. Compl. ¶ 13. The fact that some money flowed in and out of New York accounts is not nearly enough to support a conclusion that Fannie Mae suffered a loss in New York. This is a usual case in which we will apply the usual rule: Fannie Mae is located in the District of Columbia; Fannie Mae allegedly lost money to fraud; therefore, Fannie Mae allegedly lost money to fraud in the District of Columbia.

3. Federal Extender Statutes

When the FDIC takes over a bank or the NCUA a credit union, federal law automatically extends the statute of limitations on any cause of action held by the bank or credit union. See 12 U.S.C. § 1787(b)(14) (2012) (NCUA); § 1821(d)(14) (FDIC). In this regard, the parties disagree on four points:

• whether a three-year or six-year extension applies to plaintiffs' claims for unjust enrichment;

• whether the NCUA's extension runs from the date of its appointment as a credit union's liquidating agent, if the NCUA already managed the credit union as a conservator;

• whether the length of the extended limitations period is the fixed length provided by federal law, or the greater of the federal fixed length and the applicable state law; and

• how, if at all, the doctrine of American Pipe applies to the extender statutes. (We discuss this issue within our discussion of American Pipe. See infra at 353.)

3.1. Text and History

Until 1989, both the FDIC and the NCUA were subject to the extender statute that generally applies to federal agencies, 28 U.S.C. § 2415. See, e.g., FDIC v. Bank One, Waukesha, 881 F.2d 390, 393 (7th Cir. 1989) (interpreting section 2415 in a case brought by the FDIC). This statute provided (and still provides, as to most federal agencies) a minimum limitations period that runs, in most cases, from the accrual date of an action.

During the savings and loan (S&L) crisis of the late 1980s, Congress passed the Financial Institutions Reform, Recovery, and Enforcement Act of 1989 (FIRREA), Pub. L. No. 101-73, 103 Stat. 183. FIRREA's main purpose was to wind down insolvent S&Ls through the newly-established Resolution Trust Corporation (RTC). Additionally, FIRREA established more liberal extender statutes for the FDIC, NCUA, and other similar entities (such as the RTC), running from the later of the date when the cause of action accrued or the date when the agency was appointed. The rationale of this provision is obvious: Congress did not want one of its federal resolution agencies to lose a claim simply because a defunct financial institution slept on its rights.

FIRREA's original text largely survives in the NCUA's extender statute:

(A) Notwithstanding any provision of any contract, the applicable statute of limitations with regard to any action brought by the Board [i.e., the NCUA] as conservator or liquidating agent shall be—

(i) in the case of any contract claim, the longer of—

(I) the 6-year period beginning on the date the claim accrues; or
(II) the period applicable under State law; and

(ii) in the case of any tort claim, the longer of—

(I) the 3-year period beginning on the date the claim accrues; or
(II) the period applicable under State law.

(B) For purposes of subparagraph (A), the date on which the statute of limitations begins to run on any claim described in such subparagraph shall be the later of—

(i) the date of the appointment of the Board [i.e., the NCUA] as conservator or liquidating agent; or
(ii) the date on which the cause of action accrues.

The FDIC's extender statute (which also applied to the RTC, see 12 U.S.C. § 1441a(b)(4)(A) (2006), repealed, Dodd-Frank Wall St. Reform & Consumer Prot. Act, Pub. L. No. 111-203, § 364(b), 2010 U.S.C.C.A.N. (124 Stat.) 1555) differs in three ways from the NCUA's.

The first minor difference is that the FDIC's statute naturally refers to the "Corporation as conservator or receiver" instead of the "Board as conservator or liquidating agent." This difference is immaterial, because a bank receivership is similar to a credit union liquidation.

The second minor difference is that the FDIC extender statute excludes tort claims subject to 12 U.S.C. § 1441a(b)(14) (2006). This cross-reference is no longer meaningful because section 1441a has been repealed. Formerly, this exclusion referred to certain tort claims brought by the RTC that were, for a time, treated even more permissively by the RTC's own organic statute.

The major difference is that an extra paragraph in the FDIC's extender statute "revives" some claims that expired within five years before the FDIC's appointment:

(i) In general— In the case of any tort claim described in clause (ii) for which the statute of limitations applicable under State law with respect to such claim has expired not more than 5 years before the appointment of the Corporation [i.e., the FDIC] as conservator or receiver, the Corporation may bring an action as conservator or receiver on such claim without regard to the expiration of the statute of limitations applicable under State law.
(ii) Claims described— A tort claim referred to in clause (i) is a claim arising from fraud, intentional misconduct resulting in unjust enrichment, or intentional misconduct resulting in substantial loss to the institution."

This provision serves to revive any of the FDIC's claims for fraud, tortious interference, or unjust enrichment that expired before the FDIC's appointment. This provision does not revive any expired claims for negligent misrepresentation, because negligent misrepresentation is neither fraud nor intentional misconduct. Cf. FDIC v. Henderson, 61 F.3d 421, 424-25 (5th Cir. 1995) (gross negligence and breach of fiduciary duty); W Holding Co. v. AIG Ins. Co., Civ. No. 11-2271 (GAG), 2014 WL 3699383, at *4, 2014 U.S. Dist. LEXIS 102080, at *19-20 (D.P.R. July 24, 2014) (gross negligence).

With this text and history in mind, we turn to the parties' disagreements.

3.2. Classification of Unjust Enrichment

Both extender statutes provide minimum limitations periods of three years for "tort" claims and six years for "contract" claims. The parties appear to agree that causes of action for fraud, negligent misrepresentation, tortious interference, and violations of the California Unfair Competition Law are "tort" claims, while causes of action for breach of contract, including breach of the implied covenant of good faith and fair dealing, are "contract" claims. However, the parties disagree how to characterize plaintiffs' unjust enrichment claims as "tort" or "contract."

In urging us to classify unjust enrichment as "contract," plaintiffs largely rely on cases interpreting 28 U.S.C. § 2415, which generally prescribes the statute of limitations for actions brought by the United States and its agencies (though not for the FDIC or the NCUA). In relevant part, section 2415 provides that "every action for money damages . . . which is founded upon any contract express or implied in law or fact, shall be barred unless the complaint is filed within six years . . .," § 2415(a) (emphasis added), while "every action for money damages . . . which is founded upon a tort shall be barred unless the complaint is filed within three years . . .," § 2415(b) (emphasis added).

The majority of courts have held that any unjust enrichment action, including one based on a tort theory, "lands in the 'contract' cubbyhole," for purposes of section 2415. FDIC v. Bank One, Waukesha, 881 F.2d 390, 393 (7th Cir. 1989); accord United States v. Moore, 968 F.2d 1099, 1100-01 (11th Cir. 1992); United States v. P/B STCO 213, ON 527 979, 756 F.2d 364, 374-77 (5th Cir. 1985); United States v. Neidorf, 522 F.2d 916, 919 (9th Cir. 1975); contra Blusal Meats, Inc. v. United States, 638 F. Supp. 824, 831-32 (S.D.N.Y. 1986); United States v. Vicon Constr. Co., 575 F. Supp. 1578, 1579-80 (S.D.N.Y. 1983). Courts following the majority rule often apply historical reasoning "that the common law allowed the victim of a tort to elect the quasi-contract claim with its longer period of limitations." Bank One, 881 F.2d at 393; accord Neidorf, 522 F.2d at 919. Because Congress intended "to put the United States as plaintiff on the same footing as private litigants," Bank One, 881 F.2d at 393 (citing letter to Congress from the Attorney General), the same rule should apply to section 2415. Additionally, subsection 2415(a) explicitly refers to "any contract . . . implied in law," a phrase that is traditionally understood to encompass all actions for restitution. See Neidorf, 522 F.2d at 919.

At least one court has, without significant discussion, applied the section 2415 cases to the extender statutes of FIRREA. See FDIC v. Wabick, 214 F. Supp. 2d 864, 872 n.15 (N.D. Ill. 2002). We conclude, however, that the extender statutes' text, history, and context point to a different result.

Most importantly, neither extender statute includes the phrase "contract . . . implied in law" in its "contract" provision. Because Congress was drafting these extenders, at least in part, to displace section 2415, the choice to omit "implied in law" contracts from the contract provisions suggests that Congress did not mean to extend the six-year limitations period to all "implied in law" contracts.

Two amendments to FIRREA, passed in the early 1990s, also demonstrate that Congress understood tort-based restitution to fall on the tort side. Both the FDIC revival provision (passed in 1994) and the since-repealed RTC revival provision (passed in 1993) refer to "intentional misconduct resulting in unjust enrichment" as a variety of "tort."

Finally, there is no good reason to insist on an archaic bright-line rule when the text points elsewhere. Even Bank One recognized that, "[o]n first principles . . . the period of limitations actions should track that of the wrong." 881 F.2d at 392. "[B]ecause recovering 'unjust enrichment' is simply a way to recoup the loss caused by the tort, it seems strange to think that a different caption on the pleadings, a demand against a person farther removed from the wrong, means a longer period of limitations." Id. The present case is an excellent example of such perversity. Following plaintiffs' theory, a direct fraud claim against a panelist entity might be barred after three years, while a restitution claim against an affiliated counterparty would not.

We therefore conclude that, when an unjust enrichment action operates as a remedy for a tort, the extender statutes' three-year "tort" periods apply. This is a double-edged sword for defendants, as the FDIC's revival provision applies only to tort claims. Thus, if any of the FDIC's claims expired under state law before the FDIC was appointed as receiver, those claims are nevertheless timely so long as the FDIC filed suit within three years of its receivership.

3.3. Successive Appointments

The NCUA extender statute operates from the "date of the appointment of [the NCUA] as conservator or liquidating agent." It appears that the NCUA often acts as conservator for some period of time, and then (once the attempted conservation fails) acts as the same credit union's liquidating agent. Indeed, the NCUA received successive appointments as conservator and as liquidating agent of each of the failed credit unions in this case. Defendants argue that, when this occurs, the extended limitations period runs from the first appointment, and that the second appointment has no effect on the statute of limitations. The NCUA argues that its limitations period begins anew with each separate appointment, so that the time in this case should be measured from the appointments as liquidating agent.

Some context regarding conservators and liquidating agents may be useful before plunging into the details of FIRREA. The NCUA has the power to place a credit union into either conservatorship or involuntary liquidation, which are to some extent analogous to Chapter 11 and Chapter 7 bankruptcies. Conservatorship can result in the credit union emerging as a going concern in the controls of its members, merging with another credit union, or liquidating. Involuntary liquidation results in the end of the credit union. The NCUA generally may appoint another entity (such as a state regulator) as conservator or liquidating agent, but must appoint itself to liquidate a bankrupt or insolvent federal credit union. See 12 U.S.C. §§ 1786(h), 1787(a)(1)(A), (3).

Defendants rely on a case from the District of Kansas reasoning that the NCUA "satisfied the first possible date when it was appointed as conservator . . .; thus that date must be used in applying the Extender Statute's three-year limitations period." NCUA v. Credit Suisse Securities (USA), LLC, 939 F. Supp. 2d 1113, 1125 (D. Kan. 2013), abrogated on other grounds, NCUA v. Barclays Capital Inc., 785 F.3d 387, 392-95 (10th Cir. 2015).

The NCUA cites appellate cases dealing with FIRREA's provision allowing an appointed "conservator or receiver" to repudiate certain contracts "within a reasonable period following such appointment." 12 U.S.C. § 1821(e)(2). Two circuit courts (including ours) have held that a receiver may repudiate contracts after its appointment as receiver, even if the receiver had previously been appointed as conservator of a financial institution. See 1185 Ave. of the Americas Assocs. v. RTC, 22 F.3d 494, 497-98 (2d Cir. 1994); RTC v. CedarMinn Bldg., L.P., 956 F.2d 1446, 1450-55 (8th Cir. 1992). Both courts emphasized the distinct roles of conservators (maintaining a going concern) and receivers (liquidating a failed concern), and the Eighth Circuit pointed out that Congress carefully chose which rights could be exercised by a conservator, a receiver, or both. See 1185 Ave. of the Americas, 22 F.3d at 497; CedarMinn, 956 F.2d at 1451, 1453.

The "repudiation" cases are persuasive. Congress defined distinct roles for the NCUA as conservator and the NCUA as liquidating agent, and it is the NCUA's appointment as liquidating agent that now defines its duties and privileges under federal law. Just as a prudent liquidating agent (or receiver) may wish to reject more contracts than a prudent conservator would, a liquidating agent may also wish to file suits that a conservator would not have filed. It would be unreasonable to hold the NCUA in its capacity as liquidating agent to decisions made under different circumstances in its capacity as conservator.

Furthermore, the NCUA need not appoint itself as conservator of a credit union, and nothing in FIRREA or the Federal Credit Union Act suggests that Congress intended to make the NCUA's liquidation rights contingent on appointing some other entity as conservator (or leaving the credit union in the hands of its management until the last possible moment). The NCUA has considerable expertise in conserving credit unions and we do not think that Congress, in creating a generous limitations rule, intended to discourage the NCUA from choosing to appoint itself.

3.4. Accrual Date

Finally, the parties disagree as to when each part of the extended statute of limitations begins. In both extender statutes, both the "contract" and "tort" branches of subparagraph (A) take the same form. The statute of limitations is defined as "the longer of" (I) a fixed-length period "beginning on the date the claim accrues" or (II) "the period applicable under State law." Subparagraph (B) then sets the "date on which the statute of limitations begins to run" as the later of the accrual date or the appointment date, "[f]or purposes of subparagraph (A)." Defendants appear to believe that the start date of subparagraph (B) applies only to provision (I) of subparagraph (A). Plaintiffs appear to believe that the start date of subparagraph (B) applies to both provisions (I) and (II) of subparagraph (A).

This dispute makes a difference when the state-law limitations period (provision (II)) is longer than the fixed-length period (provision (I)) and the appointment date is significantly later than the accrual date. In such a case, defendants' reading means that the agencies only benefit from the fixed-length period following the appointment date, while plaintiffs' reading allows the agencies to benefit from the longer state-law period following the appointment date.

