UNITED STATES DISTRICT COURT
SOUTHERN DISTRICT OF NEW YORK
IN RE:
SSA BONDS ANTITRUST LITIGATION
This Document Relates To All Actions
1:16-cv-03711-ER
CLASS ACTION PLAINTIFFS’ CONSOLIDATED OPPOSITION
TO DEFENDANTS’ MOTIONS TO DISMISS
FOR FAILURE TO STATE A CLAIM
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 1 of 155
i
TABLE OF CONTENTS
TABLE OF CONTENTS ................................................................................................................. i
TABLE OF AUTHORITIES ...........................................................................................................v
PRELIMINARY STATEMENT .....................................................................................................1
ARGUMENT ...................................................................................................................................6
I. THE COMPLAINT PLEADS A PLAUSIBLE CONSPIRACY CLAIM
AGAINST EACH DEFENDANT .......................................................................................7
A. Defendants Mischaracterize the Relevant Legal Standards .....................................7
B. The SAC Contains Direct Evidence of an Agreement Among Defendants
Not to Compete, Including Direct Evidence of Price-Fixing ..................................9
1. The direct evidence obtained to date is clear-cut and incriminating ...........9
2. Defendants’ attempts to mischaracterize the chats fail ..............................13
3. The chats are not innocuous exchanges of “market color” ........................15
C. The SAC Also Contains Abundant Circumstantial Evidence and “Plus
Factors” ..................................................................................................................19
1. Defendants engaged in parallel conduct ....................................................19
2. Defendants shared a common motive to conspire .....................................21
3. ............21
4. Defendants engaged in acts against economic self-interest .......................23
5. The anomalous behavior of bid-ask spreads and yields on USD
SSA bonds during the conspiracy period is further evidence of a
conspiracy ..................................................................................................24
6. Government investigations and adverse personnel decisions also
constitute “plus” factors .............................................................................26
D. Defendants’ Miscellaneous Attacks Fail ...............................................................28
1. Plaintiffs allege a plausible mechanism to implement the
conspiracy ..................................................................................................28
2. That not every Defendant was on every chat makes no difference ...........33
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 2 of 155
ii
3. Defendants’ speculation that they may have had held different
positions in USD SSA bonds is irrelevant .................................................35
4. The existence of non-Defendant USD SSA bond dealers is
immaterial ..................................................................................................36
5. The precise length of each Defendant’s participation in the
conspiracy is a question for discovery .......................................................38
6. Defendants’ conspiracy was in effect by at least January 1, 2009 .............39
II. NAMED PLAINTIFFS HAVE PLAUSIBLY PLED THEY WERE INJURED
BY THE CONSPIRACY NOT TO COMPETE ................................................................42
A.
.....................................................................................................42
B. As Predicted by the Academic Literature, the Reduction of Competition
Demonstrably Widened Spreads Beyond the Specific Transactions in the
Chats to Date ..........................................................................................................47
1. A summary of the SAC’s statistical analyses ............................................48
2. The SAC is sufficiently detailed ................................................................54
3. The analyses reasonably relate to Plaintiffs’ claims ..................................57
4. The time periods used in the analyses are consistent and proper ..............58
5. The analyses all reasonably relate to the presence of a conspiracy ...........63
6. Defendants’ remaining arguments are unavailing .....................................67
III. PLAINTIFFS HAVE ANTITRUST STANDING AS DEFENDANTS’ DIRECT
CUSTOMERS....................................................................................................................69
A. Plaintiffs Present a Clear Case of Antitrust Injury.................................................70
B. Plaintiffs Are Efficient Enforcers of the Antitrust Laws .......................................73
1. Plaintiffs’ injuries are direct.......................................................................73
2. No more efficient enforcers exist ...............................................................74
3. Plaintiffs’ injuries are not speculative........................................................75
4. There will be no duplicate recoveries or complex apportionment of
damages......................................................................................................76
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 3 of 155
iii
IV. AMPLE EVIDENCE LINKS EACH DEFENDANT TO THE CONSPIRACY ..............77
A. The Barclays Defendants .......................................................................................79
1. The SAC identifies the specific role of each Barclays Defendant .............79
2. Barclays cannot explain away the evidence of its wrongdoing .................80
B. The BNP Paribas Defendants.................................................................................84
1. The SAC identifies the specific role of each BNP Paribas
Defendant ...................................................................................................84
2. The evidence of BNP Paribas’ involvement in the conspiracy is
clear ............................................................................................................85
C. The Citi Defendants ...............................................................................................86
1. The SAC identifies the specific role of each Citi Defendant .....................86
2. Absence from certain collusive communications does not imply
non-participation in the conspiracy ............................................................88
D. Crédit Agricole Corporate & Investment Bank. ....................................................92
1. Crédit Agricole’s role in the conspiracy ....................................................93
2. The evidence of Credit Agricole’s involvement in the conspiracy is
clear ............................................................................................................93
E. The Credit Suisse Defendants ................................................................................97
1. The SAC identifies the specific role of each Credit Suisse
Defendant ...................................................................................................98
2. Credit Suisse cannot explain away the evidence alleged in the SAC ........99
F. The Nomura Defendants ......................................................................................103
1. The SAC identifies the specific role of each Nomura Defendant ............103
2. NIPLC’s apparent expectation that Plaintiffs should already have
all the pertinent evidence is misplaced ....................................................104
3. Contrary to NIPLC’s assertion, the SAC alleges multiple instances
where NIPLC traded a manipulated bond ................................................105
4. NIPLC’s attempts to discount the powerful evidence against it fail .......107
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 4 of 155
iv
G. The RBC Defendants ...........................................................................................111
1. The SAC identifies the specific role of each RBC Defendant .................111
2. The RBC Defendants participated in the conspiracy ...............................113
3. Pleading pre-July 2009 actions is proper .................................................116
H. The TD Bank Defendants ....................................................................................116
1. The SAC identifies the specific role of each TD Bank Defendant ..........116
2. The TD Bank Defendants’ improper attempt to explain away the
evidence fails ...........................................................................................117
I. Individual Defendant Bhardeep Heer ..................................................................122
1. The SAC includes direct evidence of Heer engaging in
anticompetitive conduct ...........................................................................122
2. The SAC sufficiently alleges circumstantial evidence of Heer’s
participation in the conspiracy .................................................................128
J. Individual Defendant Amandeep Singh Manku ..................................................131
K. Individual Defendant Gary McDonald ................................................................132
1. Absence from certain collusive communications does not imply
non-participation in the conspiracy ..........................................................132
2. The SAC sufficiently alleges direct evidence of McDonald
engaging in anticompetitive conduct .......................................................133
3. McDonald’s inactive status on the FCA register supports a
plausible inference of wrongdoing...........................................................139
4. Plaintiffs have standing to sue McDonald ...............................................139
L. Individual Defendant Shailen Pau .......................................................................140
CONCLUSION ............................................................................................................................141
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 5 of 155
v
TABLE OF AUTHORITIES
Page
Cases
7 W. 57th St. Realty Co. v. Citigroup, Inc.,
2015 WL 1514539 (S.D.N.Y. Mar. 31, 2015) .......................................................................... 25
Alaska Elec. Pension Fund v. Bank of Am. Corp.,
175 F. Supp. 3d 44 (S.D.N.Y. 2016) ................................................................ 25, 26, 57, 64, 75
All Corp. v. Cast Iron Soil Pipe Inst.,
851 F.2d 478 (1st Cir. 1988) ................................................................................................... 135
Allen v. Int’l Bus. Machs. Corp.,
1997 WL 34501372 (D. Del. Dec. 18, 1997) .......................................................................... 66
Am. Column & Lumber Co. v. United States,
257 U.S. 377 (1921) .................................................................................................................. 16
Anderson News, L.L.C. v. Am. Media, Inc.,
680 F.3d 162 (2d Cir. 2012) .............................................................................................. passim
Apex Oil Co. v. DiMauro,
822 F. 2d 246 (2nd Cir. 1987) ........................................................................................... passim
Arista Records LLC v. Doe 3,
604 F.3d 110 (2d Cir. 2010) ....................................................................................... 26, 40, 101
Bell Atlantic Corp. v. Twombly,
550 U.S. 544 (2007) ............................................................................................................ 6, 7, 8
Beltz Travel Serv., Inc. v. Int’l Air Transp. Ass’n,
620 F.2d 1360 (9th Cir. 1980) .................................................................................................. 88
Bigelow v. RKO Radio Pictures,
327 U.S. 251 (1946) .................................................................................................................. 76
Bilinski v. Keith Haring Found., Inc.,
96 F. Supp. 3d 35 (S.D.N.Y. 2015) ........................................................................................ 128
Blomkest Fertilizer, Inc. v. Potash Corporation of Saskatchewan,
203 F.3d 1028 (8th Cir. 2000) ................................................................................................ 134
Blue Shield of Virginia v. McCready,
457 U.S. 465 (1982) ...................................................................................................... 70, 73, 74
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 6 of 155
vi
Branca v. Nordstrom, Inc.,
2015 WL 10436858 (S.D. Cal. Oct. 9, 2015) ........................................................................... 59
Broomfield v. Craft Brew Alliance,
2017 WL 3838453 (N.D. Cal. Sept. 1, 2017) ........................................................................... 62
Buchanan v. Tata Consultancy Services, Ltd.,
2017 WL 6611653 (N.D. Cal. Dec. 27, 2017) .......................................................................... 66
Burgis v. N.Y.C. Dep’t of Sanitation,
798 F.3d 63 (2d Cir. 2015) ....................................................................................................... 68
Chasins v. Smith, Barney & Co.,
438 F.2d 1167 (2d Cir. 1970) ................................................................................................. 120
Chubirko v. Better Business Bureau of South Piedmont, Inc.,
763 F. Supp. 2d 759 (W.D.N.C. 2011) ..................................................................................... 31
Cinema Village Cinemart, Inc. v. Regal Entertainment Group,
2016 WL 5719790 (S.D.N.Y. Sept. 29, 2016)........................................................................ 109
City of Royal Oak Retirement System v. Juniper Networks, Inc.,
2013 WL 2156358 (N.D. Cal. May 17, 2013) .......................................................................... 55
Colliton v. Cravath, Swaine & Moore LLP,
2008 WL 4386764 (S.D.N.Y. Sept. 24, 2008).......................................................................... 62
Concord Assocs., L.P. v. Entm’t Props. Trust,
2014 WL 1396524 (S.D.N.Y. Apr. 9, 2014) .................................................................... 78, 120
Connolly v. Wood-Smith,
2012 WL 7809099 (S.D.N.Y. May 14, 2012) ........................................................................ 110
Cont’l Ore Co. v. Union Carbide & Carbon Corp.,
370 U.S. 690 (1962) .............................................................................................................. 8, 34
Crimpers Promotions Inc. v. Home Box Office, Inc.,
724 F.2d 290 (2d Cir. 1983) ..................................................................................................... 72
Delgado v. Ocwen Loan Servicing, LLC,
2014 WL 4773991 (E.D.N.Y. Sept. 24, 2014) ......................................................................... 79
Deutsch v. Novartis Pharms. Corp.,
768 F. Supp. 2d 420 (E.D.N.Y. 2011) ...................................................................................... 66
Dover v. British Airways, PLC (UK),
2014 WL 317845 (E.D.N.Y. Jan. 24, 2014) ....................................................................... 54, 59
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 7 of 155
vii
Engineering Contractors Association of South Florida Inc. v. Metropolitan Dade
County,
122 F.3d 895 (11th Cir. 1997) .................................................................................................. 57
Five Smiths, Inc. v. National Football League Players Association,
88 F. Supp. 1042 (D. Minn. 1992) ............................................................................................ 18
FrontPoint Asian Event Driven Fund, L.P. v. Citibank, N.A.,
2017 WL 3600425 (S.D.N.Y. Aug. 18, 2017) .......................................................................... 67
FrontPoint Asian Event Driven Fund, L.P. v. Citibank, N.A.,
2018 WL 4830087 (S.D.N.Y. Oct. 4, 2018) ....................................................................... 28, 67
Gelboim v. Bank of Am. Corp.,
135 S. Ct. 897 (2015) ................................................................................................................ 62
Gelboim v. Bank of Am. Corp.,
823 F.3d 759 (2d Cir. 2016) ............................................................................. 21, 25, 62, 70, 73
Glob. Network Commc’ns, Inc. v. City of New York,
458 F.3d 150 (2d Cir. 2006) ............................................................................................... 28, 43
Harris v. City of N.Y.,
186 F.3d 243 (2d Cir. 1999) ..................................................................................................... 61
Harry v. Total Gas & Power North America, Inc.,
244 F. Supp. 3d 402 (S.D.N.Y. 2017) ...................................................................................... 58
Harry v. Total Gas & Power North America, Inc.,
889 F.3d 104 (2d Cir. May 4, 2018) ........................................................................................ 73
Heckler & Koch, Inc. v. German Sport Guns GmbH,
2014 WL 12756372 (S.D. Ind. May 15, 2014) ......................................................................... 59
Hinds Cty., Miss. v. Wachovia Bank N.A.,
700 F. Supp. 2d 378 (S.D.N.Y. 2010) .............................................................................. 12, 132
Hinds Cty., Miss. v. Wachovia Bank N.A.,
708 F. Supp. 2d 348 (S.D.N.Y. 2010) ............................................................................... 22, 68
IHS Dialysis Inc. v. Davita, Inc.,
2013 WL 1309737 (S.D.N.Y. Mar. 31, 2013) ................................................................ 106, 122
Illinois Brick Co. v. Illinois,
431 U.S. 720 (1977) ............................................................................................................ 69, 74
In re Actos End-Payor Antitrust Litig.,
848 F.3d 89 (2d Cir. 2017) ..................................................................................................... 120
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 8 of 155
viii
In re Air Cargo Shipping Servs. Antitrust Litig.,
2010 WL 10947344 (E.D.N.Y. Sept. 22, 2010) ....................................................................... 42
In re Air Cargo Shipping Servs. Antitrust Litig.,
2014 WL 7882100 (E.D.N.Y. Oct. 15, 2014) ........................................................................... 66
In re Aluminum Warehousing Antitrust Litig.,
833 F.3d 151 (2d Cir. 2016) ............................................................................................... 71, 72
In re Chocolate Confectionary Antitrust Litig.,
801 F.3d 383 (3d Cir. 2015) ..................................................................................................... 22
In re Commodity Exch. Inc. Silver Futures and Options Trading Litig.,
2013 WL 1100770 (S.D.N.Y. Mar. 18, 2013) .................................................................... 57, 67
In re Commodity Exchange, Inc. Antitrust Litig.,
213 F. Supp. 3d 631 (S.D.N.Y. 2016) .......................................... 24, 26, 42, 54, 57, 64, 96, 139
In re Commodity Exchange, Inc., Gold Futures & Options Trading Litig.,
328 F. Supp. 3d 217 (2018) ...................................................................................................... 25
In re Credit Default Swaps Antitrust Litig.,
2014 WL 4379112 (S.D.N.Y. Sep. 4, 2014) ................................................................. 73, 74, 79
In re DDAVP Direct Purchaser Antitrust Litig.,
585 F.3d 677 (2d Cir. 2009) ..................................................................................................... 75
In re Domestic Airlines Travel Antitrust Litig.,
221 F. Supp. 3d 46 (D.D.C. 2016) ............................................................................................ 25
In re Elec. Books Antitrust Litig.,
859 F. Supp. 2d 671 (S.D.N.Y. 2012) ................................................ 13, 14, 32, 42, 78, 89, 113
In re Elevator Antitrust Litigation,
2006 WL 1470994 (S.D.N.Y. May 30, 2006) ........................................................................ 113
In re Elevator Antitrust Litig.,
502 F.3d 47 (2d Cir. 2007) ....................................................................................................... 83
In re Flat Glass Antitrust Litig.,
385 F.3d 350 (3d Cir. 2004) ............................................................................................. 21, 128
In re Foreign Exch. Benchmark Rates Antitrust Litig.
74 F. Supp. 3d 581 (S.D.N.Y. 2015) ............................................................ 7, 8, 21, 70, 74, 106
In re Foreign Exch. Benchmark Rates Antitrust Litig.,
2016 WL 5108131 (S.D.N.Y. Sept. 20, 2016)................................................................... passim
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 9 of 155
ix
In re High Fructose Corn Syrup Antitrust Litig.,
295 F.3d 651 (7th Cir. 2002) .................................................................................................... 32
In re High-Tech Emp. Antitrust Litig.,
856 F. Supp. 2d 1103 (N.D. Cal. 2012) ........................................................................ 20, 34, 88
In re Iconix Brand Group, Inc.,
2017 WL 4898228 (S.D.N.Y. Oct. 25, 2017) ........................................................................... 28
In re Ins. Brokerage Antitrust Litig.,
618 F.3d 300 (3d Cir. 2010) ............................................................................................... 9, 129
In re Interest Rate Swaps Antitrust Litig.,
261 F. Supp. 3d 430 (S.D.N.Y. 2017) .......................................................................... 28, 44, 73
In re LIBOR-Based Fin. Instruments Antitrust Litig. (LIBOR I),
935 F. Supp. 2d 666 (S.D.N.Y. 2013) ...................................................................................... 54
In re LIBOR-Based Fin. Instruments Antitrust Litig.,
27 F. Supp. 3d 447 (S.D.N.Y. 2014) ........................................................................................ 62
In re Lithium Ion Batteries,
2014 WL 309192 (N.D. Cal. Jan. 21, 2014) ............................................................................. 39
In re London Silver Fixing Ltd., Antitrust Litig.,
213 F. Supp. 3d 530 (S.D.N.Y. 2016) .................................................................... 24, 26, 54, 64
In re Nasdaq Market-Makers Antitrust Litig.,
894 F. Supp. 703 (S.D.N.Y. 1995) ........................................................................................... 25
In re Nasdaq Market-Makers Antitrust Litig.,
169 F.R.D. 493 (S.D.N.Y. 1996) .............................................................................................. 34
In re NYSE Specialists Securities Litig.,
503 F.3d 89 (2d Cir. 2007) ....................................................................................................... 55
In re Optical Disk Drive Antitrust Litig.,
2011 WL 3894376 (N.D. Cal. Aug. 3, 2011) ........................................................................... 35
In re Pfizer Inc. Secs. Litig.,
584 F. Supp. 2d 621 (S.D.N.Y. 2008) ...................................................................................... 59
In re Polyurethane Foam Antitrust Litig.,
152 F. Supp. 3d 968 (N.D. Ohio 2015)..................................................................................... 88
In re Propranolol Antitrust Litig.,
249 F. Supp. 3d 712 (S.D.N.Y. 2017) ...................................................................................... 42
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 10 of 155
x
In re Tableware Antitrust Litig.,
363 F. Supp. 2d 1203 (N.D. Cal. 2005) .................................................................................... 26
In re TFT-LCD (Flat Panel) Antitrust Litig.,
267 F.R.D. 583 (N.D. Cal. 2010) .................................................................................. 20, 34, 97
In re Vitamins Antitrust Litig.,
2000 WL 1475705 (D.D.C. May 9, 2000) .................................................................................. 8
In re Zinc Antitrust Litig.,
155 F. Supp. 3d 337 (S.D.N.Y. 2016)........................................................................................... 78
J. Truett Payne Co., Inc. v. Chrysler Motors Corp.,
451 U.S. 557 (1981) .................................................................................................................. 76
John v. Whole Foods Mkt. Grp., Inc.,
858 F. 3d 732 (2d Cir. 2017) .................................................................................................... 59
Kelley v. Crosfield Catalysts,
135 F.3d 1202 (7th Cir. 1998) .................................................................................................. 62
LaFlamme v. Societe Air France,
702 F.Supp.2d 136 (E.D.N.Y. 2010) .............................................................................. 106, 122
Lexmark Int’l, Inc. v. Static Control Components, Inc.,
134 S. Ct. 1377, 1392 (2014) .................................................................................................... 75
Lotes Co. v. Hon Hai Precision Indus. Co.,
753 F.3d 395 (2d Cir. 2014) ............................................................................................... 73, 74
Major League Baseball Props., Inc. v. Salvino, Inc.,
542 F.3d 290 (2d Cir. 2008) ..................................................................................................... 17
Mayor & City Council of Baltimore v. Citigroup, Inc.
709 F.3d 129 (2d Cir. 2013) ................................................................................... 7, 8, 125, 135
Merced Irrigation District v. Barclays Bank PLC,
165 F. Supp. 3d 122 (S.D.N.Y. 2016) .................................................................................... 122
Mooney v. AXA Advisors, L.L.C.,
19 F. Supp. 3d 486 (S.D.N.Y. 2014) .............................................................................. 137, 138
Munoz v. Orr,
200 F.3d 291 (5th Cir. 2000) .................................................................................................... 66
Ochre LLC v. Rockwell Architecture Planning & Design, P.C.,
2012 WL 6082387 (S.D.N.Y. Dec. 3, 2012) ............................................................................ 79
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 11 of 155
xi
Palm Beach Strategic Income, LP v. Salzman,
457 F. App’x 40 (2d Cir. 2012) ................................................................................................ 62
Palmer v. BRG of Georgia, Inc.,
498 U.S. 46 (1990) .................................................................................................................... 17
Perry v. NYSARC Inc.,
424 Fed. App’x 23 (2d Cir. 2011) ............................................................................................ 62
Precision Assocs., Inc. v. Panalpina World Transp. (Holding) Ltd.,
2011 WL 7053807 (E.D.N.Y. Jan. 4, 2011) ............................................................................. 78
Prime Healthcare Services, Inc. v. Service Employees International Union,
2013 WL 3873074 (S.D. Cal. July 25, 2013), .......................................................................... 20
Rabin v. NASDAQ OMX PHLX LLC,
182 F. Supp. 3d 220 (E.D. Pa. 2016) ...................................................................................... 100
RxUSA Wholesale, Inc. v. Alcon Labs., Inc.,
661 F. Supp. 2d 218 (E.D.N.Y. 2009) .................................................................................... 123
S.E.C. v. Diversified Corp. Consulting Grp.,
378 F.3d 1219 (11th Cir. 2004) .............................................................................................. 129
SD3, LLC v. Black & Decker (U.S.) Inc.,
801 F.3d 412 (4th Cir. 2015) .................................................................................................. 129
Shields v. Citytrust Bancorp, Inc.,
25 F.3d 1124 (2d Cir. 1994) ..................................................................................................... 62
Sky Angel U.S., LLC v. Nat’l Cable Satellite Corp.,
947 F. Supp. 2d 88 (D.D.C. 2013) ............................................................................................ 22
Sonterra Capital Master Fund Ltd. v. Credit Suisse Group AG,
277 F. Supp. 3d 521 (S.D.N.Y. 2017) ...................................................................... 35, 129, 132
Starr v. Sony BMG Music Entm’t,
592 F.3d 314 (2d Cir. 2010) ................................................................................. 23, 26, 42, 130
State of New York v. Cedar Park Concrete Corp.,
665 F. Supp. 238 (S.D.N.Y. 1987) ................................................................................... 96, 139
State of New York v. Hendrickson Bros., Inc.,
840 F.2d 1065 (2d Cir. 1988) ................................................................................................... 70
Sullivan v. Barclays PLC,
2017 WL 685570 (S.D.N.Y. Feb. 21, 2017) ............................................................... 20, 34, 132
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 12 of 155
xii
Superior Offshore Int’l, Inc. v. Bristow Grp. Inc.,
738 F. Supp. 2d 505 (D. Del. 2010) ................................................................................ 108, 125
Texaco Inc. v. Dagher,
547 U.S. 1 (2006) ...................................................................................................................... 17
Todd v. Exxon Corp.,
275 F.3d 191 (2d Cir. 2001) ..................................................................................................... 16
Ulrich v. Moody’s Corp.,
2014 WL 12776746 (S.D.N.Y. Mar. 31, 2014) ...................................................................... 138
United States v. Container Corp. of Am.,
393 U.S. 333 (1969) ............................................................................................................ 16, 71
United States v. Huezo,
546 F.3d 174 (2d Cir. 2008) ..................................................................................................... 78
United States v. SKW Metals & Alloys, Inc.,
195 F.3d 83 (2d Cir. 1999) ..................................................................................................... 138
United States v. U.S. Gypsum Co.,
438 U.S. 422 (1978) .......................................................................................................... 18, 135
Universal Grading Service v. eBay, Inc.,
2012 WL 70644 (N.D. Cal. Jan. 9, 2012) ................................................................................. 38
Other Authorities
3 Moore’s Federal Practice
§ 15.17 (3d ed. 2018) ................................................................................................................ 62
54 Am. Jur. 2d, Monopolies and Restraints of Trade
§ 45 (2017) ................................................................................................................................ 16
FED. TRADE COMM’N & DEP’T OF JUSTICE,
Antitrust Guidelines for Collaborations Among Competitors 15-16 (2000) ............................ 16
Phillip Areeda and Herbert Hovenkamp,
Antitrust Law: An Analysis of Antitrust Principles and Their Application (4th ed.