Many cases favor plaintiffs' reading. See, e.g., FDIC v. Cameron, 986 F. Supp. 2d 1337, 1343-44 (N.D. Ga. 2013) (compiling cases); RTC v. Fiala, 870 F. Supp. 962, 975-76 (E.D. Mo. 1994); RTC v. S & K Chevrolet, 868 F. Supp. 1047, 1055-56 (C.D. Ill. 1994); see also FDIC v. McSweeney, 976 F.2d 532, 536 (9th Cir. 1992) (implementing plaintiffs' analysis without discussion); FDIC v. Shea & Gould, No. 95-cv-5491 (JFK), 1997 WL 401822, at *7-8, 1997 U.S. Dist. LEXIS 10263, at *18-21 (S.D.N.Y. July 17, 1997) (same). We agree that subparagraph (B), by its own terms, provides the starting point for the limitations period defined in subparagraph (A) to be the greater of the two time periods. If anything, it might plausibly be argued (though defendants do not so argue) that subparagraph (B) applies to provision (II) only, because provision (I) sets a starting point ("the date the claim accrues") that is arguably inconsistent with subparagraph (B).

3.5. Choice of State Law

The parties disagree as to which state's law supplies "the period applicable under State law," §§ 1787(b)(14)(A)(i)(II), 1821(d)(14)(A)(i)(II), (ii)(II). See Joint Limitations Spreadsheet at 97. Plaintiffs' theory appears to be that the "period applicable under State law" is the period that would apply in the state where the failed institution was located. Defendants' theory appears to be that this period is the period that would apply in the absence of the extender statute, under the forum state's conflicts principles.

We agree with defendants. Neither the text nor the history of this statute evinces any intention to alter the limitations that would otherwise apply in a particular case, except by extending the period to a certain uniform minimum.

Furthermore, in some circumstances, plaintiffs' interpretation would lead to the absurd result of shortening the limitations period. Suppose a failed bank is located in a state with a four-year limitations period for a tort that occurred just before the bank failed, and suppose that the FDIC sues in a state that applies its own five-year limitations period for torts. According to plaintiffs' interpretation of the extender statute, the limitations period would be only four years (the maximum of the bank's home state and the federal three-year period), even though a five-year limitations period would have applied if the bank had sued without the benefit of the extender.

4. Accrual

The parties generally agree on the basic rules governing when causes of action accrue. In each relevant state, a contract claim accrues when a party to the contract commits a breach (including a breach of an implied covenant); unjust enrichment accrues when a defendant is enriched through a wrongful act at plaintiff's expense, most commonly when money passes from plaintiff to defendant; and a tort claim accrues when all elements, most especially the plaintiff's injury, are complete.

In most states, if not all, some form of a "discovery rule" delays the commencement of the limitations period for at least some claims until the plaintiff discovers his cause of action, either actually or constructively. The states apply many variations on the "discovery" theme, so we must analyze the discovery rules of each state separately. Additionally, plaintiffs' complaints demonstrate to varying degrees that plaintiffs were on notice of LIBOR manipulation at different times, so we must apply the discovery rule to each plaintiff separately.

Plaintiffs also contend that defendants manipulated LIBOR continuously, so that the statute of limitations did not run until the end of manipulation. We have no need to examine each state's continuing violation doctrine separately, because plaintiffs all sue on a series of separate injuries. In this context, the continuing violation doctrine simply has no application.

4.1. Discovery Rule

4.1.1. Generally

A "discovery rule" postpones accrual of an action until a plaintiff discovers (or until a plaintiff can discover) facts related to his claim. Different states recognize different discovery rules, and some states even recognize different discovery rules for different causes of action. Before launching into a review of the relevant states, we catalog the types of discovery rules:

Actual discovery: The clock starts when a plaintiff actually discovers facts sufficient to state a claim.

Weak inquiry notice: The clock starts when the plaintiff discovers, or when a reasonably diligent plaintiff would discover, facts sufficient to state a claim. Plaintiff has a duty to inquire once he learns of suspicious circumstances, but is only held to account for his failure to do so if a reasonably diligent inquiry would have uncovered the relevant facts.

Intermediate inquiry notice: The clock starts when the plaintiff discovers, or when a reasonably diligent plaintiff would discover, facts sufficient to state a

claim. But, if the plaintiff fails to conduct an investigation, then the limitations period begins when the plaintiff should have commenced a reasonable investigation.

Strong inquiry notice: The clock starts when a plaintiff is on inquiry notice.

Ascertainability: The clock starts when a cause of action is reasonably discoverable, regardless of a plaintiff's personal knowledge of suspicious circumstances.

No discovery rule: The clock starts when the cause of action arises, regardless of plaintiff's knowledge.

4.1.2. By Jurisdiction

4.1.2.1. Federal

We have already held in LIBOR I that an "intermediate inquiry notice" rule applies to the Commodities Exchange Act. 935 F. Supp. 2d at 698.

The Supreme Court has adopted what we call the "weak inquiry notice" rule for private federal securities claims. See Merck & Co. v. Reynolds, 559 U.S. 633, 651-53 (2010) (rejecting both intermediate and strong inquiry notice rules); see also City of Pontiac Gen. Emps.' Ret. Sys. v. MBIA, Inc., 637 F.3d 169, 175 (2d Cir. 2011) ("[A] fact is not deemed 'discovered' until a reasonable diligent plaintiff would have sufficient information about that fact to adequately plead it."). As the Court recognized, "terms such as 'inquiry notice' and 'storm warnings' may be useful to the extent that they identify a time when the facts would have prompted a reasonably diligent plaintiff to begin investigating. But the limitations period does not begin to run until . . . a reasonably diligent plaintiff would have discovered 'the facts constituting the violation,' including scienter—irrespective of whether the actual plaintiff undertook a reasonably diligent investigation." Merck, 559 U.S. at 653 (quoting 28 U.S.C. § 1658(b)(1)).

In LIBOR I, we discussed post-Merck Second Circuit precedent holding that an intermediate inquiry notice rule applies whenever a federal statute "is silent on the issue." See LIBOR I, 935 F. Supp. 2d at 698 (quoting Koch v. Christie's Int'l PLC, 699 F.3d 141, 148-49 (2d Cir. 2012) (RICO)). Following the same reasoning, we will also apply an "intermediate inquiry notice" rule to both the tort and the contract branches of the federal extender statutes.

4.1.2.2. California

4.1.2.2.1. Torts

The statute of limitations for torts (including the statutory claims for unfair business practices) begins when a plaintiff "suspects or should suspect that her injury was caused by wrongdoing, that someone has done something wrong to her." Jolly v. Eli Lilly & Co., 44 Cal. 3d 1103, 1110, 751 P.2d 923, 927 (1988). "Wrong" and "wrongdoing" are not used in a technical legal sense, and the statute runs even though the plaintiff is unaware of a specific legal theory upon which to base his claims. Norgart v. Upjohn Co., 21 Cal. 4th 383, 397-98, 981 P.2d 79, 88 (1999) (quoting Jolly, 44 Cal. 3d at 1110 n.7, 751 P.2d at 923 n.7).

This is an inquiry notice rule, because detailed knowledge of facts or certainty about the facts is not necessary to start the limitations period. See Fox v. Ethicon Endo-Surgery, Inc., 35 Cal. 4th 797, 807, 110 P.3d 914, 920 (2005). Further, it is a strong inquiry notice rule, because the limitations period does not wait for either the actual plaintiff or a hypothetical reasonable plaintiff to conduct an investigation.

This inquiry notice rule differs from others in that a plaintiff—apparently even a highly sophisticated plaintiff—is not on inquiry notice until he actually possesses information sufficient to trigger a duty to inquire. "[P]ublic awareness of a problem through media coverage alone" does not "create[] constructive suspicion for purposes of discovery." Unruh-Haxton v. Regents of the Univ. of Cal., 162 Cal. App 4th 343, 364, 76 Cal. Rptr. 3d 146, 163 (4th Dist. 2008) (citing Nelson v. Indevus Pharm., Inc., 142 Cal. App. 4th 1202, 1206, 48 Cal. Rptr. 3d 668, 671 (2d Dist. 2006)); see also Eidson v. Medtronic, Inc., 40 F. Supp. 3d 1202, 1220 (N.D. Cal. 2014).

4.1.2.2.2. Contract and Unjust Enrichment

California courts have applied a more limited rule that accrual of a contract claim is delayed when "[t]he injury or the act causing the injury, or both, [are] difficult for the plaintiff to detect," and where the defendant was "in a far superior position to comprehend the act and the injury." April Enters., Inc. v. KTTV, 147 Cal. App. 3d 805, 831, 195 Cal. Rptr. 421, 436 (2d Dist. 1983). For example, in April Enterprises, the defendant had erased certain video tapes in violation of a contract. The tapes were (in accordance with the contract) in the defendant's possession, so the limitations period was tolled until the plaintiff was found to have been in a position to detect the breach.

California's Supreme Court has not expressly adopted April Enterprises, see El Pollo Loco, Inc. v. Hashim, 316 F.3d 1032, 1039 n.9 (9th Cir. 2003), but has favorably cited it to support a narrower point that the discovery rule applies to a claim against a professional. See Samuels v. Mix, 22 Cal. 4th 1, 9, 989 P.2d 701, 706 (1999). Despite the California Supreme Court's silence, we take April Enterprises as a fair statement of California law, as have other courts. See, e.g., El Pollo Loco, 316 F.3d at 1039.

This rule is more limited than the discovery rule for torts. As plaintiffs point out, the tort rule tolls a claim until the plaintiff personally learns information sufficient to trigger inquiry notice. By contrast, the April Enterprises rule is an "ascertainability" rule: the clock starts when the breach is no longer "difficult . . . to detect." This distinction is consistent with the idea that a discovery rule is an exception to the usual contract rule, and should therefore be construed narrowly.

An unjust enrichment claim related to a contract follows the same doctrine. See CMACO Automotive Sys., Inc. v. Wanxiang Am. Corp., 589 F.3d 235, 248 (6th Cir. 2009) (applying California law).

4.1.2.2.3. Blue Sky

The statute of limitations in California's blue sky law runs for "two years after the discovery by the plaintiff of the facts constituting the violation." Cal. Corp. Code § 25506(b) (West 2006).

Plaintiffs argue that this is an "actual discovery" rule, and rely on one of the alternative holdings in Eisenbaum v. W. Energy Res., Inc., 218 Cal. App. 3d 314, 267 Cal. Rptr. 5 (4th Dist. 1990), which interpreted similar language in a related statute of limitations. Defendants argue for the "intermediate inquiry notice" rule that we applied in LIBOR I, and rely on Deveny v. Entropin, Inc., 139 Cal. App. 4th 408, 42 Cal. Rptr. 3d 807 (4th Dist. 2006) (referring to "inquiry notice").

We agree with neither side, and predict that the California Supreme Court would follow the "weak inquiry notice" rule of Merck.

Regarding plaintiffs' approach, we are not persuaded by Eisenbaum's statement that "discovery . . . of the facts" clearly means actual discovery. The meaning of this phrase is not plain. Merck interpreted a similar phrase in the U.S. Code to include constructive discovery, and Eisenbaum's reading appears to ignore section 19 of the California Civil Code: "Every person who has actual notice of circumstances sufficient to put a prudent man upon inquiry as to a particular fact, has constructive notice of the fact itself in all cases in which, by prosecuting such inquiry, he might have learned such fact."

There is, however, no support for defendants' assumption that the Devery court intended any particular variant on the inquiry notice concept when is referred to "inquiry notice." Deveny was simply distinguishing its rule from the manifestly incorrect "actual knowledge" rule of Eisenbaum. Our view finds support in the California Civil Code, quoted above, which expresses a "weak inquiry notice" standard, and from the Ninth Circuit, which has applied the "weak inquiry notice" rule of Merck to California's blue sky law. According to the Ninth Circuit, the limitations period "does not begin to run until plaintiff discovers the facts constituting the violation or in the exercise of reasonable diligence should have . . . ." Kramas v. Sec. Gas. & Oil Inc., 672 F.2d 766 (9th Cir. 1982).

Finally, there is no reason to think that the California Supreme Court interpreting its own law would be less solicitous of California investors than the United States Supreme Court interpreting federal law. Accordingly, we apply the same "weak inquiry notice" rule to California blue sky law as to federal securities law.

4.1.2.3. Connecticut

As plaintiffs appear to concede, see Joint Limitations Spreadsheet at 46-51, Connecticut does not apply a discovery rule, even for torts. See Conn. Gen. Stat. Ann. §§ 52-576, -577 (West 2013); Farnsworth v. O'Doherty, 85 Conn. App. 145, 150, 856 A.2d 518, 521 (2004) (negligence) (quoting Collum v. Chapin, 40 Conn. App. 449, 451, 671 A.2d 1329, 1331 (1996) (tortious interference)).

4.1.2.4. District of Columbia

In the District of Columbia, a claim accrues upon actual or inquiry notice of "(1) an injury; (2) its cause in fact; and (3) some evidence of wrongdoing." Diamond v. Davis, 680 A.2d 364, 379 (D.C. 1996) (citing Bussineau v. Pres. & Dirs. of Georgetown Coll., 518 A.2d 423, 425 (D.C. 1986)). The "wrongdoing" here is not wrongdoing by "someone" (as in California), but wrongdoing of a particular defendant. See Smith v. Brown & Williamson Tobacco Corp., 108 F. Supp. 2d 12, 16 (D.D.C. 2000); Diamond, 680 A.2d at 380 (citing D.C. Cir. cases).

Inquiry notice as to these facts exists when a person would have discovered the facts with ordinary diligence. See Diamond, 680 A.2d at 375 (quoting P.H. Sheehy Co. v. E. Importing & Mfg. Co., 44 App. D.C. 107, 112 (D.C. Cir. 1915) ("[T]he statute of limitations only beg[i]n[s] to run from the time when plaintiff, by the exercise of ordinary diligence under all the circumstances of the case, ought to have ascertained the fact of the breach.")). This is essentially a "weak inquiry notice" rule, under which the clock when plaintiff is aware of the essential facts.

D.C. applies its discovery rule with equal force to contract claims as to tort claims. See P.H. Sheehy Co. (breach of warranty); Fred Ezra Co. v. Psych. Inst. of Wash., D.C., 687 A.2d 587, 592 (D.C. 1996) (applying Diamond to breach of contract).

4.1.2.5. Iowa

Even for fraud, Iowa does not apply a discovery rule to claims for money damages. See Bob McKiness Excavating & Grading, Inc. v. Morton Bldgs., Inc., 507 N.W.2d 405, 411 (Iowa 1993) (citing Koppes v. Pearson, 384 N.W.2d 381, 386 (Iowa 1986), abrogated on other grounds, Christy v. Miulli, 692 N.W.2d 694, 701 (Iowa 2005)); but cf. Iowa Code Ann. § 614.4 (West 1999).