2017) ................................................................................................. 16, 17, 69, 71, 89, 133, 137
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 13 of 155
1
PRELIMINARY STATEMENT
The fallout from Defendants’ collusion to rig the trading of U.S.-dollar denominated SSA
bonds to investors continues to grow. Just recently, the European Commission issued Statements
of Objections to four banks regarding their participation in this conspiracy.1 All four targeted
banks are defendants in this case, and one is reportedly actively cooperating with the
Commission in the hopes of avoiding a fine. Plaintiffs have already entered into proposed
settlements in this case with three Defendants for nearly $100 million. For these and many other
reasons, Defendants’ conspiracy is not simply “plausible” but highly probable.
In its order dismissing Plaintiffs’ prior complaint, the Court concluded that Plaintiffs had
not pled the standing of the named Plaintiffs with sufficient particularity. The Court
“highlight[ed] the precise defects of Plaintiffs’ pleading” as a failure to plausibly allege that the
named Plaintiffs “themselves were injured by the alleged conspiracy.” Dkt. No. 495 (“Order”) at
18, 19. Specifically, the Court observed that “plaintiffs failed to allege any specific transactions
that they entered into that harmed them through the defendants’ misconduct” or any “statistical
analysis” showing that Defendants’ collusion would have impacted their trades. Id. at 15, 17. In
accordance with the Court’s analysis, the Second Consolidated Amended Class Action
Complaint (Dkt. No. 506) (“SAC” or “amended complaint”) now does both.
First, the SAC adds a plaintiff whose victimhood is directly reflected in the pre-discovery
chats.
1 Declaration of Daniel L. Brockett (“Brockett Decl.”), Ex. 1, European Commission, Press Release,
Antitrust: Commission sends Statement of Objections in US Dollar supra-sovereign, sovereign and agency bond
trading cartel (Brussels, 20 December 2018), http://europa.eu/rapid/press-release_IP-18-6895_en.htm.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 14 of 155
2
A trade that Pau made the very next day confirms that ADR was harmed.
2 The lower the bond’s yield, the higher its price, and vice versa. SAC ¶ 141.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 15 of 155
3
Second, the SAC describes 16 different statistical studies demonstrating that Defendants’
conspiracy impacted the entire Class—including named Plaintiffs—in the form of less-favorable
prices (higher spreads). As the Court invited Plaintiffs to do, these analyses all confirm that the
“unsurprising” prediction made in the academic literature—that a conspiracy not to compete
would lead to higher spreads across the board—was in fact “borne out” in this market. The data
shows such widespread impact during the core conspiracy period even after controlling for
macroeconomic changes across time.
Plaintiffs have cured the only deficiency the Court identified in the precise manner
implied by the Court’s reasoning. Nonetheless, Defendants’ briefs regurgitate the same
arguments they made previously. Notably, these arguments appear to have been copied from the
defendants’ playbook in the In re Foreign Exchange Benchmark Rates Antitrust Litigation, 2016
WL 5108131, at *3 (S.D.N.Y. Sept. 20, 2016) (“FX”), where Judge Schofield roundly rejected
them.
Defendants’ primary argument is once again to mischaracterize the SAC. Defendants fail
to address what the SAC actually alleges, and instead devote pages to attacking a straw man.
Specifically, they treat the SAC as alleging and supporting only a series of discrete and
disconnected acts to fix, rig, or manipulate the individual price of “every single” USD SSA bond
transaction that occurred anywhere around the world. Having set up that straw man, Defendants
3 The SAC also alleges other newly-added facts regarding transactions involving named Plaintiffs. See
Section II.A, infra.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 16 of 155
4
then argue it is implausible. How, they ask, could Defendants literally fix the price of every
single USD SSA bond transaction anywhere around the globe?
Of course, the SAC does not allege that Defendants fixed the price of every single USD
SSA bond transaction. Instead, the SAC alleges that Defendants agreed not to compete with
each other on their own USD SSA bond transactions. There is nothing implausible about that
agreement, which is amply supported by the mountain of evidence
and 16 statistical analyses. Nor is there anything mysterious about how
Defendants put this agreement into effect:
Defendants’ repeated suggestion that the SAC alleges some sort of “global” or
“sprawling” conspiracy is just as disingenuous. This case is not “sprawling.” It is about those
specific USD SSA bond transactions in which the cartel members engaged. The proposed class
of Plaintiffs here does not comprise every USD SSA bond investor on the planet. It is limited to
those unfortunate U.S.-based investors who transacted directly with these Defendants. Contrary
to Defendants’ claims that this is some implausibly sprawling or amorphous case, it is squarely
focused on those purchasers and sellers of USD SSA bonds who dealt directly with the cartel
members.
In FX, Judge Schofield rejected the same attempt to transform the case into a series of
supposedly one-off mini-conspiracies. There, the defendants argued—in the context of an over-
the counter-market—that “[t]he facts . . . can at most support an inference that different
combinations of traders at different groups of banks sporadically carried out a series of small
conspiracies . . .” FX, 2016 WL 5108131 at *3. Judge Schofield dismissed this argument:
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 17 of 155
5
The [complaint] adequately alleges a Section 1 conspiracy among Defendants in
the FX market based on, among other things, the sharing of market-sensitive, non-
public information in chat rooms and instant messages . . . In support of its
allegation of a broader ‘overarching’ conspiracy . . ., the [complaint] alleges that
each Defendant . . . participated in chat rooms discussing multiple currencies.
“Defendants’ top-level traders” used chat rooms, instant messages and emails to
coordinate their conspiracy.
Id.
So too here.
When combined with Plaintiffs’ statistical analyses,
the damning chat examples discussed in the complaint make an even stronger case for the
plausibility of an “overarching” conspiracy than in FX.
Beyond attacking a straw man, Defendants yet again spend pages trying to thrust
innocent interpretations upon the damning chats by asking the Court to draw every inference in
their favor. That request is plainly improper; it is Plaintiffs who must be given every reasonable
inference. But it is also striking how fruitless Defendants’ efforts are. Their strained attempts to
portray their chats as innocuous “market color” fall flat in every instance. The most
straightforward reading of the chats support the SAC’s allegations.
where
Judge Schofield again rejected the same tactic. FX, 2016 WL 5108131 at *4 (“[A] fair reading
of the chats quoted in the [complaint] . . . suggests that more than a mere harmless exchange of
information was occurring. For example, a chat where a [defendant] trader asks others what they
believe a spread should be plausibly suggests that the traders in the room are colluding . . .”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 18 of 155
6
Defendants also attempt to downplay the investigations into their USD SSA bond trading
by the U.S. Department of Justice and the European Commission. But the Court may take
judicial notice of the fact that, just six weeks ago, the Commission announced that it had issued
Statements of Objection to four banks expressing “its preliminary view that they have breached
EU antitrust rules by colluding, in periods from 2009 to 2015, to distort competition in secondary
market trading in the [European Economic Area] of” USD SSA bonds. They did so by
“exchang[ing] commercially sensitive information and coordinat[ing] on prices . . . mainly
through online chatrooms.”4 Under European competition law, a Statement of Objections carries
great weight—it is issued only after an “in-depth investigation confirms the Commission’s
competition concerns.”5 The four infringing banks are reported to be Defendants Deutsche
Bank, Bank of America, Crédit Agricole, and Credit Suisse. Deutsche Bank has been
proactively cooperating with the Commission in the hopes of avoiding a financial penalty.
In the end, it is more than plausible that Defendants agreed not to compete, and it is more
than plausible that their agreement harmed their customers. That was, after all, the very reason
they conspired. The evidence in the SAC is voluminous, but also just the tip of the iceberg,
The SAC should be upheld and this case should
proceed to discovery so the full scope of this conspiracy can be adjudicated.
ARGUMENT
A complaint must be sustained if it alleges “enough facts to state a claim to relief that is
plausible on its face.” Bell Atlantic Corp. v. Twombly, 550 U.S. 544, 570 (2007). In making this
determination, the court must accept factual allegations as true and draw all reasonable
4 Brockett Decl., Ex. 1.
5 Brockett Decl., Ex. 2, EUROPEAN COMMISSION, Procedures in anticompetitive agreements,
http://ec.europa.eu/competition/antitrust/procedures_101_en.html.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 19 of 155
7
inferences and resolve all conflicts and ambiguities in favor of the plaintiffs. Anderson News,
L.L.C. v. Am. Media, Inc., 680 F.3d 162, 168 (2d Cir. 2012). Thus, even if “an innocuous
interpretation of the defendants’ conduct may be plausible, that does not mean that the plaintiff’s
allegation that the conduct was culpable is not also plausible,” and “it is not the province of the
court to dismiss the complaint on the basis of the court’s choice among competing plausible
alternatives.” Id. at 190. There is no heightened pleading standard in antitrust cases, and at the
pleading stage, plaintiffs need only “raise a reasonable expectation that discovery will reveal
evidence of illegality.” Mayor & City Council of Baltimore v. Citigroup, Inc. (“Mayor”), 709
F.3d 129, 140 (2d Cir. 2013).
I. THE COMPLAINT PLEADS A PLAUSIBLE CONSPIRACY CLAIM AGAINST
EACH DEFENDANT
A. Defendants Mischaracterize the Relevant Legal Standards
A complaint alleging a Section 1 violation need only contain “enough factual matter
(taken as true) to suggest that an agreement was made.” Twombly, 550 U.S. at 556. “Asking for
plausible grounds to infer an agreement does not impose a probability requirement at the
pleading stage; it simply calls for enough fact[s] to raise a reasonable expectation that discovery
will reveal evidence of illegal agreement.” Id.; see also Anderson News, 680 F.3d at 184 (“[T]o
present a plausible claim at the pleading stage, the plaintiff need not show that its allegations
suggesting an agreement are more likely than not true or that they rule out the possibility of
independent action.”).
“A complaint asserting a claim under Section 1 ‘must allege enough facts to support the
inference that a conspiracy existed’ by either direct or circumstantial evidence, because ‘in many
antitrust cases’ a ‘smoking gun’ can be hard to come by, especially at the pleading stage.” In re
Foreign Exch. Benchmark Rates Antitrust Litig. (“FX II”), 74 F. Supp. 3d 581, 591 (S.D.N.Y.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 20 of 155
8
2015) (quoting Mayor, 709 F.3d 129 at 136). Because “conspiracies are rarely evidenced by
explicit agreements,” they “nearly always must be proven through inferences that may fairly be
drawn from the behavior of the alleged conspirators.” Anderson News, 680 F.3d at 183.
Importantly, a motion to dismiss is not an occasion to resolve the full and precise scope
of an alleged conspiracy; such issues can only be resolved with the benefit of discovery. See FX,
2016 WL 5108131 at *4 (“Questions as to each Defendant’s participation in the conspiracy and
the conspiracy’s scope may be raised later in litigation, but do not merit dismissal at this
phase.”); In re Vitamins Antitrust Litig., 2000 WL 1475705, at *11 (D.D.C. May 9, 2000)
(“[P]laintiffs here have alleged a broader conspiracy and at this point, the Court is bound to
accept plaintiffs’ allegations as true. It would be improper for the Court to reach the merits of
plaintiffs’ claims at this time.”). This is consistent with the Supreme Court’s admonition that the
“character and effect of a conspiracy are not to be judged by dismembering it and viewing its
separate parts,” but must be read “as a whole.” Cont’l Ore Co. v. Union Carbide & Carbon
Corp., 370 U.S. 690, 699 (1962).
The allegations in this case are clear that “Defendants agreed not to compete and instead
to cooperate to manipulate the market for [USD] SSA bonds and maximize their own profits at
the expense of their customers.” SAC ¶ 9. These allegations—laid out in SAC paragraphs 159-
361—include direct evidence of an illegal agreement; circumstantial evidence of an illegal
agreement; and numerous “plus factors,” including a shockingly high level of inter-firm
communications and coordination. These allegations are not only plausible—they are
overwhelming.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 21 of 155
9
B. The SAC Contains Direct Evidence of an Agreement Among Defendants Not
to Compete, Including Direct Evidence of Price-Fixing
Unlike the complaint found lacking in Twombly, the SAC here does not seek to draw an
inference of an agreement based merely on passive parallel behavior and inaction. Cf. Twombly,
550 U.S. at 564-65. Rather, the SAC contains direct, detailed, concrete evidence of Defendants’
agreement not to compete in the secondary market for USD SSA bonds,
These detailed
allegations require no inference and are “independently adequate” to state a Section 1 claim. See
In re Ins. Brokerage Antitrust Litig., 618 F.3d 300, 323-24 (3d Cir. 2010) (noting that
“[a]llegations of direct evidence of an agreement, if sufficiently detailed, are independently
adequate” to state a claim pursuant to Section 1).
1. The direct evidence obtained to date is clear-cut and incriminating
The direct evidence shows that Defendants engaged in at least four techniques to
cooperate in the USD SSA bond market.
Defendants agreed to refrain from acting when doing so could hurt a co-conspirator.
One concrete way in which Defendants implemented their conspiracy was by refraining from
buying or selling particular USD SSA bonds when doing so could hurt one of their co-
conspirators. SAC ¶¶ 164-72. This blatantly anticompetitive conduct would not have occurred
absent a conspiracy.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 22 of 155
10
Contrary to Defendants’ assertion, this is not an innocuous or “unconnected” discussion
between two market-makers.6 Dkt. No. 525 (“Merits Mem.”) at 30.
Another concrete
way in which Defendants cooperated, rather than competed, was by
6 Defendants reflexively paint every factual allegation—no matter how specific, incriminating, or directly
supported—as “vague” or “conclusory.” See, e.g., Merits Mem. at 8, 14, 15 n.21. The Second Circuit has rejected
this tactic, which runs counter to the requirement that courts accept reasonable inferences supported by a
complaint’s factual allegations. See Anderson News, 680 F.3d at 189 (holding “the district court erred in
characterizing the [complaint’s] factual allegations as conclusory and in refusing to accept as true the reasonable
inferences that could be drawn from those allegations”).
The evidence
contained in the SAC is extensive and represents just the “tip of the iceberg.”
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 23 of 155
11
Defendants also pursued their
agreement not to compete by
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 24 of 155
12
7
Defendants’ efforts to suggest that these four patently anticompetitive techniques should
be overlooked
See, e.g.,
Hinds Cty., Miss. v. Wachovia Bank N.A., 700 F. Supp. 2d 378, 395 (S.D.N.Y. 2010) (finding
plausible plaintiffs’ allegation that financial services firm participated in conspiracy to rig bids in
municipal derivatives market where complaint contained “one taped conversation” where a
person “stated that he gave [the defendant] the opportunity to lower its bid on a transaction”).
7 A bond priced at 18 basis points of yield above Treasuries is more expensive than if it were priced at 19
basis points because SSA bond yields are inversely related to bond prices. SAC ¶ 141.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 25 of 155
13
2. Defendants’ attempts to mischaracterize the chats fail
Faced with damning direct evidence, Defendants offer fanciful explanations that are
contradicted by the plain meaning of their own words. As a legal matter, Defendants’ strategy is
improper. At this juncture, the Court must “accept as true the factual allegations of the
complaint, and construe all reasonable inferences that can be drawn from the complaint in the
light most favorable to the plaintiff.” Anderson News, 680 F.3d at 185 (emphasis added). “The
question at the pleading stage is not whether there is a plausible alternative to the plaintiff’s
theory; the question is whether there are sufficient factual allegations to make the complaint’s
claim plausible.” Id. at 189; see also In re Elec. Books Antitrust Litig., 859 F. Supp. 2d 671,
687-88 (S.D.N.Y. 2012) (denying motion to dismiss and rejecting defendants’ attempts to offer
“legitimate alternative explanations for their parallel actions”).
But
the attempt fails miserably, as do similar efforts by Defendants in their individual briefs as
discussed in Section IV below.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 26 of 155
14
This is direct, overt coordination no matter how one slices it. It is frankly stunning that
Defendants would even suggest it is not.
That is true—and it is also illegal price fixing. Competitors are not allowed
to coordinate before providing price quotes in the market. Elec. Books, 859 F. Supp. 2d at 682
(noting that agreements on prices are unlawful per se).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 27 of 155
15
In short, Defendants’ insistence that the cherry-picked exchanges above are legitimate
only strengthens the plausibility of Plaintiffs’ claims. If Defendants genuinely believed these
types of communications were acceptable, one can only imagine how many more of them will be
revealed by comprehensive discovery into their communications.
3. The chats are not innocuous exchanges of “market color”
Defendants argue that certain chats show nothing more than the exchange of “market
color” “among traders attempting to make markets.” Merits Mem. at 9, 33. As an initial matter,
Defendants’ assertion that their position as market-makers meant “traders did at times need to
share information regarding price and bond availability,” id. at 32, is just made up from whole
cloth. It finds no support in the SAC. The SAC’s alleges only that dealers “provide liquidity or
‘make markets’ by being willing to buy and sell SSA bonds.” SAC ¶ 54. It nowhere says that
Defendants’ traders needed to share sensitive information with each other to serve that function.
In fact, the SAC pleads the exact opposite—that in a market “where dealers actually competed
against each other, no dealer would forfeit its competitive advantage by disclosing such sensitive
information to another dealer.” Id. ¶ 129.
In any event, the law is clear that the sharing of information “relating to price, output,
costs, or strategic planning, . . . current operating and future business plans, . . . [and] individual
company data” among competitors is powerful evidence of an illegal conspiracy. FED. TRADE
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 28 of 155
16
COMM’N & DEP’T OF JUSTICE, Antitrust Guidelines for Collaborations Among Competitors 15-16
(2000) (“FTC Guidelines”); see also 54 Am. Jur. 2d, Monopolies and Restraints of Trade § 45
(2017) (“[G]enuine competitors do not ordinarily provide [reports of specific transactions with
identified customers] to their rivals, but only those united in an agreement, express or implied, to
act together and pursue a common purpose to restrict production and increase prices.”); Am.
Column & Lumber Co. v. United States, 257 U.S. 377, 410 (1921) (stating that “[g]enuine
competitors do not make daily, weekly, and monthly reports of the minutest details of their
business to their rivals”). Traders are not exempted from these basic principles. See FX, 2016
WL 5108131 at *4 (“[T]he sharing of information between competitors constitutes
circumstantial evidence of an antitrust conspiracy and is sufficient at the pleading stage.”).
The communications in this case go far beyond anything that could be characterized as
“market color.” They implicate every “competitive concern” identified by the Federal Trade
Commission and Department of Justice as being indicative of an antitrust conspiracy. FTC
Guidelines. at 15.
Not only do Defendants’ exchanges “support an
inference of a price-fixing agreement,” Todd v. Exxon Corp., 275 F.3d 191, 198 (2d Cir. 2001),
See United States v. Container Corp. of Am., 393 U.S. 333, 334-38 (1969) (finding
defendants’ “exchange of information concerning specific sales to identified customers” to be an
illegal price-fixing agreement because “interference with the setting of price by free market
forces is unlawful per se”); Phillip Areeda and Herbert Hovenkamp, Antitrust Law: An Analysis
of Antitrust Principles and Their Application (4th ed. 2017) (“Areeda & Hovenkamp”) ¶ 2111g
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 29 of 155
17
(“Ad hoc competitor-to-competitor ‘exchange’ of particularized price information, such as the
price offered or made to a particular customer, should ordinarily be considered a naked or nearly
naked restraint, which can be condemned without a structural inquiry unless the defendant offers
a compelling justification.”); cf. Major League Baseball Props., Inc. v. Salvino, Inc., 542 F.3d
290, 335 (2d Cir. 2008) (“Indeed, the mere agreement among competitors to exchange price
information is a per se price-fixing violation.”) (Sotomayor, J., concurring).8
Beyond the conclusory assertion that the cartel’s discussions were necessary to “make
markets,” Defendants offer no specific, compelling explanation of why they needed to share their
customers’ most competitively sensitive information in order to generate market liquidity.
Defendants were supposed to be pricing USD SSA bonds individually. That is plainly what their
customers expected, as evidenced by their customer’s efforts to place them in competition with
each other. Nothing gave Defendants the right to secretly collaborate to subvert those efforts.
Defendants would not have taken such extraordinary steps to hide the
innocent exchange of “market color.”
8 For this reason, Defendants’ contention that, “[t]o the extent Plaintiffs purport to assert Section 1 claims
based merely on exchanges of information, they are required to plead their claims under the rule of reason,” Merits
Mem. at 41, is misplaced. Plaintiffs do not need to plead under the rule of reason because Defendants’ agreement
not to compete—for which their exchange of confidential information was just one manifestation—is illegal per se.
See, e.g., Palmer v. BRG of Georgia, Inc., 498 U.S. 46, 48-50 (1990) (agreement between bar review companies not
to compete in certain states was “unlawful on its face”). Texaco Inc. v. Dagher, 547 U.S. 1 (2006), cited by
Defendants, is not to the contrary. That case did not involve a claim based on a conspiracy not to compete that was
further fueled by improper exchanges of information among competitors. Rather, it confronted the question of
whether the rule of reason applied to an agreement to unify gasoline pricing by two joint-venture partners who were
not competitors in the relevant geographical market. Id. at 4-6.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 30 of 155
18
Defendants rely heavily on a 27-year old case from the District of Minnesota, Five
Smiths, Inc. v. National Football League Players Association, 788 F. Supp. 1042, 1051 (D.
Minn. 1992). Merits Mem. at 33. But in Five Smiths, the plaintiffs made just one vague factual
allegation of concerted action, i.e., that various player-agents and their association shared
information about NFL players’ salaries. Five Smiths, 788 F. Supp. at 1046. The plaintiffs did
not allege which specific competitors made agreements, what competitively sensitive
information was shared, and how the goals of the conspiracy were accomplished. Id.
Defendants’ appeal to Apex Oil Co. v. DiMauro, 822 F. 2d 246 (2nd Cir. 1987) is equally
unavailing. There was no allegation in Apex that the defendants exchanged commercially
sensitive-information in furtherance of the alleged conspiracy to manipulate a market or
otherwise. Nor was there any analysis of whether such communications constitute a per se
Section 1 violation.
Finally, Defendants cite the 40-year-old case of United States v. U.S. Gypsum Co., 438
U.S. 422, 441 n.16 (1978), for the proposition that “[t]he exchange of price data and other
information among competitors does not invariably have anticompetitive effects.” Merits Mem.
at 32 n. 46. But here, the SAC explains in detail how Defendants’ sharing of information came
at the expense of customers. Defendants did not just exchange general information about the
number and types of bonds traded. Far from it.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 31 of 155
19
see also Areeda & Hovenkamp ¶ 2111e (“[S]uppose that a firm in the process of
bidding on a contract calls a rival to inquire as to the amount of its bid on that contract. One is
hard-pressed to see how such a scheme could be profitable except on the premise that the two
firms, or the two firms plus others that must be engaging in similar behavior, had the power to
charge a particular price or had made a collateral agreement to rig bids.”).
C. The SAC Also Contains Abundant Circumstantial Evidence and “Plus
Factors”
1. Defendants engaged in parallel conduct
The SAC alleges that Defendants engaged in parallel conduct. Apex, 822 F.2d at 253
(“Parallel conduct can be probative evidence bearing on the issue of whether there is an antitrust
conspiracy.”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 32 of 155
20
Defendants argue that Plaintiffs fail to allege parallel conduct because they “conclusorily
allege a few isolated instances of like pricing or other similar conduct by two or three traders.”9
Merits Mem. at 35.
See, e.g.,
Sullivan v. Barclays PLC, 2017 WL 685570, at *24 (S.D.N.Y. Feb. 21, 2017) (holding that the
plaintiffs had sufficiently pled a conspiracy even though “[v]irtually all substantive allegations
touch on communications involving [only two individual defendants]”); FX, 2016 WL 5108131
at *3 (rejecting argument that the complaint “at most support ‘a series of small conspiracies’ that
cannot sustain a ‘grand conspiracy theory’”); In re High-Tech Emp. Antitrust Litig., 856 F. Supp.
2d 1103, 1118 (N.D. Cal. 2012) (denying motion to dismiss and rejecting defendants’ argument
that “multiple bilateral agreements do not make up an overarching conspiracy”); In re TFT-LCD
(Flat Panel) Antitrust Litig., 267 F.R.D. 583, 607 (N.D. Cal. 2010) (rejecting “defendants’
attempt to divide the alleged conspiracy into smaller pieces”).10
9 Notably, in its supplemental memorandum of law, Credit Suisse actually admits that the SAC pleads
parallel conduct. See Dkt. No. 524, Supplemental Memorandum of Law in Support of the Credit Suisse Defendants’
Motion to Dismiss the Second Consolidated Amended Class Action Complaint for Failure to State a Claim (“Credit
Suisse Mem.”) at 7 .
10 Defendants’ comparison of this case to Prime Healthcare Services, Inc. v. Service Employees
International Union, 2013 WL 3873074 (S.D. Cal. July 25, 2013), is inapt. In Prime Healthcare, the court found
that the plaintiffs could not prove the existence of a conspiracy based on parallel conduct that consisted of one
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 33 of 155
21
2. Defendants shared a common motive to conspire
The SAC also alleges that Defendants shared “a common motive to conspire.” Apex, 822
F.2d at 254 (holding that a “common motive to conspire” is indicative of conspiracy); see also In
re Flat Glass Antitrust Litig., 385 F.3d 350, 360 (3d Cir. 2004) (observing that “evidence that the
defendant had a motive to enter into a price fixing conspiracy” is a “plus factor” indicative of
conspiracy). As the SAC explains, “Defendants’ overarching objective was to ensure that cartel
members could transact with their investor clients at prices that were more favorable for the
conspiring dealers—and thus worse for their customers—than would have been achieved in the
absence of the collusion.” SAC ¶ 143.
see also FX II, 74 F. Supp. 3d at 594 (“The logic [of the
conspiracy] is at least facially simply and intuitive—price-fixing to maximize profits at the
expense of consumers.”). And, as discussed below, the SAC’s statistical analyses confirm that
Defendants’ motive was rewarded by the desired results.
3.
“isolated action during an alleged fifteen year conspiracy.” Id. at *9. As discussed herein, the SAC alleges parallel
conduct far beyond one “isolated action.” Id.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 34 of 155
22
11
In response, Defendants engage in misdirection by asserting that the SAC fails to plead
sufficient “social ties” among all of the “twelve traders defined as ‘primary players’ . . .” Merits
Mem. at 37. This misses the forest for the trees.
11 Defendants’ cited cases are easily distinguished because each involves only the opportunity for
communication or social interaction, without more. See, e.g., Hinds Cty. v. Wachovia Bank N.A., 708 F. Supp. 2d
348, 362 (S.D.N.Y. 2010) (“mere presence at industry associations and meetings” insufficient to infer agreement);
Sky Angel U.S., LLC v. Nat’l Cable Satellite Corp., 947 F. Supp. 2d 88, 102 (D.D.C. 2013) (mere presence of
company representatives sitting on board of directors insufficient to infer agreement); In re Chocolate Confectionary
Antitrust Litig., 801 F.3d 383, 406 (3d Cir. 2015) (mere social introduction alone was insufficient to infer
agreement). In contrast, the SAC directly quotes from many actual communications.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 35 of 155
23
12
4. Defendants engaged in acts against economic self-interest
The SAC alleges that Defendants frequently engaged in actions that would have been
against their economic self-interest absent an agreement to conspire. See Starr v. Sony BMG
Music Entm’t, 592 F.3d 314, 327 (2d Cir. 2010) (reversing district court decision dismissing
complaint where “plaintiffs have alleged behavior that would plausibly contravene each
defendant’s self-interest in the absence of similar behavior by rivals”); FX, 2016 WL 5108131 at
*4 (“The [complaint] alleges that sharing of confidential customer information . . is against each
bank’s economic self-interest as a competitor absent collusion.”). The SAC alleges that
Defendants used a wide variety of tactics that would make no sense absent a collusive motive:
12
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 36 of 155
24
Banks and traders spend huge amounts of manpower and money to
gather such sensitive commercial information. Surrendering the fruits of these efforts to a
competitor makes no economic sense.13 Indeed, this sharing only made sense insofar as
Defendants were colluding and operating as a single, unitary “super-desk,” rather than as arms’-
length competitors. Id. ¶ 16. Defendants have no serious response to this point.
5. The anomalous behavior of bid-ask spreads and yields on USD SSA bonds
during the conspiracy period is further evidence of a conspiracy
The SAC also includes compelling statistical analyses that are probative of collusion. Id.
¶¶ 508-51. Courts have repeatedly found well-pled antitrust conspiracies based solely on
statistical evidence. For example, in In re Commodity Exchange, Inc. Antitrust Litigation
(“Gold”), 213 F. Supp. 3d 631 (S.D.N.Y. 2016) and In re London Silver Fixing Ltd., Antitrust
Litigation (“Silver”), 213 F. Supp. 3d 530 (S.D.N.Y. 2016) Judge Caproni relied on statistical
analyses of anomalous data patterns to uphold the plaintiffs’ Section 1 claims—despite no
‘whistleblower,’ or chat quotes, or any other of the ‘smoking gun’ evidence present here. Gold,
13 Defendants’ contention that this assertion is “conclusory,” Merits Mem. at 33, n. 48, ignores that (a) the
SAC explains the logical reasons why customers would not want their information shared
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 37 of 155
25
213 F. Supp. 3d at 645; Silver, 213 F. Supp. 3d at 561-64. Likewise, Judge Furman in the
ISDAfix case upheld a complaint built largely around analysis of economic data. See Alaska
Elec. Pension Fund v. Bank of Am. Corp. (“ISDAfix”), 175 F. Supp. 3d 44, 54-55 (S.D.N.Y.
2016). These are just a few of the many cases, including others in this District, where courts
have upheld antitrust complaints built from economic data.14
As discussed in Section II.B below, the data here show that the prediction in the
academic literature that spreads would widen during a conspiracy that removed competition was
borne out during the core conspiracy period in the USD SSA market. Though included primarily
to confirm that Defendants’ cartel in fact harmed Plaintiffs, the existence of these pricing
abnormalities also confirms the plausibility of the conspiracy in the first place. As discussed
below, it is only the conspiracy’s presence and eventual breakup that explains the full range of
results seen across the multiple tests run.
Defendants contend that “Plaintiffs fail to plausibly tie” these economic analyses “to any
conduct by Defendants.” Merits Mem. at 39.
14 See, e.g., Gelboim, 823 F.3d at 772, 778 (relying in part on statistical analyses to find a conspiracy to
suppress LIBOR plausibly pled); In re Domestic Airlines Travel Antitrust Litig., 221 F. Supp. 3d 46, 63-65 (D.D.C.
2016) (upholding complaint that used averages in a variety of data analyses); In re Nasdaq Market-Makers Antitrust
Litig., 894 F. Supp. 703, 712-13 (S.D.N.Y. 1995) (plaintiffs “sufficiently pled” an antitrust conspiracy based on
analysis of fluctuations in average Nasdaq spreads).
15 Defendants’ reliance on In re Commodity Exchange, Inc., Gold Futures & Options Trading Litigation
(“Gold II”), 328 F. Supp. 3d 217 (2018) is thus misplaced. There, the Gold II court upheld the core claims for
benchmark manipulation against the panel of banks who set the benchmark. The court dismissed the claims against
an additional bank that was not on the panel, because the complaint contained no alleged facts plausibly linking that
non-benchmark-panel bank to the conspiracy carried out through the panel’s setting of the benchmark.
Defendants’ citation to 7 W. 57th St. Realty Co. v. Citigroup, Inc., 2015 WL 1514539 (S.D.N.Y. Mar. 31, 2015),
fares no better. There, the studies showed that LIBOR was suppressed, which contradicted the plaintiffs’ theory that
LIBOR was inflated. 7 W. 57th, 2015 WL 1514539 at *12. It says nothing of the propriety of relying on data
analyses that, as here, logically support the claim asserted.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 38 of 155
26
Defendants’ call for more
specifics is disingenuous because they know Plaintiffs have been denied access to Defendant-
specific data. SAC ¶ 509. See, e.g., ISDAfix, 175 F. Supp. 3d at 67-68 (upholding claims based
on economic data despite the fact that “perhaps the most important data—the Defendant Banks’
individual submissions to ICAP—were . . . not publicly available at all”).16 Defendants cannot
hide behind the very opacity that made their conspiracy feasible.
Defendants’ additional arguments as to why the economic analyses do not help confirm
the conspiracy as being plausibly pled—that the analyses are inconsistent and unreliable, Merits
Mem. at 40-41—fail for the reasons discussed in Section II.B below.
6. Government investigations and adverse personnel decisions also
constitute “plus” factors
The ongoing regulatory investigations by the DOJ and European Commission, SAC
¶¶ 375, 381-86, also constitute a “plus factor.” See, e.g., Starr, 592 F.3d at 324-25 (conspiracy
was plausibly alleged where the DOJ “launched two new investigations” into the very conduct
plaintiffs were suing over); In re Tableware Antitrust Litig., 363 F. Supp. 2d 1203, 1205 (N.D.
Cal. 2005) (“A plaintiff may surely rely on governmental investigations . . .”).
Defendants argue that the SAC does not allege any adverse action thus far against them
resulting from these investigations. But the Court may take judicial notice of the fact that, after
16 See also Gold, 213 F. Supp. 3d at 666 (holding that “disregarding” statistical analyses “would
effectively foreclose Plaintiffs’ ability to state an antitrust claim for manipulation . . . unless they had direct
evidence, which is generally not required at the pleading stage.” ); Silver, 213 F. Supp. 3d at 563 (same); FX, 2016
WL 5108131 at *20 (excusing the plaintiffs’ failure to plead information concerning “specific dates, times, currency
pairs and customers discussed” in traders’ chat messages because that information was “exclusively in Defendants’
control”). See generally Arista Records LLC v. Doe 3, 604 F.3d 110, 120 (2d Cir. 2010) (the plausibility standard
“does not prevent a plaintiff from pleading facts alleged upon information and belief where the facts are peculiarly
within the possession and control of the defendant”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 39 of 155
27
the SAC was filed, the European Commission issued Statements of Objection to four Defendants
concerning their trading of USD SSA Bonds—Bank of America, Crédit Agricole, Credit Suisse,
and Deutsche Bank.17 The Statements express the Commission’s “preliminary view that they
have breached EU antitrust rules by colluding, in periods from 2009 to 2015, to distort
competition in secondary market trading in the EEA of supra-sovereign, sovereign and agency
(SSA) bonds denominated in US dollars.”18 Notably:
The Commission has concerns that at different periods between 2009 and 2015,
the four banks exchanged commercially sensitive information and coordinated on
prices concerning US dollar denominated supra-sovereign, sovereign and agency
bonds, known as “SSA bonds”. These contacts would have taken place mainly
through online chatrooms.19
Under European competition law, a Statement of Objections carries great weight—it is issued
only after an “in-depth investigation confirms the Commission’s competition concerns.”20
Deutsche Bank has admitted it has “proactively cooperated with the European Commission in
this matter” in exchange for prosecutorial immunity.21
The SAC also pleads that, subsequent to the start of these investigations, four of the
primary players in the cartel were suspended from further trading and/or fired from their jobs.
SAC ¶¶ 376-80. All five Individual Defendants, and many of their colleagues, have been listed
as “inactive” on the FCA register for traders since either late 2015 or early 2016. Id. ¶¶ 377-80.
For instance, Gudka has been inactive since November 2015, Manku has been inactive since
17 Brockett Decl., Ex. 3, Aoife White, et al., Deutsche Bank, Credit Suisse Targeted by EU Bond Probe,
BLOOMBERG (Dec. 20, 2018), https://www.bloomberg.com/news/articles/2018-12-20/four-banks-targeted-by-eu-
antitrust-objections-over-ssa-bonds.
18 Brockett Decl., Ex. 1.
19 Id.
20 Brockett Decl., Ex. 2.
21 Brockett Decl., Ex. 4, DEUTSCHE BANK, Statement on antitrust investigation of the European
Commission (Dec. 20, 2018), https://www.db.com/newsroom_news/2018/statement-on-antitrust-investigation-of-
the-european-commission-en-11752.htm.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 40 of 155
28
December 2015, Pau has been inactive since February 2016, and Heer has been inactive since
March 2016. Id.; see FrontPoint Asian Event Driven Fund, L.P. v. Citibank, N.A., 2018 WL
4830087 (S.D.N.Y. Oct. 4, 2018) (holding the fact that “133 traders attempted to manipulate
SIBOR and SOR, and were disciplined for their infractions . . . satisfied the ‘plus factors’ to state
a claim for antitrust conspiracy”).
Defendants assert that allegations of employee termination or suspension are an
insufficient “plus” factor, relying on In re Iconix Brand Group, Inc., 2017 WL 4898228
(S.D.N.Y. Oct. 25, 2017). But this case did not involve allegations of adverse personnel
decisions, as the SAC does. The plaintiffs in In re Iconix alleged senior executives had
resigned—not that they had been suspended or terminated.22 2017 WL 4898228 at *19. Iconix
is thus inapposite.
D. Defendants’ Miscellaneous Attacks Fail
1. Plaintiffs allege a plausible mechanism to implement the conspiracy
Defendants argue that the conspiracy is implausible because “there is no benchmark or
similar rate-setting mechanism through which Defendants could have reliably and systematically
affected prices across SSA bonds.” Merits Mem. at 26-27 & n. 41. Without such a benchmark,
Defendants claim it is implausible that “a handful of traders pervasively affected competition and
22 Like they did when moving to dismiss the CAC, Defendants once again seek to contradict the SAC by
relying on extrinsic evidence—this time a purported change to FCA registration requirements. Merits Mem. at 39,
n.55. But taking judicial notice of such extrinsic evidence is considered improper. See Glob. Network Commc’ns,
Inc. v. City of New York, 458 F.3d 150, 156 (2d Cir. 2006) (holding that taking judicial notice of extrinsic material
was improper where the court “relied on those materials to make a finding of fact that controverted the plaintiff’s
own factual assertions”); In re Interest Rate Swaps Antitrust Litig. (“IRS”), 261 F. Supp. 3d 430, 480 (S.D.N.Y.
2017) (“[O]n a motion to dismiss, the Court is not at liberty to second-guess plaintiffs’ well-pled § 1 claim on the
basis of the ‘market realities’ defendants derive from secondary sources.”). Even if the law did not compel
disregarding such evidence, Defendants do not contest Plaintiffs’ allegations that the conspiring traders faced
dismissal or removal from their jobs as a result of their illegal actions. Nor do Defendants explain how a purported
change “effective March 7, 2016” accounts for Gudka’s inactivation since November 2015, Manku’s inactivation
since December 2015, and Pau’s inactivation since February 2016. SAC ¶¶ 377-79.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 41 of 155
29
prices for all USD SSA bonds for seven years based solely on ad hoc communications.” Id. at
25.
Initially, even Defendants themselves admit the SAC identifies more than “a handful of
traders.” Merits Mem. at 1, 25.
23 Given the limited
documents available to Plaintiffs thus far, and the admittedly close connections between Dealer
Defendants’ traders and salespersons, see, e.g., id. ¶¶ 114, 130, 161, 390-93, it is plausible that
many additional employees participated in the conspiracy.
Defendants also mischaracterize the nature and scope of the conspiracy. Plaintiffs do not
allege that Defendants expressly fixed the price of every USD SSA bond traded in the market.
Rather, the SAC alleges that, by virtue of their agreement not to compete, Defendants were able
to charge supracompetitive prices to their own customers. The SAC defines the Class as “[a]ll
persons or entities who . . . directly entered into U.S.-dollar denominated SSA bond transactions
with Defendants . . . .” Id. ¶ 561 (emphasis added). The SAC also explicitly pleads that “this
case is limited to those investors who had the misfortune to trade directly with the cartel
members—that is, the Dealer Defendants.” Id. ¶ 493 n. 53 (emphasis added).24
To effectuate their conspiracy, the 12 or more conspiring traders at the Dealer Defendants
did not need to manipulate a benchmark. Instead, they used the old-fashioned method of
23 Defendants’ focus on a “handful of traders” is strange given they concede Plaintiffs have already
identified 12 individuals who participated in the scheme. Merits Mem. at 24 (discussing the “twelve individual
traders”). These traders are Hiren Gudka, Shailen Pau, Amandeep Singh Manku, Bhardeep Heer, Gary McDonald,
24 The SAC’s economic analyses show that Defendants’ agreement not to compete had broader effects in
the market, beyond harming their customers. But this case is not about those other trades in which Defendants did
not participate, but instead is about those USD SSA bond transactions conducted by the cartel members.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 42 of 155
30
agreeing not to compete with each other. As the SAC states, Defendants “agreed instead to
cooperate to maximize their own profits at the expense of their customers, many of whom they
shared in common (and for whose business they were supposed to be competing).” Id. ¶ 143.
Contrary to Defendants’ assertion that the instances of collusion were “ad hoc” and
“isolated,” Merits Mem. at 25-27, 31, 35,
25
25
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 43 of 155
31
26
Defendants’ core arguments are remarkably similar to those rejected by Judge Schofield
in the FX case. There, the defendant banks (including some Defendants) were alleged to have
manipulated both the primary benchmark in the foreign exchange market and bid-ask spreads on
spot transactions. FX, 2016 WL 5108131 at *4. In their motion to dismiss, the defendants
argued that the plaintiffs’ allegations:
(1) at most support “a series of small conspiracies” that cannot sustain a “grand
conspiracy theory”; (2) do not create a plausible inference that any [defendant]
joined a global conspiracy that other banks may have participated in; [and] (3)
misrepresent chat transcripts discussing spreads.
Id. at *3.
The court rejected each of these arguments. As to the first argument that the plaintiffs
pled only “that different groups of banks sporadically carried out a series of small conspiracies,”
the court observed the complaint “adequately alleges a Section 1 conspiracy among Defendant in
26 See Brockett Decl., Ex. 1. Defendants’ reliance on Chubirko v. Better Business Bureau of South
Piedmont, Inc., 763 F. Supp. 2d 759 (W.D.N.C. 2011) is misguided. Merits Mem. at 27. There, a pro se plaintiff
failed to plead any facts supporting his allegation that 16 “disparate entities” from entirely different markets were
colluding “for purposes of destroying small businesses and business owners.” Id. at 765-66. In contrast, in this
case, Plaintiffs have pled that Defendants are from the same market, worked together in the past,
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 44 of 155
32
the FX market based on, among other things, the sharing of market-sensitive, nonpublic
information in chat rooms and instant messages.” Id. With respect to the second argument, the
court noted that, “[w]hile it is true that not every chat transcript contained in the [complaint]
includes each individual Defendant or a discussion of each possible currency pair or benchmark
rate, every [Defendant] is alleged as a participant in at least one illegal communication.” Id. at
*4. As to the Defendants’ third argument that the evidence “amounts to no more than claims of
sporadic information exchanges that do not constitute per se antitrust violations,” the court
disagreed:
As an initial matter, a fair reading of the chats quoted in the [complaint] as to each
[defendant] suggests that more than a harmless exchange of information was
occurring. For example, a chat where a [defendant] trader asks others what they
believe a spread should be plausibly suggests that the traders in the room are
colluding to fix the spread of the currency pair they are discussing. . . . Even if
these conversations themselves did not set spreads or fixes, the sharing of
information between competitors constitutes circumstantial evidence of an
antitrust conspiracy and is sufficient at the pleading stage. . . .
Id.
Numerous additional courts have upheld Section 1 claims based on agreements not to
compete implemented via direct communications and meetings among the conspirators. See,
e.g., In re High Fructose Corn Syrup Antitrust Litig., 295 F.3d 651, 662 (7th Cir. 2002)
(reversing grant of summary judgment for defendants where plaintiffs provided evidence of
defendant communications that referenced “an understanding within the industry not to undercut
each other’s prices” and an “understanding between the companies that . . . causes us not to . . .
make irrational decisions”); Elec. Books, 859 F. Supp. at 682-83 (holding that complaint
sufficiently alleged an agreement to fix eBook prices where it “describes specific conversations
from which it is fair to infer that the [defendants] had agreed among themselves to adopt a joint
strategy to force an increase in the price of eBooks”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 45 of 155
33
2. That not every Defendant was on every chat makes no difference
Defendants argue the conspiracy could not have worked unless all participants
communicated with each other in a single chatroom at the same time. But there is no such
requirement. To the contrary, the cases—including those on which Defendants rely—uniformly
hold that communications among subsets of conspirators must be viewed as a whole and can
plausibly evidence a broader conspiracy. This makes sense for multiple reasons.