4.1.2.6. Kansas

Kansas applies a "weak inquiry notice" rule to fraud claims. See Kan. Stat. Ann. § 60-513(a)(3) (West 2008); Kan. Wastewater, Inc. v. Alliant Techsystems, Inc., 257 F. Supp. 2d 1344, 1349-50 (D. Kan. 2003); Waite v. Adler, 239 Kan. 1, 6, 716 P.2d 524, 527 (1986).

Other tort claims with a two-year statute of limitations (such as tortious interference and negligent misrepresentation) accrue when "the fact of injury becomes reasonably ascertainable to the injured party." § 60-513(b); Hutton v. Deutsche Bank AG, 541 F. Supp. 2d 1166, 1169-71 (D. Kan. 2008); Davidson v. Denning, 259 Kan. 659, 914 P.2d 936, 948 (1996) ("reasonably ascertainable" does not mean "actual knowledge"). This is an "ascertainability" rule.

Kansas applies no discovery rule to contract and unjust enrichment claims. See § 60-511 to -512.

4.1.2.7. New Jersey

Plaintiffs do not argue that New Jersey applies a discovery rule to contract or unjust enrichment claims.

As to tort claims, plaintiffs argue for a rule that a claim accrues when "the injured party actually discovers or should have discovered through reasonable diligence the fact[s] essential to the cause of action." R.A.C. v. P.J.S., Jr., 192 N.J. 81, 98, 927 A.2d 97, 106 (2007). This quotation is a passing description of the discovery rule in the context of a holding that no discovery rule applies to the statutory limitations period of New Jersey's Parentage Act. We acknowledge this language as a rough description of New Jersey's "weak inquiry notice" rule, but we also apply refinements from other cases.

Under New Jersey law, inquiry notice is triggered when a reasonable person in plaintiff's circumstances would have undertaken an investigation. See Martinez v. Cooper Hosp.-Univ. Med. Ctr., 163 N.J. 45, 56, 747 A.2d 266, 272 (2000). The limitations period then begins to run when a reasonable person would have discovered that his "injury is due to the fault of another." Cetel v. Kirwan Fin. Grp., Inc., 460 F.3d 494, 513 (3d Cir. 2006). We conclude that New Jersey's "weak inquiry notice" rule starts the limitations period when plaintiff's investigation would lead to the information set out in Cetel.

4.1.2.8. New York

New York's statute of limitations for fraud runs for six years from accrual or two years from discovery, whichever is later. See N.Y. C.P.L.R. 213(8). Normally, discovery is imputed when a plaintiff is "possessed of knowledge of facts from which [the fraud] could be reasonably inferred." Sargiss v. Magarelli, 12 N.Y.3d 527, 532, 909 N.E.2d 573, 576 (2009) (quoting Erbe v. Lincoln Rochester Trust Co., 3 N.Y.2d 321, 326, 144 N.E.2d 78, 80 (1957)); see also N.Y. C.P.L.R. 203(g) (Consol. 2008). "The fraud" may include all of the elements, including scienter. See Allstate Ins. Co. v. Credit Suisse Sec. (USA) LLC, Index No. 650547/2011, 42 Misc. 3d 1220(A), 986 N.Y.S.2d 864 (Table), 2014 WL 432458, at *4, 2014 N.Y. Misc. LEXIS 428, at *12-13 (Sup. Ct. N.Y. Cty. 2014) (citing Merck); Phoenix Light SF Ltd. v. ACE Sec. Corp., Index No. 650422/2012, 39 Misc. 3d 1218(A), 975 N.Y.S.2d 369 (Table), 2013 WL 1788007, at *5, 2013 N.Y. Misc. LEXIS 1723, at *12-13 (Sup. Ct. Apr. 24, 2013) (same).

However, "where the circumstances are such as to suggest to a person of ordinary intelligence the probability that he has been defrauded, a duty of inquiry arises, and if he omits that inquiry when it would have developed the truth, and shuts his eyes to the facts which call for investigation, knowledge of the fraud will be imputed to him." Gutkin v. Siegal, 85 A.D.3d 687, 688, 926 N.Y.S.2d 485, 486 (1st Dep't 2011) (quoting Armstrong v. McAlpin, 699 F.2d 79, 88 (2d Cir. 1983) (quoting Higgins v. Crouse, 147 N.Y. 411, 416, 42 N.E. 6, 7 (1895))).

This is a quintessential "intermediate inquiry notice" rule, and is the ultimate source of the federal rule stated in LIBOR I. See 935 F. Supp. 2d at 712 (quoting Koch v. Christie's Int'l PLC, 699 F.3d 141, 151 (2d Cir. 2012) (quoting Lentell v. Merrill Lynch & Co., 396 F.3d 161, 168 (2d Cir. 2005) (quoting LC Capital Partners, LP v. Frontier Ins. Grp., Inc., 318 F.3d 148, 154 (2d Cir. 2003) (citing Armstrong (quoting Higgins))))).

Plaintiffs do not argue that a discovery rule applies to any of their non-fraud claims in New York.

4.1.2.9. Pennsylvania

The Pennsylvania Supreme Court has characterized its discovery rule as a "narrow[er] . . . rule than [in] most other jurisdictions." Gleason v. Borough of Moosic, 609 Pa. 353, 362, 15 A.3d 479, 484 (2011) (citing Wilson v. El-Daief, 600 Pa. 161, 178, 964 A.2d 354, 364 (2009)). The rule is an inquiry-notice rule, as plaintiff's obligation to "direct diligence in the channel in which it would be successful," Wilson, 600 Pa. at 178-79, 15 A.3d at 364, arises with "storm warnings of possible fraud." BPP Ill., LLC v. Royal Bank of Scotland Grp., PLC, No. 13-cv-638 (JMP), 2013 WL 6003701, at *6, 2013 U.S. Dist. LEXIS 161761, at *23 (S.D.N.Y. Nov. 13, 2013) (quoting Ciccarelli v. Gichner Sys. Grp., Inc., 862 F. Supp. 1293, 1301 (M.D. Pa. 1994)), vac'd in part on other grounds, 603 F. App'x 57 (2d Cir. 2015).

It is difficult, however, to tell from the leading cases precisely what sort of inquiry-notice rule Pennsylvania applies. It appears that the rule is "weak inquiry notice," but is "narrow" in comparison to the Merck rule in that a Pennsylvania action accrues when a reasonable plaintiff would have discovered "at least some form of significant harm and of a factual cause linked to another's conduct" even without knowing all the elements of a cause of action. Wilson, 600 Pa. at 178, 964 A.2d at 364.

Pennsylvania applies its discovery rule, in appropriate circumstances, to contract claims as well as tort. See Crouse v. Cyclops Indus., 560 Pa. 394, 403-04, 745 A.2d 606, 611 (2000) (applying discovery rule to promissory estoppel); Pagano v. Flakey Jake's of Greater Phila., Inc., Civ. No. 94-7777, 1995 WL 650241, at *2, 1995 U.S. Dist. LEXIS 16284, at *6 (E.D. Pa. Oct. 31, 1995) (citing Amodeo v. Ryan Homes, Inc., 407 Pa. Super. 448, 595 A.2d 1232 (1991), and A.J. Aberman, Inc. v. Funk Bldg. Corp., 278 Pa. Super. 385, 420 A.2d 594 (1980)); but cf. Crouse, 560 Pa. at 407 n.1, 745 A.2d at 613 n.1. (Saylor, J., dissenting in part) (questioning applicability of discovery rule to contract claims).

4.1.2.10. Texas

Texas defers accrual of a claim when the wrong is "inherently undiscoverable." Via Net v. TIG Ins. Co., 211 S.W.3d 310, 315 (Tex. 2006). This is an "ascertainability" rule.

Texas applies its discovery rule to contract cases, but only rarely, because due diligence in a typical contract case "may include asking a contract partner for information needed to verify contractual performance." Id. at 314.

4.1.2.11. Virginia

A claim for fraud in Virginia accrues "when such fraud . . . is discovered or by the exercise of due diligence reasonably should have been discovered." Schmidt v. Household Fin. Corp. II, 276 Va. 108, 117, 661 S.E.2d 834, 838-39 (2008) (quoting Va. Code § 8.01-249(1)). This is a "weak inquiry notice" rule, as Schmidt considered a plaintiff who apparently engaged in no investigation despite being armed with facts sufficient to "suspect that something was amiss." Even so, the plaintiff could have prevailed by showing that a reasonable investigation would not have revealed fraud until within two years of his filing. Id., 276 Va. at 118, 661 S.E.2d at 839.

This doctrine has no application to other causes of action because section 8.01-249(1) applies only to fraud. See Mid-Atl. Telecom, Inc. v. Long Distance Servs., Inc., 32 Va. Cir. 75, 79 (Fairfax Cty. 1993). Virginia precedent holds that "[s]tatutes of limitations are strictly enforced and exceptions thereto are narrowly construed. Consequently, a statute [of limitations] should be applied unless the General Assembly clearly creates an exception, and any doubt must be resolved in favor of the enforcement of the statute." Arrington v. Peoples Sec. Life Ins. Co., 250 Va. 52, 55, 458 S.E.2d 289, 290-91 (1995).

4.1.3. Application

In order to decide when each statute of limitations began to run, we must address two questions. First, when were plaintiffs placed on inquiry notice of LIBOR manipulation? Second, for each of the relevant discovery rules, at which point did inquiry notice cause each limitations period to begin?

4.1.3.1. Inquiry Notice

Except as to California state law, we consider investors to have been on constructive notice of information of which a reasonable investor would keep himself aware, including well-publicized financial news. See, e.g., Pension Trust Fund for Operating Eng'rs v. Mortg. Asset Securitization Transactions, Inc., 730 F.3d 263, 276-77 (3d Cir. 2013) (listing law suits, news articles, prospectuses, and financial reports as examples of public documents that may lead a diligent investor to inquire); DeBenedictis v. Merrill Lynch & Co., 492 F.3d 209, 216 (3d Cir. 2007) (finding inquiry notice on the basis of information in public filings); LC Capital Partners, LP v. Frontier Ins. Grp., Inc., 318 F.3d 148, 155 (2d Cir. 2003) (same); In re Merrill Lynch & Co., Inc. Research Reports Sec. Litig., 289 F. Supp. 2d 416, 425 (S.D.N.Y. 2003) (finding inquiry notice on the basis of a "plethora of public information"); In re Ultrafem Inc. Sec. Litig., 91 F. Supp. 2d 678, 692 (S.D.N.Y. 2000) (dismissal "appropriate when the facts from which knowledge may be imputed are clear from . . . the public disclosures themselves"); cf. Newman v. Warnaco Grp., Inc., 335 F.3d 187, 193-95 (2d Cir. 2003) (examining public filings as a basis for inquiry notice); Dodds v. Cigna Sec., Inc., 12 F.3d 346, 350-51 (2d Cir. 1993) (finding inquiry notice on the basis of prospectus received by investor, regardless of whether investor read it).

This was the rule we followed in LIBOR I, and we now make explicit our holding that (except under California law), exchange-based plaintiffs were on constructive notice of news articles relating to LIBOR by May 29, 2008. Eurodollar contracts are liquid, speculative instruments tied directly to LIBOR. In this context, a reasonable Eurodollar trader would have sought out LIBOR-related news, just as a trader of Toyota stock would seek out news of Camry inventories or a corn trader would seek out news of drought in Nebraska.

We have less confidence that a reasonable OTC or fixed-income investor would have noticed news about LIBOR in Spring 2008. We understand that at least some institutional swap and bond investors buy an investment with the expectation of holding the position for a long time (perhaps to maturity), simply receiving or sending payments as they come due. In this context, it is conceivable that even a sophisticated investor would not have sought out or closely attended news affecting the technical details of a long-established investment position. Cf. BPP Ill., LLC v. Royal Bank of Scotland Grp. PLC, 603 F. App'x 57 (2d Cir. 2015) (vacating dismissal of OTC LIBOR claim on limitations grounds). Of course, if the pleadings reveal that a particular plaintiff was actually aware of "smoke signals" on a particular date, then we will not hesitate, even at the pleadings stage, to consider that plaintiff on inquiry notice of manipulation. At a later stage, we will also entertain evidence of what reasonable investors would have made themselves aware of.

For cases involving the California inquiry notice rule or OTC and bond plaintiffs, we examine the pleadings to learn when (if ever) plaintiffs admit to being on notice of LIBOR fraud. Few plaintiffs admit reading the LIBOR-related articles in Spring 2008, but several allege relying on responses that the BBA and banks published through August 5, 2008. After reading the BBA's vociferous, self-interested statements regarding LIBOR, a reasonable investor would have asked what the BBA was responding to, and would have almost immediately discovered the barrage of news articles criticizing LIBOR. Therefore, we consider these plaintiffs to have been on inquiry notice as of August 5, 2008, at the latest. Some plaintiffs also allege reliance on class actions, including the Metzler action filed on April 15, 2011. These plaintiffs were surely on inquiry notice by April 15, 2011.

The District of Columbia holds that inquiry notice is triggered only when circumstances suggest an injury inflicted by a particular defendant. Accordingly, in applying D.C. law, we apply the inquiry notice rule only to claims against panel bank entities, because it was implicit in the news articles that the panel banks were responsible for submitting low LIBOR bids.

As to jurisdictions besides the District of Columbia, we apply the more coherent rule that a plaintiff has a duty to investigate warnings that he has been injured, regardless of whether the warning signs point to a particular defendant. The plaintiff's inquiry is intended to identify the cause of his injury, and so the defendant-specific part of our accrual analysis comes when we ask whether a reasonable investigation would have revealed that a particular defendant was responsible for injuring plaintiff.

4.1.3.2. Effect of Inquiry Notice

For states that apply an "ascertainability" or "strong inquiry notice" rule, the statute of limitations began to run on the inquiry notice date, as to claims that arose before then. No plaintiff alleges undertaking a meaningful investigation, so this conclusion holds for "intermediate inquiry notice" rules as well.

In states with "weak inquiry notice" rules, the limitations period commenced when a reasonable plaintiff would have discovered the breach or tort. We have not previously considered how long a reasonable investigation would have taken, because we have not previously applied a "weak inquiry notice" rule. We believe that it would have taken one year, at the very most, for a sophisticated investor to discover that he had been injured by the panel banks' LIBOR suppression. All that a prospective plaintiff needed to do was to download the banks' LIBOR submissions (which were publicly available) and compare the banks' reported credit spreads to public data regarding the banks' credit, or to compare published LIBOR to other indices. It would have also been simple to verify (as had been publicly reported) that many banks' submissions experienced an uncanny jump immediately after the BBA threatened to monitor submissions more closely.