First, Defendants’ basic agreement not to compete was not something that had to be
repeated over and over among all the conspirators once the agreement was reached.
Second, Defendants could implement their overarching agreement through bilateral,
trilateral, or sometimes multilateral communications, without the need for all conspirators to
communicate at one time. This is because “[a]n investor interested in buying or selling an SSA
bond generally contacted no more than a few dealers.” SAC ¶ 133.
Finally, Judge Schofield rejected a similar argument by the defendants in the FX case.
There, the complaint referenced a number of chat transcripts
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 46 of 155
34
in which the defendants manipulated spreads and
benchmark rates by sharing “market-sensitive price and customer information and trading
positions.” FX, 2016 WL 5108131 at *3. Many of the chats included only a small subset of the
defendants, and they argued (as here) that the chats “amount[ed] to no more than claims of
sporadic information exchanges that do not constitute per se antitrust violations as a matter of
law.” Id. at *4. The Court flatly rejected this argument:
While it is true that not every chat transcript contained in the [complaint] includes
each individual Defendant or a discussion of each possible currency pair or
benchmark rate, every [non-settling defendant] is alleged as a participant in at
least one illegal communication. Taken as a whole, the [complaint] sufficiently
pleads an antitrust conspiracy to fix spot prices and benchmark rates in the FX
market.
Id. at *4. See also Sullivan, 2017 WL 685570 at *23-24 (holding that the plaintiffs had
sufficiently pled a conspiracy even though “[v]irtually all substantive allegations touch on
communications involving [only two individual defendants]”); In re NASDAQ Market-Makers
Antitrust Litig., 169 F.R.D. 493, 518 (S.D.N.Y. 1996) (rejecting defendants’ argument that
“Plaintiffs’ complaint alleges not a single conspiracy but thousands of mini-conspiracies”);
High-Tech, 856 F. Supp. 2d at 1118 (denying motion to dismiss and rejecting defendants’
argument that “multiple bilateral agreements do not make up an overarching conspiracy”); TFT,
267 F.R.D. at 607 (rejecting “defendants’ attempt to divide the alleged conspiracy into smaller
pieces”); see also Cont’l Ore Co., 370 U.S. at 699 (“The character and effect of a conspiracy are
not to be judged by dismembering it and viewing its separate parts, but only by looking at it as a
whole.”).27
27 Defendants’ reliance on Sonterra Capital Master Fund Ltd. v. Credit Suisse Group AG, 277 F. Supp. 3d
521 (S.D.N.Y. 2017), and In re Optical Disk Drive Antitrust Litigation, 2011 WL 3894376 (N.D. Cal. Aug. 3,
2011), is unavailing. In the former, the complaint included robust allegations of inter-defendant collusion with
respect to only one defendant, Royal Bank of Scotland (“RBS”). Sonterra, 277 F. Supp. 3d at 556-57. In the latter,
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 47 of 155
35
3. Defendants’ speculation that they may have had held different positions in
USD SSA bonds is irrelevant
Defendants next attack the plausibility of their common motive by positing that they
would have “continuously found themselves on both sides of SSA bond trades.” Merits Mem. at
35-36. From this unsupported premise, they contend there are “no facts suggesting that the
purported cartel members would have overlapping interests regarding what direction the price of
a particular bond should be moved at any given time.” Id. at 35. This misses the mark because,
as Defendants repeatedly emphasize, this is not a “benchmarking” case, where manipulation of a
published figure would instantly alter cash flows for everyone’s instruments that were
contractually tied to that benchmark. This, by contrast, is a case involving an agreement not to
compete, which led to higher spreads—meaning, the dealers won when they bought and when
they sold. This is borne out by the SAC’s statistical analyses, which show that bid-ask spreads
were consistently wider during the Class Period than before or afterwards. SAC ¶¶ 508-51.
Defendants were thus—in transaction after
transaction—able to extract advantages for their cartel at the expense of their customers,
irrespective of whether they held overlapping or antagonistic positions. And even in those
instances where Defendants agreed on a course of action that was against the economic interest
the plaintiffs identified potential bid rigging on only “three specifically-identified occasions” involving “only a
small subset of defendants.” Optical, 2011 WL 3894379, at *9. In contrast, the SAC contains specific evidence of
participation in the conspiracy by each Defendant. SAC ¶¶ 159-361.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 48 of 155
36
of one conspirator, they understood the conspirator would be made whole in due course.
28
4. The existence of non-Defendant USD SSA bond dealers is immaterial
Defendants argue the conspiracy is implausible because “any customer could obtain a
competitive quote simply by requesting one from one of the many dealers that are not alleged to
have been part of a price-fixing conspiracy.” Merits Mem. at 29. This argument is both
factually incorrect and legally irrelevant.
First, that Defendants’ customers could theoretically “solicit quotes from non-Defendant
SSA bond dealers” makes no difference, not least because it ignores that customers would have
been unaware that Defendants were conspiring. Merits Mem. at 29. The SAC is brought only on
behalf of customers who directly traded USD SSA bonds with Defendants during the Class
Period. SAC ¶ 561. Those investors who were able to avoid the (secret) conspiracy by trading
with other dealers are not members of the proposed class.
Second, the argument contradicts the well-pled allegations. The SAC alleges that the
USD SSA bonds market is opaque because it relies on over-the-counter trading and there is often
no transparency into trading volumes, prices, or other useful commercial information. Id.
¶¶ 129-34. It also alleges customers are often required to communicate with dealers via phone or
chat messages, which makes it impractical to contact more than a few dealers at a time. Id.
28 Defendants argue there are no facts supporting the notion that there would be such payback. Merits
Mem. at 36. This is simply false.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 49 of 155
37
¶¶ 131-33. This means that customers, even sophisticated ones, are not only forced to rely on
dealers for pricing information, id. ¶ 132; they also have no way of knowing they are being
exploited by the very dealers on whom they rely, id. ¶¶ 129-34. This observation is reinforced
by the fact “that Defendants were dominant players in the overall USD SSA bond market,”29
likely controlling at least 60% of the market. Id. ¶ 505.
This made it even harder for customers, who were unaware of
Defendants’ scheme, to find better bargains elsewhere. Defendants’ argument to the contrary
rests on an unsupported factual assertion, which must be disregarded at the pleading stage.
Third, Defendants’ argument rests on an assumption they simply make up, i.e., that
Defendants did not know which dealers were being approached by customers for bids and offers.
Finally, Defendants’ argument assumes that a conspiracy is implausible unless it works
perfectly in every instance. But the mere hypothetical possibility that the conspiracy would not
work every time does not mean Defendants would have forsaken the opportunity to maximize
their profit most of the time. In any event, the SAC’s economic analyses bear out the
29 The SAC explains that “Defendants have been estimated to have been responsible for as much as 60%
of the underwriting for primary issuances of USD SSA bonds” and that the “market share of dealers trading in the
secondary market is closely correlated with that of lead underwriters in the primary market, as the syndicate banks
that underwrite bond issuances are typically among the most active traders of those bonds post-issuance.” SAC
¶ 505. While Defendants might disagree with these logical factual allegations, they are by no means speculative, as
Defendants posit. Merits Mem. at 29 n.43.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 50 of 155
38
conspiracy’s plausibility by empirically demonstrating that Defendants’ collusion had the effect
of skewing prices and yields during the Class Period. Id. ¶¶ 508-51.30
5. The precise length of each Defendant’s participation in the conspiracy is a
question for discovery
Defendants assert that “Plaintiffs still cannot connect the Dealer Defendants to the
alleged conspiracy across its purported length.” Merits Mem. at 29. But Plaintiffs have
plausibly alleged that each Defendant participated in the conspiracy during the Class Period, and
questions of “the conspiracy’s scope may be raised later in litigation, but do not merit dismissal
at this phase.” FX, 2016 WL 5108131, at *4. Indeed, the European Commission’s Statementa of
Objection is focused on precisely the same period as the SAC: “periods between 2009 and
2015.”31
Relying on pre-discovery evidence collected to date, the SAC pleads that, by 2009, “at
least nine of the individual cartel members were employed on the USD SSA trading desks of
Defendant Dealers: Hiren Gudka and were working at Bank of America; Shailen
Pau was working at RBC; Amandeep Manku was working at HSBC; Bhardeep Heer was
working at Nomura; Gary McDonald was working at TD Bank; was working
at BNP Paribas; was working at Nomura; and was working at
Barclays.” SAC ¶¶ 144-57.
30 Universal Grading Service v. eBay, Inc., 2012 WL 70644 (N.D. Cal. Jan. 9, 2012) does not aid
Defendants’ cause. There, the court dismissed a complaint that made “no economic sense” because the allegedly
anticompetitive practice would have caused the defendant to lose money. 2012 WL 70644 at *5. In contrast, as
discussed above, because of the over-the-counter nature of the SSA bonds market, “[a]n investor interested in
buying or selling an SSA bond generally contacted no more than a few dealers.” SAC ¶ 133. This made it relatively
easy for Defendants to coordinate pricing (among other anticompetitive activities) with little fear of being thwarted
by traders not part of the cartel. Indeed, the extensive evidence that Defendants did coordinate shows they believed
in the economic utility of the conspiracy.
31 Brockett Decl., Ex. 1.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 51 of 155
39
Id. ¶¶ 159-492.
32 and the SAC specifically identifies how each Defendant so participated. Id.
¶¶ 44-114. The SAC’s statistical analyses further confirm that the conspiracy’s effects began in
the beginning of 2009, when bid-ask spreads trended markedly higher than would be predicted
by standard economic variables. Id. ¶¶ 523-48. That Plaintiffs do not yet—pre-discovery—have
evidence that each Defendant participated in the conspiracy throughout the entire Class Period
says nothing about its plausibility. It simply means there has been no discovery. See FX, 2016
WL 5108131 at *4 (“Questions as to each Defendant’s participation in the conspiracy and the
conspiracy’s scope may be raised later in litigation, but do not merit dismissal at this phase.”); In
re Lithium Ion Batteries, 2014 WL 309192, at *2 (N.D. Cal. Jan. 21, 2014) (accord).
6. Defendants’ conspiracy was in effect by at least January 1, 2009
See Arista Records, 604 F.3d at
120 (the plausibility standard “does not prevent a plaintiff from pleading facts alleged upon
32
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 52 of 155
40
information and belief where the facts are peculiarly within the possession and control of the
defendant”); FX, 2016 WL 5108131 at *20 (excusing the plaintiffs’ failure to plead information
concerning “specific dates, times, currency pairs and customers discussed” in traders’ chat
messages because that information was “exclusively in Defendants’ control”).
That the conspiracy commenced by at least January 2009 is not surprising given that “at
least nine of the individual cartel members were employed on the USD SSA trading desks of
Defendant Dealers” by then. Id. ¶ 158.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 53 of 155
41
Additional support for beginning the class period in January rather than July 2009 is
provided by the economic data discussed in the SAC. As discussed in Section II below, the data
show a break in behavior at the beginning, not middle, of 2009.33
Plaintiffs’
allegations that they occurred are sufficiently plausible to survive a motion to dismiss. See
generally Starr, 592 F.3d at 325 (“[P]laintiffs were not required to mention a specific time, place
or person involved in each conspiracy allegation.”); In re Air Cargo Shipping Servs. Antitrust
33 Though Defendants cite to Silver as an example where the court rejected statistical allegations, that case
is inapposite. There, in largely denying the defendants’ motion to dismiss the plaintiffs’ Section 1 claim, Judge
Caproni conformed the Class Period to the timeframe supported by the plaintiffs’ statistical analyses. 213 F. Supp.
3d at 558, n. 19. Here, Plaintiffs’ statistical analyses demonstrate higher bid-ask spreads in USD SSA bonds, and
other anomalous pricing behavior, from January 1, 2009, through the end of the Class Period. See, e.g., SAC
¶¶ 516-51. This analysis further underscores that the conspiracy was in effect by at least January 2009.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 54 of 155
42
Litig., 2010 WL 10947344, at *11 (E.D.N.Y. Sept. 22, 2010) (“[S]pecific dates of [defendants’
collusive] discussions and meeting are not needed to state a claim that is plausible on its face.”);
In re Propranolol Antitrust Litig., 249 F. Supp. 3d 712, 718 (S.D.N.Y. 2017) (same); Elec.
Books, 859 F. Supp. 2d at 687 (same).
II. NAMED PLAINTIFFS HAVE PLAUSIBLY PLED THEY WERE INJURED BY
THE CONSPIRACY NOT TO COMPETE
This Court previously found that the Plaintiffs had sufficiently pled injury to provide
Article III standing. Order at 11. However, it also found that the evidence of Defendants’
misconduct in the then-quoted chats was, “by itself, insufficient” to infer that “Plaintiffs’
individually negotiated transactions with the Dealer Defendants during [the conspiracy period]
must have been likewise tainted” for purposes of antitrust standing. Id. at 18. This was because
(1) id. at
7, and (2) Plaintiffs did not provide any “analysis” showing that the “unsurprising theory” that
less competition leads to wider spreads “was actually borne out” here, id. at 17.
As explained below, Plaintiffs further support their claims in both of the ways
contemplated by the Court’s order.
The SAC also includes numerous USD SSA-specific analyses
showing that Plaintiffs’ allegation that Defendants’ conspiracy to remove competition from the
market widened spreads was indeed “borne out” in the data.
A.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 55 of 155
43
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 56 of 155
44
34
35
34
35 See generally, e.g., Glob. Network Commc’ns, Inc., 458 F.3d at 156 (holding that taking judicial notice
of extrinsic material was improper where the court “relied on those materials to make a finding of fact that
controverted the plaintiff’s own factual assertions”); IRS, 261 F. Supp. 3d at 481 (“[O]n a motion to dismiss, the
Court is not at liberty to second-guess plaintiffs’ well-pled § 1 claim on the basis of the ‘market realities’ defendants
derive from secondary sources.”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 57 of 155
45
36
36
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 58 of 155
46
37 38
37
Sharing current pricing information among horizontal competitors is not mere “market
color.” See Section I.B.3, above.
38
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 59 of 155
47
B. As Predicted by the Academic Literature, the Reduction of Competition
Demonstrably Widened Spreads Beyond the Specific Transactions in the
Chats to Date
Academic literature supports the conclusion that “Defendants’ collusion harmed their
customers because it greatly diminished the amount of dealer competition in the market.” SAC
¶ 496. In fact, “[t]he academic literature has long established that more competition among
dealers reduces spreads.” Id. ¶ 497. Studies found that trading costs decline rapidly the more
dealers are actively (independently) providing quotes. Id. ¶¶ 498-502. Here, Defendants are all
huge players in their own right—even more so when taken together. Id. ¶ 505. A conspiracy
that, among other things, led such big players to refrain from providing bids and deliberately
bidding-to-miss would thus naturally result in wider spreads.
Plaintiffs have now linked their theory to the USD SSA bond market using economic
analyses. Id. ¶¶ 508-51. That the conspiracy had statistically significant impacts measurable
even in market-wide summary data confirms the plausibility of the allegation that the conspiracy
impacted far Put
another way, the analyses confirm that the academic literature’s prediction that a reduction of
competition would widen spreads holds true for this market. The analyses thus confirm that
Plaintiffs have plausibly pled that they were harmed by the conspiracy,
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 60 of 155
48
1. A summary of the SAC’s statistical analyses
The SAC explains that, although the USD SSA bond market is relatively opaque, there is
some data that allow for analysis of market-wide prices. Id. ¶ 509. The best publicly available
data is from Bloomberg, which provides three data series tracking daily bid-ask spreads and
yields for bonds classified as “Supranational, Sub-Sovereign, and Agency.” Id.39 While the
Bloomberg data does not allow for the identification of individual bids, offers, or prices, it does
provide pricing information for a diverse range of thousands of USD SSA bonds, and takes into
account over 1.8 million pricing data points. Id. Plaintiffs retained three separate economic
expert consultants to analyze the Bloomberg data. Id. ¶ 510. The consultants operated
independently from each other and independently decided which tests to run. Id. The
consultants ultimately conducted 16 analyses—each of which confirms the plausibility of the
allegation that the Defendants’ reduction in competition harmed Plaintiffs. Id. ¶¶ 512-51.
First, a regression model tests for the impact of Defendants’ conspiracy on the bid-ask
spreads for USD SSA bonds. Id. ¶¶ 512-22. A regression model is an accepted method to detect
the existence and impact of a price-fixing conspiracy. Id. ¶ 512. This regression model uses all
three of the USD SSA data series available from Bloomberg. Id. ¶ 513. The model also uses a
broad set of relevant variables derived from the academic literature to control for things that
would validly impact bid-ask spreads, such as movements in related instruments, or changes in
the time to maturity for a given bond. Id. ¶ 514. Another variable was used to indicate whether
the pricing information was being drawn from the core conspiracy period—i.e., January 2009 to
39 Contrary to Defendants’ suggestion, Merits Mem. at 15, the SAC identifies the data used for the
analyses. Defendants—which include some of the world’s most sophisticated financial actors—cannot realistically
claim to be ignorant of how Bloomberg works, particularly when they use data from the same data series for their
own purposes. See Merits Mem. at 11. Though Defendants already well-know this, and it is irrelevant, the three
data series are referred to in Bloomberg as “BGN,” “CBBT,” and “BVAL.”
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 61 of 155
49
early December 2015. Id. ¶ 515-17. The role of this “Collusion Indicator” is to detect if bid-ask
spreads were higher (or lower) during the core conspiracy period than before or after, controlling
for the factors identified above that can legitimately cause spreads to vary. Id.
Across all three Bloomberg data series, the Collusion Indicator is positively associated
with the actual bid-ask spreads for the bonds to a statistically significant degree. Id. ¶ 519. This
indicates that bid-ask spreads were higher during the core conspiracy period than they were
either before or after, even after controlling for economic factors that can legitimately cause
changes in spreads. Id. ¶ 520. Notably, this result holds to a statistically significant degree even
after further controlling for any lingering effect of the European debt crisis. Id. ¶ 521.
A second regression model uses recognized, legitimate economic factors, but does not
include a “Collusion Indicator” variable. Id. ¶¶ 523-24. Instead, the model predicts what bid-ask
spreads “should” be over time, again using a broad set of control variables supported by
academic literature. Id. The model accurately estimates the actual bid-ask spreads before
January 2009 and after December 2015 using these standard factors. Id. ¶ 525. But during the
core conspiracy period, these factors are unable to explain the observed bid-ask spreads. Instead,
the actual bid-ask spreads were considerably higher than can be accounted for by the control
variable set. Id. This confirms that, during the core conspiracy period, bid-ask spreads were
higher than they would have been if not for the conspiracy. Id. These results are statistically
significant.
Third, yet another regression model is used to predict yields, rather than spreads, again
using a variety of standard economic factors indicated by the academic literature. Id. ¶ 526. The
results of this predictive model are again compared to real-world yield data provided by
Bloomberg. Id. Again, the model does very well for the periods before January 2009 and after
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 62 of 155
50
December 2015—meaning that the standard economic factors, which are capable of predicting
yields with high accuracy, explain 96% of the yield movements during those years. Id. ¶ 527. In
contrast, during the core conspiracy period, the model performed much worse—explaining only
85% of the movements during those years. The difference in performance is statistically
significant. Id.
Fourth, the same regression model, estimated in the pre- and post-periods and then
applied to the conspiracy period, is used to measure the “excess volatility” of yields for USD
SSA bonds. Id. ¶ 528. The result is consistent with the other studies: excess volatility is
significantly larger during the core conspiracy period. Id. ¶ 529. This result indicates that
Defendants were pushing yields to both artificially high and artificially low levels, causing yields
to bounce around more than what the regression model can account for during the core
conspiracy period. Id. These results are statistically significant.
Fifth, still another regression was run, with this one trying to predict bid-ask spreads for
USD SSA bonds by using the then-current level of volatility in the USD SSA bond market and
the bid-ask spreads for sovereign (not SSA) bonds. Id. ¶ 530. The relevance of these distinct
control variables is confirmed by the fact they are all related, to a statistically significant degree,
to the actual bid-ask spread for the studied USD SSA bond as seen in the Bloomberg data. Id.
However, as with the other studies above, the model breaks down during the core conspiracy
period. During those years, actual bid-ask spreads were much higher than the model predicts.
Id. ¶¶ 530-31. This difference is statistically significant. Id. This again shows that the market
was being made artificial, to a measurable degree, throughout the conspiracy period.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 63 of 155
51
Sixth, the data show that both the mean and median bid-ask spreads for the studied bonds
were, in fact, higher during the core conspiracy period than during the periods before and after.
Id. ¶ 533. These results are statistically significant.
Seventh, performing a similar analysis but comparing the results to spreads for U.S.
Treasuries finds that there is a gap between spreads for USD SSA bonds and spreads for
Treasuries that is inexplicably wider during the core conspiracy period. Id. ¶ 534. In other
words, the spreads for USD SSA bonds during the core conspiracy period were too high, even
after using U.S. Treasuries as a control variable to account for underlying movements in the
macroeconomic environment. Id. These results are statistically significant.
Eighth, performing a similar analysis but comparing the results to spreads on sovereign
debt issued by Germany, France, and the United Kingdom finds that there is a gap between
spreads for USD SSA bonds and spreads for those sovereign instruments that is inexplicably
wider during the core conspiracy period. Id. ¶ 535. In other words, the spreads for USD SSA
bonds during the core conspiracy period were too high, even after using sovereign European debt
to account for underlying movements in the macroeconomic environment. Id. These results are
statistically significant.
Ninth, performing a similar analysis but comparing the results to spreads to a larger
basket of sovereign debt still finds that there is a gap between spreads for USD SSA bonds and
spreads for those sovereign instruments that is inexplicably wider only during the core
conspiracy period. Id. ¶ 536. In other words, the spreads for USD SSA bonds during the core
conspiracy period were too high, even after using a large basket of sovereign European debt to
account for underlying movements in the macroeconomic environment. Id. These results are
statistically significant.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 64 of 155
52
Tenth, a conspiracy to maintain artificially high bid-ask spreads can also manifest itself in
spreads being relatively stable. Id. ¶ 538. One way to detect this is by testing how well the
spread from one day on a given bond did at predicting the spread on that same bond the next day.
Id. ¶ 539. A higher level of predictability is consistent with a conspiracy to keep spreads
artificially high. Id. That is just what the data show: the level of predictability from one day to
the next is much higher during the core conspiracy period than for either the before or after
periods. Id. These results are statistically significant. Id.
Eleventh, another way to see if the conspiracy made bid-ask spreads more stable is to
measure the “coefficient of variation” among the bid-ask spreads for all bonds traded on a given
day. Id. ¶ 540. A higher average coefficient of variation indicates that spreads were more
different from each other, whereas a lower coefficient of variation indicates that spreads were
more clustered together. Id. Such a clustering would be consistent with a conspiracy to keep
spreads artificially high. Id. ¶ 541. Again, that is just what the data show: spreads for USD
SSA bonds were more clustered together during the core conspiracy period than they were
during the before or after periods. Id. Though not mentioned expressly in the SAC, to confirm,
these results are statistically significant.
Twelfth, the same coefficient of variation analysis run using U.S. Treasuries as a control
for underlying macroeconomic changes still finds that spreads for USD SSA bonds were more
clustered together during the core conspiracy period than they were after. Id. ¶ 542. These
results are statistically significant.
Thirteenth, a direct study of the variability of SSA bond yields confirms what the fourth
analysis also found: the conspiracy to keep bid-ask spreads at artificially high levels also
introduced extra volatility into the yields of USD SSA bonds. Id. ¶ 544. That is because a
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 65 of 155
53
conspiracy that manifested itself most strongly when Defendants were pushing yields artificially
low when selling, then pushing yields artificially high when buying, would cause yields to
bounce around more than they otherwise would have. Id. For the majority of the USD SSA
bonds studied, this analysis shows yields being much more volatile during the core conspiracy
period than they were during the before and after periods. Id. ¶ 545. Moreover, on average
across all bonds studied, yield volatility was higher during the core conspiracy period. These
results are statistically significant.
Fourteenth, the same yield volatility analysis run using U.S. Treasuries as a control still
finds that yields for USD SSA bonds were more volatile during the core conspiracy period than
they were after. Id. ¶ 546. These results are statistically significant.
Fifteenth, the same yield volatility analysis run using sovereign debt issued by Germany,
France, and the United Kingdom as a control still finds that yields for USD SSA bonds were
more volatile during the core conspiracy period than they were before and after. Id. ¶ 547.
These results are statistically significant.
Finally, given that they share similar features, the yields of different bonds from the same
issuer would be expected to move in similar ways in response to various stimuli. Id. ¶ 549.
However, if the conspiracy put artificial pressure on a particular bond that a particular conspiring
Defendant was interested in buying or selling on that particular day, then the correlation in yields
between that bond and related bonds would be reduced. Id. ¶ 550. That is just what the data
show. Id. ¶ 551. Thus, again, the data show yields acting in a way consistent with the operation
of a conspiracy impacting the market. Id. These results are statistically significant.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 66 of 155
54
2. The SAC is sufficiently detailed
In response to the phalanx of analyses linking the academic literature’s expectations of
widening spreads with observations of spread and pricing behavior in the USD SSA market,
Defendants initially try to paint the SAC as insufficiently detailed. Each such argument fails.
Defendants say that “much” of the analysis consists of “conclusory opinions.” Merits
Mem. at 15. To the contrary, as seen above, the SAC describes the methodology and results of
each of 16 analyses. Allegations such as these, about how yields and spreads moved over time,
are “factual analysis regarding pricing and other economic data, which courts generally accept at
the pleading stage.” Silver, 213 F. Supp. 3d at 563.40 The cases cited by Defendants are not to
the contrary. In re NYSE Specialists Securities Litigation, 503 F.3d 89, 95 (2d Cir. 2007), had
nothing to do with data analyses in a complaint—the court there merely stated that legal
conclusions need not be presumed true. And City of Royal Oak Retirement System v. Juniper
Networks, Inc., 2013 WL 2156358, at *7 (N.D. Cal. May 17, 2013), concerned the irrelevant
issue of whether a plaintiff may submit a separate expert affidavit to oppose a motion to dismiss.
All the analyses here are described in the SAC.
Tellingly, despite charging that “much” of the SAC is conclusory, Defendants only
actually complain that a single regression analysis did not provide specific “numerical results.”
Merits Mem. at 15. But that regression model is fully detailed, as are the results. SAC ¶¶ 512-
22. The SAC explains the results in layman’s terms: the standard factors seen in the academic
literature were found to actually have a statistically significant relationship to bid-ask spreads
40 See also Gold, 213 F. Supp. 3d at 666 (same); In re LIBOR-Based Fin. Instruments Antitrust Litig., 935
F. Supp. 2d 666, 679-80, 716-17 (S.D.N.Y. 2013) (accepting plaintiffs’ proffered analyses comparing whether
banks’ rates were grouped and comparing LIBOR to other data); Dover v. British Airways, PLC (UK), 2014 WL
317845, at *2 (E.D.N.Y. Jan. 24, 2014) (noting that plaintiff’s statistical analysis of prices “is a factual allegation
that the Court must credit”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 67 of 155
55
“no matter which of the three Bloomberg series were used.” Id. ¶ 518. Similarly, the “Collusion
Indicator” was also found to be related “to a statistically significant degree—even at the 99
percent confidence level—across all three alternative Bloomberg data series.” Id. ¶ 519.
Knowing that the adjusted R-squared for the default premium variable in the BGN data series
was (hypothetically) 0.1720 or some other multi-decimal figure, would add nothing.
Defendants next argue that the analyses must be discarded because Plaintiffs allege that
the USD SSA market is “opaque.” Merits Mem. at 15. What matters for making particular
trading decisions is real-time, granular, actionable information—which is not made available.
SAC ¶¶ 509, 553-54. That the market is thus “opaque” for purposes of transacting at a particular
moment is a different issue from whether the publicly available data is sufficient to analyze for
long-term trends. The daily, summary data provided by Bloomberg—over 1.8 million data
points used here—is more than “reliable” for those distinct purposes, as evidenced by the fact
that the same data are used in the finance research literature to study market-wide phenomenon
like those at issue here. Id. ¶ 513 n.69.
Finally, Defendants contradict themselves by simultaneously arguing that Plaintiffs used
not enough data but also too much. They do this by arguing initially that the failure to use
literally all of the Bloomberg data constitutes nefarious modification of the data. Merits Mem. at
15. To the contrary, all that was done to the Bloomberg data was to reduce noise in the data by
eliminating a small number of outlier bonds for which sufficient data points did not exist; a small
number of dates on which an insufficient number of bonds traded; and a small number of
observations for which the data appeared unreliable. Id. ¶¶ 508-10, 509 n.67.
But then Defendants turn to complain that an analysis on a dataset as large as that which
Plaintiffs did use is per se meaningless because it includes a “variety of unrelated types of
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 68 of 155
56
bonds.” Merits Mem. at 21. The bonds used in Plaintiffs’ analysis are all grouped by
Bloomberg under the category “Supranational, Sub-Sovereign, and Agency (SSA),” SAC
¶ 509—a well-recognized asset class of bonds that share common economic characteristics and
that were traded on Defendants’ SSA bond trading desks. The suggestion that all 16 tests using
1.8 million pricing data points and multiple control variables all happened to be the result of
“statistical anomalies” caused by differences amongst USD SSA bonds, Merits Mem. at 21, is
absurd.41
The far more plausible conclusion is that the data consistently show a difference between
the before, during, and after periods because there was a difference between those periods.
Unsurprisingly then, other courts have refused to dismiss complaints incorporating multi-faceted
economic analyses that involved averages. For instance, in Gold, Judge Caproni relied on
comparisons of average price differences before and after a conspiracy period of similar duration
in upholding plaintiffs’ allegations of a conspiracy. 213 F. Supp. 3d at 646. Similarly, in
ISDAfix, Judge Furman accepted the plausibility of a complaint that presented averages across
multi-year periods. 175 F. Supp. 3d at 55.
Defendants instead direct the Court’s attention to a silver manipulation case before Judge
Patterson, which is distinct from the data-driven one approved by Judge Caproni. In that other
silver case, unilateral price manipulation was supported by only a few “specific days” studies. In
re Commodity Exch. Inc. Silver Futures and Options Trading Litig., 2013 WL 1100770, at *4 n.4
(S.D.N.Y. Mar. 18, 2013). Here, Plaintiffs have analyzed data on over 1.8 million SSA bond
41 Indeed, the main explanation Defendants give for their argument that “Plaintiffs improperly rely on
aggregated data” is that the volume of a bond could fluctuate day to day because of a new issuance. Merits Mem. at
21. But if anything, this confirms the power of Plaintiffs’ multi-year, 1.8 million-data point analysis. While
Defendants’ hypothetical might explain why one bond might see changes from one specific day to the next, such a
phenomenon would not explain why USD SSA bonds overall showed measurable differences during the seven-year
conspiracy period than before or after.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 69 of 155
57
transactions over the course of several years, and have shown the prevalence of collusion during
the class period through multiple statistical studies. See id. (suggesting plaintiff should have
instead presented an “averaged longer term basis” study). In other words, Plaintiffs here did the
exact type of analysis called for by the very case Defendants use to attack the use of
“averages.”42
3. The analyses reasonably relate to Plaintiffs’ claims
Defendants argue that the analyses must be disregarded because they do not relate to
Plaintiffs’ specific transactions. Merits Mem. at 16. Multiple studies, across multiple years,
across a large swath of the USD SSA bonds that were ever traded during the relevant period, find
prices acting anomalously. This buttresses the plausibility of the allegation that the conspiracy
not to compete was so insidious and continuous that named Plaintiffs were harmed,
This plausibility support holds true even if it
were the case that the publicly available data did not happen to capture information on a
particular bond at a particular point in time.43 Even so, to confirm what was thought to be
obvious: a study utilizing almost all of the publicly available data on the USD SSA market did,
in fact, cover nearly all the bonds Plaintiffs traded.
Defendants also argue that Plaintiffs’ analyses do not relate to their claims because there
was no “control” for the activity of other USD SSA market-makers. Merits Mem. at 20-21. As
42 Engineering Contractors Association of South Florida Inc. v. Metropolitan Dade County, 122 F.3d 895,
919 n.4 (11th Cir. 1997), also cited by Defendants, Merits Mem. at 21 n.34, is even further afield. There, after a
bench trial involving testimony from multiple experts, the court merely noted issues that should be considered when
determining how much weight to give aggregated versus disaggregated data on individual employment statistics.
43 Defendants ask whether the analyses “included bonds and pricing information from interdealer trades.”
Merits Mem. at 16. But Defendants well know their question makes no sense. Bloomberg’s historical data used by
Plaintiffs is available on a daily basis and does not report individual “trades.” SAC ¶ 509. There thus are no
“interdealer trades” that were analyzed, because the publicly available data contains no specific trades at all. The
market-wide, daily data is nonetheless relevant for the reasons discussed above. Even if that were not the case, the
supposed distinction between the interdealer space and the client space is also irrelevant here, as discussed in
Section III.A below.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 70 of 155
58
an initial matter, Plaintiffs’ analyses could not be more granular because Defendants do not make
their own trade data publicly available. SAC ¶ 509. As discussed in Section I.C.5 above, courts
recognize the unfairness of holding plaintiffs to impossible standards in such situations. But
even that aside, an analysis of the Bloomberg data is still relevant despite not being Defendant-
specific.
The impact of that conspiracy is predicted by the academic literature to be
wider bid-ask spreads. Accordingly, on the issue of standing, the analyses are performing a
relatively small function: confirming a well-established conspiracy caused the academically
predicted result of impacting customers 44
Market-wide data is more than up to that task. Indeed, if anything, using market-wide, rather
than Defendant-specific, spread information would be expected to understate the impact of the
conspiracy. That the conspiracy can be observed in the market-wide data despite its limitations
is thus, if anything, a testament to the strength of Plaintiffs’ allegations.
4. The time periods used in the analyses are consistent and proper
Defendants argue in various ways that the analyses should be discarded because Plaintiffs
should have chosen different time periods to study. Merits Mem. at 16-17. But a motion to
dismiss “is not the proper stage to determine” either “the accuracy” of plaintiffs’ allegations or
the “methodology” that underlies them. John v. Whole Foods Mkt. Grp., Inc., 858 F. 3d 732, 737
(2d Cir. 2017); see also Dover, 2014 WL 317845, at *2 (an allegation based on data “is a factual
44 By contrast, in Harry v. Total Gas & Power North America, Inc., 244 F. Supp. 3d 402 (S.D.N.Y. 2017),
the defendants allegedly manipulated the price of natural gas commodities at four regional trading hubs, but the
plaintiffs did not trade with the defendants, or on any of the regional hubs. Id. at 407-09. Rather, plaintiffs traded
futures contracts on a national exchange, which were not even based on the prices at any of the regional hubs. Id.
This disconnect between the manipulated market and plaintiffs’ actual trades is why analyses of pricing data from
the regional hubs was insufficient to allege an actual injury. Id. at 413. This says nothing of the relevance of
Bloomberg’s market-wide daily data on USD SSA bonds, to a case about USD SSA bonds.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 71 of 155
59
allegation the Court must credit” as true on a motion to dismiss); In re Pfizer Inc. Secs. Litig.,
584 F. Supp. 2d 621, 635 (S.D.N.Y. 2008) (a motion to dismiss “is not an appropriate vehicle for
determination as to the weight of the evidence, expert or otherwise”). Thus, courts reject as
“premature” methodological challenges to data analyses in a complaint. Branca v. Nordstrom,
Inc., 2015 WL 10436858, at *7 (S.D. Cal. Oct. 9, 2015); In re Ambac Fin. Grp., Inc. Sec. Litig.,
693 F. Supp. 2d 241, 272 (S.D.N.Y. 2010) (“To evaluate whether or not” plaintiffs’ data analyses
“are adequate would require the Court to delve into complex factual disputes that cannot be
resolved on a motion to dismiss.”); see also Heckler & Koch, Inc. v. German Sport Guns GmbH,
2014 WL 12756372, at *11 (S.D. Ind. May 15, 2014) (“calling into question [an expert’s] . . .
methodology . . . [would be] premature”).45
It is worth noting that Plaintiffs’ general selection of 2009-2015 as the time period was
prescient—it is the same window in which the European Commission has charged some
Defendants with collusion. If anything, it confirms that Plaintiffs’ refinement of the Class Period
was not arbitrary, but was backed by solid evidence.
It is proper to start the analyses prior to January 1, 2009. Defendants mislead by saying
it was “cherry picking” to have “none” of the “start dates for the analysis” be “consistent with
Plaintiffs’ allegations.” Merits Mem. at 16-17. That the analyses do not begin at January 1,
45 Defendants suggest they are interested in mounting yet another premature methodological challenge in a
footnote, where they query whether Plaintiffs initially set the “parameters” for their regressions using data from
within the conspiracy period or data outside the conspiracy period. Merits Mem. at 16 n.22. But statisticians find
reasons to consider both approaches depending on the circumstances. Notably, initially setting the parameters using
seed data that includes the conspiracy period has the potential disadvantage of understating the impact of any
conspiracy by forcing the model to accept some of the tainted price movements as if they were clean. It is no
wonder that Defendants suggest (incorrectly) that is the only acceptable approach. In any event, some of the SAC’s
regressions used one approach, while others used the other—for instance, this is a major differentiator between the
first and second regressions. That all of the analyses reached the same result—a detectible conspiracy during the
alleged core conspiracy years—thus further confirms the irrelevance of Defendants’ premature methodological
challenge. For similar reasons, that all of the analyses found signs of the conspiracy despite the 2008 economic
crises sometimes being within the measurement period, and sometimes not, confirms the robustness of Plaintiffs’
results.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 72 of 155
60
2009, is a feature, not a bug. Studying earlier years allows for a comparison of behavior during
the core conspiracy years—confirming that spreads were not just ‘wide’ but abnormally wide in
ways that could not be explained by macroeconomic factors. It thus would have made no sense
to start all of the analyses at January 1, 2009, if the data allow a particular analyses to go back
further. And as for the question of how many early years to include in the comparative period for
each study, that was not the result of “cherry-picking,” either, but rather was driven by the
availability of data. SAC ¶¶ 509-10, 516 n.74.
It is proper to end the core conspiracy period in December 2015. Defendants also argue
that the analyses’ choice to measure before and after December 2015 is incorrect because of
chatroom restrictions in 2013 and news stories in 2014. Merits Mem. at 17-18. The suggestion
that those earlier events “should” have stopped the conspiracy and thus lead Plaintiffs to use an
earlier break point in their analyses cannot be squared with the fact that the SAC contains
evidence suggesting that Defendants were actually still colluding through 2015.
20 (DOJ investigation launched at end of 2015); 145-158
(conspiring traders mostly left their banks in late 2015 or early 2016); 381 (first reports of
government investigations into collusion in SSA market in December 2015). Plainly then, the
other events Defendants hypothesize should have stopped the conspiracy earlier, did not.46 They
thus cannot show, particularly at the pleading stage, that the analyses should have used a
different date for the end of the core conspiracy period.
The pre- and post-periods are sufficiently long to provide meaningful results.
Defendants also label the non-conspiracy control periods “short.” Merits Mem. at 18. But many
46 It should be noted that Defendants relatedly posit that Plaintiffs’ analyses is relevant only if one
presumes the conspiracy “abruptly” started and stopped. Merits Mem. at 17. It does not, as discussed in Section
II.B.5 below.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 73 of 155
61
of the studies go back to 2006, id. ¶¶ 524, 29, 33-35, 39, 41, 45, 51, and one goes all the way
back to 2002, id. ¶ 536. That means there were over three years prior to 2009 that were studied,
as well as three years after. As discussed above, the length of those studies were dictated by the
available data—including, most obviously that Plaintiffs could only use so much data after
December 2015 because only so much time has passed since then. More basically, Defendants
present no statistical reason why pre- and post-period control periods must be the same length as
an alleged conspiracy. In fact, there is no such reason. What matters is whether sufficient data
exists in each period for statistical methods to establish, with statistical significance, whether
there were differences between them. That is what Plaintiffs’ analyses have established here.
The data cannot be ignored merely because the Class Period has changed. Defendants
also argue that Plaintiffs’ use of the years prior to 2009 as control periods is inconsistent with
Plaintiffs’ prior allegation in the CAC that the conspiracy was in place as early as 2005. Merits
Mem. at 16-17. But “[i]t is well-established that an amended pleading supersedes the original
pleading,” Kelley v. Crosfield Catalysts, 135 F.3d 1202, 1205 (7th Cir. 1998), and facts from the
original pleading that are not incorporated into the amended pleading “cannot be considered by
the court on a motion to dismiss the amended complaint,” id.47
Nor is this an instance of “contradiction,” as Defendants posit. This is because the data
are what they are, and Plaintiffs made no allegations about the data in the prior consolidated
complaint—the CAC allegation about the length of the conspiracy was based entirely on when
the individual cartel members became employed by the Dealer Defendants. See Dkt. No. 388 at
47 See also Harris v. City of N.Y., 186 F.3d 243, 249 (2d Cir. 1999); Shields v. Citytrust Bancorp, Inc., 25
F.3d 1124, 1128 (2d Cir. 1994) (“an amended complaint ordinarily supersedes the original, and renders it of no legal
effect”); 3 Moore’s Federal Practice § 15.17[3] (3d ed. 2018) (“An amended pleading that is complete in itself and
does not reference or adopt any portion of the prior pleading supersedes the prior pleading.”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 74 of 155
62
57-59.48 There is also no contradiction because it is sensible that the conspiracy was in effect
earlier but grew over time. And if the conspiracy did exist in some weaker form prior to 2009,
that would only lessen the effects of the conspiracy shown in Plaintiffs’ analysis during the core
years of the conspiracy (2009-2015). That Plaintiffs found such a sizable impact during the
Class Period, despite the possibility that the conspiracy pre-dated the Class Period in a weaker
form, only reinforces Plaintiffs’ analyses.
This case thus shares nothing with those Defendants use to argue that “contradictions” in
prior pleadings can justify ignoring otherwise well-pled allegations. Merits Mem. at 17. Those
cases involved plaintiffs who reversed course on simple facts that had always been in their
possession. In Colliton v. Cravath, Swaine & Moore LLP, 2008 WL 4386764 (S.D.N.Y. Sept.
24, 2008), the plaintiff originally alleged he was employed as an attorney—as evidenced by a
letter-agreement—but then amended to allege he was not employed as an attorney. Id. at *6. In
Palm Beach Strategic Income, LP v. Salzman, 457 F. App’x 40, 44 (2d Cir. 2012), the plaintiff
litigated the case for two years under the terms of one contract, only to try to argue that a
completely different agreement was at issue in her breach of contact case. Id. Such cases do not
support ignoring factual allegations derived from economic investigations that were not yet
performed at the time of the CAC, so as to artificially deny Plaintiffs the benefit of the facts as
they actually exist in the data.49
48 Defendants also attempt to manufacture contradictions by referring to a pre-consolidation complaint.
Merits Mem. at 17 n.26, 18 n.27. But the consolidation process could not function if class members were boxed in
by pre-consolidation allegations. Thus, a consolidated complaint in a multi-district litigation that “merg[es] the
discrete actions” necessarily “supersede[s] prior individual pleadings.” Gelboim v. Bank of America Corp., 135 S.
Ct. 897, 904 n.3 (2015); see also Broomfield v. Craft Brew Alliance, 2017 WL 3838453, at *4 (N.D. Cal. Sept. 1,
2017) (refusing to compare allegations in original complaints to those in consolidated complaint); In re LIBOR-
Based Fin. Instruments Antitrust Litig., 27 F. Supp. 3d 447, 484 (S.D.N.Y. 2014) (consolidated complaint in MDL
“superseded the previous complaints”).
49 Defendants’ citation to Perry v. NYSARC Inc., 424 Fed. App’x 23 (2d Cir. 2011), Merits Mem. at 40-41,
also misses the mark. In that case, the complaint did not contain any factual allegations that actually supported the
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 75 of 155
63
5. The analyses all reasonably relate to the presence of a conspiracy
Defendants argue there are other innocent explanations for why spreads widened to an
unusual degree during the core conspiracy period. Merits Mem. at 18-21. But the presence of
the conspiracy is well-pled. See Section I, supra. And the academic literature predicts, sensibly,
that the resulting reduction in competition would cause spreads to be wide. See Section II.B,
supra. The data then show that spreads were, by 16 different measures, behaving in a way that is
consistent with that prediction. See Section II.B, supra. Plaintiffs have thus more than met their
burden of plausibly pleading they were impacted by the conspiracy, regardless of what
counterfactuals Defendants might dream up to try to explain away the data. See Anderson News,
680 F.3d at 185 (“The choice between two plausible inferences that may be drawn from factual
allegations is not a choice to be made by the court on a Rule 12(b)(6) motion.”). Courts have
consistently rejected attempts to explain away the results of data analyses even in cases where,
unlike here, the data were almost the only facts used to show the conspiracy existed in the first
place. See Gold, 213 F. Supp. 3d at 665 (alternative explanations “unpersuasive at this stage of
the litigation” because “the Court need not find that Plaintiffs’ theory is the only plausible
explanation” for price patterns); Silver, 213 F. Supp. 3d at 561-64 (rejecting attempts to explain
away economic trends); ISDAfix, 175 F. Supp. 3d at 55 (upholding plaintiffs’ claims even though
“Defendants offer plausible non-collusive explanations for many of the facts alleged”).