Some plaintiffs argue that they would not have been able to plead all the elements of their claims (especially scienter) after discovering their injuries, and so (under at least some discovery rules, such as the one in Merck) the clock did not then start on their claims. It is true that scienter is often a private plaintiff's most significant pleading challenge, and that scienter is an element of nearly all of plaintiffs' claims. But this is an unusual case, in which scienter was fairly easy to plead once a plaintiff learned of his injury and its factual cause. Scienter may be pleaded with facts sufficient to show a motive and opportunity. Here, the reputational motive was clear and was thoroughly explained in at least some of the LIBOR-related articles in 2008. The opportunity would have been equally clear, once a diligent plaintiff educated himself about the LIBOR-setting process. Thus, we do not distinguish between a rule that the clock starts when a plaintiff discovers his injury and its cause, and a rule that the clock starts when a plaintiff can survive a motion to dismiss.

It is not, however, clear when a reasonable plaintiff would have discovered the alleged complicity of non-panel-bank entities (bank affiliates and the BBA) in LIBOR manipulation, so we cannot hold that a reasonable investigation would have concluded as to non-panel-bank entities on any particular date. This inability prevents us from holding that the limitations period ran as to these defendants under a "weak inquiry notice" rule, although it has no effect on the stronger inquiry notice rules.

4.2. Continuing Violation Doctrines

The label "continuing violation doctrine" can be used to refer to two separate doctrines. See generally White v. Mercury Marine, Div. of Brunswick, Inc., 129 F.3d 1428, 1430 (11th Cir. 1997). One doctrine (the "pure" continuing violation doctrine, or simply the continuing violation doctrine) holds that a cause of action does not accrue until a defendant ceases a continuous course of wrongful conduct. When this doctrine applies, the plaintiff may sue for the entire course of conduct, including wrongful acts that would otherwise be untimely. The other doctrine (the "modified" continuing violation doctrine or "theory of continuous accrual") holds that a plaintiff may recover for a harm committed within the limitations period, even though the plaintiff cannot recover for similar harms committed outside the limitations period.

4.2.1. Pure CVD

The pure continuing violation doctrine applies when a plaintiff's injury is the product of a sequence of small harms. For example, in the Title VII context, a hostile work environment claim "occurs over a series of days or perhaps years and . . . a single act of harassment may not be actionable on its own." Nat'l R.R. Passenger Corp. v. Morgan, 536 U.S. 101, 115 (2002). By contrast, a Title VII claim based on a discrete act (for example, termination or refusal to extend a raise) accrues at the moment of the discrete act, regardless of subsequent discrimination or retaliation. See id. at 114-15; Ledbetter v. Goodyear Tire & Rubber Co., 550 U.S. 618 (2007), abrogated on other grounds, Lilly Ledbetter Fair Pay Act of 2009, Pub. L. No. 111-2, 123 Stat. 5 (Jan. 29, 2009) (amending scattered sections of 29 and 42 U.S.C.).

Some cases suggest that the doctrine applies when the related nature of harms is not apparent until some later time. E.g., Yanowitz v. L'Oreal USA, Inc., 36 Cal. 4th 1028, 1058, 116 P.3d 1123, 1141 (2005) (applying doctrine to employer's pattern of retaliatory acts when the retaliatory nature was not readily apparent). This holding is more logically characterized as an application of the discovery rule that an action accrues only when the plaintiff "suspects or should suspect that her injury was caused by wrongdoing." Jolly v. Eli Lilly & Co., 44 Cal. 3d 1103, 1110, 751 P.2d 923, 927 (1988).

The pure continuing violation doctrine does not apply to the facts of these cases, because plaintiffs suffered discrete, actionable harms each time they paid too much or received too little on a LIBOR-based transaction. Plaintiffs cite no case in any state in which a court applied the pure doctrine to allow a plaintiff to sue over one faulty payment within a series of payments. To the contrary, plaintiffs' cases involving discrete payments actually apply the modified doctrine. See ABKCO Music & Records Inc. v. Chimeron LLC, 517 F. App'x 3, 5-6 (2d Cir. 2013); Shelton v. Elite Model Mgmt., Inc., 11 Misc. 3d 345, 360-61, 812 N.Y.S.2d 745, 757-58 (Sup. Ct. N.Y. Cty. 2005), abrogated on other grounds, Rhodes v. Herz, 84 A.D.3d 1, 920 N.Y.S.2d 11 (1st Dep't 2011).

Separately plaintiffs argue for application of Pennsylvania's "continuing contract" doctrine, which tolls a contract claim until the termination of a "continuous" contractual relationship. See, e.g., Thorpe v. Schoenbrun, 202 Pa. Super. 375, 378, 195 A.2d 870, 872 (1963) (quoting 22 Pa. Law Encyc., Limitation of Actions § 56). The "continuing contract" doctrine, however, applies only when the agreement "does not fix any certain time for payment." If the agreement does fix certain times for payments, then a claim relating to each one of those payments accrues when the payment is to be made. See Franklin v. SKF USA Inc., 126 F. Supp. 2d 911, 930 (E.D. Pa. 2000).

4.2.2. Continuous Accrual

The continuous accrual doctrine applies in a variety of circumstances. See, e.g., Hanover Shoe, Inc. v. United Shoe Machinery Corp., 392 U.S. 481, 502 (1968) (permitting plaintiff to sue for damages within limitations period caused by a continuous antitrust violation dating back to 1912). This doctrine easily applies here to allegations that a defendant manipulated LIBOR or wrongfully received the fruits of LIBOR manipulation. A defendant that manipulated LIBOR in 2008 is not immune simply because it also manipulated LIBOR in 2007 or 2006.

Notwithstanding this doctrine, the fact that a plaintiff suffered injury within the limitations period does not salvage a claim that a defendant committed a misrepresentation or omission about the nature of LIBOR before the limitations period. In this case, the wrongful act occurred at the moment of the misrepresentation or omission, and only the injurious effects continued into the limitations period. See, e.g., SEC v. Kelly, 663 F. Supp. 2d 276, 288 (S.D.N.Y. 2009) (holding that continuous accrual applies "only to 'continual unlawful acts, not continual ill effects from a single violation'" (quoting SEC v. Jones, No. 05-cv-7044 (RCC), 2006 WL 1084276, at *4, 2006 U.S. Dist. LEXIS 22800, at *10 (S.D.N.Y. Apr. 25, 2006))).

5. Fraudulent Concealment

Every relevant state recognizes a fraudulent concealment doctrine, which tolls the statute of limitations "where a defendant, through deceptive conduct, has caused a claim to grow stale." Aryeh v. Canon Bus. Solutions, Inc., 55 Cal. 4th 1185, 1192, 292 P.3d 871, 875 (2013); see also Conn. Gen. Stat. Ann. § 52-595 (West 2013); SEC v. Wyly, 788 F. Supp. 2d 92, 103-04 (S.D.N.Y. 2011); Luksch v. Latham, 675 F. Supp. 1198, 1203 (N.D. Cal. 1987); DGB, LLC v. Hinds, 55 So. 3d 218, 224 (Ala. 2010); Christy v. Miulli, 692 N.W.2d 694, 700-01 (Iowa 2005); O'Keeffe v. Snyder, 83 N.J. 478, 498, 416 A.2d 862, 872-73 (1980); Baselice v. Franciscan Friars Assumption BVM Province, Inc., 2005 PA Super 246 ¶ 20; Grimes v. Suzukawa, 262 Va. 330, 332, 551 S.E.2d 664, 646 (2001).

Concomitantly, the states hold with equal consistency that a plaintiff may not claim the protection of this doctrine if he was on notice of his claims or if he was not reasonably diligent in investigating evidence of fraud. See Koch v. Christie's Int'l PLC, 699 F.3d 141, 157 (2d Cir. 2012) (New York law); Martinelli v. Bridgeport Roman Catholic Diocesan Corp., 196 F.3d 409, 420 (2d Cir. 1999) (Connecticut law); Lampliter Dinner Theater, Inc. v. Liberty Mut. Ins. Co., 792 F.2d 1036, 1042-43 (11th Cir. 1986) (Alabama law); Herremans v. BMW of N. Am., LLC, No. CV 14-2363 (MMM), 2014 WL 5017843, at *7, 2014 U.S. Dist. LEXIS 145957, at *20 (C.D. Cal. Oct. 3, 2014); Perelman v. Adams, 945 F. Supp. 2d 607, 617 (E.D. Pa.), aff'd sub nom. Perelman v. Perelman, 545 F. App'x 142 (3d Cir. 2013); Lucas v. Henrico Cty. Sch. Bd., No. 3:11-cv-5, 2012 WL 1665428, at *6, 2012 U.S. Dist. LEXIS 67060, at *21-22 (E.D. Va. Apr. 12, 2012), adopted in relevant part, 2012 WL 1665427, 2012 U.S. Dist. LEXIS 66569 (E.D. Va. May 11, 2012); Gonzales v. Nat'l Westminster Bank PLC, 847 F. Supp. 2d 567, 572 (S.D.N.Y. 2012); New Amsterdam Project Mgmt. Humanitarian Found. v. Laughrin, No. C 07-935 (JF), 2009 WL 1513390, at *7, 2009 U.S. Dist. LEXIS 48298, at *22-23 (N.D. Cal. May 29, 2009), aff'd, 400 F. App'x 250 (9th Cir. 2010); Farmers Coop. Co. v. Swift Pork Co., 602 F. Supp. 2d 1095, 1113-14 (N.D. Iowa 2009); Snapp & Assocs. Ins. Servs., Inc. v. Robertson, 96 Cal. App. 4th 884, 890-91, 117 Cal. Rptr. 2d 331, 335 (4th Dist. 2002), disapproved on other grounds, Aryeh, 55 Cal. 4th at 1194, 292 P.3d at 877; Christy, 692 N.W.2d at 702; Trinity Church v. Lawson-Bell, 394 N.J. Super. 159, 172-73, 925 A.2d 720, 728-29 (App. Div. 2007); cf. Antonios A. Alevizopoulos & Assocs., Inc. v. Comcast Int'l Holdings, Inc., 100 F. Supp. 2d 178, 185-86 (S.D.N.Y. 2000) (allowing U.S. plaintiff to rely on fraudulent concealment when relevant disclosures were made in Brazil).

Applying federal law in LIBOR I, we rejected a fraudulent concealment argument because the Exchange-Based Plaintiffs there had been on notice of potential claims by May 29, 2008, and because each panel bank's LIBOR submission was published daily. We reaffirmed this analysis in LIBOR III, holding that the banks' public reassurances were insubstantial in light of a continuing stream of troublesome articles about LIBOR.

Having been presented with no persuasive reason to deviate from these conclusions, we decline to apply the fraudulent concealment doctrine in the cases before us. 6. Class-Action Tolling (American Pipe)

6.1. Generally

It is well-established that a federal statute of limitations is tolled while a putative class action is pending. The Supreme Court established this principle in American Pipe, where an individual class member sought to intervene in a timely class action after the statute of limitations had run out on the individual's claim. The Court later applied this rule in the more common situation in which a class member files a separate suit after the putative class action is denied certification. See Crown, Cork & Seal Co. v. Parker, 462 U.S. 345 (1983). Most federal courts have further extended this rule to cases in which a class member seeks to file an individual suit before certification is denied in the associated class action.

The individual plaintiffs assert (mostly) state-law claims, and rely for class-action tolling on their membership from 2011 onward in various federal class actions. Because most of plaintiffs' claims arise under state law, we must look to the state-law analogs of American Pipe tolling. See Casey v. Merck & Co. (Casey I), 653 F.3d 95, 100 (2d Cir. 2011) (certifying questions of class-action tolling to the Virginia Supreme Court), answered, 283 Va. 411, 722 S.E.2d 842 (Casey II), conformed to, 678 F.3d 134 (2d Cir. 2012) (Casey III). States generally recognize some form of class-action tolling, but the precise contours of their doctrines vary. See, e.g., Staub v. Eastman Kodak Co., 320 N.J. Super. 34, 50-51, 726 A.2d 955, 963-64 (App. Div. 1999) (surveying state-court cases); Tigg v. Pirelli Tire Corp., 232 S.W.3d 28, 33-34 (Tenn. 2007) (same).

Several legal questions arise regarding the scope of each state's tolling law (and, with respect to some questions, regarding federal law). The three major questions, cutting across many plaintiffs, are as follows:

Cross-jurisdictional tolling: Whether each tolling doctrine applies when the predicate class action was filed in federal court.

Cross-claim tolling: Whether each tolling doctrine applies when the individual plaintiff asserts causes of action that are distinct from those raised by the putative class.

Cross-defendant tolling: Whether each tolling doctrine applies when the individual plaintiff asserts claims against a defendant who was not named in the predicate class action.

Additionally, the parties raise three questions that are particular to certain plaintiffs:

• Whether class-action tolling applies to the FDIC's and the NCUA's extended statutes of limitations, and, if so, how.

• Whether American Pipe tolling applies to a federal statute of repose.

• Whether class-action tolling applies to the time during which the Schwab Plaintiffs' first set of individual cases was under consideration.

Finally, after considering certain minor issues raised by the parties, we review the essential data of the class actions on which plaintiffs rely.

6.2. Legal Analysis

6.2.1. Tolling Across Jurisdictions

Because each plaintiff relies on a federal class action to toll claims arising under state law, we must decide whether each relevant state's class-action tolling doctrine applies when the predicate class action was filed in a different jurisdiction's court.