Each of Defendants’ alternative explanations—which often nitpicks at only a small part
of the 16 different studies—also fails on its own terms.
plaintiff’s claim for workplace discrimination. Id. at 25-26. To the contrary, many of the facts alleged instead
showed that her employer’s actions were “a reasonable effort to accommodate [her] health condition.” Id. As
discussed above, here the SAC is both internally consistent and replete with factual allegations supporting Plaintiffs’
claims.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 76 of 155
64
For instance, Defendants argue that the analyses fail to account for macroeconomic
factors. Merits Mem. at 18-19. Not so. The first major regression uses as inputs every relevant
factor identified in the academic literature as driving the bid-ask spread of a bond over time.
SAC ¶ 514. The test was even specifically broken up to isolate any alleged effects of the so-
called European debt crisis. Id. ¶ 521. The regressions used in the second, third, and fourth
studies similarly used a wide variety of relevant factors supported by the academic literature. Id.
¶¶ 523-29. The fifth regression tested whether volatility from one day accurately predicted
volatility the next; thus, by definition the absolute level of volatility shifting was being accounted
for. Id. ¶ 530. Even the other, more straightforward studies repeatedly showed how the results
held even when incorporating other data points as “controls,” in that they would be expected to
move in response to “macroeconomic factors” but not in response to the conspiracy, including
U.S. Treasuries, debt issued by other specific countries, and even a larger basket of sovereign
bonds. Id. ¶¶ 534-36, 542, 546-47.
Defendants ignore all this, simply positing that the SAC “fails to control” for anything,
before quibbling with certain of the expert’s choices. Merits Mem. at 18-20. Specifically,
Defendants argue certain controls might have been inappropriate if they were denominated in a
different currency. Merits Mem. at 19.50 They then in a footnote inappropriately posit it is “well
known” that certain countries experienced different macroeconomic events differently than
others. Id. at 19 n.30.
Plaintiffs allege Defendants’ conspiracy impacted all of their USD SSA trades. One valid
way of isolating the effect of the conspiracy is to compare USD SSA data across time, when the
50 Because the analyses were done in terms of interest rate yields, not currency units, there was no
currency-conversion step that might have introduced unrelated noise. In other words, because the yields of bonds
denominated in Euros are directly comparable to the yields for bonds denominated in USD, this was an apples-to-
apples comparison.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 77 of 155
65
conspiracy was in different phases—which Plaintiffs did. Another valid way of isolating the
effect of the conspiracy is to compare movements of USD SSA instruments to other instruments
at the same point in time—which Plaintiffs did as well. That Plaintiffs could not find a control
group of un-manipulated SSA bonds denominated in U.S. dollars and traded during the core
conspiracy period to perfectly isolate the effects of the conspiracy, is only because no such
control group exists. The “failure” to do the impossible is nothing to fault Plaintiffs for, and
nothing that demonstrates the robust findings here, across multiple years and tests, fall short of
making Plaintiffs’ claims plausible. Defendants do not and cannot argue that courts must
disregard results of economic analyses unless a perfect control group is found, let alone that
courts should do so at the pleading stage where mere plausibility is the question.51 To the
contrary, even at the Daubert phase and beyond, courts have held that disagreements over which
control variables should have been used are not a basis to strike an expert’s findings.52
51 Defendants argue that Plaintiffs’ analyses were flawed because they did not deploy all potential controls
at the same time. Merits Mem. at 19-20. Undoubtedly, no matter what controls Plaintiffs used, Defendants would
find some combination of controls that Plaintiffs had not used. But simply noting that another set of controls could
have been used in no way indicates doing so would have been preferable, let alone necessary. Defendants do not
claim, and cannot establish at the pleading stage, that only statistical tests with a combination of all the controls used
across all the tests can produce meaningful results. Similarly unavailing is Defendants’ criticism of one
supplemental test for splitting the conspiracy period into two sub-periods. Id. at 20 n.31. That was done—after the
already control-laden regression model had shown the impact on spreads during the conspiracy period as a whole—
to even further control for Defendants’ supposedly all-explaining event: the European debt crisis. SAC ¶ 521. That
the result remained the same—i.e., the collusion indicator was again found to be statistically significant at the 99-
percent confidence level—further confirms the impact of Defendants’ conspiracy, regardless of whether
“macroeconomic conditions were perfectly constant” in one time period or another. Id. ¶¶ 518-22.
52 See, e.g., In re Air Cargo Shipping Servs. Antitrust Litig., 2014 WL 7882100, at *14-15 (E.D.N.Y. Oct.
15, 2014) (holding that the court was “unconcerned” about “minutiae” such as which specific controls were used,
because “expert statisticians may differ on these issues”); Deutsch v. Novartis Pharms. Corp., 768 F. Supp. 2d 420,
456 (E.D.N.Y. 2011) (denying motion to exclude based on fact that the expert “concluded it was not necessary to
control for” certain factors); Allen v. Int’l Bus. Machs. Corp., 1997 WL 34501372, at *26 (D. Del. Dec. 18, 1997)
(recognizing that “no study is perfect or completely free of possible confounding,” and that even studies that “do not
control for confounding factors” may be considered at summary judgment); see also Buchanan v. Tata Consultancy
Servs., Ltd., 2017 WL 6611653, at *10 (N.D. Cal. Dec. 27, 2017) (holding that economic literature does not say that
“studies which fail to control for certain . . . variables are inherently flawed or unreliable,” but rather “simply
encourages the researcher to try to control for such variables when sufficient data is available”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 78 of 155
66
Defendants also assert that a “simple comparison” of one of Plaintiffs’ studies against
publicly-available reports shows that Plaintiffs did not properly control for the impact of the
European debt crisis. Merits Mem. at 20. It is true that for that one study the gap between
expected and actual spreads was widest in early 2012, when Defendants posit was the European
debt crises. SAC ¶ 524. Even within that single study, this observation fails to explain why the
gap was in existence at all before that point, and after—
and closing right when the DOJ announced its investigation in December
2015. And, again, picking at one study in no way reasonably calls into question the combined
effect of all Plaintiffs’ studies, including those that added additional “control” factors into their
analysis.
This case is thus nothing like those cited by Defendants for the proposition that the
analyses should be ignored. For instance, Munoz v. Orr, 200 F.3d 291 (5th Cir. 2000) involved a
Daubert motion where an expert’s report was excluded after he admitted in depositions that he
failed to consider key explanatory variables and did not perform a multiple regression analysis.
Id. at 301. Though irrelevant and unnecessary at the pleading stage, Plaintiffs here have already
cleared those bars: as discussed above, the consultants did consider the key explanatory
variables and performed multiple regression analyses. And in FrontPoint Asian Event Driven
Fund, L.P. v. Citibank, N.A., 2017 WL 3600425 (S.D.N.Y. Aug. 18, 2017), the only economic
evidence supporting the allegation that the SIBOR and SOR benchmark rates were manipulated
was a simple comparison showing that the difference between those two rates became more
pronounced. Id. at *11. That comparison failed because plaintiffs failed to account for the
different inputs for those two rates, and thus failed to explain why the rates should be the same in
the first place. Id.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 79 of 155
67
Here, the existence of the conspiracy is not based on a single study of two “dissimilar
data sets,” And the 16 studies
showing that the conspiracy had a market-wide impact focused directly on the market that was
impacted, and explained what controls were used, and why.53
6. Defendants’ remaining arguments are unavailing
Defendants note that some of the 16 analyses do not expressly allege that their results are
statistically significant. Merits Mem. at 21 n.33. The SAC does so for many. See, e.g., SAC
¶¶ 23, 24, 513, 518, 519, 522, 523, 527, 530-31, 539. As summarized in Section II.B.1 above,
though not expressly alleged as such, in fact all 16 of the analyses produced statistically
significant results.
Relatedly, Defendants argue that Plaintiffs do not allege that any of the results are “large
enough to have a meaningful impact on anyone.” Merits Mem. at 21 n.33. But, again, Plaintiffs
allege that Defendants’ collusion harmed all Class members on all their USD SSA bond
transactions during the Class Period. See, e.g., SAC ¶¶ 439. Given the massive size of the USD
SSA bond market, and Defendants’ dominance of that market, this means that many billions of
dollars of transactions were impacted. To the extent Defendants are demanding that Plaintiffs
quantify class-wide damages, that is obviously not required at the pleading stage, and the cases
that Defendants cite are not to the contrary.54 In any event, the economic analyses, which are
supported by academic literature, make clear that the conspiracy had a meaningful impact.
53 Defendants’ citation to an early ruling in the silver manipulation case before Judge Patterson, In re
Commodity Exch. Inc. Silver Futures and Options Trading Litigation, 2012 WL 6700236, at *12-13 (S.D.N.Y. Dec.
21, 2012), is similarly unavailing. There, the court merely held that a comparison of silver and gold prices was
insufficient to show that silver prices were manipulated, without any explanation of why the two markets were
comparable in the first place.
54 Burgis v. N.Y.C. Dep’t of Sanitation, 798 F.3d 63, 69 (2d Cir. 2015), cited by Defendants, merely held
that, for civil rights claims, statistical evidence of racial discrimination must be of a level sufficient to show
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 80 of 155
68
As a final gambit, Defendants resort to labeling the analyses “nonsensical.” Merits Mem.
at 22. To support such a sweeping charge, Defendants merely observe that one of the many
studies found that yields were abnormally high for a bit before January 2009, rather than starting
right then. Id. (referring to SAC ¶ 529). But the combined effect of numerous analyses
comparing the before, during, and after periods using statistical methods is not all trumped by
eyeballing one movement in one chart for one study. This is particularly true because the
common-sense relevance of the analyses does not turn on a presumption that the conspiracy
started on one specific day, then ended at another.
True, to make any statistical comparison between two periods, one must select some
specific date that divides those periods. But it is the overall difference across numerous bonds
across many years that demonstrates a unique pattern that occurred much more strongly during
the core conspiracy years of 2009 to 2015. This is true even if one allows for the possibility that
the conspiracy existed in some weaker form before 2009, after 2015, or both; in such a scenario,
the pre- and post-conspiracy periods would only appear more similar to the core conspiracy
period, thus dampening any effect shown in empirical tests. That Plaintiffs’ statistical tests find
such a strong pattern, even despite the possibility that the conspiracy existed in some form
outside the core period, makes those results all the more powerful.
Defendants similarly miss the mark by observing that bid-ask spreads were wider after
December 2015 than they were before January 2009. Merits Mem. at 22. Plaintiffs do not allege
that the pre-January 2009 spreads are the “correct” width for all time; of course, economic
factors can cause them to change over time. The relevance of those studies is that spreads got
discriminatory intent. And in Hinds County, 708 F. Supp. 2d at 360, the court upheld the validity of the data
analyses involving several defendants.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 81 of 155
69
wider then went back down, no matter how the data was looked at. SAC ¶ 533-37. That spreads
were widest during period of 2009 to 2015 helps make plausible the allegation that the
conspiracy— —was causing
spreads to be wide through the removal of competition, just as the academic literature predicted
would occur. This is particularly true given the SAC is not reliant on these comparatively simple
analyses at all; in fact, the regression analyses presented in the SAC demonstrate that even after
controlling for other economic factors that legitimately cause spreads to fluctuate over time,
spreads during the Class Period were significantly wider than those either before or after. The
bar graphs Defendants criticize make up just a small part of the analyses demonstrating the
uniqueness of behavior during the core conspiracy period.
III. PLAINTIFFS HAVE ANTITRUST STANDING AS DEFENDANTS’ DIRECT
CUSTOMERS
Defendants do not contest that this direct-purchaser class has antitrust standing under the
law. Nor could they: It is black letter law that consumers who purchase a product subject to
anticompetitive collusion directly from the cartel members have antitrust standing.55 Rather,
Defendants focus on the factual question of whether any named Plaintiffs were actually harmed.
See Merits Mem. at 8, 23 & n.27. As discussed above in Section II.A, Plaintiffs have more than
plausibly alleged that named Plaintiffs were harmed. The few legal arguments that Defendants
muster do not change the necessary conclusion that Plaintiffs have antitrust standing.
55 See Areeda & Hovenkamp ¶ 391b (“Direct purchasers from cartels have long had standing to recover
any collusive overcharges.”). Indeed, the Supreme Court has held that such plaintiffs are the “preferred” antitrust
plaintiffs because they have the most direct incentive to vindicate the antitrust laws. Illinois Brick Co. v. Illinois,
431 U.S. 720, 746 (1977) (“elevating direct purchasers to a preferred position as private attorneys general”). See
also Blue Shield of Virginia v. McCready, 457 U.S. 465, 474 (1982) (direct purchasers are the plaintiffs “most likely
to press their claims with the vigor that the § 4 treble-damages remedy was intended to promote”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 82 of 155
70
A. Plaintiffs Present a Clear Case of Antitrust Injury
“To demonstrate antitrust injury, a plaintiff must show (1) an injury-in-fact; (2) that has
been caused by the violation; and (3) that is the type of injury contemplated by the statute.”
Order at 14.
Plaintiffs easily meet this standard. Again, as discussed in Section III.A above, Plaintiffs
have plausibly alleged they suffered an injury-in-fact. Plaintiffs’ injuries were the direct result of
Defendants’ agreement not to compete. SAC ¶ 12. And Plaintiffs’ injuries—paying supra-
competitive prices, SAC ¶ 4—is precisely the type of injury contemplated by the antitrust laws.
This should end the Court’s analysis of antitrust injury, as “a consumer’s injury of having to pay
supra-competitive prices as a result of a horizontal price-fixing conspiracy is the quintessential
antitrust injury.” FX II, 74 F. Supp. 3d at 598.56
56 See also Gelboim, 823 F.3d at 772 (“Generally, when consumers, because of a conspiracy, must pay
prices that no longer reflect ordinary market conditions, they suffer injury of the type the antitrust laws were
intended to prevent and that flows from that which makes defendants’ acts unlawful.”); State of New York v.
Hendrickson Bros., Inc., 840 F.2d 1065, 1079 (2d Cir. 1988) (“In general, the person who has purchased directly
from those who have fixed prices at an artificially high level in violation of the antitrust laws is deemed to have
suffered the antitrust injury within the meaning of § 4 of the Clayton Act . . . .”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 83 of 155
71
57
Second, Defendants’ assertion that trading among dealers, and trading between dealers
and their customers, occur in two separate “markets” has no basis in the SAC. Nowhere in the
SAC do Plaintiffs allege there are distinct USD SSA bond markets for dealer-to-dealer
transactions and dealer-to-client transactions for purposes of this case. In Aluminum, the issue
became ripe at the pleading stage because, in a case involving an “exceedingly complex”
aluminum warehousing issue, the plaintiffs—commercial entities that purchased semi-fabricated
aluminum to manufacture aluminum products, and consumers who purchased finished aluminum
products—expressly “disavow[ed] participation in the relevant markets.” 833 F.3d. at 155.
Here, Plaintiffs concede no such thing. Unlike the Aluminum plaintiffs—whose injuries were
simply “too remote” because they never “engaged in futures trades with any of the defendants
57 There is no doubt that harm caused by the sharing of pricing information among horizontal competitors
is the type of injury contemplated by the antitrust laws. See Container Corp., 393 U.S. at 334-38 (finding
defendants’ “exchange of information concerning specific sales to identified customers” to be an illegal price-fixing
agreement because “interference with the setting of price by free market forces is unlawful per se”); Areeda &
Hovenkamp ¶ 2111g (“Ad hoc competitor-to-competitor ‘exchange’ of particularized price information, such as the
price offered or made to a particular customer, should ordinarily be considered a naked or nearly naked restraint,
which can be condemned without a structural inquiry unless the defendant offers a compelling justification.”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 84 of 155
72
[nor] purchased aluminum that was ever present in any of the defendants’ warehouses,” id. at
156, 163—Plaintiffs participated in the same market in which Defendants operate: the market
for USD SSA bonds. See SAC ¶¶ 1, 3, 36, 38, 40.
Third, even if there were separate “markets” here, Aluminum would weigh in Plaintiffs’
favor. That is because the Second Circuit realized that a plaintiff can have antitrust standing
despite being in a different “market” from the wrongful activity, if the plaintiff’s market “was the
target of the alleged scheme” and the plaintiff “directly suffered harm” in that market.
Aluminum, 833 F.3d at 159; see also Crimpers Promotions Inc. v. Home Box Office, Inc., 724
F.2d 290 (2d Cir. 1983) (trade show organizer alleging group boycott by cable television
networks had standing because “[t]he plaintiff’s injury . . . was the precisely intended
consequence of defendants’ boycott”).
58
58 Citing Harry v. Total Gas & Power North America, Inc., 889 F.3d 104, 115 (2d Cir. May 4, 2018),
Defendants argue in a footnote that because Plaintiffs allege that Defendants’ agreement not to compete caused
prices to be manipulated both higher and lower, then it is just as plausible that Plaintiffs benefitted from price
artificiality as it is that they were harmed by it. Merits Mem. at 14 n.19. But as Defendants repeatedly point out,
this is not a benchmark case. This is about a conspiracy not to compete, the primary effect of which is to widen
spreads for the benefit of the dealers (and to the detriment of Plaintiffs and class members) when buying and selling.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 85 of 155
73
B. Plaintiffs Are Efficient Enforcers of the Antitrust Laws
Plaintiffs are “efficient enforcers.” In determining whether a plaintiff is an efficient
enforcer, courts generally consider four factors: (1) the “directness or indirectness of the
asserted injury,” (2) the “existence of more direct victims of the alleged conspiracy,” (3) the
extent to which the damages claim is “highly speculative,” and (4) the importance of avoiding
“either the risk of duplicate recoveries on the one hand, or the danger of complex apportionment
of damages on the other.” Gelboim, 823 F.3d at 778.
1. Plaintiffs’ injuries are direct
Determining whether an injury is “direct” entails what is “essentially a proximate cause
analysis.” In re Credit Default Swaps Antitrust Litig., 2014 WL 4379112, at *8 (S.D.N.Y. Sep.
4, 2014) (“CDS”). The question is ultimately whether the claimed injury is a “type of loss” the
conspiracy’s conduct “would be likely to cause,” Blue Shield of Virginia v. McCready, 457 U.S.
465, 479 (1982), i.e., whether the injury is “within the scope of the risk created by” the
defendants’ conduct, Lotes Co. v. Hon Hai Precision Indus. Co., 753 F.3d 395, 412 (2d Cir.
2014). For the same reasons they satisfy the factors to establish antitrust injury discussed above,
see Section III.A., Plaintiffs readily meet this “scope of risk” standard.
Defendants argue that Plaintiffs’ injuries are “only indirectly related to the primary
violation asserted,” and that they “depend on some complex chain of events whereby the
allegedly manipulated prices of a few transactions . . . impacted the price of all USD SSA
bonds.” Merits Mem. at 23. That is not true. Plaintiffs traded USD SSA bonds directly with
Defendants at artificial prices. Defendants were able to sell for more and buy for less only
because of their agreement not to compete, which reduced competition in the relevant market.
SAC ¶¶ 13-19. The harm suffered by Plaintiffs “is not some far-flung consequence of” the
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 86 of 155
74
conduct challenged in the SAC; rather, it is “‘precisely the type of loss that [defendants’ conduct]
would be likely to cause.’” Lotes, 753 F.3d at 412.59
2. No more efficient enforcers exist
Defendants’ conduct here was targeted at their customers—that is, the proposed class of
Plaintiffs here—and was intended to deprive those customers of the benefits of competition. The
Supreme Court has recognized that plaintiffs like these are the preferred entities to vindicate the
antitrust laws. See llinois Brick Co. v. Illinois, 431 U.S. 720, 746 (1977) (“elevating
direct purchasers to a preferred position as private attorneys general”); Blue Shield, 457 U.S. at
474 (direct purchasers are the plaintiffs “most likely to press their claims with the vigor that the
§ 4 treble-damages remedy was intended to promote”).
Merits Mem. at 24. But then, citing a small subset of the SAC’s allegations, Defendants
nonetheless argue Plaintiffs are not efficient enforcers
Id. (citing SAC ¶¶ 221-87).
as discussed in Section II.A above.
Plaintiffs
are all investors who directly entered into USD SSA bond transactions with Defendants, and who
59 See also In re Interest Rate Swaps Antitrust Litig., 261 F. Supp. 3d 430, 491 (S.D.N.Y. 2017) (direct
purchasers of interest rate swaps had standing to pursue claims for damages caused by conspiracy to exclude
competitors); FX II, 74 F. Supp. 3d at 598 (“[H]aving to pay supra-competitive prices as a result of a horizontal
price-fixing conspiracy is the quintessential antitrust injury.”); CDS, 2014 WL 4379112, at *9 (direct purchasers of
credit default swaps had standing to pursue claims for damages caused by conspiracy aimed at blocking exchange
trading).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 87 of 155
75
have plausibly alleged they paid supra-competitive prices as a result of Defendants’ conspiracy
as discussed in Section II.B above. No more efficient enforcers than Plaintiffs exist.
3. Plaintiffs’ injuries are not speculative
Defendants argue that Plaintiffs’ injuries are unduly speculative and would require the
Court to “guess” how Plaintiffs’ USD SSA bond trades would have been priced but-for
Defendants’ collusion. Merits Mem. at 21. But this argument relates to damages calculations,
which are routine in antitrust cases and do not render the injuries Plaintiffs suffered unduly
speculative at the pleading stage. See FX, 2016 WL 5108131 at *8 (rejecting defendants’
argument that damages were unduly speculative where “the defendants engaged in a price-fixing
conspiracy spanning more than a decade and involving innumerable combinations of currency
pairs for almost every country in the world”); ISDAfix, 175 F. Supp. 3d at 61 (rejecting argument
that damages were too speculative for antitrust standing where plaintiffs “alleged that they were
directly harmed by Defendants’ anticompetitive conduct by having to pay higher prices (or
earning lower profits) from instruments tied to ISDAfix”); see also Lexmark Int’l, Inc. v. Static
Control Components, Inc., 134 S. Ct. 1377, 1392 (2014) (stating that “potential difficulty in
ascertaining and apportioning damages is not . . . an independent basis for denying standing”)
(emphasis in original).
In In re DDAVP Direct Purchaser Antitrust Litigation, 585 F.3d 677, 689 (2d Cir. 2009),
for example, the Second Circuit rejected the defendants’ argument that the plaintiffs’ claimed
damages rested on “tenuous assumptions” about what impact increased competition would have
had on the prices paid by the plaintiff class. The Second Circuit expressed “little doubt that
those effects can be sufficiently estimated and measured” and that this was “especially so when
‘[t]he most elementary conceptions of justice and public policy require that the wrongdoer shall
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 88 of 155
76
bear the risk of the uncertainty which his own wrong has created.’” Id. (quoting Bigelow v. RKO
Radio Pictures, 327 U.S. 251, 265 (1946)).60
As the SAC notes, there are a multitude of ways that damages could potentially be
quantified in this case. One such way is to compare the bid-ask spreads paid by class members
in the actual world with the spreads paid on comparable instruments after the period of collusion
ended, while controlling for other (non-collusive) factors. SAC ¶¶ 513-43. In another example
noted in the SAC, a benchmark of what spreads and profit margins would have looked like in a
non-collusive world could be constructed based on the profit margins and spreads on similar
types of bonds or investment vehicles in a competitive market with a similar number of dealers
over a similar period. Id. ¶ 507. Of course, these proposed methods pre-date discovery, and
Plaintiffs will put forth an expert report analyzing the fruits of discovery and provide abundant
grounds upon which to estimate damages.
4. There will be no duplicate recoveries or complex apportionment of
damages
Defendants make no argument that the Court would have any difficulty apportioning
damages, or that Plaintiffs’ claims present some danger of duplicative recovery. And there is no
such difficulty in this case. Plaintiffs seek damages for over- and under-pricing on the USD SSA
bond transactions, and there are no entities besides Plaintiffs among which those damages will
need to be apportioned.
60 See also J. Truett Payne Co., Inc. v. Chrysler Motors Corp., 451 U.S. 557, 566-67 (1981) (holding that a
wrongdoer is in a poor position “to insist upon specific and certain proof of the injury which it has itself inflicted,”
especially when the “vagaries of the marketplace usually deny us sure knowledge of what plaintiff’s situation would
have been in the absence of the defendant’s antitrust violation”); Bigelow, 327 U.S. at 265 (“The wrongdoer may not
object to the plaintiff’s reasonable estimate of the cause of injury and of its amount, supported by the evidence,
because [of the lack of] more accurate data which the wrongdoer’s misconduct has rendered unavailable.”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 89 of 155
77
The harm arising from one USD SSA bond transaction is not duplicative in any way of
the harm arising out of another USD SSA bond transaction, even if the conspiracy impacting
both was put into motion through similar means.
would have the obvious—and negative—effect of leaving the vast majority
of the conspiracy’s victims unable to recover for their losses. It is thus undisputed that this factor
weighs in favor of finding Plaintiffs to be “efficient enforcers.”
IV. AMPLE EVIDENCE LINKS EACH DEFENDANT TO THE CONSPIRACY
In supplemental memoranda of law, several Defendants argue separately that Plaintiffs
have not adequately pled their participation in the conspiracy. In reality, these memoranda are
little more than attempts to offer self-serving interpretations of damning evidence.61 Defendants’
requests for the Court to draw inferences in their favor are plainly improper. See Anderson
News, 680 F.3d at 189 (“The question at the pleading stage is not whether there is a plausible
alternative to the plaintiff’s theory; the question is whether there are sufficient factual allegations
to make the complaint’s claim plausible.”). Regardless, the SAC alleges ample facts to support
the inference that each Defendant joined and took steps in furtherance of the unlawful agreement
not to compete. See Apex, 822 F.2d at 257 (“[O]nce a conspiracy is shown, only slight evidence
is needed to link another defendant with it”).62
61 Every non-settling Defendant except the BNP Paribas Defendants submitted a supplemental
memorandum of law in support of their motions to dismiss. See Dkt. Nos. 530 (Barclays), 544 (Citi), 528 (Crédit
Agricole), 524 (Credit Suisse), 543 (Heer), 538 (Manku), 534 (McDonald), 539 (NIPLC), 529 (NSII), 541 (Pau),
531 (RBC), 532 (TD Bank).
62 See also Elec. Books., 859 F. Supp. 2d at 690 (“[A] conspirator may join a conspiracy at any time that it
is ongoing; there is no requirement that a conspirator join a conspiracy from its inception. . . . [T]he fact that [a
defendant was] involved in only a portion of it[] does not undermine the existence of the conspiracy itself or [the
defendant’s] role as a participant.”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 90 of 155
78
Defendants also assert—both in their consolidated memorandum of law and in several
supplemental memoranda—that Plaintiffs have engaged in “group pleading.” See, e.g., Merits
Mem. at 43. Not so. The touchstone under Rule 8 is notice, and the SAC provides each
Defendant with more than fair notice of its role in the conspiracy. Plaintiffs do not need to
“show that the defendant knew all of the details of the conspiracy” or even “prove that the
defendant knew the identities of all of the other conspirators.” United States v. Huezo, 546 F.3d
174, 180 (2d Cir. 2008). Rather, “a single act may be sufficient for an inference of [a
defendant’s] involvement” in a conspiracy. Id.; see also Precision Assocs., Inc. v. Panalpina
World Transp. (Holding) Ltd., 2011 WL 7053807, at *24 (E.D.N.Y. Jan. 4, 2011) (finding
sufficient connection to the conspiracy where defendant was on a mailing list “where the
defendants exchanged information about the progress of the conspiracy”).
Here, Plaintiffs have alleged sufficient facts connecting each Defendant to this
conspiracy. The SAC identifies how each Defendant participated in the conspiracy, including its
role, each Defendants’ employees and representatives who participated in the conspiracy, and
evidence showing that each trader (and the banks they represent) agreed not to compete.63 See,
e.g., SAC ¶¶ 159-374. No more detail is required or can reasonably be expected at this early
stage. Delgado v. Ocwen Loan Servicing, LLC, 2014 WL 4773991, at *7 (E.D.N.Y. Sept. 24,
63 The specificity of the SAC’s allegations distinguishes this case from Defendants’ cited cases (e.g.,
Merits Mem. at 43-45). C.f., In re Zinc Antitrust Litig., 155 F. Supp. 3d 337, 384 (S.D.N.Y. 2016) (finding group
pleading where allegations against certain defendants were “sparse to the point of near non-existence or [were]
grouped together with specific allegations relating to their affiliated but legally separate entities,” and where the
complaint ascribed the conspiracy to generic defendant groups without providing further identifying information,
such as specific entity names or affiliations); Concord Assocs., L.P. v. Entm’t Props. Trust, 2014 WL 1396524, at *6
(S.D.N.Y. Apr. 9, 2014) (finding group pleading where complaint alleged only that a group of defendants entered
into “agreements, combinations and conspiracies,” and gave scant details on what those specific agreements entailed
or how each defendant participated in the combination); Ochre LLC v. Rockwell Architecture Planning & Design,
P.C., 2012 WL 6082387, at *7 (S.D.N.Y. Dec. 3, 2012) (finding group pleading where complaint relied on
conclusory allegations as to various groups of defendants, without specifying any individual conduct, which served
to “force the various defendants to guess at the nature of [plaintiff’s] claims”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 91 of 155
79
2014) (“While some paragraphs in the FAC refer to Defendants collectively, there is more than
enough detail of each Defendant’s individual conduct . . . to give each fair notice of what
[each] . . . claim is and the grounds upon which it rests.”).64
A. The Barclays Defendants
The four Barclays Defendants—Barclays Bank plc (“BBPLC”), Barclays Capital Inc.
(“BCI”), Barclays Services Limited (“BSL”), and Barclays Capital Securities Limited (“BCSL”)
(collectively, “Barclays”)—argue that the SAC does not plausibly allege they participated in the
conspiracy. Dkt. No. 530 (“Barclays Mem.”) at 1. The argument is meritless.
1. The SAC identifies the specific role of each Barclays Defendant
Defendants BBPLC, BSL, and BCSL all employed conspiring trader , who
is listed in FCA records as a registered adviser for each entity. SAC ¶¶ 51, 55, 56.
As the London-based entities that set the marketing, pricing, and other decisions for the
group’s SSA operations, Defendants BBPLC, BSL, and BCSL “purposefully directed and
controlled the trading of . . . USD SSA bonds at artificial prices by Barclays Capital Inc. and/or
other Barclays affiliates, knowing and intending such transactions were with Class members in
64 In CDS, the court rejected the argument that the term “Dealer-Defendant” insufficiently identified the
defendants, where the plaintiffs specifically listed each defendant representative alleged to have participated in the
collusion. 2014 WL 4379112, at *10 (“While no single list [of conspiratorial acts] includes all Dealer-Defendants,
each Dealer-Defendant is included on at least one of the lists.”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 92 of 155
80
the United States, and booked the profit from such trades.” Id. ¶¶ 51, 55, 56. BBPLC executed
collusive USD SSA bond trades with Plaintiffs ADR and APFC. Id. ¶ 52.
Similarly, as the Barclays Defendants’ U.S.-based broker-dealer, Defendant BCI, which
took directions from its overseas counterparts and served as the counterparty to most
transactions, knowingly “executed collusive USD SSA bond transactions with Plaintiffs Alaska
Department of Revenue, Alaska Permanent Fund Corporation, and Iron Workers during the
Class Period.” Id. ¶ 54. BCI employed the U.S.-based salespeople and traders who relayed and
executed the collusive trades with U.S. customers, Id. ¶ 396, which were essential steps in the
conspiracy. Each Barclays Defendant thus participated in the conspiracy.
2. Barclays cannot explain away the evidence of its wrongdoing
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 93 of 155
81
65
Barclays both paints the conspiracy too narrowly and ignores the facts.
65
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 94 of 155
82
Far from being mere “industry scuttlebutt,” id. at 5, these information exchanges enabled
Barclays and its co-conspirators to avoid competition, to and seek and obtain advantageous deals
at the expense of customers and class members.
Barclays’ reliance on In re Elevator Antitrust Litigation, 502 F.3d 47 (2d Cir. 2007) is
misplaced. Barclays Mem. at 3. The In re Elevator complaint was dismissed because it made
only vague allegations of “basically every type of conspiratorial activity that one could
66 Barclays contends that trader is alleged to have
Id. But, as discussed above, because
investors typically only approached a small set of dealers at any one time, the fact that
See Section II, supra.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 95 of 155
83
imagine . . . in entirely general terms without any specification of any particular activities by any
particular defendant.” 502 F.3d at 50. The SAC’s allegations, including with respect to Barclays
specifically, are nothing of the sort.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 96 of 155
84
B. The BNP Paribas Defendants
The BNP Paribas Defendants—BNP Paribas S.A. and BNP Paribas Securities Corp.
(collectively, “BNP Paribas”)—did not file a separate supplemental memorandum of law.67
1. The SAC identifies the specific role of each BNP Paribas Defendant
Defendant BNP Paribas S.A. employed conspiring trader
, who was registered as an adviser to perform controlled functions or
otherwise act on BNP Paribas S.A.’s behalf during the Class Period. SAC ¶ 59.
As the London-based entity that set the marketing, pricing, and other decisions for the
group’s SSA operations, BNP Paribas S.A. “purposefully directed and controlled the trading
of . . . USD SSA bonds at artificial prices by BNP Securities Corp. and/or other BNP Paribas
affiliates, knowing and intending such transactions were with Class members in the United
States, and booked the profit from such trades.” Id. ¶ 59.
Similarly, as the BNP Paribas Defendants’ broker-dealer, BNP Paribas Securities Corp.,
which took directions from its overseas counterparts and served as the counterparty to most
transactions, knowingly “executed [manipulated] USD SSA bond trades with Plaintiffs Alaska
Department of Revenue and Alaska Permanent Fund Corporation during the Class Period,” and
its activities were often performed “at the direction of and with the knowledge and consent of
67 BNP Paribas joined the Dealer Defendants’ consolidated motion to dismiss on the grounds of subject-
matter jurisdiction and failure to state a claim. See Merits Mem. at 1 n.1. BNP Paribas S.A. joined the Foreign
Dealer Defendants’ motion to dismiss on the grounds of lack of personal jurisdiction and improper venue. See Juris.
Mem. at 1 n.1.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 97 of 155
85
BNP Paribas S.A.” Id. ¶ 61; Dkt. No. 522-3 (Decl. of Stephanie Gyetvan) ¶ 7. BNP Paribas
Securities Corp. employed the U.S.-based salespeople and traders who relayed and executed the
collusive trades with U.S. counterparties, SAC ¶ 396, which were essential steps in the
conspiracy. Each BNP Paribas Defendant thus participated in the conspiracy.
2. The evidence of BNP Paribas’ involvement in the conspiracy is clear
As discussed above, BNP Paribas blatantly fixed prices with other cartel members.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 98 of 155
86
All these allegations represent direct evidence of BNP’s collusion with its horizontal
competitors.
C. The Citi Defendants
The Citi Defendants—Citigroup Inc. (“Citigroup”), Citibank N.A. (“Citibank”),
Citigroup Global Markets Inc. (“CGMI”), and Citigroup Global Markets Limited (“CGML,” and
collectively, “Citi”)—also argue that the SAC does not plausibly allege they participated in the
conspiracy. Dkt. No. 544 (“Citi Mem.”) at 1. The Court should reject this argument.
1. The SAC identifies the specific role of each Citi Defendant
Pre-discovery evidence already shows that Citi participated in the conspiracy through at
least two traders, Gary McDonald and
SAC ¶¶ 149, 155. Defendants Citigroup, Citibank, and CGML all employed
conspiring trader Gary McDonald (September 2010 to March 2016), who is listed in FCA
records as a registered adviser for Citibank and CGML.68 Id. ¶¶ 65, 69.
68 The SAC specifically alleges that McDonald was employed by Defendant Citigroup, SAC ¶ 64, an
allegation , but also by McDonald’s own initial
declaration filed on July 14, 2017. See Dkt. No. 257, Decl. of Gary McDonald ¶ 2 (“I have worked for Citigroup,
Inc. (“Citigroup”) since September 2010.”). After Plaintiffs filed the CAC and SAC, McDonald filed declarations
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 99 of 155
87
Not surprisingly, FINRA records state that he was terminated from his employment
at Citi because of “[a]llegations . . . relating to certain communications between and
another trader regarding European Government Bond trading.” Id. ¶ 155.
As the London-based entity that set the marketing, pricing, and other decisions for the
group’s SSA operations, CGML “purposefully directed and controlled the trading of . . . USD
SSA bonds at artificial prices by Citigroup, Citibank, CGMI, and/or other Citi affiliates, knowing
and intending such transactions were with Class members in the United States, and booked the
profit from such trades.” Id. ¶ 69.
Similarly, as the Citi Defendants’ U.S.-based broker-dealers, Citigroup, Citibank, and
CGMI. took directions from its overseas counterparts and served as the counterparty to most
transactions, and “purposely transacted in USD SSA bonds in the United States with Class
members at artificial prices at the direction of and with the knowledge and consent of Citigroup
Global Markets Limited, who knew such transactions were with U.S. counterparties.” Id. ¶ 64-
65, 68. Citibank executed collusive USD SSA bond trades with Plaintiff APFC during the Class
Period. Id. ¶ 66. Similarly, Defendant CGMI knowingly “executed collusive USD SSA bond
trades with Plaintiffs Alaska Department of Revenue and Alaska Permanent Fund Corporation
claiming (differently) that his previous representation was incorrect, and that he has “worked for Citigroup Global
Markets Limited (“CGML”) since September 2010.” Dkt. Nos. 364 & 536, Decl. of Gary McDonald ¶ 2.
Discovery should reveal whether he was employed by Citigroup, CGML, or both.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 100 of 155
88
during the Class Period.” Id. ¶ 68. CGMI employed the U.S.-based salespeople and traders who
relayed and executed the collusive trades with U.S. counterparties, id. ¶ 396, which were
essential steps in the conspiracy. Likewise, CGML knowingly “executed collusive USD SSA
bond trades with Plaintiff Alaska Department of Revenue during the Class Period.” Id. ¶ 70.
Each Citi Defendant thus participated in the conspiracy.
2. Absence from certain collusive communications does not imply non-
participation in the conspiracy
See Beltz Travel Serv., Inc. v. Int’l Air Transp. Ass’n, 620 F.2d 1360, 1367 (9th Cir. 1980)
(“Participation by each conspirator in every detail in the execution of the conspiracy is
unnecessary to establish liability, for each conspirator may be performing different tasks to bring
about the desired result.”); In re Polyurethane Foam Antitrust Litig., 152 F. Supp. 3d 968, 979
(N.D. Ohio 2015) (finding a conspiracy existed because “a single conspiracy [does not] fragment
into multiple conspiracies because a member does not ‘know every other member’ or ‘know of
or become involved in all of the activities in furtherance of the conspiracy.’”); High-Tech, 856 F.
Supp. 2d at 1118 (noting that a “co-conspirator need not know of the existence or identity of the
other members of the conspiracy or the full extent of the conspiracy”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 101 of 155
89
See Areeda & Hovenkamp
¶ 2111g (“Ad hoc competitor-to-competitor ‘exchange’ of particularized price information, such
as the price offered or made to a particular customer, should ordinarily be considered a naked or
nearly naked restraint, which can be condemned without a structural inquiry unless the defendant
offers a compelling justification.”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 102 of 155
90
69
Courts may only “take judicial notice of the complete contents of articles cited in the
Complaint” to determine “what statements they contained.” Elec. Books, 859 F. Supp. 2d at 686.
69
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 103 of 155
91
70
70
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 104 of 155
92
Despite the detailed evidence to the contrary, Citi insists that is somehow not
connected to the conspiracy, claiming there is “no plausible connection” between and any
Citi Defendant. Citi Mem. at 6. Citi asserts he “never worked for any of the four Citi
Defendants” and that he was employed instead at Citibank Global Markets Japan. Citi Mem. at 6
(citing Decl. of Thomas Reich ¶ 3). But, as alleged in the SAC, FINRA records show was
“registered to perform controlled functions or otherwise act on CGMI’s behalf during the Class
Period.” SAC ¶ 67. Citi insists this does not amount to an assertion that was “employed by
or performed work for CGMI (or for any other Citi Defendant).” Citi Mem. at 6. But Citi
misses the point. FINRA records—which Citi presumably ensured were accurate—afford a
reasonable basis to believe that was acting on CGMI’s behalf in trading USD SSA bonds.71
Even assuming that was based in Japan, as Citi claims, Citi Mem. at 7, such a fact
would not exonerate Citi. Merely because many of the conspiring traders were located in
London does not mean that all the conspirators were located in London. Regardless of where
was physically located, the SAC’s detailed allegations show that engaged in
anticompetitive activities with co-conspirator Deutsche Bank. The SAC sufficiently alleges Citi
participated in the conspiracy through , McDonald, and other employees.
D. Crédit Agricole Corporate & Investment Bank.
Crédit Agricole Corporate & Investment Bank (“Crédit Agricole”) also separately argues
the SAC does not plausibly allege that it participated in the conspiracy. Dkt No. 528 (“Crédit
Agricole Mem.”) at 1. The argument is unpersuasive, particularly in light of the European
Commission’s recent decision to issue a Statement of Objections to Crédit Agricole, alleging it
71
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 105 of 155
93
exchanged “commercially sensitive information and coordinated on prices concerning US dollar
denominated supra-sovereign, sovereign and agency bonds.”72
1. Crédit Agricole’s role in the conspiracy
Crédit Agricole participated in the conspiracy through at least four traders: Shailen Pau
(July 2009 to March 2010), Amandeep Singh Manku (January 2013 to December 2015),
As the overseas-based entity that set the marketing, pricing, and other decisions for the
group’s SSA operations, Crédit Agricole “purposefully directed and controlled the trading of . . .
USD SSA bonds at artificial prices by Crédit Agricole affiliates, knowing and intending such
transactions were with Class members in the United States, and booked the profit from such
trades.” Id. ¶ 73.
2. The evidence of Credit Agricole’s involvement in the conspiracy is clear
72 Brockett Decl., Exs. 1 & 2.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 106 of 155
94
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 107 of 155
95
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 108 of 155
96
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 109 of 155
97
E. The Credit Suisse Defendants
The four Credit Suisse Defendants—Credit Suisse AG (“CSAG”), Credit Suisse
Securities (USA) LLC (“CSUSA”), Credit Suisse Securities (Europe) Ltd. (“CSSEL”), and
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 110 of 155
98
Credit Suisse International (“CSI,” and collectively, “Credit Suisse”)—also argue that the SAC
does not plausibly allege they participated in the conspiracy. Dkt. No. 524 (“Credit Suisse
Mem.”) at 1. For similar reasons as discussed with respect to the other Defendants, Credit
Suisse’s argument fails because of the And
insofar as Credit Suisse assert that Plaintiffs have suffered no antitrust injury and thus lack
standing, Credit Suisse Mem. at 3, that assertion fails for the additional reasons stated below and
in Sections I and II, above.
1. The SAC identifies the specific role of each Credit Suisse Defendant
74
As the London-based entities that set the marketing, pricing, and other decisions for the
group’s SSA operations, CSAG, CSSEL, and CSI “purposefully directed and controlled the
74 CSSEL’s December 21, 2018 declaration admits that it employed Pau. See Dkt. No. 522-11, Decl. of
Damian Bisseker ¶ 8. Credit Suisse argues that Plaintiffs made a “judicial admission” in the prior complaint that
Pau was employed only by CSSEL, but not the other Credit Suisse entities. Credit Suisse Mem. at 8. Not so.
Plaintiffs have been consistent in alleging that according to FCA and FINRA records, Pau was registered as an
adviser to act on behalf of CSAG, CSSUSA, CSSEL, and CSI. See CAC ¶¶ 59, 60, 62. These assertions also exist
in the current complaint, SAC ¶¶ 75-79.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 111 of 155
99
trading of . . . USD SSA bonds at artificial prices by Credit Suisse Securities (USA) LLC and/or
other Credit Suisse affiliates, knowing and intending such transactions were with Class members
in the United States, and booked the profit from such trades.” Id. ¶¶ 75, 78-79.
Similarly, as the Credit Suisse Defendants’ U.S.-based broker-dealer, Defendant
CSSUSA, which took directions from its overseas counterparts and served as the counterparty to
most transactions, knowingly “executed collusive USD SSA bond trades with Plaintiffs Alaska
Department of Revenue, Alaska Permanent Fund Corporation, and Iron Workers during the
Class Period.” SAC ¶ 77. CSSUSA employed the U.S.-based salespeople and traders who
relayed and executed the collusive trades with U.S. counterparties, id. ¶ 396, which were
essential steps in the conspiracy. Each Credit Suisse Defendant thus participated in the
conspiracy.
2. Credit Suisse cannot explain away the evidence alleged in the SAC
Credit Suisse argues that “Plaintiffs do not plausibly allege that the Credit Suisse
Defendants participated in the” conspiracy. Credit Suisse Mem. at 4. Credit Suisse largely
repeats the same five main arguments raised in Defendants’ joint memorandum, equally
unpersuasively.
First, like the other Defendants, Credit Suisse repeatedly mischaracterizes the SAC as
alleging that Defendants “manipulated the price of every USD SSA bond traded during the Class
Period.” Id.. at 1 (emphasis added). It then asserts that Plaintiffs fail to “allege any mechanism
that would allow the Credit Suisse Defendants to” do so. Id. at 5. However, as discussed in
Sections II and III, Plaintiffs do not allege Defendants expressly fixed the price of every USD
SSA bond, whether through a financial benchmark or otherwise. Rather, the SAC alleges
Defendants entered into a broad agreement not to compete against each other when trading USD
SSA bonds with their customers, see, e.g., SAC ¶ 143. This agreement was implemented
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 112 of 155
100
through
75 , and had the predictable result of hurting Defendants’
customers (which was, of course, the very point). These allegations comport perfectly with the
European Commission’s recent issuance of a Statement of Objections to Credit Suisse for
colluding “to distort competition in secondary market trading” in USD SSA bonds by
exchanging “commercially sensitive information and coordinat[ing] on prices . . . mainly through
online chatrooms.”76
Second, Credit Suisse attempts to spin the evidence in the SAC as “sporadic chat
excerpts” that show merely the exchange of “market color.” Credit Suisse Mem. at 4-5. But
there is nothing “sporadic” or benign about Credit Suisse’s behavior.77
All of these activities served to facilitate the cartel’s ability to
operate on the same page and harm outside customers, see id. ¶¶ 317-25, 327, 332-34, 337-39,
342-44, 351, 354-61. Far from permissible and necessary behavior, this conduct constitutes per
se antitrust violations. See Section I.B.3 (citing cases and discussing per se violations).