States are split, and few of the states relevant to the cases before us have controlling law. Compare Dow Chem. Corp. v. Blanco, 67 A.3d 392, 395 (Del. 2013) (adopting cross-jurisdictional tolling), Lee v. Grand Rapids Bd. of Educ., 148 Mich. App. 364, 370, 384 N.W.2d 165, 168 (1986) (same), 801 S.W.2d 382, 389 (Mo. Ct. App. W. Dist. 1990) (same), Stevens v. Novartis Pharm. Corp., 358 Mont. 474, 486-91, 247 P.3d 244, 253-56 (2010) (same), Staub, 320 N.J. Super. at 58, 726 A.2d at 967 (App. Div. 1999) (same), Vaccariello v. Smith & Nephew Richards, Inc., 94 Ohio St. 3d 380, 382-83, 763 N.E.2d 160, 163 (2002) (plurality of two out of seven justices) (same); id., 94 Ohio St. 3d at 390, 763 N.E.2d at 168-69 (partial concurrence of two additional justices) (same), with Portwood v. Ford Motor Co., 183 Ill. 2d 459, 465-67, 701 N.E.2d 1102, 1104-05 (1998) (rejecting cross-jurisdictional tolling), Maestas v. Sofamor Danek Grp., Inc., 33 S.W.3d 805, 808 (Tenn. 2000) (same), Bell v. Showa Denko K.K., 899 S.W.2d at 758 (Tex. App. Amarillo 1995), and Casey II, 283 Va. at 416, 722 S.E.2d at 845 (same); see also Seaboard Corp. v. Marsh Inc., 295 Kan. 384, 284 P.3d 314 (2012) (interpreting Kan. Stat. Ann. § 60-518 (West 2008)) (putative class member may file individual action within six months of failure of out-of-state class action).

The cases rejecting cross-jurisdictional tolling (particularly those that were decided early in the development of this doctrine) tend to emphasize a risk that a state will attract individual out-of-state plaintiffs after a failed federal class action. See Portwood, 183 Ill. 2d at 465, 701 N.E.2d at 1104; Maestas, 33 S.W.3d at 808. Offsetting this consideration, the cases adopting cross-jurisdictional tolling point out that rejecting cross-jurisdictional tolling creates a risk that individual plaintiffs will file duplicative "placeholder" suits to preserve their rights while a class action is pending elsewhere. See Dow, 67 A.3d at 395.

The cases adopting cross-jurisdictional tolling also emphasize the underlying reasons why courts originally developed class-action tolling: a class complaint gives fair notice of claims to defendants; a putative class member acts reasonably when he relies on a class action to vindicate his rights; and run-of-the-mill individual suits are disfavored when a class action is viable because too many individual suits would subvert the modern class-action mechanism. See Am. Pipe, 414 U.S. at 550-51.

Additionally, in performing an Erie analysis, we must consider any relevant factors that are specific to a given state. This includes a review of any state-specific non-binding sources, such as opinions of lower state courts, opinions of federal courts applying state law, and dicta of the state's high court. In the context of cross-jurisdictional tolling, we also consider whether the state has a class action rule similar to federal Rule 23, whether the state allows cross-jurisdictional tolling when a person files in state court after an individual case elsewhere has been dismissed, and the state's general policies regarding statutes of limitations and regarding class actions. See In re Enron Corp. Sec., Derivative & "ERISA" Litig., 465 F. Supp. 2d 687, 719 (S.D. Tex. 2006) (reviewing Texas and Ohio law).

Absent state-specific considerations, we believe the best prediction is that a state would recognize cross-jurisdictional tolling. The most salient consideration is that the reasoning of American Pipe and analogous state-court cases applies with equal force regardless of whether a class action is filed in the same jurisdiction as the subsequent individual action or in a different jurisdiction. A class action gives defendants just as much notice of an individual's claims regardless of where each case happens to be filed, and denying cross-jurisdictional tolling would, at the margin, encourage parallel, individual litigation while a viable class action is pending. States that are receptive to these concerns in the context of intra-jurisdictional tolling should be equally receptive in the context of cross-jurisdictional tolling.

The parties' parades of docket-control horribles are unpersuasive. Most states have a general "borrowing rule" that prevents an out-of-state plaintiff from filing if his suit would be barred in his home state. Such a rule should suffice to prevent a state from becoming a magnet for litigation after a foreign court denies class certification. If, after denial of class certification, a plaintiff's individual suit would be timely in the plaintiff's home state, then the plaintiff would have no reason to sue elsewhere; if his individual suit would be time-barred at home, then it would be time-barred elsewhere too. Personal jurisdiction may also bar many claims filed by non-resident plaintiffs against non-resident defendants. Moreover, neither side has offered any evidence that states have become swamped with cases after adopting or rejecting cross-jurisdictional tolling.

Addressing the Erie question empirically, we note that state courts are evenly divided. Cross-jurisdictional tolling may even be the majority rule among state courts that have decided the question, and the trend is in favor of tolling. Since 2010, only Virginia (which, unusually, lacks a class-action procedure and which has a general policy of construing its statutes of limitations strictly) has rejected cross-jurisdictional tolling. Several states have accepted cross-jurisdictional tolling, including two with special reasons to worry about attracting out-of-state plaintiffs: Montana, which lacks a borrowing rule that would independently bar out-of-state plaintiffs, and Delaware, which has general personal jurisdiction over many large businesses incorporated there.

Some federal courts seem to have applied a presumption against cross-jurisdictional tolling out of a concern over federal interference with state policy. See, e.g., Clemens v. DaimlerChrysler Corp., 534 F.3d 1017, 1025 (9th Cir. 2008) (predicting California law); Soward v. Deutsche Bank AG, 814 F. Supp. 2d 272, 281-82 (S.D.N.Y. 2011) (predicting New York law). But a proper respect for state judiciaries does not require such timidity. Our duty is to predict accurately what the high court of a particular state would do in the same circumstance, and we fail equally in this duty when we erroneously dismiss a case that that the state courts would sustain as we do when we erroneously sustain a case that a state courts would dismiss.

For these reasons, we will predict that each state would accept cross-jurisdictional tolling unless state-specific factors suggest otherwise.

6.2.1.1. California

In general, California courts have recognized class-action tolling in lower court holdings and Supreme Court dicta, and we can comfortably predict that the California Supreme Court would accept class-action tolling generally. See Jolly v. Eli Lilly & Co., 44 Cal. 3d 1103, 751 P.2d 923 (1988) (citing American Pipe favorably, but declining to apply tolling to a mass tort case when the predicate class action was patently unlikely to be certified); Becker v. McMillin Constr. Co., 226 Cal. App. 3d 1493, 1501, 277 Cal. Rptr. 491, 496 (4th Dist. 1991) (accepting American Pipe when "the substantive class and individual claims were sufficiently similar to give [defendant] notice of the litigation").

Regarding cross-jurisdictional tolling, the two most relevant cases come from the Ninth Circuit. In these cases, the Ninth Circuit has made two opposite Erie predictions by treating "American Pipe tolling" and "equitable tolling" as separate doctrines with different results as to cross-jurisdictional application. Compare Hatfield v. Halifax PLC, 564 F.3d 1177, 1188 (9th Cir. 2009) (applying "equitable tolling" cross-jurisdictionally), with Clemens v. DaimlerChrysler Corp., 534 F.3d 1017, 1025 (9th Cir. 2008) (declining to apply California's analogue of American Pipe tolling cross-jurisdictionally).

We disagree with the view of Hatfield that these tolling principles are distinct. The first indication that Hatfield reached in incorrect result is that the panel recognized, but refused to lend credence to, state-court cases that plainly treat class-action tolling as a form of equitable tolling. See, e.g., Becker; S.F. Unified Sch. Dist. v. W.R. Grace & Co., 37 Cal. App. 4th 1318, 44 Cal. Rptr. 2d 305 (1st Dist. 1995) (permitting tolling under Jolly after examining and rejecting defendant's equitable defenses to tolling). "Equitable tolling" is simply broader than "class-action tolling." It is significant that class-action tolling is entirely consistent with the California Supreme Court's examples of equitable tolling in McDonald v. Antelope Valley Community School District, 45 Cal. 4th 88, 100, 194 P.3d 1026, 1032 (2008) (for example, "where a first action, embarked upon in good faith, is found to be defective").

Second, Hatfield relies on a controversial view that the American Pipe doctrine is "legal" rather than "equitable." There is no reason to impute this view to California courts that clearly favor the opposite approach when this question is not even settled in federal law. Compare Bright v. United States, 603 F.3d 1273, 1282, 1285-86 (Fed. Cir. 2010) ("statutory"), Joseph v. Wiles, 223 F.3d 1155, 1167 (10th Cir. 2000) ("legal"), and N.J. Carpenters Health Fund v. Residential Capital, LLC, 288 F.R.D. 290, 294 (S.D.N.Y. 2013) ("legal"), with Bridges v. Dep't of Md. State Police, 441 F.3d 197, 211 (4th Cir. 2006) ("equitable"), Veltri v. Bldg. Serv. 32B-J Pension Fund, 393 F.3d 318, 322-23 (2d Cir. 2004) ("equitable" in dicta), and In re Lehman Bros. Sec. & Erisa Litig., 800 F. Supp. 2d 477, 482 (S.D.N.Y. 2011) ("equitable").

Third, the Hatfield court, having distinguished class-action tolling from equitable tolling, grounded its "equitable tolling" analysis on Jolly and Becker, which in turn relied heavily on the "class action" case American Pipe. This is a strange turn indeed, if American Pipe is truly different from California's "equitable tolling."

The bottom line is that a single doctrine is in play here, and no California state court has ever suggested otherwise. We will therefore engage with both Hatfield and Clemens to discern whether California would accept cross-jurisdictional tolling.

Hatfield predicted that California would recognize cross-jurisdictional tolling ("equitable tolling") for the benefit of its own residents. In Hatfield, the predicate class action was filed in New Jersey state court, where it was ultimately dismissed as to non-New Jersey plaintiffs for lack of personal jurisdiction. The non-New Jersey plaintiffs immediately re-filed in California federal court. Citing Jolly and Becker, Hatfield accepted as timely the claims of California residents because California would want its residents to be on equal footing with the New Jersey plaintiffs whose claims were permitted to go forward in the original class action. Hatfield recognized, however, that California limitations law was generally more solicitous of its own residents than of non-residents, and declined to apply cross-jurisdictional tolling for the benefit of non-residents. See 564 F.3d at 1189-90 (citing Cal. Civ. Proc. Code § 361).

Clemens, on the other hand, predicted that California would not recognize cross-jurisdictional class-action tolling. Without citing to California cases, Clemens relied on the fact (possibly true then, but no longer so) that a majority of state courts have rejected cross-jurisdictional tolling, and the fact (still true now) that "several federal courts have declined to import the doctrine into state law where it did not previously exist." 534 F.3d at 1025.

In this District, one case has followed Hatfield and Clemens on their own terms. See Vincent v. Money Store, 915 F. Supp. 2d 553, 569-71 (S.D.N.Y. 2013). Vincent accepted Hatfield's view that equitable tolling and class-action tolling are distinct doctrines in California. Relying on Clemens (and a subsequent district court case from the same circuit, Centaur Classic Convertible Arbitrage Fund Ltd. v. Countrywide Fin. Corp., 878 F. Supp. 2d 1009, 1017 (C.D. Cal. 2011)), Vincent concluded that California's class-action tolling was not cross-jurisdictional. The court next declined to apply equitable tolling because the plaintiff filed the second case in the same court where the underlying class action had been filed. This situation does not occur in our case, where the class actions were generally filed in New York and the relevant individual actions in California.

We find Hatfield, which relies on California cases favoring class-action tolling, to be more persuasive than Clemens in its conclusion that California would favor cross-jurisdictional tolling as to resident plaintiffs. We will go a step further and predict that California would accept cross-jurisdictional tolling as to non-resident plaintiffs as well. Although such plaintiffs may benefit from class-action tolling under California law, California's borrowing rule (Cal. Civ. Proc. Code § 361) may independently bar non-resident plaintiffs' claims. Contrary to Hatfield, this borrowing rule does not exemplify a general policy of disfavoring non-residents, but simply prevents California from becoming a magnet for time-barred litigation in which California has no interest. This policy is sufficiently vindicated by the borrowing rule itself, without limiting California's class-action tolling doctrine.

6.2.1.2. Connecticut

The Connecticut Supreme Court has recognized class-action tolling, and one federal case has predicted that Connecticut would recognize cross-jurisdictional tolling. See Grimes v. Hous. Auth. of New Haven, 242 Conn. 236, 242-43, 698 A.2d 302, 306 (1997); Primavera Familienstifung v. Askin, 130 F. Supp. 2d 450, 515-16 (S.D.N.Y. 2001). The later opinion has limited precedential force, because it relied in part on the since-rejected view that American Pipe can toll a state-law claim heard in federal court regardless of state tolling law.

Connecticut has applied cross-jurisdictionally its tolling statute (Conn. Gen. Stat. § 52-592 (West 2013)) that generally allows time to file a second suit after the "accidental failure" of a first suit. See Daoust v. McWilliams, 49 Conn. App. 715, 722, 716 A.2d 922, 927 (1998).

It appears that no general borrowing rule would protect Connecticut from becoming a magnet for non-resident plaintiffs, it is unlikely that Connecticut would have personal jurisdiction over a great number of non-residents' claims. In any case, we need not decide whether Connecticut would allow non-residents to take advantage of cross-jurisdictional tolling because the relevant entities (certain assignors of claims in Salix) are all Connecticut residents.

Accordingly, we predict that Connecticut would recognize cross-jurisdictional class-action tolling.

6.2.1.3. Iowa

The Iowa Supreme Court recognizes class-action tolling (see Lucas v. Pioneer, Inc., 256 N.W.2d 167, 180 (Iowa 1977)), but the doctrine has produced very little subsequent case law.

Iowa has, however, applied its ordinary tolling statute cross-jurisdictionally. See Iowa Code Ann. § 614.10 (West 1999) (allowing time to file a second suit after the procedural failure of a first suit); Doe v. Hartz, 52 F. Supp. 2d 1027, 1041-44 (N.D. Iowa 1999) (first suit in federal court); Grimm v. US West Commc'ns, Inc., 644 N.W.2d 8, 17 (Iowa 2002) (same); Beilke v. Droz, 316 N.W.2d 912 (Iowa 1982) (first suit in foreign state court).

Iowa also has a general borrowing rule that would independently protect it from becoming a magnet for non-resident plaintiffs. § 614.7.

Accordingly, we predict that Iowa would recognize cross-jurisdictional class-action tolling.

6.2.1.4. Kansas

Kansas does not recognize class-action tolling, or any tolling during the pendency of an action. See Waltrip v. Sidwell Corp., 234 Kan. 1059, 1064, 678 P.2d 128, 133 (1984). Instead, Kansas permits a putative class member to file a new complaint within six months after a denial of certification, including denial of certification in the court of another jurisdiction. See Seaboard Corp. v. Marsh Inc., 295 Kan. 384, 284 P.3d 314 (2012); Waltrip, 234 Kan. at 1065, 678 P.2d at 133; cf. Chardon v. Fumero Soto, 462 U.S. 650, 661 (1983) (upholding a similar Puerto Rico statute as an acceptable substitute for American Pipe tolling). This follows from Kansas's savings statute: "If any action be commenced within due time, and the plaintiff fail in such action otherwise than upon the merits, and the time limited for the same shall have expired, the plaintiff . . . may commence a new action within six (6) months after such failure." Kan. Stat. Ann. § 60-518 (2008).