Third, recognizing the weakness of its positions, Credit Suisse asserts that even if there
was a conspiracy, its involvement was “limited” because it (a) “participated only through Pau,
who was employed by one of the Credit Suisse Defendants, CSSEL, for only a portion of the
75
76 Brockett Decl., Ex. 2.
77 Unlike the other Defendants, Credit Suisse at very least admits it engaged in “parallel activity.” Credit
Suisse Mem. at 7.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 113 of 155
101
Class Period,” and (b)
Credit Suisse Mem. at 5-6. For the reasons explained in Section I.D, supra, this
makes no difference.78
Given that additional information about the extent of Credit Suisse’s participation in the
conspiracy is uniquely within its control, it is highly plausible that it extended far beyond Pau (to
at least those traders Pau supervised at Credit Suisse)
See Arista Records, 604 F.3d at 120 (the plausibility standard
“does not prevent a plaintiff from pleading facts alleged upon information and belief where the
facts are peculiarly within the possession and control of the defendant”).
Fourth, Credit Suisse improperly advances several unconvincing explanations of the
incriminating chats summarized in the SAC.
78 Relying on Rabin v. NASDAQ OMX PHLX LLC, 182 F. Supp. 3d 220 (E.D. Pa. 2016), Credit Suisse
argues that it cannot be held responsible for periods in the conspiracy when Pau was not an employee. Id. at 5.
However, Rabin is a securities fraud case that involved a defendant that “did not itself directly injure Plaintiff” and
indeed “did not exist until after the transactions that give rise to Plaintiff’s own claims.” Id. at 229. That is simply
not the case here.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 114 of 155
102
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 115 of 155
103
79
Finally, Credit Suisse contends—like the other Defendants—that a conspiracy “cannot be
established by showing that they engaged in social activities or are currently under
investigation.” Credit Suisse Mem. at 7. But, as discussed above in Section I.C, supra, Plaintiffs
are not relying solely on these factors—they are additional corroborating evidence.
F. The Nomura Defendants
Filing separate supplemental memoranda, the Nomura entities—Nomura Securities Inc.
(“NSII”) and Nomura International plc (“NIPLC”) (collectively, “Nomura”)—argue that the
allegations in the SAC do not plausibly allege they participated in the conspiracy. Dkt. No. 539
(“NIPLC Mem.”) at 1; Dkt. No. 529 (“NSII Mem.”) at 1. Plaintiffs address both entities’
arguments here.
1. The SAC identifies the specific role of each Nomura Defendant
79
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 116 of 155
104
As the London-based entity that set the marketing, pricing, and other decisions for the
group’s SSA operations, NIPLC “purposefully directed and controlled the trading of . . . USD
SSA bonds at artificial prices by NSII and/or other Nomura affiliates, knowing and intending
such transactions were with Class members in the United States, and booked the profit from such
trades.” Id. ¶ 94. The pricing of all SSA bond trades had to go through NIPLC’s trading desk
(Heer and Moulle), thus directly implicating it in the conspiracy. Id.
Similarly, as the Nomura Defendants’ U.S.-based broker-dealer, Defendant NSII, which
took directions from its overseas counterparts and served as the counterparty to most
transactions, knowingly “executed a collusive USD SSA bond trade with Plaintiff Iron Workers
during the Class Period.” Id. ¶ 93. NSII employed the U.S.-based salespeople and traders who
relayed and executed the collusive trades with U.S. counterparties, id. ¶ 396, which were
essential steps in the conspiracy. Nonetheless, because NIPLC had ultimate authority over all
USD SSA trades, NSII’s trading activities were performed “at the direction of and with the
knowledge and consent of Nomura International plc.” Id. ¶ 92. Both NIPLC and NSII thus
played essential roles in the conspiracy.
2. NIPLC’s apparent expectation that Plaintiffs should already have all the
pertinent evidence is misplaced
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 117 of 155
105
See FX II, 74 F. Supp. 3d at 591
(noting that “smoking gun” evidence of conspiracy is unnecessary at the pleading stage).
3. Contrary to NIPLC’s assertion, the SAC alleges multiple instances where
NIPLC traded a manipulated bond
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 118 of 155
106
80
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 119 of 155
107
4. NIPLC’s attempts to discount the powerful evidence against it fail
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 120 of 155
108
81
81
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 121 of 155
109
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 122 of 155
110
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 123 of 155
111
G. The RBC Defendants
Royal Bank of Canada, RBC Capital Markets LLC (“RBCCM”), and RBC Europe
Limited (“RBCEL”) (collectively, “RBC”) argue that the SAC does not plausibly allege they
participated in the conspiracy. Dkt. No. 531 (“RBC Mem.”) at 1. They are wrong.
1. The SAC identifies the specific role of each RBC Defendant
Defendants Royal Bank of Canada and RBCEL employed conspiring trader
, who is listed as a registered adviser for both entities
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 124 of 155
112
according to FCA records. SAC ¶¶ 97, 100.
As the overseas-based entities that set the marketing, pricing, and other decisions for the
group’s SSA operations, Royal Bank of Canada and RBCEL “purposefully directed and
controlled the trading of . . . USD SSA bonds at artificial prices by RBCCM and/or other RBC
affiliates, knowing and intending such transactions were with Class members in the United
States, and booked the profit from such trades.” SAC ¶ 100.
Similarly, as the RBC Defendants’ U.S.-based broker-dealer, Defendant RBCCM, which
took direction from its overseas counterparts and served as the counterparty to most transactions,
knowingly “executed collusive USD SSA bond trades with Plaintiffs Alaska Department of
Revenue, Alaska Permanent Fund Corporation, and Iron Workers during the Class Period.” Id.
¶ 99. RBCCM employed the U.S.-based salespeople and traders who relayed and executed the
collusive trades with U.S. counterparties, id. ¶ 396, which were essential steps in the conspiracy.
Contrary to RBC’s claim of “group pleading,” these particularized allegations explain the role of
each RBC Defendant in the conspiracy.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 125 of 155
113
2. The RBC Defendants participated in the conspiracy
RBC’s lead argument is that Plaintiffs’ allegations of a conspiracy are implausible as to
it. Specifically, RBC claims that its employment of conspirators Shailen Pau (May 1999 to May
2009) and is insufficient to link RBC to the
conspiracy. RBC Mem. at 3. Citing In re Elevator, RBC argues that, even if Pau and
engaged in collusive activities during their stints at other banks, this does not allow the Court to
infer they engaged in such activities while employed by RBC. As noted earlier, however, the
complaint in In re Elevator contained “nothing in the way of specific transactions or patterns of
transactions . . . indicating possible conspiratorial collusion.” In re Elevator Antitrust Litig.,
2006 WL 1470994, at *3 (S.D.N.Y. May 30, 2006). Instead, the In re Elevator plaintiffs offered
only a “list of theoretical possibilities” about potential wrongdoing that “one could postulate
without knowing any facts whatever.” Id. In stark contrast, Plaintiffs have pled specific facts
showing that RBC traders—including Pau and —engaged in collusion both while
employed by RBC and at other banks.
Insofar as RBC suggests that its link to the conspiracy is somehow peripheral and might
cover less than a year of the Class Period, RBC Mem. at 3, that argument fails. A conspirator
need not be present for the entire period of the conspiracy to be liable. See Apex, 822 F.2d at 257
(“[O]nce a conspiracy is shown, only slight evidence is needed to link another defendant with
it”); see also Elec. Books, 859 F. Supp. 2d at 689-90 (“[A] conspirator may join a conspiracy at
any time that it is ongoing; there is no requirement that a conspirator join a conspiracy from its
inception. . . . [T]he fact that [a defendant was] involved in only a portion of it[] does not
undermine the existence of the conspiracy itself or [the defendant’s] role as a participant.”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 126 of 155
114
82
82
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 127 of 155
115
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 128 of 155
116
3. Pleading pre-July 2009 actions is proper
H. The TD Bank Defendants
In a supplemental memorandum, The Toronto-Dominion Bank and TD Securities (USA)
LLC (collectively, “TD Bank”) argue that Plaintiffs have not adequately pled that TD
participated in the conspiracy. Dkt. No. 532 (“TD Bank Mem.”) at 1. This contention withers
under scrutiny.
1. The SAC identifies the specific role of each TD Bank Defendant
TD Bank participated in the conspiracy through at least two traders, Gary McDonald
(December 2007 to July 2010) and SAC ¶ 149.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 129 of 155
117
As the overseas-based entity that set the marketing, pricing, and other decisions for the
group’s SSA operations, The Toronto-Dominion Bank “purposefully directed and controlled the
trading of . . . USD SSA bonds at artificial prices by TD Securities (USA) LLC and/or other TD
affiliates, knowing and intending such transactions were with Class members in the United
States, and booked the profit from such trades.” Id. ¶ 103.
Similarly, as the TD Bank Defendants’ U.S.-based broker-dealer, Defendant TD
Securities (USA) LLC (“TDS USA”), which took directions from its overseas counterparts and
served as the counterparty to most transactions, “executed collusive USD SSA bond trades with
Plaintiffs Alaska Department of Revenue, Iron Workersduring the Class Period.” SAC ¶ 105.
TDS USA employed the U.S.-based salespeople and traders who relayed and executed the
collusive trades with U.S. counterparties, id. ¶ 396, which were essential steps in the conspiracy.
These particularized allegations implicate each of the TD Bank Defendants in the conspiracy.
2. The TD Bank Defendants’ improper attempt to explain away the evidence
fails
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 130 of 155
118
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 131 of 155
119
83
83
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 132 of 155
120
see In re Actos End-Payor Antitrust Litig., 848 F.3d 89, 99
(2d Cir. 2017) (court dismisses plaintiffs’ “wishful” suggesting that generic drug manufacturers
might be made aware of false patent descriptions, which was not alleged in the complaint); see
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 133 of 155
121
also Concord Assocs., L.P. v. Entm’t Properties Tr., 2014 WL 1396524, at *24 (S.D.N.Y. Apr.
9, 2014) (Ramos, J.) (dismissing as “speculative” plaintiffs’ assertion that individual was
involved in antitrust conspiracy because financing for casino site would “likely” come from a
“silent” third party).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 134 of 155
122
I. Individual Defendant Bhardeep Heer
Defendant Bhardeep Heer—
—claims that the SAC does not
plausibly allege his participation in the conspiracy. Dkt. No. 543 (“Heer Mem.”) at 1. His
attempts are futile.
1. The SAC includes direct evidence of Heer engaging in anticompetitive
conduct
Heer argues there is no direct or circumstantial evidence linking him to the conspiracy.
Heer Mem. at 15. This argument is based on Heer’s attempt to impose pleading requirements
that do not exist and
Id. at
3-4, 15-22. Heer’s self-serving counter-interpretations are both improper at the pleading stage
and fanciful.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 135 of 155
123
86
87
88
84 Heer’s invocation of LaFlamme fails for the same reason as NIPLC’s. See note 80, supra.
85
86
87 Heer cites Merced Irrigation District v. Barclays Bank PLC, 165 F. Supp. 3d 122 (S.D.N.Y. 2016),
arguing that Plaintiffs “do not explain how these allegations “suggest that there was no independence of action.”
But Merced is inapposite because the plaintiff electricity buyer in that case had alleged that Barclays had sought to
restrain trade in the electricity market, but had failed to identify a co-conspirator or to allege any “concerted action
between at least two legally distinct economic entities.” Id. at 139.
88 Heer cites RxUSA Wholesale, Inc. v. Alcon Labs., Inc., 661 F. Supp. 2d 218, 238 (E.D.N.Y. 2009) for
the proposition that “[t]he Court need not accept plaintiffs’ conclusory allegation as to the reasoning behind Mr.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 136 of 155
124
Heer’s alleged action.” But RxUSA is inapplicable because it involved a speculative allegation by plaintiff (a
secondary pharmaceutical wholesaler) that defendant wholesalers were restraining trade by refusing to deal—a
purely “parallel action” that is insufficient to infer a conspiracy. Id. at 238.
89 Contrary to Heer’s supposition, nothing in the SAC suggests that the counterparty was a dealer. Heer
Mem. at 19, n.12.
.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 137 of 155
125
90
91
90
91
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 138 of 155
126
92
92 Similarly, Heer cites Superior, 738 F. Supp. 2d at 515, a case where the court dismissed a claim that
relied on a defendant’s statement expressing “concern about new entrants coming into the market.” Heer Mem. at
18 n.10. The Superior complaint offered only vague allegations that failed to “exclude the possibility that
Defendants independently priced their services,” Superior, 738 F. Supp. 2d at 515-16.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 139 of 155
127
93
93
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 140 of 155
128
2. The SAC sufficiently alleges circumstantial evidence of Heer’s
participation in the conspiracy
Second, the SAC plainly alleges a plausible motive for Heer to conspire. Apex, 822 F.2d
at 254 (“common motive to conspire” is indicative of conspiracy). See also Flat Glass, 385 F.3d
at 360 (“evidence that the defendant had a motive to enter into a price fixing conspiracy” is
indicative of conspiracy). Heer was motivated to refrain from competing to maximize the
group’s profits at the expense of outside customers. SAC ¶ 9.
Heer argues that Plaintiffs’ motive allegations are “implausible” because
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 141 of 155
129
94
94 Heer cites two cases to suggest he somehow did not benefit from his role in the conspiracy. But neither
set of plaintiffs in those two cases alleged a plausible theory on how the alleged conspirators would benefit. See
Bilinski v. Keith Haring Found., Inc., 96 F. Supp. 3d 35, 44 (S.D.N.Y. 2015) (conspiracy implausible where “the
plaintiffs have not stated in a non-conclusory fashion any facts indicating what benefit [defendants] derive” from
participating in the alleged conspiracy); Sonterra, 277 F. Supp. 3d at 556 (motive to manipulate Swiss franc LIBOR
implausible because no reason to believe all conspirators would benefit from the manipulation).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 142 of 155
130
95
Fourth, the SAC alleges that Heer frequently engaged in actions that would contravene
his economic self-interest absent an agreement to conspire. See Starr, 592 F.3d at 327 (reversing
district court decision dismissing complaint where “plaintiffs have alleged behavior that would
plausibly contravene each defendant’s self-interest in the absence of similar behavior by rivals”).
95
Heer also invokes several cases to argue that his activities constituted mere innocuous “market
making.” Heer Mem. at 19-20; see S.E.C. v. Diversified Corp. Consulting Grp., 378 F.3d 1219, 1222 n.7 (11th Cir.
2004) (“A market maker in the over-the-counter market is a dealer who, with respect to a security, routinely enters
quotations in an interdealer communication system or otherwise and is willing to buy and sell securities for the
dealer’s own account.”); Chasins v. Smith, Barney & Co., 438 F.2d 1167 (2d Cir. 1970) (“A market maker in the
over-the-counter market is a dealer who, with respect to a security, routinely enters quotations in an interdealer
communication system or otherwise and is willing to buy and sell securities for the dealer’s own account.”); In re
Ins. Brokerage Antitrust Litig., 618 F.3d 300 (3d Cir. 2010) (noting that traders had “obvious reasons . . . to share
this information . . . that have nothing to do with preexisting agreements of any kind.”).
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 143 of 155
131
96
J. Individual Defendant Amandeep Singh Manku
Defendant Amandeep Singh Manku filed a separate memorandum directed to his
personal jurisdiction defense and joined the Dealer Defendants’ joint merits memorandum, but
did not separately address the merits of the allegations against him. Dkt. No. 538 (“Manku
Mem.”) at 1 n.1. This is presumably because the SAC contains extensive evidence of Mr.
Manku’s deep participation in the conspiracy.
96
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 144 of 155
132
K. Individual Defendant Gary McDonald
Defendant Gary McDonald separately argues that the allegations in the SAC do not
plausibly allege that he participated in the conspiracy. Dkt. No. 534 (“McDonald Mem.”) at 1.
But McDonald’s attempt to avoid liability fares no better than that of his co-conspirators.
1. Absence from certain collusive communications does not imply non-
participation in the conspiracy
97
98
97
98
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 145 of 155
133
See Sullivan, 2017 WL 685570 at *23
(holding that the plaintiffs had sufficiently pled a conspiracy even though “[v]irtually all
substantive allegations touch on communications involving [only two individual defendants]”).
2. The SAC sufficiently alleges direct evidence of McDonald engaging in
anticompetitive conduct
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 146 of 155
134
99 McDonald cites an Eighth Circuit case, Blomkest Fertilizer, Inc. v. Potash Corporation of
Saskatchewan, 203 F.3d 1028 (8th Cir. 2000), for the proposition that sharing price information about completed
sales does not necessarily support a Section 1 claim. McDonald Mem. at 15. There, the court found that there was
“no evidence to support the inference that the verifications [of prices between the defendants] had an impact on
price increases.” Id. at 1034.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 147 of 155
135
100
100
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 148 of 155
136
101
101
102
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 149 of 155
137
103 See United States v. SKW Metals & Alloys, Inc., 195 F.3d 83,
93 (2d Cir. 1999) (Newman, J., concurring) (“Once a seller agrees to fix prices, he . . . at least
has the fixed price in mind and uses it as a point of departure for himself in deciding what
slightly different price to quote . . . As long as the seller has the fixed price in mind when he
decides by how much to depart from it when quoting a price, his final sale price has been
affected by the price-fixing agreement.”).
103
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 150 of 155
138
104
104 McDonald’s reliance on Ulrich v. Moody’s Corp., 2014 WL 12776746 (S.D.N.Y. Mar. 31, 2014) and
Mooney v. AXA Advisors, L.L.C., 19 F. Supp. 3d 486 (S.D.N.Y. 2014) is misplaced. McDonald Mem. at 17. In
Ulrich, the complaint was based on the plaintiff’s “vague suspicions and limited anecdotal evidence” based on
“searching LinkedIn and cold-calling” potential victims. Ulrich, 2014 WL 12776746 at *27.
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 151 of 155
139
3. McDonald’s inactive status on the FCA register supports a plausible
inference of wrongdoing
McDonald argues—with the help of his lawyer’s own declaration—that his identification
as inactive on the FCA register is not because of any wrongdoing, but because under an
interpretation of British law, “he is no longer required” to be registered. McDonald Mem. at 9;
see Dkt. No. 535 (“Roberts Decl.”) ¶ 5. But this is simply an improper attempt to contradict the
SAC’s factual allegations at the pleading stage. The SAC specifically alleges that McDonald’s
status was changed to inactive by virtue of his misconduct in the USD SSA bonds market. SAC
¶¶ 376, 380 n.39. When the SAC’s factual allegations are credited, and all inferences are drawn
in Plaintiffs’ favor, the highly suspicious timing of McDonald’s deactivation from the FCA
register within months of the other traders being deactivated between November 2015 and March
2016, SAC ¶¶ 376-80, is strongly suggestive of wrongdoing.
4. Plaintiffs have standing to sue McDonald
McDonald also contends that Plaintiffs “lack standing” to assert any claims against him
because they “never allege that they traded any bonds with Mr. McDonald or that they even
bought the same bonds that Mr. McDonald was allegedly trading.” McDonald Mem. at 21. But,
as noted above, Plaintiffs need not show that each Defendant was involved in a trade with a
named Plaintiff. See State of N.Y. v. Cedar Park Concrete Corp., 665 F. Supp. 238, 247
(S.D.N.Y. 1987) (“[A]n overt act need not be pleaded against each defendant, because a single
overt act by just one of the conspirators is enough to sustain a conspiracy claim even on the
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 152 of 155
140
merits.”); cf. Gold, 213 F. Supp. 3d at 652 (holding that complaint adequately alleged injury
where Plaintiffs “allege that they were harmed by being forced to sell their gold investments at
depressed prices as a result of defendant banks’ manipulation” of the gold benchmark).
Moreover, the SAC alleges that, during the Class Period, (a) ADR and APFC traded USD
bonds with Citi, which was “acting at least through McDonald,” SAC ¶¶ 64-70, and (b) ADR
and Iron Workers traded USD SSA bonds with TD Bank, which was “acting at least through
McDonald,” id. ¶¶ 103-05. Finally, insofar as McDonald claims Plaintiffs have suffered no
antitrust injury and thus lack standing, McDonald Mem. at 21-24, that argument fails for the
reasons in Section I above.
L. Individual Defendant Shailen Pau
Like Defendants Gudka and Manku, Shailen Pau filed an individual memorandum
addressing his personal jurisdiction defense and joined the Dealer Defendants’ merits
memorandum, but did not separately address the merits of the allegations against him. Dkt. No.
541 (“Pau Mem.”) at 1.
105
105
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 153 of 155
141
CONCLUSION
For the foregoing reasons, Plaintiffs respectfully request that the Court deny Defendants’
motions to dismiss the Second Consolidated Amended Complaint in their entirety.
DATED: New York, New York
February 11, 2019
ROBBINS GELLER RUDMAN & DOWD
LLP
QUINN EMANUEL URQUHART &
SULLIVAN, LLP
By: /s/ David W. Mitchell____ By: /s/ Daniel L. Brockett____
David W. Mitchell
Brian O. O’Mara
Carmen A. Medici
655 West Broadway, Suite 1900
San Diego, CA 92101
Telephone: (619) 231-1058
Fax: (619) 231-7423
davidm@rgrdlaw.com
bomara@rgrdlaw.com
cmedici@rgrdlaw.com
Daniel L. Brockett
Sascha N. Rand
Steig D. Olson
Thomas J. Lepri
Christopher M. Seck
51 Madison Avenue, 22nd Floor
New York, New York 10010
Telephone: (212) 849-7000
Fax: (212) 849-7100
danbrockett@quinnemanuel.com
sascharand@quinnemanuel.com
steigolson@quinnemanuel.com
thomaslepri@quinnemanuel.com
christopherseck@quinnemanuel.com
Jeremy D. Andersen
Adam B. Wolfson
865 South Figueroa Street, 10th Floor
Los Angeles, California 90017
Telephone: (213) 443-3000
Fax: (213) 443-3100
jeremyandersen@quinnemanuel.com
adamwolfson@quinnemanuel.com
Counsel for Plaintiffs and the Proposed Class
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 154 of 155
142
CERTIFICATE OF SERVICE
I hereby certify that on February 11, 2019, I filed and therefore caused the foregoing
document to be served in the United States District Court for the Southern District of New York.
Copies were emailed on all Defendants in the above-captioned matter.
/s/ Daniel L. Brockett____
Daniel L. Brockett
Case 1:16-cv-03711-ER Document 578 Filed 02/19/19 Page 155 of 155