We have not denied certification to any class or dismissed any class action "otherwise than upon the merits." Accordingly, Kansas's savings statute has no application to the individual plaintiffs' complaints.

6.2.1.5. New Jersey

A persuasive opinion from the Appellate Division has recognized class-action tolling, both within New Jersey and across jurisdictions. See Staub v. Eastman Kodak Co., 320 N.J. Super. 34, 50-51, 726 A.2d 955, 963-64 (App. Div. 1999). Defendants themselves cite Staub in their opening brief, and do not raise this as a bar in the Joint Limitations Spreadsheet. We follow Staub in recognizing cross-jurisdictional class-action tolling.

6.2.1.6. New York

Both the Second Circuit and many state courts have recognized class-action tolling under New York law. See Cullen v. Margiotta, 811 F.2d 698, 721 (2d Cir. 1987) (collecting cases), abrogated on other grounds, Agency Holding Corp. v. Malley Duff & Assocs., Inc., 483 U.S. 143 (1987).

One recent case in this Court appears to apply tolling cross-jurisdictionally without discussing the issue. See Sapirstein-Stone-Weiss Found. v. Merkin, 950 F. Supp. 2d 621, 625-26 (S.D.N.Y. 2013). However, other cases predict that New York would reject cross-jurisdictional tolling.

In Soward v. Deutsche Bank AG, 814 F. Supp. 2d 272, 280-82 (S.D.N.Y. 2011), Judge Scheindlin noted that "[p]redicting how New York courts would rule on the issue of cross-jurisdictional tolling would be difficult," id. at 281, but concluded, "I cannot say that New York would adopt cross-jurisdictional tolling and decline to import the doctrine into New York's law." Id. at 282. Relying on Soward (and on other federal cases dealing with non-New York law), Judge Sweet more recently found "compelling" the concern that a state "will invite into its courts a disproportionate share of suits which the federal courts have refused to certify as class actions" by accepting cross-jurisdictional tolling. In re Bear Stearns Cos., Inc. Sec. Derivative, and ERISA Litig., 995 F. Supp. 2d 291, 312 (S.D.N.Y. 2014) (quoting Vincent v. Money Store, 915 F. Supp. 2d, 553, 569-70 (S.D.N.Y. 2013)).

We respectfully disagree with our fellow judges' predictions of New York law. Specifically, we disagree with our colleagues' prediction that New York would reject cross-jurisdictional tolling to the extent that those decisions are predicated on floodgate concerns. As with Iowa (discussed above), a general borrowing rule (N.Y. C.P.L.R. 202 (Consol. 2008)) protects New York from burdensome litigation, and New York has applied cross-jurisdictionally the statute (N.Y. C.P.L.R. 205(a)) that generally extends time after an individual suit is dismissed. See, e.g., Stylianou v. Inc. Vill. of Old Field, 23 A.D.3d 454, 805 N.Y.S.2d 573 (2d Dep't 2005).

Accordingly, we conclude that New York would apply cross-jurisdictional class-action tolling as to both residents and non-residents.

6.2.1.7. Ohio

Ohio recognizes cross-jurisdictional class-action tolling. See Vaccariello v. Smith & Nephew Richards, Inc., 94 Ohio St. 3d 380, 382-83, 390, 763 N.E.2d 160, 163, 168-69 (2002).

6.2.1.8. Pennsylvania

The Pennsylvania Supreme Court has approved class-action tolling, but has not decided the question of cross-jurisdictional application. See Cunningham v. Ins. Co. of N. Am., 515 Pa. 486, 489-90, 530 A.2d 407, 408-09 (1987); Alessandro v. State Farm Mut. Auto. Ins. Co., 478 Pa. 247, 279 n.9, 409 A.2d 347, 350 n.9 (1979). This holding follows from Pennsylvania's rule that a putative class member is considered a party to an action until denial of certification. See Alessandro; Explanatory Cmt. to Pa. R. Civ. P. 1701(a).

One appellate court has disapproved of cross-jurisdictional tolling in the class context. "It is noteworthy that of [the jurisdictions that have adopted cross-jurisdictional class-action tolling], Michigan and New Jersey permit cross-jurisdictional tolling for individual actions as well, whereas Pennsylvania does not." Ravitch v. Pricewaterhouse, 2002 PA Super 49 at ¶¶ 12-13. Consistent with Ravitch, a federal district court in Pennsylvania assumed that cross-jurisdictional tolling would be available when the original class action was filed in a district court in Pennsylvania. Huber v. Taylor, 519 F. Supp. 2d 542, 572-73 (W.D. Pa. 2007), rev'd on other grounds, 532 F.3d 237 (3d Cir. 2008).

As noted in both cases, Pennsylvania law allows an individual to re-file a case dismissed from federal district courts "embracing any part of this Commonwealth." 42 Pa. Cons. Stat. § 5103(b). We therefore conclude that Pennsylvania would reject class-action tolling when the class action was filed in a federal court outside Pennsylvania. As none of the LIBOR class actions were filed in Pennsylvania, we will not apply class-action tolling under Pennsylvania law.

6.2.1.9. Virginia

Virginia rejects cross-jurisdictional class-action tolling. See Casey II, 283 Va. at 411, 722 S.E.2d at 842 (2012).

6.2.1.10. Conclusion

For the foregoing reasons, we will apply class-action tolling cross-jurisdictionally to the limitations laws of California, Connecticut, Iowa, New Jersey, New York, and Ohio, but not Pennsylvania and Virginia. As to Kansas, we hold that the savings rule has no application.

6.2.2. Tolling Across Claims

Many of the individual plaintiffs state different causes of action than were pleaded in the class complaints. The parties dispute whether class-action tolling operates to preserve plaintiffs' new claims. Defendants argue that tolling would prejudice them. Plaintiffs counter that an anti-tolling policy would require individuals to file duplicative cases in state courts in order to preserve state-law claims, thus defeating the purpose of a class action.

Federal precedent strongly favors plaintiffs:

[W]e do not regard the fact that the state court [class] action was premised on different legal theories as a reason not to apply American Pipe tolling to save the claims of class members . . . . It is not a flaw under that doctrine that the first action did not alert the defendant to have its lawyers research the applicability of a particular statute or rule of common law. Indeed, limiting American Pipe tolling to the identical "causes of action" asserted in the initial class action would encourage and require absent class members to file protective motions to intervene and assert their new legal theories prior to class certification, thereby producing . . . "court congestion, wasted paperwork and expense."

Available state-court precedent also tends to support plaintiffs' view. The Connecticut Supreme Court has specifically held that individual plaintiffs may raise new legal theories on the same general facts pleaded in the class complaint. See Grimes v. Hous. Auth. of New Haven, 242 Conn. 236, 246-52, 698 A.2d 302, 307-10 (1997) (relying on Cullen and on the rule of Pavelka v. St. Albert Society Branch No. 30, 82 Conn. 146, 147, 72 A. 725, 725 (1909), that "[t]he facts which establish . . . [a plaintiff's] right and [a defendant's] delict constitute the cause of action"). Likewise, New York's Appellate Division has applied class-action tolling when the underlying class action was for a declaratory judgment and the later individual action was for money damages. Clifton Knolls Sewerage Disposal Co. v. Aulenbach, 88 A.D.2d 1024, 451 N.Y.S.2d 907 (3d Dep't 1982).

Courts do not, however, permit tolling when a plaintiff raises a new factual theory. As Justice Powell stated in his concurrence to Crown, Cork, "[t]he rule should not be read as leaving a plaintiff free to raise different or peripheral claims following denial of class status. . . . [T]he district court should take care to ensure that the suit raises claims that concern the same evidence, memories, and witnesses as the subject matter of the original class suit." Crown, Cork & Seal Co. v. Parker, 462 U.S. 345, 355 (1983); accord Jolly v. Eli Lilly & Co., 44 Cal. 3d 1103, 1124, 751 P.2d 923, 937 (1988) (quoting Justice Powell's concurrence).

These conclusions operate in harmony with the amendment and relation-back rules of the Federal Rules of Civil Procedure. Rule 15 liberally allows a class representative to amend his complaint, and then permits the amended complaint to relate back to a prior pleading if "the amendment asserts a claim . . . that arose out of the conduct, transaction, or occurrence set out . . . in the original pleading." Fed. R. Civ. P. 15(c)(1)(B). Defendants cannot plausibly claim prejudice when a class member sues individually on a new legal theory that the class representatives could still sue on in the predicate class action.

Defendants are correct, however, that plaintiffs' tortious interference claims are insufficiently related to the claims stated by the predicate classes. The classes assert two broad types of viable fact patterns: (1) that plaintiffs' counterparties dealt improperly with plaintiffs with respect to particular contracts or investments; and (2) that panel banks manipulated LIBOR and concealed their manipulation without regard to particular relationships between plaintiffs and counterparties. The first category of claims includes contract claims, unjust enrichment claims, and fraud claims based on a failure to tell counterparties about LIBOR manipulation. The second category includes fraud claims based on false public reassurances, fraud claims based on false LIBOR quotes, and antitrust claims. Tortious interference is in a different factual category, in that a viable tortious interference claim alleges that a panel bank entity intended to disrupt a specific contract. This factual theory was not suggested in the class complaints, so American Pipe tolling does not apply to tortious interference claims. Cf. In re Coastal Plains, Inc., 179 F.3d 197, 216 (5th Cir. 1999) (tortious interference claims in amended complaint did not relate back to original complaint's claims involving a contract between plaintiff and defendant); Constr. Interior Sys., v. Donohoe Cos., 813 F. Supp. 29, 36 (D.D.C. 1992) (similar).

We also note that the early class complaints alleged only persistent suppression, rather than trader-based manipulation. Trader-based claims rely on very different "evidence, memories, and witnesses" than suppression claims—different incentives, leading to qualitatively different misconduct committed by an overlapping but different set of people. Therefore, class-action tolling for trader-based claims must rely on class complaints that set forth trader-based allegations.

6.2.3. Tolling As to New Defendants

Defendants argue that class-action tolling is inappropriate as to defendants who were not named in a predicate class action. Plaintiffs contend that all relevant defendants had practical notice of the action, and some cases favor plaintiffs' view that a new defendant's awareness of a suit is the only relevant factor.

The better view is that a defendant should not be charged with tolling unless the class action gave the newly added defendant notice that it would be exposed to liability. Awareness that a class action is progressing against other defendants is insufficient. See Wachovia Bank & Trust Co. v. Nat'l Student Mktg. Corp., 461 F. Supp. 999, 1012 (D.D.C. 1978), rev'd on other grounds, 650 F.2d 342 (D.C. Cir. 1980).

This conclusion is not inequitable to plaintiffs. If a class member believes that the putative class representatives have failed to raise claims against a particular entity, then the class member may address his concerns with class counsel or may move to intervene in order to name the missing defendant. This would not lead to a deluge of intervenors or separate filings, because only one dissatisfied class member needs to intervene and state class claims against the missing defendant in order to toll claims against that defendant as to all class members.

The one exception is that claims may be tolled against a new defendant if the class complaint put that defendant on notice that the defendant should have been sued but for a mistake in the pleading. Cf. Fed. R. Civ. P. 15(c)(1)(C) (permitting an amended pleading against a new defendant to relate back in such a circumstance). This exception applies to certain of the predicate class complaints in this MDL, which misstated Rabobank's lengthy official name or misspelled various entities' names.

6.2.4. Application to Extender Statutes

The FDIC's and the NCUA's extender statutes adopt the longer of two limitations periods—a federal statutory term whose length depends on the type of action and the applicable term under state law. The first of these is a purely federal rule, and the federal American Pipe doctrine applies with full force to the federal statutory term. See NCUA v. Bear, Stearns & Co., No. 12-2781 (JWL), 2013 WL 4736247, at *8, 2013 U.S. Dist. LEXIS 125238, at *24 (D. Kan. Sept. 3, 2013). The second of these is a state rule incorporated into federal law. Absent any reason to believe otherwise, we assume that Congress intended to apply state-law tolling principles to the state-law limitations period. Cf. Hardin v. Straub, 490 U.S. 536, 539 (1989) ("'In virtually all statutes of limitations the chronological length of the limitation period is interrelated with provisions regarding tolling, revival, and questions of application.' Courts thus should not unravel state limitations rules unless their full application would defeat the goals of the federal statute at issue." (quoting Johnson v. Ry. Express Agency, Inc., 421 U.S. 454, 464 (1975))).

There is no merit to defendants' argument that the class-action tolling rule should not be combined with the extender statutes' accrual and revival rules. This contention is essentially a complaint that Congress selected an overly generous policy towards federal agencies. Congress passed and amended these extender statutes to help federal agencies recoup money that had been lost by failed financial institutions in the 1980s, much of it under scandalous circumstances. We find nothing in the legislative history to suggest that Congress wished to adopt parsimonious tolling rules to offset the statutes' liberal accrual and revival rules.

6.2.5. Application to Statute of Repose

In this Circuit, American Pipe does not toll a statute of repose. See Police & Fire Ret. Sys. of Detroit v. IndyMac MBS, Inc., 721 F.3d 95, 109 (2d Cir. 2013) (citing Lampf, Pleva, Lipkind, Prupis & Petigrow v. Gilbertson, 501 U.S. 350, 363 (1991)), cert. dismissed as improvidently granted sub nom. Pub. Emps.' Ret. Sys. of Miss. v. IndyMac MBS, Inc., ___ U.S. ___, 135 S. Ct. 42 (2014); contra Joseph v. Wiles, 223 F.3d 1155, 1167-68 (10th Cir. 2000).

We have already held, in the course of denying a remand motion, that the Schwab Plaintiffs' new Securities Act claims are untimely under that Act's three-year statute of repose. See supra at 25. Now that the Supreme Court has dismissed its review of Police & Fire, the Schwab Plaintiffs accept that these claims are barred and have chosen to withdraw their federal Securities Act claim without prejudice in the hopes that the Supreme Court or the Second Circuit en banc will eventually adopt the rule of Joseph.

At least some of the Schwab Plaintiffs' Securities Exchange Act claims are also barred by that Act's five-year statute of repose. See infra at 380, 423.

6.2.6. Application to Schwab

The Schwab Plaintiffs first filed LIBOR complaints in August 2011, in the Northern District of California, several months after the first of the LIBOR class actions was filed. See supra at 43, infra at 380. After the JPML transferred all three Schwab actions to this Court, the Schwab Plaintiffs strategically abandoned all of their securities claims in order to bolster their arguments against dismissing their RICO claims. Indeed, at oral argument, Schwab counsel stated, regarding the Exchange Act claims:

If I was standing here today arguing . . . in support of a 10b-5 case, I could not do it credibly. . . . There's no predicate act here that gives rise to a 10b-5 case. . . .

The month after LIBOR I, plaintiffs filed the present Charles Schwab action in California state court, re-asserting claims under the Securities Act and California state law but not under the Securities Exchange Act. After removal to federal court, plaintiffs amended their complaint in October 2014 to add Exchange Act claims that they had previously dismissed.

In this context, the Schwab Plaintiffs may not rely on class-action tolling to save their securities claims (including those under California's blue sky laws). Plaintiffs had a fair opportunity to present securities claims in their original set of cases, but made a conscious, strategic decision not to do so, even as an inconsistent alternative to their RICO theory. No equitable principle, including American Pipe, requires us to alleviate the consequences of plaintiffs' own strategic decision simply because a class action happened to be progressing in parallel with plaintiffs' own cases.

The state-law claims that LIBOR I dismissed without prejudice all effectively relate back to the original complaints of August 2013. See 28 U.S.C. § 1367(d) (2012). Furthermore, American Pipe tolling does apply to these claims for the brief period before the Schwab Plaintiffs originally filed in August 2013.

6.2.7. Other Arguments

Defendants criticize some pleadings for failing to allege facts supporting class-action tolling. This criticism is misplaced because the statute of limitations is an affirmative defense; rebuttal of the statute is not an element that must be pleaded in a complaint. Plaintiffs were therefore free to wait until defendants have raised the limitations argument before alleging their own facts in support of tolling. See Abbas v. Dixon, 480 F.3d 636, 640 (2d Cir. 2007).

Defendants also object that some of the underlying classes are too broad to have provided adequate notice. For example, the original Metzler complaint treated "all persons . . . that transaction in LIBOR-based derivatives" as class members. It is possible that these classes were too broad to achieve certification, but that fact does not prevent us from applying American Pipe, which (in its most common applications) pre-supposes that certification has been denied. What matters is that these classes were sufficiently clear to put defendants on notice, and that it was plausible that the class claims would survive in some form, perhaps after sub-classing.

Finally, neither party addresses whether tolling based on Gelboim continued after we dismissed that case in LIBOR I. Federal courts have consistently held that American Pipe tolling stops at the moment of decertification, even if a decertification order might ultimately be reversed or vacated on reconsideration or appeal. See, e.g., Giovanniello v. ALM Media, LLC, 726 F.3d 106, 115-19 (2d Cir. 2013). The same reasoning holds when a case is dismissed on the merits before a certification decision. An individual plaintiff should not, in such a case, rely on the outside chance that a dismissal will be reversed on appeal any more that an individual plaintiff should rely on the chance that a certification decision will be. Accordingly, no statute of limitations is tolled on the basis of Gelboim for the period after LIBOR I (dated March 29, 2013) dismissed all of Gelboim's claims.

6.3. Predicate Cases

6.3.1. OTC

Most of the individual plaintiffs assert claims related to over-the-counter derivatives. The following class complaints toll at least some of plaintiffs' claims:

6.3.2. Exchange-Based

The Amabile Plaintiffs assert claims related to over-the-counter derivatives. The following class complaints toll at least some of their claims:

6.3.3. Lenders

Some plaintiffs assert claims related to bonds or other debt instruments. The following class complaints toll at least some of plaintiffs' claims:

7. Relation Back

7.1. Defendants Added in Amended Complaints

When nearly all of the individual plaintiffs amended their complaints in late 2014, a few added claims against new defendants.

An amended complaint that "changes the party or the naming of the party against whom a claim is asserted" relates back to the date of a previous complaint only if:

(1) "the amendment asserts a claim or defense that arose out of the conduct, transaction, or occurrence set out—or attempted to be set out—in the original pleading";

(2) the new defendant, within 120 days after the original complaint was filed, "received such notice of the action that it will not be prejudiced in defending on the merits"; and

(3) the new defendant, within 120 days after the original complaint was filed, "knew or should have known that the action would have been brought against it, but for a mistake concerning the proper party's identity."

7.1.1. Principal

The original complaints in the Principal Cases named certain entities as "Panel Bank Defendants," and alleged that these defendants committed antitrust violations, committed fraud (by making representations about the quality of LIBOR and by making false LIBOR quotes), breached their swap agreements with plaintiffs, and received unjust enrichment. See Compl. ("Principal Funds Orig. Compl."), Principal Funds, Aug. 1, 2013, ECF No. 1; Compl. ("Principal Fin. Grp. Orig. Compl."), Principal Fin. Grp., Aug. 1, 2013, ECF No. 1.

The Original Principal Financial Group Complaint listed examples of adjustable-rate bonds that plaintiffs owned, issued by the Royal Bank of Scotland Group PLC, "Bear Stearns," "Bank of America Capital," and "Chase Capital II"; asset-backed securities, including one labeled "Citigroup Global Markets Inc."; and interest rate swaps, including ones traded with "Bank of America" and "JP Morgan Chase."

The Original Principal Funds Complaint also listed examples of adjustable-rate bonds issued by "Bank of America," "Royal Bank of Scotland," and "JP Morgan Chase"; and asset-backed securities including ones labeled "Citigroup Mortgage Loan," "Deutsche Alt-A Securities," "JP Morgan Alternative Loan Tr," and "Bear Stearns Asset-backed Security." The Original Principal Funds Complaint provided no specific examples of interest rate swaps.

The Amended Complaints both added several affiliates as defendants. The Principal Financial Group Amended Complaint pleaded that some of these new defendants traded swaps with plaintiffs, while the Principal Funds Amended Complaint did not. Compare Principal Fin. Grp. Am. Compl. ¶ 203 (alleging specific counterparties) with Principal Funds Am. Compl. ¶ 201 (alleging swaps with "Bank of America, Barclays, Citigroup, Credit Suisse, Deutsche Bank, JPMorgan, RBS, and UBS"). Both amended complaints alleged that plaintiffs held specific bonds that were issued or underwritten by specific defendants. See Principal Funds Am. Compl., Ex. A; Principal Fin. Grp. Am. Compl., Ex. A.

Plaintiffs argue that the new defendants knew that the original complaints should have named them because the new defendants were counterparties to contracts identified in plaintiffs' original complaints. This argument fails because the original complaints completely failed to identify any contracts with enough precision that an unnamed party would have known that it was the intended defendant. For the most part, the original complaints speak in generalities of plaintiffs' "numerous variable rate bonds," "numerous asset-backed securities," "numerous loans," "numerous interest rate swaps," and "ISDA Master Agreements." When plaintiffs did list specific securities or contracts, they provided little to no identifying information.

Notwithstanding the general failure of Principal's original complaints to place new defendants on notice, there are two instances in which the Principal Financial Group Plaintiffs provided enough information to identify a security or contract. First, the Principal Financial Group Original Complaint listed an asset-backed security associated with "Citigroup Global Markets Inc." This is sufficient to inform Citigroup Global Markets, Inc., that it was associated in some way with a security that was the subject of the Principal Financial Group Plaintiffs' suit. See Principal Fin. Grp. Orig. Compl., line 5 of table at ¶ 136.

However, even in these two instances when the Original Complaints adequately identified certain contracts, a reasonable reader would have understood that panel banks—and not non-panel counterparties—were the target of the suit. "Defendants," said the complaints, were "the British Bankers Association, . . . the sixteen banks that comprised the U.S. Dollar Libor panel . . ., and the interdealer money brokers who the panel banks consulted." Principal Funds Orig. Compl. ¶ 12; Principal Fin. Grp. Orig. Compl. ¶ 17. Moreover, the complaints referred throughout to all of the defendants (except for some interdealer brokers and the BBA) as "Panel Bank Defendants," even when Count Four stated a claim against some of the "Panel Bank Defendants" for breach of contracts.

Significantly, this is not a case in which plaintiffs made a factual mistake as to who their counterparties were or were ignorant of their identity. Rather, any error was a strategic decision to sue panel banks in their capacity as panel banks, rather than counterparties in their capacity as affiliates of panel banks. Rule 15 does not save plaintiffs from this kind of "mistake."

7.1.2. Salix

Salix's original first amended complaint both alleged claims against "Credit Suisse Group AG" in its ostensible capacity as "a member of the USD Libor panel." Salix Compl. ¶ 26; Salix Am. Compl. ¶ 27(d). In fact, Credit Suisse AG was a member of the LIBOR panel. Salix has corrected this error in its second amended complaint by naming Credit Suisse AG as a new defendant. Salix Second Am. Compl. 29(e). Rule 15 is meant to offer a plaintiff reprieve for precisely this kind of factual mistake. The two Credit Suisse entities are closely allied, they have similar names, the first complaint made it clear that Salix intended to sue whichever entity was a panel member, and both defendant entities presumably knew which one was the panel member.

We cannot, and do not, resolve the factual question whether Credit Suisse AG actually knew of Salix's suit and knew that Credit Suisse Group AG had been named as a defendant panelist. On the circumstances presented here, it is entirely plausible that Credit Suisse AG did know this. This is enough for Salix to survive a motion to dismiss.

7.1.3. NCUA

The NCUA originally filed claims against Credit Suisse Group AG, Barclays Bank PLC, and entities associated with the Bank of Tokyo, Barclays, HBOS, JPMorgan, Lloyds, Norinchukin, Rabobank, RBC, RBS, Société Générale, WestImmo, and UBS. Its amended complaint adds claims against "Credit Suisse Group International," Barclays Capital, Inc., and entities associated with Bank of America, Citigroup, Deutsche Bank, and HSBC.

The new claims against Bank of America, Citigroup, Deutsche Bank, and HSBC clearly do not relate back, because none of these entities is remotely associated with any of the original defendants.

According to defendants, the entity "Credit Suisse Group International" does not exist. Moreover, no such entity is mentioned in the amended complaint except in the caption. Therefore, we simply dismiss the complaint against Credit Suisse Group International for failure to state a claim without reaching the relation-back issue.

It appears that Barclays Capital, Inc., was added because some of the unscrupulous swap traders mentioned in Barclays settlement documents worked for Barclays Capital, Inc. The original NCUA complaint is replete with allegations that Barclays's LIBOR submitters modified their submissions at the behest of Barclays's swap traders, and it is likely that at least some of these swap traders worked for Barclays Capital, Inc. Because the NCUA's allegations derive from Barclays's (both Barclays Bank's and Barclays Capital's) own admissions to regulators and prosecutors, Barclays Capital, Inc., was in a position to know that many of the NCUA's allegations referred to misconduct of its own employees.

As with Salix and Credit Suisse, we do not resolve whether Barclays Capital actually knew of the NCUA's suit and knew that the NCUA's suit alleged misconduct by Barclays Capital's employees. It is enough to survive a motion to dismiss that Barclays Capital may plausibly have known this.

7.2. Schwab

As described above, the Schwab Plaintiffs filed a group of cases in 2011. Plaintiffs dropped some of their claims in an amended complaint. Their federal (and state antitrust) claims were dismissed with prejudice in 2013, and their remaining state-law claims were dismissed without prejudice. Approximately one month later, plaintiffs re-filed their dismissed state-law claims along with some new state and federal claims. Under 28 U.S.C. § 1367(d) (2012), the dismissed state-law claims are timely in this second case so long as the dismissed state-law claims were timely in the first set of cases.

Plaintiffs go one step further and ask us to hold that their other claims also somehow "relate back" to their 2011 complaints.

This argument fails. When a federal court declines to exercise supplemental jurisdiction over a timely state-law claim, as we did in LIBOR I, subsection 1367(d) allows such a claim to be re-filed within 30 days. But subsection 1367(d) plainly applies only to pendent state-law claims that are dismissed in this way (or voluntarily dismissed after a dismissal of predicate federal claims). Subsection 1367(d) cannot be read to revive claims that were never filed in the first case or claims that were withdrawn from the first case before a judicial dismissal of federal claims.

Nor does Rule 15 permit relation back. Rule 15 plainly applies to an "amendment to a pleading" and an "original pleading," not to two pleadings in different cases. See Fed. R. Civ. P. 15(c)(1).

8. Application

In this section, we briefly explain how our legal conclusions apply to each of the individual suits. As each case is situated differently, each subsection should be read separately; the charts, defined groups of claims, and analyses of each subsection refer only to the particular case under discussion in that subsection.

We do not analyze every legal and factual claim by every Individual Plaintiff against every defendant in complete detail, but rather indicate in broad strokes how this opinion affects each action. To the extent that we omit to discuss some particular claim, the foregoing discussion nevertheless applies with full force.

Claims that we refer to as "untimely" or "time-barred" in the following discussion fail as time-barred, and this opinion dismisses those claims entirely. However, we use the words "timely" and "not time-barred" to mean only that a claim survives this stage of motion practice—not that a claim will ultimately prove to be timely. We have reserved judgment for a later stage on important factual issues relating to inquiry notice and discovery, and recognize that the parties have declined to thoroughly brief certain aspects of some states' limitations laws.

Some specific claims that we refer to as timely may even be time-barred on the basis of this opinion. For example, we note below that a CEA claim in Amabile arising on April 15, 2009, against Bank of America Corp., is timely on the basis of American Pipe tolling. This comment is generally true, but only if the CEA claim overlaps factually with the original Metzler complaint that provides the basis for tolling. Thus, a trader-based CEA claim would in fact be time-barred, because the Metzler complaint alleged persistent suppression only. The parties should therefore understand that we use "timely" as a shorthand, and that the following discussion must be read in light of the legal analysis above.

We use the word "arise" to indicate the time when a claim came into existence, even if some discovery rule or tolling principle delayed the legal accrual until some later date.

Finally, we assume that each relevant jurisdiction applies a rule that a complaint filed on a Monday is timely even if the statute of limitations expired on the previous weekend. See, e.g., Fed. R. Civ. P. 6(a)(1)(C); Cal. Code Civ. P. § 12(b); Iowa Code Ann. § 4.1(34); N.Y. Gen. Constr. Law § 25-a(1).

8.1. Amabile

The Amabile plaintiffs filed suit in New York on March 13, 2013. Federal statutes of limitations govern their federal claims, while New York law governs their unjust enrichment claim. The dates of plaintiffs' exchange-based claims range from 2005 to 2010.

CEA: The statute of limitations runs for two years, so claims accruing on or after March 13, 2011, are indisputably timely.

The following class-action tolling periods based on the Metzler complaints apply to the groups of claims defined in the following table:

This set of tolling periods imply that, without applying a discovery rule, the following claims are timely through American Pipe tolling: Group B claims arising on or after April 15, 2009; and Group C claims arising on or after April 28, 2010.

Because these plaintiffs traded Eurodollar futures and options, each was on inquiry notice by May 29, 2008, of all claims arising on or after August 9, 2007. The manager of one plaintiff (Michael Rane, for plaintiff 303 Proprietary Trading LLC), has even stated in an affidavit (ECF No. 874) that he was personally aware of news articles regarding LIBOR in the spring of 2008. Plaintiffs were not, however, on inquiry notice of claims arising before August 9, 2007. Therefore, the following claims are time-barred: claims against Group A/B defendants arising between August 9, 2007, and April 14, 2009, or after May 2010; and claims against Group C defendants arising between August 9, 2007, and April 27, 2010 or after May 2010.

Unjust enrichment: The statute of limitations runs for three years, so claims arising on or after March 13, 2010, are indisputably timely.

Class-action tolling applies equally to plaintiffs' unjust enrichment claims as to plaintiffs' CEA claims. However, New York does not apply a discovery rule to unjust enrichment claims, so the following claims are time-barred: claims against group A/B defendants arising before April 15, 2008; and claims against Group C defendants arising before April 28, 2009.

8.2. BATA

BATA filed suit in California on Monday, March 31, 2014, and the parties agree that California law governs. BATA's OTC claims range from August 2007 to May 2010.

Fraud: The statute of limitations runs for three years, so all claims are untimely in the absence of class-action tolling or a discovery rule.

Class-action tolling assists claims against every defendant except Citigroup Financial Products, Inc., the British Bankers' Association, BBA Enterprises Ltd., and BBA Libor Ltd. At least some of BATA's fraud claims were tolled as far back as the following dates on the basis of complaints in Metzler, Baltimore, and other OTC cases:

BATA's complaint (¶¶ 307-11) implies that BATA was on inquiry notice by August 5, 2008. Therefore, claims against Group A defendants are timely, claims against Group B defendants arising on or after April 28, 2009, are timely, and claims against Group C defendants are time-barred.

Breach of contract and unjust enrichment: The statute of limitations runs for four years, so claims accruing on or after March 29, 2010, are indisputably timely.

Class-action tolling applies to at least some claims, extending the limitations period as far back as April 15, 2007.

By May 29, 2008, BATA's injuries were no longer "difficult for the plaintiff to detect," so BATA's contract claims are time-barred to the extent that class-action tolling does not extend the limitations period back to the later of the accrual date or May 29, 2008.

Applying the above chart on class-action tolling, claims against Group A and Group B defendants are timely, while claims against Group C defendants are time-barred.

Tortious interference: The statute of limitations runs for two years, so all claims are untimely in the absence of class-action tolling or a discovery rule. Class-action tolling does not apply to tortious interference claims, and BATA was on inquiry notice of its claims by August 5, 2008. Thus, all tortious interference claims are time-barred.

8.3. California Consolidated

The California Consolidated Plaintiffs filed suits in California between January 9, 2013, and November 13, 2013, and the parties agree that California law governs. The dates of their claims range from 2005 to March 2011.

Fraud and negligent misrepresentation: The statute of limitations runs for three years, so claims accruing on or after the following dates are indisputably timely.

For OTC claims, essentially the same tolling applies to these plaintiffs as to BATA, above. And, like BATA, plaintiffs plead information showing that they were on inquiry notice by August 5, 2008. As a result, essentially the same OTC claims are timely for the California Consolidated Plaintiffs as for BATA.

Plaintiffs' fraud claims relating to bonds and notes were possibly tolled by the original Gelboim complaint (depending on whether each issuer was a Fortune 500 company and whether each bond was underwritten by a defendant), but not by the amended Gelboim complaint (because that complaint excluded bonds issued by a defendant) or by the Berkshire Bank or Directors Financial complaints (because none of the California Consolidated Plaintiffs are financial institutions). Therefore, plaintiffs' bond-related claims are tolled for at most 81 days (from February 9 to April 30, 2012), and the following are time-barred:

Unjust enrichment: The statute of limitations runs for three years, so claims arising on or after the dates listed above for fraud are untimely. The same class-action tolling analysis also applies.

By May 29, 2008, plaintiffs' Period One and Two injuries were no longer "difficult for the plaintiff to detect," so no discovery rule saves plaintiffs' unjust enrichment claims after that date. Thus, unlike fraud claims arising before August 2007, unjust enrichment claims arising before then are all time-barred.

Breach of contract: The statute of limitations runs for four years, so claims accruing on or after the following dates are indisputably timely.

Class-action tolling makes the OTC claims and the following bond claims timely:

By May 29, 2008, plaintiffs' Period One and Two injuries were no longer "difficult for the plaintiff to detect," so no discovery rule saves plaintiffs' unjust enrichment claims after these dates.

Tortious interference: The statute of limitations runs for two years, so claims accruing on or after the following dates are indisputably timely.

Class-action tolling does not apply to tortious interference claims, and plaintiffs were on inquiry notice of Period One and Two claims by August 2008, so all of plaintiffs' tortious interference claims accruing between August 2007 and the above dates are time-barred.

8.4. CEMA

Defendants' Notice of Motion did not seek to dismiss any claims on timeliness grounds, so we will not do so. See Def. Master App.; see also Joint Limitations Spreadsheet (omitting arguments for or against timeliness of CEMA claims).

In any event, many of CEMA's claims, at least against the Royal Bank of Scotland Group PLC, would probably survive, because Ohio recognizes cross-jurisdictional class-action tolling.

8.5. Darby

The Darby Plaintiffs, Darby Financial and Capital Ventures, filed their suit in New York on November 21, 2013. The dates of their claims range from at least August 2007 to the end of 2010.

Both plaintiffs were on inquiry notice of their claims by April 15, 2011, when they allege relying on class actions. See Darby Am. Compl. ¶ 383. This implies that New York's two-year limitations period for fraud began to run by April 15, 2011, and that Pennsylvania's limitations periods began to run by April 15, 2012, as to panel bank defendants, when a diligent inquiry would have uncovered plaintiffs' claims.

8.5.1. Darby Financial

Because Darby Financial's principal place of business is in Pennsylvania, the parties agree that New York's borrowing rule requires Darby Financial's claims to be timely under both New York and Pennsylvania law.

Unjust enrichment: The New York statute of limitations runs for three years and the Pennsylvania statute runs for four, so claims accruing on or after November 21, 2010, are indisputably timely.

At this stage, Pennsylvania's discovery rule makes all claims timely under Pennsylvania law, so New York is the limiting factor. No discovery rule applies to unjust enrichment in New York, so claims arising before November 21, 2010, are time-barred unless preserved by class-action tolling.

The New York statute of limitations was tolled by the OTC class actions, making the following claims time-barred:

Contract: As with unjust enrichment, Pennsylvania's discovery rule makes New York's six-year period the limiting factor at this point. Applying the same class-action tolling as above, the following claims are time-barred:

Fraud: The New York statute of limitations runs for the longer of six years from injury or two years from discovery (imputed for Period One and Two claims as of April 15, 2011), while the Pennsylvania statute runs for two years from discovery (imputed for Period One and Two claims as of April 15, 2012).

Applying the same class-action tolling as above, the following claims are time-barred:

Tortious interference: The New York statute of limitations runs for three years from injury, while the Pennsylvania statute runs for two years from discovery (imputed for Period One and Two claims as of April 15, 2012). As with fraud, it follows that New York's statute of limitations controls the result. Because class-action tolling is unavailable for tortious interference, all claims arising before Nov. 21, 2010, are time-barred.

8.5.2. Capital Ventures

Because Capital Ventures' principal place of business is in the Cayman Islands, the parties agree that New York's borrowing rule requires Capital Ventures' claims to be timely under both New York and Caymanian law. We have not analyzed Caymanian law because defendants have not argued that Caymanian law renders any claims untimely.

Unjust enrichment and tortious interference: The New York analysis for Capital Ventures is the same as for Darby Financial, since New York is the limiting factor for both plaintiffs. Accordingly, the same set of claims are time-barred for Capital Ventures as for Darby.

Contract: It is not clear whether defendants seek to dismiss Capital Ventures' contract claims on timeliness grounds. Defendants' Master Appendix cites only Pennsylvania law, which is not relevant to Capital Ventures, and the Joint Limitations Spreadsheet makes no mention of these claims. Accordingly, we will not dismiss any of Capital Ventures' contract claims.

8.6. Fannie Mae

Fannie Mae filed suit in New York on October 31, 2013. Applying New York's borrowing rule, Fannie Mae's claims must be timely under both New York and District of Columbia law. The dates of Fannie Mae's claims range from at least August 2007 to 2010.

Fannie Mae has not advocated class-action tolling as a defense to the statute of limitations, and the parties have not briefed the District of Columbia's tolling law. Accordingly, we do not consider whether class-action tolling applies to Fannie Mae's claims.

Fraud: New York's statute of limitations runs for the longer of six years from injury or two years from discovery, while the District of Columbia's runs for three years from discovery. Thus, claims arising on or after October 31, 2010, are indisputably timely.

It is difficult to believe that an institutional entity tasked with purchasing and guaranteeing residential mortgages did not inform itself of readily available information regarding a critical ingredient of many of the adjustable-rate mortgages in its portfolio. Nevertheless, at the present stage, we cannot discern from the pleadings whether Fannie Mae learned of articles relating to LIBOR before October 31, 2010, so none of Fannie Mae's claims are time-barred.

Contract: New York's statute of limitations runs for six years, and the District of Columbia's for three. Thus, claims arising on or after October 31, 2010, are indisputably timely.

New York does not apply a discovery rule to contract claims, so claims accruing before October 31, 2007, are time-barred by the New York limitations rule.

Unjust enrichment: New York's and the District of Columbia's statutes of limitations both run for three years, so that claims arising on or after October 31, 2010, are timely.

New York does not apply a discovery rule to unjust enrichment claims, so claims arising before October 31, 2010, are time-barred.

8.7. FDIC

The FDIC filed suit in New York on March 14, 2014, in its capacity as receiver for thirty-eight failed banks in several states and Puerto Rico. The parties agree that the FDIC's extender statute (12 U.S.C. § 1821(d)(14) (2012)) governs. Each failed bank was based outside New York so, to the extent that state law is relevant, we apply the more restrictive limitations doctrine of New York or the bank's home state. The relevant appointment dates and home states are as follows, organized by state:

The FDIC complaint does not contain a clear, simple statement of the range of time during which the FDIC alleges that defendants manipulated LIBOR. It appears that the FDIC's claims relate to persistent suppression between "summer of 2007" (which we take to mean August 9, 2007) and 2011. See FDIC Am. Compl. ¶ 122.

To assess whether a particular claim by a particular failed bank against a particular defendant is timely, we take the following steps:

1. If the claim is for negligent misrepresentation (not an intentional tort), and the claim arose before the FDIC's appointment date, then we assess whether the claim was still timely on the appointment date under state law. If not, then we conclude our analysis and consider the claim time-barred. Otherwise, we continue.

2. We mark the starting date for the statute of limitations as the later of the accrual date or the appointment date.

3. We calculate whether the claim is timely under the extender statute's fixed-length limitations period. To do so, we apply the federal "intermediate inquiry notice" rule and American Pipe tolling. If the claim is timely according to the fixed-length limitations period, then we conclude our analysis and consider the claim timely. Otherwise, we continue to step 4.

4. We calculate whether the claim is timely under both New York law and the law of the failed bank's home state. To do so, we apply each state's discovery rule and class-action tolling doctrine separately. If the claim is timely under both laws, then we consider the claim timely; if the claim is untimely under either law, then we consider the claim untimely.

Revival: Each failed bank's intentional tort claims (including its unjust enrichment claims) were timely within five years of the corresponding appointment date. Accordingly, every claim on behalf of a failed bank is deemed to have accrued no later than the FDIC's corresponding appointment date.

Class-action tolling: The FDIC states OTC-related claims, so several class actions are relevant to the tolling analysis. Class-action tolling does not apply to tortious interference, to other factual claims outside the scope of class complaints, or to the limitations law of certain states. To the extent that class-action tolling applies, the original Metzler complaint and the first amended Baltimore complaint toll at least some claims in the following time ranges:

Discovery rule: The limitations period began by May 29, 2008, in California as to unjust enrichment, in Kansas as to torts other than fraud, and in Texas, because LIBOR manipulation was ascertainable by then.

The FDIC's amended complaint (¶¶ 298-301) implies that each failed bank knew of the BBA's denials of LIBOR manipulation between April and August 5, 2008, so that each failed bank was on inquiry notice of Period One and Two claims by August 5, 2008. This knowledge is sufficient to start the clock on August 5, 2008, under an "intermediate inquiry notice" rule (federal and New York), and under a "strong inquiry notice" rule (California tort claims). This is also enough to start the limitations by August 5, 2009 (one year after inquiry notice), in states that apply "weak inquiry notice" rules, such as Kansas (as to fraud). In the absence of specific contrary briefing on the discovery rules of each bank's home state, we assume at this stage that each of the remaining jurisdictions would also apply the plaintiff-friendly "weak inquiry notice" rule.

Fraud: New York's six-year statute of limitations for fraud (the longer of six years from injury or two years from discovery) is unusually long, and therefore the home state's statute is the limiting factor for most claims. The one exception occurs when a home state's discovery rule is more permissive than New York's and would preserve a claim for over six years. In that case, because each bank was on inquiry notice in New York by August 5, 2008, no claim survives more than six years under New York law. To reiterate, all of the FDIC's fraud claims are time-barred after six years regardless of the home state's discovery rule, unless saved by class-action tolling.

Applying the borrowing rule, the extender statute, and each state's discovery rule, we calculate that the effective length of time that applies to banks based in each state:

Applying these effective lengths and the foregoing discussion of class-action tolling, we conclude that the following claims are time-barred: