IRS Antitrust MDL Plaintiffs v. IRS Antitrust MDL Defendants

Filing 49

OPINION AND ORDER. For the reasons set out above: (1) The Court grants defendants' motion to dismiss class plaintiffs' claims of a Sherman Act § 1 conspiracy to the extent covering the period 2008-2012. The Court denies defendants' ; motion to dismiss plaintiffs' § 1 claims to the extent covering the period 2013-2016. (2) The Court grants defendants' motion to dismiss the claims of plaintiffs Javelin and Tera of tortious interference with business relations. (3) The Court denies defendants' motion to dismiss the claims of plaintiffs Javelin and Tera under the Donnelly Act, but limits these claims to the period 2013-2016. (4) The Court grants defendants' motion to dismiss the claims of plaintiffs Ja velin and Tera of unjust enrichment. The Court denies defendants' motion to dismiss the class plaintiffs' claims of unjust enrichment, but limits these claims to the period 20132016. (5) The Court grants the motions to dismiss all claims ag ainst defendants HSBC, ICAP, and Tradeweb. (6) All other motions to dismiss are denied. An order will follow shortly as to next steps in this case. The Clerk of Court is respectfully directed to terminate all pending motions. SO ORDERED. re: 15 JO INT MOTION to Dismiss All Claims. filed by IRS Antitrust MDL Defendants, 18 MOTION to Dismiss of defendant ICAP Capital Markets LLC filed by IRS Antitrust MDL Defendants, 24 JOINT MOTION to Dismiss the Second Consoli dated Amended Class Action Complaint and the Javelin and TeraExchange Plaintiffs' Second Consolidated Amended Complaint Filed on Behalf of The Dealer Defendants filed by IRS Antitrust MDL Defendants, 28 MOTION to Dismiss of defendants ICAP Capital Markets LLC's and ICAP SEF (US) LLC'S filed by IRS Antitrust MDL Defendants. (Signed by Judge Paul A. Engelmayer on 7/28/2017) (rjm).

ICAP Capital Markets, LLC (“ICAP”); and one provider of electronic trading services for IRSs,
Tradeweb Markets LLC (“Tradeweb”).
Pending are defendants’ motions, under Federal Rule of Civil Procedure 12(b)(6), to
dismiss plaintiffs’ claims. For the reasons that follow, these motions are granted in part and
denied in part.
I.

Factual Background
The following facts are drawn from the Second Consolidated Amended Class Action

Complaint (“SAC”), Dkt. 142 in No. 16-MD-2704, and Javelin/Tera’s Second Consolidated
Amended Complaint (“JTSAC”), Dkt. 145 in No. 16-MD-2704 (together, the “SACs”). In
resolving the motions to dismiss, the Court assumes all well-pleaded facts to be true and draws
all reasonable inferences in favor of the plaintiffs. See Koch v. Christie’s Int’l PLC, 699 F.3d
141, 145 (2d Cir. 2012).2
An important preface to the long factual allegations that follow is that the IRS market
underwent major changes during the period (2007–2016) covered by the alleged conspiracy. The

2

On a motion to dismiss for failure to state a claim, the Court must limit itself to facts stated in
the complaint or in documents attached as exhibits or incorporated by reference; the Court may
also consider matters of which judicial notice may be taken under Federal Rule of Evidence 201.
Kramer v. Time Warner, Inc., 937 F.2d 767, 773 (2d Cir. 1991).

The parties have supplied the Court with voluminous documents cited in the SACs. These are
attached as exhibits to the declarations of John S. Playforth, Dkt. 161, Stacey Anne Mahoney,
Dkts. 167, 207, and Michael J. Garvey, Dkt. 171. A court may consider materials where
incorporated by reference or “integral” to the SACs, see Chambers v. Time Warner, Inc., 282
F.3d 147, 152–53 (2d Cir. 2002), materials “that the plaintiffs either possessed or knew about
and upon which they relied in bringing the suit,” Rothman v. Gregor, 220 F.3d 81, 88–89 (2d
Cir. 2000) (citation omitted), and materials that are matters of public record, Pani v. Empire Blue
Cross Blue Shield, 152 F.3d 67, 75 (2d Cir. 1998).
As discussed later, the parties dispute whether the Court can consider statements embedded in
these materials for the truth of the matters asserted therein. See infra IV.B.2. For avoidance of
doubt, the only materials upon which the Court has relied are those cited herein.
2

infrastructure necessary to support the electronic trading platforms which the investor plaintiffs
claim were denied them as a result of defendants’ collusion—platforms enabling anonymous
“all-to-all” exchange trading of IRSs—evolved. There was also a major regulatory development:
the Dodd-Frank statute, passed July 21, 2010 in response to the 2008 financial crisis. DoddFrank’s mandates, as implemented by ensuing regulations, reshaped the IRS market. They
facilitated the emergence, in 2013–2014, of electronic “all to all” platforms for anonymous IRS
trading, like those of Javelin and Tera, accessible to investors.
In considering plaintiffs’ claims, it is, therefore, important to focus on the distinct goal, at
each point, of the alleged conspiracy. Through 2012, plaintiffs allege concerted action among
the Dealer Defendants and others to inhibit the emergence of electronic trading platforms
accessible to investors that threatened to erode the Dealers’ profit margins on IRS trades. From
2013 on, after such platforms emerged, plaintiffs allege a boycott aimed at destroying three such
new platforms, including Javelin’s and Tera’s.
A.

Interest Rate Swaps

An IRS is a financial derivative. It permits two parties to trade interest-rate-based cash
flows on a specific amount of money over a fixed time period. Typically, one party to an IRS
pays cash flows based on a fixed interest rate, while the counterparty pays cash flows based on a
floating interest rate, keyed to a benchmark or reference point such as the London Interbank
Offered Rate, or “LIBOR.” For example, a party might agree to pay a fixed interest rate on a
$10 million notional amount for 10 years in exchange for the other party paying a floating rate
(such as one tied to LIBOR) for that same 10-year period. The party paying a fixed rate is
typically referred to as the “buyer,” and the party making payments at the floating rate is known

3

as the “seller.” The value of the contract to each side moves (in opposite directions) depending
on changes in interest rates. SAC ¶¶ 1–2; JTSAC ¶ 2.
IRSs are used by an array of investors to manage risk and protect themselves against
fluctuating interest rates. For example, a municipality that issued a floating rate bond to pay for
a clean-water project might later use an IRS to convert floating-rate payments on the bond into
fixed payments, so as to hedge against interest-rate increases. Investors in interest-rate swaps
include pension funds, asset managers, endowment funds, corporations, insurance companies,
municipalities, and hedge funds. The IRS market has grown exponentially over the last three
decades, with billions of dollars in notional quantity traded daily. In 2006, the outstanding
notional quantity of IRSs was approximately $230 trillion. By 2014, it was approximately $381
trillion. SAC ¶¶ 3, 72; JTSAC ¶ 3.
B.

The Evolution of IRS Trading Practices and Platforms
1.

Early Years to 2013

In the early years of IRS trading, IRS contracts were not standardized. They typically
were negotiated and documented on a trade-by-trade basis, imposing high transaction costs.
Over time, IRS trading became more standardized. In 1987, the International Swaps and
Derivatives Association (“ISDA”) created the ISDA Master Agreement. It set out standardized
terms to govern over-the-counter (“OTC”) derivatives transactions, and became widely used.
By 2000, the material terms of most IRSs were standardized. These included the tenor (the
maturity or term of the swap), the fixed rate, the reference index used to calculate floating
payment, the payment frequency, and the timing of payments. This resulted in lower trading
costs and higher trading volumes. SAC ¶¶ 70–71; JTSAC ¶¶ 67–68.

4

As demand for IRSs spread, the investment banks who dealt in interest swaps positioned
themselves as the exclusive market makers or liquidity providers in the IRS market. As market
makers, these banks became the sellers of IRSs (the “sell side”), offering fixed and floating-rate
cash flows to their customers (the “buy side”). In this role, these Dealer Defendants profited
from the “spread” between the “bid” and the “ask” for IRSs. The “bid” and the “ask” were
typically set as follows: Because the floating rate is usually keyed to LIBOR, the key variable
that is negotiated when entering into an IRS is generally the fixed rate that will be paid. A buyside customer that seeks to pay the floating rate and to receive the fixed rate will receive the
“bid” price quoted by a dealer for the fixed rate; a buy-side customer that seeks to pay the fixed
rate and to receive the floating rate will pay the “ask” or “offer” price quoted by a dealer for the
fixed rate. For example, a dealer’s “bid” to pay the customer a fixed rate on a five-year IRS
might be 2.00% whereas its “ask” as to the fixed rate it would receive from the customer for the
same IRS might be 2.05%. The wider the spread between the bid and ask prices quoted by the
dealer, the more profit the dealer will make from its IRS line of business. SAC ¶¶ 73–75;
JTSAC ¶¶ 70–73.
Historically, dealers and their buy-side customers communicated almost exclusively over
the counter (“OTC”), by telephone. This means of IRS trading was advantageous to dealers. In
practice, it meant that customers received real-time pricing information from only the dealer, or
at most from a small number of dealers, because contacting many dealers for price information
was logistically unrealistic and costly. Dealers were also advantaged, because, to obtain a price
quote, the customer was required to disclose to the dealer its identity and the direction and
notional amount of the trade it sought to make. Dealers could use this “one-way flow of
information” about upcoming trades to their trading advantage. Dealers also often required

5

customers to execute trades “on the wire,” meaning during the phone call, or else the price quote
might lapse; or to disclose whether the customer was putting one dealer “in competition” with
other dealers. These practices, too, tended to diminish price competition. SAC ¶¶ 76–78;
JTSAC ¶¶ 74–76.
Electronic trading in fixed income securities was introduced in the mid-1990s. It gained
momentum in the IRS market in the early 2000s. Such trading had potential to make IRS trading
more efficient, transparent and competitive. But, plaintiffs allege, electronic trading developed
asymmetrically. With respect to trading between dealers, the dealers utilized electronic
platforms operated by entities known as interdealer brokers (“IDBs”). These platforms allowed
dealers to access better and transparent pricing for themselves, and speedier execution, while
they responded to buy-side requests to trade. A dealer using an IDB submits its bid and ask
prices to the IDB, which then publicizes the best quotes (known as the “inside market”)
anonymously to all other dealers in the platform. A dealer can immediately enter into an IRS
contract at a quoted price without negotiations, or it can ask the IDB to attempt to negotiate a
better price. As to standardized IRS products that are actively traded, an IDBs uses electronic
trading platforms akin to electronic “order books,” which automatically match the best bids and
offers. In return for facilitating dealer-to-dealer trades, an IDB earns a commission, known as a
“brokerage fee,” on each trade. But, plaintiffs allege, the IDBs allowed only dealer-to-dealer
transactions. Plaintiffs allege, for example, that defendant ICAP, the leading IDB for IRSs, did
not open its interdealer platform to buy-side customers. SAC ¶¶ 10–13, 84–86; JTSAC ¶¶ 81–
83.
In contrast, plaintiffs allege, until 2013, the only IRS electronic trading platforms that
developed that were open to buy-side customers typically used a “request-for-quotes” (“RFQ”)

6

protocol. In it, a buy-side customer, via electronic messages, can request quotes from several
dealers. Such a customer, however, was required to supply its identity at the time of the request.
Plaintiffs allege that these platforms are inferior to the dealer-to-dealer trading platforms in
several ways. Buy-side customers using these platforms are denied streaming prices (they
instead receive limited quotes via RFQs); are required to disclose their identities (they cannot
remain anonymous); cannot participate in live markets (they are limited to RFQ responses); and
are limited in the number of dealers whom they can access. SAC ¶¶ 78–80; JTSAC ¶¶ 77–80.
Plaintiffs allege that different infrastructures and processes also developed as between
dealer-to-dealer and buy-side trading to handle the process of clearing IRS trades. For dealer-todealer trades, dealers used central clearinghouses. A clearinghouse is an entity designed to step
into the middle of a bilateral trade to reduce counterparty risk. It becomes a counterparty to both
sides. It turns the transaction into two separate trades: a sale from the seller to the clearinghouse,
and a sale from the clearinghouse to the buyer. Central clearing is common to developed
financial markets, including equity and commodities markets. It is essential to anonymous
exchange trading, because it brings buyers and sellers to a centralized platform, creates an
infrastructure and unified standards for the processing of trades, and eliminates the need for
trade- and party-specific creditworthiness assessments. Because each party faced the same
counterparty (the clearinghouse), central clearing eliminates the need for contracts between the
parties to an IRS. In contrast, in a non-cleared IRS trade, the parties to the trade face each other
directly; each party therefore bears the risk that its counterparty will default on its obligations.
Plaintiffs allege that central clearing became feasible in the IRS market by the early 2000s. By
about 2005, SwapClear, an entity controlled by the Dealers, cleared most interdealer trades.
Despite its technological feasibility, however, plaintiffs allege that the central clearing model

7

was not extended to buy-side trades, inhibiting the introduction of an all-to-all trading platform
accessible to the buy-side. As addressed below, plaintiffs claim that the lack of central clearing
for buy-side trades, before Dodd-Frank took effect, resulted from collusion among the Dealer
Defendants. SAC ¶¶ 12, 90–98; JTSAC ¶¶ 87–96, 343. The Dealers counter that this reflected
unwillingness by the buy-side to incur the start-up and ongoing costs associated with a central
clearing mechanism, including the posting of substantial collateral to secure cleared trades.
2.

Dodd-Frank

On July 21, 2010, Congress passed the Dodd-Frank Wall Street Reform and Consumer
Protection Act. Relevant here, Title VII of Dodd-Frank amended the Commodities Exchange
Act (“CEA”) to “establish a comprehensive new regulatory framework for swaps,”3 and vested
the Commodities Futures Trading Commission (“CFTC”) with exclusive jurisdiction to
implement that framework. 7 U.S.C. §§ 2(a)(1)(A), 2(h). A goal of Dodd-Frank was to increase
accountability and transparency in the financial system, including in the OTC swaps markets,
which were viewed as less transparent than the exchange-traded futures and securities markets.
SAC ¶ 22; JTSAC ¶¶ 97–98.
The CFTC has issued sweeping regulations implementing Dodd-Frank. Three categories
of these regulations are relevant here.
Trading regulations: Dodd-Frank required that certain types of swaps be traded on
“swap execution facilities,” or “SEFs.” SAC ¶ 22, 202; JTSAC ¶¶ 97–101. A SEF is defined as
“a trading system or platform in which multiple participants have the ability to execute or trade
swaps by accepting bids or offers made by multiple participants in the facility or system.”

3

Core Principles and Other Requirements for SEFs, 78 Fed. Reg. 33,476, 33,477 (June 4, 2013)
(“SEF Rule”).
8

7 U.S.C. § 1a(50); 17 C.F.R. § 1.3(rrrr). Subject to certain exceptions, swaps must trade on SEFs
if they are subject to the CFTC’s mandatory clearing rules and have been “made available to
trade,” or “MAT,” by a registered SEF with the consent of the CFTC. See 7 U.S.C. §§ 2(h)(8),
7b-3.4
Under CFTC regulations issued in 2013, trading on a SEF may occur through either an
order book or an RFQ system. See 17 C.F.R. § 37.9(a)(2).5 In order-book trading, offers to buy
and sell are entered into a matching system and matched either manually or automatically by
algorithm. See generally 17 C.F.R. § 37.3(a). In RFQ trading, by contrast, a customer can send
a price request to a number of counterparties in a particular swap tenor, the prices are relayed
back to the customer, and the customer then has a time window in which to execute. Customers
who engage in RFQ trading must send requests to trade to a minimum of three recipients. See
17 C.F.R. § 37.9(a)(3). The CFTC thus enabled, but did not require, that IRSs be traded via an
anonymous, all-to-all trading platform.
In February 2014, a CFTC mandate for certain assets “made available to trade” went into
effect. See 7 U.S.C. § 2(h)(8); 17 C.F.R. § 37.10.
Clearing regulations: Congress believed that the risk of defaults on non-cleared swaps
“played an important role in freezing up credit markets” during the 2008 financial crisis. S. Rep.

4

Swaps subject to the SEF mandate may alternatively be traded on designated contract markets
(“DCMs”), regulated boards of exchange that have many of the same powers and duties as SEFs.
See Trading Organizations, Commodity Futures Trading Comm’n, available at
http://www.cftc.gov/IndustryOversight/TradingOrganizations/index.htm.
5

RFQ trading must be offered “in conjunction with an Order Book,” 17 C.F.R.
§ 37.9(a)(2)(i)(B), i.e., in transmitting responses to an RFQ in connection with a potential trade,
the SEF must also furnish any relevant quotes resting in its order book. See 17 C.F.R.
§ 37.9(a)(3)(i).
9

111-176, at 30 (2010). Dodd-Frank therefore granted broad authority to the CFTC to mandate
clearing of swaps. See 7 U.S.C. § 2(a)(1)(A).
In 2012, the CFTC enacted a final rule for mandatory clearing of IRS trades, effective on
a rolling basis beginning in March 2013. See 17 C.F.R. § 37.701; see also Clearing Requirement
Determination Under Section 2(h) of the CEA, 77 Fed. Reg. 74,284 (Dec. 13, 2012) (“Clearing
Requirements Rule”).6 In enacting this rule, the CFTC recognized that the vast majority of new
IRS clearing volume would come from the buy-side. See, e.g., Clearing Requirements Rule, 77
Fed. Reg. 74,284 at 74,287. The CFTC delayed implementation of this mandate to 2013, in part
due to multiple requests from buy-side entities for extra time to cope with the costs and burdens
imposed by implementing mandatory clearing, which one commenter described as
“overwhelming.” See Clearing Requirements Rule at 74,320; Swap Transaction Compliance and
Implementation Schedule, 77 Fed. Reg. 44,441, 44,456 (July 30, 2012).7
Dodd-Frank used a futures commission merchant (“FCM”) model for clearing, in which
an entity called an FCM serves as an intermediary between buy-side clients and the
clearinghouse. Plaintiffs allege that the Dealer Defendants’ ownership of almost all of the
approximately 20 FCMs for IRS trading effectively made them, under the Dodd-Frank regime,
“gatekeepers for IRS clearing.” SAC ¶¶ 22, 202; JTSAC ¶¶ 97–101 & n.14.

6

The CFTC exempted certain types of swaps and certain swap “end users” from the new
requirements. See Clearing Requirements Rule at 74,300–14; End-User Exception to the Clear
Requirement for Swaps, 77 Fed. Reg. 42,560, 42,590–91 (July 19, 2012).

7

See also Clearing Requirements Rule at 74,323 (“In order to comply with required clearing
under this adopting release, market participants are likely to face certain startup and ongoing
costs relating to technology and infrastructure, new or updated legal agreements, ongoing fees
from service providers, and costs related to collateralization of their positions”); Dealer Mem.
at 8 n.10 (citing and quoting buy-side comment letters).
10

Impartial access regulations: Dodd-Frank requires that SEFs provide all market
participants with “impartial access” to their trading facilities, thereby making, for example, buyside firms free to trade on “dealer-to-dealer” platforms. See 7 U.S.C. § 7b-3(f)(2)(B)(i). This
requirement “prevent[s] a SEF’s owners or operators from using “discriminatory access
requirements as a competitive tool” against particular market participants. See SEF Rule, 78
Fed. Reg. 33,476 at 33,508.
C.

Overview of Plaintiffs’ Theories of Collusion

The following sections first recount plaintiffs’ allegations as to the means used by the
Dealer Defendants, before Dodd-Frank took effect, to discourage emergence of an anonymous
all-to-all trading platform accessible to buy-side investors. Plaintiffs allege that the different
trading platforms and related infrastructure such as clearing mechanisms that developed for interdealer versus buy-side IRS transactions result from a long-running conspiracy among the Dealer
Defendants and others such as IDBs.
The later sections set out plaintiffs’ allegations as to how the conspiracy continued after
Dodd-Frank took effect. Plaintiffs claim that, after several electronic platforms (including Tera’s
and Javelin’s SEFs) emerged with potential to enable buy-side IRS transactions to occur on an
anonymous all-to-all basis, the conspiracy pivoted to boycotting and destroying these platforms.
D.

The Dealers Prevent Tradeweb From Developing an All-to-All Platform

In 1998, Tradeweb was founded as a private, dealer-backed firm that provided an online
marketplace for fixed-income products, such as U.S. Treasuries. Tradeweb later developed a
dealer-to-client RFQ platform for IRSs on which buy-side investors could (non-anonymously)
request non-executable and non-binding quotes from dealers. It was thus an OTC platform that
facilitated the dealers’ roles as the sole IRS market makers. Because Tradeweb’s platform relied

11

on the dealers to make markets, its success depended on the dealers’ use of the platform. In
2004, the dealers sold Tradeweb to Thomson Corporation. SAC ¶¶ 100–01; JTSAC ¶¶ 273–74.
Plaintiffs allege that—with IRS trading volumes having grown and IRS products having
evolved to the point where introduction of all-to-all electronic trading market was feasible—the
Dealer Defendants perceived a threat to their position as the primary IRS market makers. In
response, and aware that other financial instruments had migrated from OTC trading, Goldman
Sachs’ Principal Strategic Investments Group (“PSI”) devised a “dealer consortium” strategy to
work with the other Dealers “to maintain control of the IRS market.” The PSI’s principal goal
was to control how markets would evolve so as to protect the “dealer community” from the
threat to profitability presented by electronic exchanges and other forms of all-to-all trading that
threatened to “disintermediate” the dealers. Specifically, the PSI recognized that, if the buy side
of the IRS market shifted to all-to-all platforms, buy-side entities would be able to trade directly
with one another. This, in turn, stood to diminish the profits earned by the Dealer Defendants
from serving as intermediaries in the dealer-to-client market. Plaintiffs allege that similar
“strategic investment” groups at other Dealers (including at Bank of America, Barclays,
Citibank, Credit Suisse, Deutsche Bank, and J.P. Morgan) existed to control market structures
and that dealers who lacked a distinct group, such as BNPP, “conduct[ed] similar strategic
activities through their trading business.” Plaintiffs allege that, between 2008 and 2016, these
groups “regularly met in private,” communicated by phone and electronically, and “secretly
collaborated,” to protect their “privileged status as dealers in markets they have historically
controlled.” SAC ¶¶ 102–09; JTSAC ¶¶ 275–82.
This collaboration, plaintiffs allege, involved regaining control of Tradeweb “to prevent
it from introducing all-to-all trading.” By late 2007, plaintiffs allege, Tradeweb was planning to

12

introduce electronic all-to-all trading to the IRS market, for which there was great demand from
the buy-side. Tradeweb had touted itself as a “real-time trading platform” and represented that it
was “well positioned to capture new business as the market migrates to more efficient electronic
trading platforms.” In response, members of three strategic investment groups—Goldman Sachs
(Brad Levy), Deutsche Bank (Stephen Wolff), and Lehman Brothers (Dexter Senft)—“devised
and implemented a scheme to eliminate the threat from Tradeweb.” SAC ¶¶ 111–12; JTSAC
¶¶ 283–84.
Specifically, plaintiffs allege, Levy, Wolff and Senft recognized that Tradeweb’s
liquidity contracts with the Dealers—which Tradeweb needed to drive liquidity to its RFQ and
eventually its all-to-all platform—were expiring. This gave the Dealers leverage over Thomson,
which “realized the banks would take their liquidity and shop it around, which would threaten
the value of TradeWeb.” To exploit this leverage, Levy, Wolff and Senft recruited other Dealers
to join an initiative they named “Project Fusion.” It was designed to enable the Dealers to take
control of Tradeweb’s IRS business to eliminate Tradeweb as a threat, but without attracting
scrutiny. SAC ¶¶ 113–14; JTSAC ¶¶ 285–86.
Project Fusion, plaintiffs allege, entailed first taking over majority ownership of a
company that would be the entity through which the Dealer Defendants controlled Tradeweb,
while making it appear that they were taking only a “minority” stake as investors. To this end,
on October 1 and 2, 2007, the Dealer Defendants and Tradeweb formed two Delaware LLCs.
The first, “Tradeweb Markets LLC,” would be “the public face of the deal and the one through
which [d]efendants would take a minority stake in exchange for paying $180 million to
Thomson[,]” Tradeweb’s then-parent. The second was “Tradeweb NewMarkets LLC”; the
Dealer Defendants paid Thomson approximately $280 million for an “overwhelming majority

13

stake (80%) in this entity,” which would “house Tradeweb’s IRS business.” Plaintiffs allege that
the Dealers gave the two new entities similar names and used the generic name “Tradeweb” in
their press releases, to conceal their control over the entity responsible for IRS trading. Plaintiffs
allege these press releases also misleadingly claimed the Dealers were investing in Tradeweb’s
business to provide “a significant opportunity for institutional investors to benefit from the
efficiencies of electronic trading,” when in fact, their purpose was to keep Tradeweb from
offering an all-to-all platform.8 Plaintiffs allege that the Dealers agreed with each other (1) that
Tradeweb would not move forward with an all-to-all trading platform or support other platforms
that would move the market toward all-to-all trading; and (2) to provide liquidity to Tradeweb’s
platforms to the exclusion of competing platforms.” SAC ¶ 115–25; JTSAC ¶¶ 287–94.
After obtaining these stakes, plaintiffs allege, the Dealer Defendants installed their senior
personnel on the boards of Tradeweb Markets, LLC (16 of 26 seats) and Tradeweb NewMarkets,
LLC (16 of 24 seats). These included named personnel from Goldman Sachs, Bank of America,
Barclays, Citibank, Credit Suisse, Deutsche Bank, J.P. Morgan, Morgan Stanley, RBS and UBS.
The CEO of both entities was a former chief operating officer for fixed income at Credit Suisse.
SAC ¶¶ 126–37; JTSAC ¶¶ 295–06.
Plaintiffs allege that the boards’ members “include some of the primary architects of the
conspiracy” and communicated in several ways. Members “met regularly under the cover of
Tradeweb’s boards and committees to plan how the Dealer Defendants could maintain control of
the IRS market and make sure they were not ‘disintermediated’ by buy-side friendly IRS trading

8

Plaintiffs allege that the other dealers originally did not allow Barclays to participate in the
Tradeweb consortium because “they had placed Barclays in the ‘penalty box’ for attempting to
launch a dealer-to-client IRS electronic trading platform without the other dealers’ permission.
When Barclays’ “penalty” expired, plaintiffs allege, the Dealers allowed it, in 2009, to join other
dealers in taking what they described as a “minority equity stake” in the business. SAC ¶ 124.
14

platforms.” There were also often-weekly board conference calls, and an annual board meeting
in Miami, Florida, which, plaintiffs claim, the Dealers used “to discuss and coordinate their
strategy for controlling the IRS market.” The Dealers also allegedly controlled Tradeweb
through governance committees that determined, inter alia, who can participate on Tradeweb’s
SEFs. Finally, the Dealers’ personnel discussed “market structure issues” at dinners at the home
of the head of interest rate swaps for Tradeweb’s interdealer SEF and at New York City
restaurants, attended by Dealer representatives. In these gatherings, plaintiffs allege, the Dealers
“coordinate[d] their strategy and “discussed their plans for keeping the market bifurcated and
preventing the natural progression to all-to-all trading.” SAC ¶¶ 138–41; JTSAC ¶¶ 307–10.
In November 2010, Tradeweb NewMarkets LLC merged into Tradeweb Markets LLC,
leaving the latter as the surviving company. The Dealers continued to hold a majority of seats on
the surviving entity’s board. SAC ¶ 142; JTSAC ¶ 311.
Plaintiffs allege that, from Project Fusion forward, Tradeweb, while holding itself out as
independent, acted for the benefit of the Dealers. Although Tradeweb’s independent (i.e., nonbank) managers recognized that it would be in Tradeweb’s interest to launch an all-to-all
anonymous electronic trading platform, the Dealers pressured them to change course; these
managers capitulated. Plaintiffs allege that the Dealers publicly touted Tradeweb as a platform
that could provide “accurate pricing information” and “greater market transparency.” In fact,
these claims were untrue as to the dealer-to-client OTC market structure, in that the “buy side
can trade on Tradeweb only via its dealer-to-client RFQ platform and cannot engage in all-to-all
trading.” The decision to forego all-to-all trading cost Tradeweb additional business and profits
from the resulting fees. SAC ¶¶ 143–46; JTSAC ¶¶ 312–15.

15

As of the date of the SACs, plaintiffs allege, Tradeweb, pursuant to its agreement with
the Dealer Defendants, “continues to play an active role in maintaining a two-tiered market
structure.” Post-Dodd-Frank, Tradeweb today operates two different SEFs: Dealerweb SEF,
which plaintiffs claim is meant only for dealers and allows anonymous competitive trading; and
Tradeweb SEF, which “is designed for non-dealer market participants and always discloses
counterparty identities.” To deal on the former, Tradeweb charges $50,000 per month and
requires that an entity pay for a minimum of a full year (i.e., $600,000). But it charges only $100
per month to trade on Tradeweb SEF. The reason for this cost differential, plaintiffs allege, is to
“establish and entrench a bifurcated dealer-to-dealer and dealer-to-customer marketplace.”
Plaintiffs also allege that while Tradeweb SEF claims to offer an “order book,” it is effectively
closed off to the buy side, because dealers do not trade on that platform, but instead trade with
each other on Dealerweb or other IDBs and limit trades with end users to the dealer-to-client
RFQ. The Tradeweb platform is thus “inactive and seen by the buy side as inaccessible.” SAC
¶¶ 147–49; JTSAC ¶¶ 316–19.
E.

The Dealers Use Trade Associations as Forums for Collusion

Plaintiffs identify Tradeweb as the “principal forum though which the Dealer Defendants
. . . colluded to coordinate their efforts to control the IRS market.” But, they allege, the Dealers
also used two trade associations as means of collusion. SAC ¶ 151; JTSAC ¶ 320.
One was ISDA. From 2008 forward, the Dealers controlled ISDA’s board, on which
employees of 10 Dealer Defendants sat. Another was the Futures Industry Association (FIA), on
whose board employees of 10 Defendant Dealers sat. Among the working groups created by
these trade associations was one devoted to drafting the “Cleared Derivatives Execution
Agreement (“CDEA”),” a form contract intended to govern the relationship between clearing

16

agents and their customers. After Javelin came to exist, its executives sought to attend a meeting
of this working group out of concern that the CDEA was being drafted only for OTC transactions
and did not anticipate IRS trades covered on a SEF; Javelin’s CEO and general counsel were
permitted to attend, but when the general counsel inquired about the CDEA, her question was not
answered. The chair of the meeting, a Credit Suisse in-house counsel, later told Javelin that its
personnel were unwelcome at future meetings. SAC ¶¶ 152–57; JTSAC ¶¶ 321–27.9
F.

The Dealers Prevent IDBs from Opening Platforms to the Buy Side

The SACs allege that, in several contexts, the Dealer Defendants acted to prevent IDBs
from opening all-to-all trading platforms to buy-side investors.
1.

Punishment of IDBs That Took Steps Towards All-to-All Platforms

Plaintiffs allege that the Dealers threatened to punish any IDB that considered opening its
platforms to the buy side. This included threatening to withdraw liquidity from such platforms,
which the Dealers referred to as placing the IDB in the “penalty box” or “pulling the line” on the
IDB. Twice, in 2009 and 2014, GFI group (“GFI”), which operates IDB platforms, attempted to
introduce anonymous trading protocols on its platform; each time, the Dealers threatened to pull
their business from GFI, which reversed course. In the 2014 episode, GFI received heated phone
calls from Credit Suisse and J.P. Morgan, and reverted to using a name-disclosed protocol. The
Dealers also threatened buy-side investors with the “penalty box” for attempting to trade on an
all-to-all platform. Such threats were effective, plaintiffs claim, because they deprived the
investor of access to primary market makers and thus from trading IRSs; the Dealers sometimes
also threatened investors with “withdrawal of key banking services.” As a result of these tactics,

9

The SACs also allege generally that the heads of trading desks at the Dealer Defendants
“maintained an ongoing dialogue” through emails and text messages and thereby discussed their
mutual desire to maintain the status quo. SAC ¶ 158.
17

plaintiffs allege, IDBs did not open their platforms to the buy side. SAC ¶¶ 159–65; JTSAC
¶¶ 328–34.
2.

The Dealers’ “Détente” With ICAP

In 2009, plaintiffs allege, defendant ICAP agreed with the Dealer Defendants to prevent
the buy side from accessing its platform, as part of an agreed “détente.” ICAP had contemplated
giving such access to the buy side in retaliation for Tradeweb and the Dealers having launched a
trading platform for mortgage bonds called “Dealerweb,” which cut into ICAP’s mortgage-bond
platform’s business. In retaliation, ICAP threatened to launch an all-to-all trading platform open
to the buy side. This threat, plaintiffs claim, was credible because ICAP was developing an
electronic-trading platform for Europe called iSwap, which ICAP could have launched in the
United States as an all-to-all anonymous IRS trading platform, and/or because ICAP could have
allowed buy-side investors to use its existing IDB platform. In response, representatives of
ICAP and the Dealer Defendants spoke in 2009 and reached a détente. The Dealers agreed not to
further expand Dealerweb into the IDB space; in exchange, ICAP agreed not to establish an allto-all anonymous trading platform. SAC ¶¶ 166–69; JTSAC ¶¶ 335–37.
Plaintiffs allege that ICAP and the Dealers have abided by that agreement, despite
iSwap’s success and the potential to ICAP of expanding iSwap to the United States. They allege
that when ICAP did expand that platform to the United States in 2013, it took steps that
effectively limited participation to the Dealers, while publicly claiming otherwise. The Dealers,
in turn, abiding by the détente, have kept Dealerweb from expanding into IRS or any other
market in a meaningful way. SAC ¶¶ 166–73; JTSAC ¶¶ 335–42.
G.

The Dealers’ Attempts to Block Buy-Side Clearing

18

Plaintiffs allege that the Dealer Defendants identified the central clearing of OTC
products as a first step towards electronic trading. They cite a 2010 report from J.P. Morgan
describing clearing as a “main concern for the Investment Banking industry” because it could
lead to “exchange/SEF trading for OTC products.” As a result, plaintiffs allege, the Dealers long
sought to control the clearing infrastructure for the financial markets they dominate, to position
themselves to “head off threats to their dominance.” In the context of the IRS markets, the
Dealers allegedly took control of the IRS clearing infrastructure, and their representatives
allegedly met to discuss “how to prevent or delay buy-side access to IRS clearing.” These
discussions occurred on Tradeweb and through an entity called OTCDerivNet.” SAC ¶¶ 174–
75; JTSAC ¶¶ 343–44.
Plaintiffs allege that the Dealers took several steps to block such clearing.
First, the Dealers agreed, through their strategic investment groups, to take control of an
IRS clearinghouse, “SwapClear,” formed in 1999 by LCH.Clearnet, an entity that builds
clearinghouses for financial instruments. To do so, eight Dealers, in October 2000, formed
OTCDerivNet; four others joined in 2001 and two others in 2009. They held out OTCDeriv.Net
as for all IRS participants, including the buy side. In fact, plaintiffs allege, OTCDerivNet was
formed as a vehicle to secure control of SwapClear. After creating the new entity, the Dealers
announced that it would “partner” with LCH.Clearnet to develop a secure, efficient and costeffective trade environment” for the OTC derivative industry. In fact, the Dealers, through
OTCDerivNet, provided 100% of the funding of SwapClear, in return for a share of profits and
governance control. Using that control, the Dealers imposed restrictions that effectively limited
clearing membership only to Dealers. These included that any new member must put up $5
billion in capital towards the clearinghouse, a sum so large only investment banks could afford it.

19

The Dealers also required unanimous approval of existing members before a new member could
join. SAC ¶¶ 176–80; JTSAC ¶¶ 345–49.
Second, plaintiffs allege, the Dealers, through their board domination of OTCDerivNet,
acted to block other buy-side clearing ventures. In the mid-2000s, the Chicago Mercantile
Exchange (“CME”), a sophisticated entity with a long history of developing clearing solutions
for financial markets, was focused on doing so for swaps markets, including IRSs and CDSs. It
announced plans to introduce a cleared IRS product called “CME Cleared Swaps,” which
promised to offer clearing of OTC swaps executed on or submitted through its trading platform
Swapstream or through its OTC platform. Given CME’s track record and technical and
commercial ability, the Dealers viewed this as a “serious threat.” In response, plaintiffs allege,
the Dealers boycotted Swapstream. “Meeting through OTCDerivNet, the Dealer Defendants
agreed to clear only interdealer IRS trades and only on SwapClear (the entity they controlled).”
Because the Dealers are a party to nearly every IRS trade, plaintiffs allege, their collective
refusal to clear IRS trades through Swapstream meaning that effectively no trades were cleared
through the platform, “starving CME of clearing volumes and revenues.” As a result, “CME’s
clearing solution for IRS[s] swiftly failed.” SAC ¶¶ 181–88; JTSAC ¶¶ 350–56.
Third, after the passage of Dodd-Frank, regulators put pressure on the Dealers to expand
clearing in derivatives markets and to open up clearing to the buy side. In response, plaintiffs
allege, the Dealers made superficial changes at SwapClear that purported to expand buy-side
access, while adopting new rules that prevented meaningful buy-side participation. For example,
they added a rule that needlessly required clearing members to contribute large amounts of
capital to the clearinghouse’s default (or “guaranty”) fund. Only large investment banks like the
Dealers could meet this requirement. SAC ¶¶ 189–90; JTSAC ¶¶ 357–58.

20

Fourth, as noted, Dodd-Frank mandated that certain IRSs move to SEFs and be centrally
cleared. Dodd-Frank’s FCM model of clearing made the FCM divisions owned by the Dealer
Defendants—most FCMs—the “gatekeepers” for IRS clearing. Plaintiffs allege that the Dealers
exploited this by causing their FCMs to refuse to clear buy-side trades on IRS trading platforms
that use all-to-all protocols. Plaintiffs allege that FCMs should be agnostic as to which trading
platforms their customers use; any trade it clears adds to its revenue. But, plaintiffs allege, the
Dealers caused their FCMs to withhold clearing services from any buy-side entity seeking to
trade on SEFs that operated all-to-all anonymous trading platforms, including Tera, Javelin, and
TrueEx. This practice was aimed at preventing IRS trades from proceeding on these platforms.
Separately, plaintiffs allege, FCMs refused to carry out pre-trade checks for any prospective
trade on an anonymous all-to-all SEF. This restricted order flow of buy-side customers onto allto-all platforms. Plaintiffs allege that the Dealers coordinated these activities through their
clearing operations, which communicate regularly and meet “under the cover of FIA board
meetings.” SAC ¶¶ 191–201; JTSAC ¶ 100.
H.

The Dealers Attempt to Block SEFs Offering All-to-All Trading

Plaintiffs allege that, after Dodd-Frank’s implementing regulations took effect, the Dealer
Defendants conspired to boycott and destroy the SEFs that emerged and that promised to offer
all-to-all anonymous trading of IRSs. As developed below, plaintiffs allege that three
companies—TeraExchange, Javelin, and TrueEx—each spent millions of dollars developing
such platforms. But, plaintiffs allege, the Dealers conspired to boycott these platforms, and to
cause their affiliated clearing entities to refuse to clear trades on them. Plaintiffs allege that the
Dealers coordinated their “joint opposition” to these SEFs through their “strategic investment
groups,” through “secret discussions,” and through a forum supplied by an IDB called

21

“Tradition,” which, between 2013 and 2015, hosted monthly meetings in New York City with
the collective heads of the Dealers’ trading desks to discuss issues relating to SEFs. Plaintiffs
further allege that, when a SEF solicited participation by the Dealers, the Dealers coordinated
their response; for example, a Goldman Sachs employee in its E-Commerce division would call
his counterparts to discuss whether to use the platform. As a result, “Goldman Sachs and the
other Dealer Defendants did not make individual decisions about which SEFs to support. They
made those decisions in collaboration with each other.” SAC ¶¶ 202–06; JTSAC ¶¶ 124–42.
The effect of the “roadblocks” implemented by the Dealers, plaintiffs allege, was that,
whereas it had been forecast that 40–50 firms could end up competing for swaps execution
business, the three SEFs that invested heavily in doing so were “largely crushed by the Dealer
Defendants’ cartel.” Each had developed the necessary technology, secured the required
regulatory approvals, and garnered sufficient support to introduce anonymous all-to-all trading
platforms to the buy side. But, plaintiffs allege, the Dealers’ tactics—including refusing to trade
on these new platforms, refusing to let their affiliates clear buy-side trades executed on the new
platforms, and threatening to retaliate against anyone who traded on the platforms—“effectively
shut down Tera Exchange and Javelin.” The Dealers allegedly used similar strategies to limit
dealer-to-client trading on TrueEx’s platform to by RFQ protocols. As a result, plaintiffs claim,
a dichotomous IRS trading market persisted, in which the buy side “continues to be shut out of
all-to-all electronic trading” but the Dealers are not. SAC ¶¶ 207–09; JTSAC ¶¶ 138, 141–42.
1.

The Boycott of TeraExchange

Tera was formed in 2010. It raised $7 million in capital. It developed all-to-all electronic
trading platforms for asset classes including IRSs—platforms which it claimed promised greater
price transparency and competition and to offer firm quotes to the buy-side. Tera developed the

22

technology necessary for such a platform, including swap data repositories, connectivity with the
various clearinghouses and credit hubs, and a system to ease the calculation and posting of
collateral. Tera offered “full market depth” and “real time pricing.” JTSAC ¶¶ 115–16.
Tera, with such features, began offering access to its electronic platform in 2011, while it
waited for the CFTC to formalize SEF registration rules. Tera was described at the time as
poised to take business from “the big banks that have dominated swaps trading.” While it
welcomed the Dealers to participate on its platform, plaintiffs allege, Tera did not depend on
them. Instead, it focused on recruiting large proprietary trading firms that were eager to trade
IRSs, 15 of which the SAC names. These firms, plaintiffs allege, “were ready, willing, and able
to provide the liquidity to the Tera platform that would allow the platform to thrive,” were eager
to trade IRSs, and “were willing to quote immediately executable bid/ask spreads” that were
tighter than those offered by traditional dealers. Plaintiffs allege that, had these proprietary
trading firms participated in Tera’s platform, then Dealers would effectively have been obliged
to use the platform as well. That is because Dealers’ clients often demand that Dealers supply
them “best execution,” meaning securing the best price available in executing a trade when
acting as the customer’s agent. Had the best price been available on Tera, plaintiffs allege,
Dealers would have been required “by regulation or contract” to obtain it, leading to execution of
trades on that platform. SAC ¶¶ 210–17; JTSAC ¶¶ 119–20.
In preparation for its launch, plaintiffs allege, Tera made its platform compliant with
regulatory rules. On September 19, 2013, the CFTC granted Tera temporary SEF certification.
And numerous market participants, 12 of which plaintiffs name, committed to provide liquidity
to the platform. By the end of 2013, Tera was valued at more than $50 million. SAC ¶¶ 217–19;
JTSAC ¶¶ 122–23.

23

Plaintiffs allege that the Dealers “conducted secret meetings, including under the cover of
Tradeweb, to discuss how to neutralize this threat,” which the Dealers termed a “Trojan Horse.”
One tactic was to attempt to jointly invest in Tera, “with the real goal of taking it over and
shutting it down.” Several Dealers reached out to Tera to arrange meetings; at these, in early
2013, Tera expected to discuss trading but were instead “met by strategic investment personnel
offering to take a stake in TeraExchange’s company.” One meeting was with two Goldman
Sachs officials; similar tactics were used by strategic investment groups at Barclays, Bank of
America, Credit Suisse, Citi, Deutsche Bank and Morgan Stanley. Tera declined the Dealers’
“investment” offers; no Dealers signed up for Tera’s platform or ever agreed to provide liquidity,
“pursuant to their agreement to starve TeraExchange of liquidity.” SAC ¶¶ 220–24; JTSAC
¶¶ 186–88.
Plaintiffs allege that the Dealers used other “aggressive tactics” to thwart Tera. One was
refusing to clear, on their FCMs, trades of any of their buy-side clients occurring on Tera; Tera
received reports from those who had signed up for it that these FCMs were refusing to clear. To
obscure this refusal, at times, these FCMs would quote “extremely high clearing fees” for trades
executed on Tera, while quoting far lower fees for clearing trades on platforms they did not view
as threats. For example, Bank of America’s FCM quoted Knight Capital Group “exorbitant
clearing fees” to Tera, while charging Knight no clearing fees for trades executed “on dealerfriendly platforms such as Tradeweb and Bloomberg.” When Tera tried to convince FCMs to
change course and clear trades, one official, at Bank of America, replied: “My bosses are never
going to let me” clear trades for TeraExchange. SAC ¶¶ 225–28; see also id. ¶¶ 229–31 (citing
“runaround” explanations for non-participation given to Tera by Barclays, Morgan Stanley,

24

Credit Suisse, HSBC, and ANZ Bank); JTSAC ¶¶ 189–210 (alleging refusals to clear by FCMs
of numerous dealers).
The Dealers’ “collective refusal” to clear trades on Tera’s platform “was a major blow”
and caused early supporters to withdraw. Plaintiffs also allege that several buy-side entities
indicated to Tera that they “were facing reprisals” from dealers for meeting with Tera. On
Friday, June 13, 2014, two large trading platforms successfully conducted the first trade on
Tera’s order book, a single trade for a notional amount of $10 million. The trade was to be
cleared through BNPP’s clearing affiliate, but BNPP’s trading desk “immediately contacted the
parties to the trade and threatened them with a loss of access to clearing and other banking
services, including execution services in other asset classes and access to general market
research, if they continued to trade on TeraExchange.” The next business day, Monday, June 16,
2014, BNPP, Citibank, J.P. Morgan, and UBS, “each separately contacted TeraExchange
demanding to ‘audit’ TeraExchange’s rulebook and stating that they would not help clear any
further trades until the audit was complete.” There was, in fact, no valid basis for such an audit,
as Tera had already been certified by the CFTC; the Dealers’ “real goal was to prevent the buy
side from trading on TeraExchange.” At around the same time, Bank of America’s FCM
“threatened buy-side clients with inflated clearing fees and liquidity boycotts if they made
markets or traded on Tera’s all-to-all trading platform. As a result, no such trades took place.
SAC ¶¶ 232–36; JTSAC ¶¶ 206–10.
Plaintiffs allege that the Dealers shut down a different means by which Tera attempted to
build its platform. Because Dodd-Frank requires certain IRS trades involving U.S. entities—
whether trading domestically or internationally—to be executed through a SEF, Tera recognized
that European IDBs serving U.S. banks had to become a SEF or execute their trades through a

25

SEF. Tera reached out to European IDBs, which lacked proprietary SEFs, to offer them its
services; it initially received favorable responses. The Dealers, however, “shut down” this effort.
“In strikingly parallel fashion, in early- to mid-2014, the Dealer Defendants informed these IDBs
they would not accept any trades reported or executed on TeraExchange.” Citibank and BNPP in
particular each told an IDB that they would not accept trades reported or executed on Tera, and
the CEO of a second-tier IDB told Tera that the major swap dealers would not allow him to sign
up with TeraExchange. SAC ¶¶ 237–42; JTSAC ¶¶ 211–16.
In the end, plaintiffs allege, these efforts succeeded. “[T]o date, no IDB has been able to
process a trade through TeraExchange’s order book, and some have since gone out of business.”
TeraExchange, in turn, shifted its focus to other lines of business, and “effectively left the IRS
market.” SAC ¶¶ 243–44; JTSAC ¶¶ 217–18.
2.

The Boycott of Javelin

Javelin was formed in 2009. It developed two all-to-all IRS trading platforms—an
anonymous RFQ platform and an anonymous order book, both offering firm pricing for swaps as
opposed to indicative quotes. In 2011, it began soliciting support for its platforms by giving
demonstrations to buy-side entities and five Dealer Defendants (BNPP, Deutsche Bank,
Goldman Sachs, J.P. Morgan, and RBS). It offered these Dealers the opportunity to receive a
portion of Javelin’s brokerage fees if they traded on the platform. Javelin’s platform, which had
a “state of the art user interface,” successfully tested both as to execution and clearing of trades.
It contained options to allow market participants to trade electronically either through that
interface or through an industry-standard third-party gateway. The development of Javelin’s
SEF and technology platform called upon “enormous industry knowledge and technological
skill,” as well as substantial investments in, among other things, office rental, data center fees,

26

and licenses for software and financial vendor platforms. By 2011, Javelin’s order book was
ready for testing and had successful test trials; by 2013, Javelin had created its own live creditcheck system, which enabled Javelin to confirm that the customer had sufficient credit with their
FCM to execute and clear a trade. On September 19, 2013, the CFTC granted Javelin temporary
registration as a SEF, and Javelin’s platforms were operational. On October 1, 2013, Javelin
successfully executed and cleared its first trade as a SEF; by late 2013, Javelin had signed up
seven second-tier dealers to trade on the platform. SAC ¶¶ 245–48; JTSAC ¶¶ 104–12.
Most Dealers, however, refused to provide liquidity to Javelin. This, plaintiffs claim,
reflected an agreement “with each other only to support platforms they could control.” Plaintiffs
allege that the Dealers recognized that if Javelin successfully launched an anonymous all-to-all
platform, this “would imperil their privileged status as market makers in a bifurcated market, as
well as the supracompetitive bid/ask spreads they extracted from the buy side.” SAC ¶¶ 247–48;
JTSAC ¶ 148.
Between 2013 and 2015, Javelin met with senior employees at each Dealer, seeking their
support. Of the Dealers, only RBS, for a short period, agreed to trade or provide liquidity to
Javelin. Some Dealers refused to use the platform; others “provided an endless stream of
pretextual excuses,” such as the need for legal reviews of Javelin’s rulebook, for refusing to do
so; one stated that it needed to see Javelin’s internal documentation but never did so. Some
refused to test the platform’s operability; others “flatly refused to engage.” Plaintiffs claim that
J.P. Morgan initially proclaimed interest in trading on Javelin, but this “was merely a ruse to
collect information on its trading platforms,” in that J.P. Morgan “burden[ed] [Javelin personnel]
with numerous pretextual requests for information.” SAC ¶¶ 249–50; JTSAC ¶¶ 143–45, 149,
151–61.

27

In September 2013, Javelin conducted a series of mock trading sessions to showcase its
all-to-all platforms to dealers and buy side entities. In advance of these sessions, Javelin asked
the Dealers to conduct pre-trade credit checks for their buy-side customers to participate in these
sessions. The Dealers’ FCMs, including J.P. Morgan’s, “largely refused” to do so, “effectively
preventing the buy side from using Javelin’s platform. Morgan Stanley similarly refused to act
when Javelin, claiming to speak for a number of Morgan Stanley’s FCM clients, asked it to
connect to the platform so enable those clients to trade. Eventually, a Morgan Stanley official
agreed to a demonstration of the platform’s functionality, but delayed attending; even though the
demonstration was successful, Morgan Stanley refused to clear trades executed on the platform.
SAC ¶¶ 251–54; JTSAC ¶¶ 146, 154.
Deutsche Bank allegedly gave Javelin “a similar runaround.” For months, it claimed that
its legal department needed to approve Javelin’s rulebook. This delay prevented seven buy-side
entities that cleared trades through Deutsche Bank from trading on Javelin, as they had wished.
Deutsche Bank also refused to test Javelin’s connection to it, claiming a lack of capacity, even
though, plaintiffs claim, testing the connection would have been simple. Plaintiffs allege that
there were conversations over a period of 15 months that reveal an evolving series of excuses by
Deutsche Bank for not doing such testing, for not completing a “legal review,” and ultimately,
for not approving trades on Javelin. SAC ¶¶ 255–61; JTSAC ¶¶ 152, 162–69. Other Dealers,
including Morgan Stanley and Barclays gave similar, allegedly pretextual, excuses for not
allowing their FCMs to clear trades executed through Javelin. JTSAC ¶¶ 170–72.
Goldman Sachs also allegedly refused to allow its FCM to clear such trades. Although its
officials did not state that Goldman Sachs would not support Javelin, its officers “demonstrated
hostility” to Javelin at a September 24, 2013 meeting, and asked for the names of buy-side

28

entities that had signed up to use the platform.” The next week, Javelin employees confronted a
Goldman Sachs official, Tony Smith, a vice president of network engineering, about its refusal to
clear trades executed on Javelin. “Smith responded that under no circumstances would Goldman
Sachs deal with Javelin” and that “if Goldman Sachs was somehow forced to clear trades
executed on Javelin, it would simply set the credit limits for all of its customers who attempted to
trade on Javelin at ‘zero,’ thereby preventing the clearing of any trades executed on its
platforms.” Thus: “Goldman Sachs’ FCM was willing to forego clearing revenues and damage
customer relationships in order to shut Javelin out of the market. SAC ¶¶ 262–63; JTSAC
¶¶ 154, 170.
Some Dealers “actively pressured their customers not to trade on Javelin.” In August
2013, Senft, by then of Morgan Stanley, asked Javelin for a list of the buy-side customers who
had expressed interest in Javelin; after Javelin gave this information, Senft distributed this list to
other Dealers. One such customer, NISA Investment Advisors LLC (“NISA”) had been eager to
trade on Javelin. But when Goldman learned this from Senft, a Goldman official notified NISA
that “if NISA traded on Javelin, Goldman Sachs would withdraw all clearing services for the
buy-side entity in any trading venue.” NISA later backed out of using Javelin’s platforms; a
NISA official later told Javelin that Goldman Sachs had forced him to stop trading on the
platform. Another buy-side entity, the Citadel Fixed Income Master Fund, started using the
platform in early 2014 but “suddenly stopped” such trading after its chief operating officer
“received calls from the Dealer Defendants telling him not to use Javelin’s platforms.” SAC
¶¶ 264–66; JTSAC ¶¶ 173–74.
Despite these obstacles, plaintiffs allege, Javelin got some trading volume from its
participants, having signed up, by late 2014, approximately 80 entities (19 named in the SAC) to

29

trade on its all-to-all platform. But, because the Dealers refused to allow Javelin-based trades to
clear through their FCMs and because of “the threats . . . described above,” many potential
customers of Javelin were unable to clear trades executed there, and, ultimately, “few were able
to make any trades.” One buy-side entity, Mitsubishi, told Javelin that it had ceased attempting
to trade on Javelin because it had “received ‘too much static’ from J.P. Morgan, [its] FCM, for
doing so.” In October 2014, “buy-side interest in Javelin and other all-to-all trading platforms
collapses as buy-side investors grew concerned the Dealer Defendants would retaliate against
them for trading on the platform.” Also in October 2014, a Deutsche Bank executive met with
and told a Javelin official that “if he continued to push for Javelin’s success, he would ‘never
work’ on Wall Street again.” SAC ¶¶ 267–70; JTSAC ¶¶ 173–78.
In sum, plaintiffs allege, the Dealers had “clearly coordinated their joint opposition to
Javelin; and proffered, in conversations with Javelin employees between August 2014 and April
2015, “nearly identical excuses for their refusal to deal, pointing to a supposed lack of buy-side
participation on Javelin as the reason for their actions.” But these excuses, plaintiffs claim, were
pretextual, because these Dealers “had gone to great lengths to ensure that buy-side firms did not
trade on Javelin.” SAC ¶¶ 271–72.
As for RBS, plaintiffs allege that it initially provided liquidity on Javelin, although not
“in a meaningful way,” and the prices that RBS streamed were “at unfavorable bid/ask spreads
that resulted in no trades with buy-side customers.” Meanwhile, RBS demanded significant
control in the operation and strategy of Javelin’s platform. After Javelin, on October 18, 2013,
filed an application with the CFTC that stood to allow a wide range of cleared IRSs to be
executed on SEF’s, plaintiffs allege, a senior IRS trader at RBS complained to Javelin’s CEO,
James Cawley, and RBS soon withdrew all its liquidity on the platform. In the ensuing days, a

30

half-dozen RBS officials pressured Cawley to retract that submission. On October 30, 2013,
Javelin filed a revised submission, narrowing the range of IRS instruments it covered. Even so,
RBS and other Dealers continued to pressure it to reduce the scope of its application. Only after
Javelin narrowed its submission did RBS resume providing Javelin with liquidity. Plaintiffs
further allege that RBS’s head IRS trader told Javelin not to discuss that RBS was a market
maker on its platform, and that RBS gradually began widening the bid/ask spread it would
stream to Javelin and otherwise distancing itself from Javelin. On April 26, 2015, RBS withdrew
as a participant on Javelin’s platform. SAC ¶¶ 273–78; JTSAC ¶¶ 158–61.
In October 2014, plaintiffs allege, buy-side interest in trading on Javelin “collapsed”
when buy-side customers learned that all-to-all trades on Javelin were not assured anonymous.
News of this spread via a Dealer-friendly IDB. JTSAC ¶ 179.
As a result of the Dealers’ actions, plaintiffs allege, Javelin today “effectively has no
revenues and facilitates no IRS trading. “Despite years of development and millions of dollars in
investment capital, Javelin has executed less than 180 trades since its launch.” SAC ¶ 278;
JTSAC ¶¶ 177–78, 182–84.
3.

The Boycott of TrueEx

TrueEx was founded by Sunil Hirani, who had previously started a successful electronic
trading platform for CDSs for an IDB called Creditex. In 2013, TrueEx sought to bring to
market a SEF offering two IRS trading platforms: an anonymous all-to-all order book and a nonanonymous dealer-to-client RFQ platform. Approved by the CFTC, the TrueEx swaps exchange
had features that caused it to be touted as having potential to increase competition and reduce
prices in derivatives markets. Hirani stated publicly that TrueEx had signed up 62 buy-side
participants. SAC ¶¶ 279–81; JTSAC ¶¶ 219–22.

31

Plaintiffs allege that the Dealer Defendants boycotted TrueEx, by refusing to trade IRSs
with buy-side investors on the TrueEx order book. They allege generally that, were a buy-side
investor to execute a trade on TrueEx’s order book, it would have to clear the trade through one
of the Dealers’ FCMs, which would alert the Dealers to this, causing them to retaliate. They
further allege that the Dealers refused to use the TrueEx platform for dealer-to-dealer trading, a
refusal which “makes no sense, except as part of [d]efendants’ conspiracy,” because TrueEx
offered better financial terms than other platforms in which the Dealers traded, such as
Tradeweb. The Dealers did provide “some limited liquidity to TrueEx’s non-anonymous dealerto-client RFQ platform,” but ensured that this platform did not “reach critical trading mass” by
refusing to trade “plain vanilla swaps”—those that trade in the highest volumes around the
world—with the buy side on TrueEx. Instead, they limited such trading to “bespoke, one-off
swaps, which trade in significantly smaller volumes.” There is no legitimate explanation for this
behavior, plaintiffs claim, given that TrueEx provides “better technology with lower fees.”
Plaintiffs claim that the Dealers’ conduct successfully neutralized TrueEx “from bringing an allto-all platform to market and becoming a competitive threat.” SAC ¶¶ 282–87; JTSAC ¶¶ 223–
25.
I.

The Dealers’ Insistence on “Name Give-Up”

Plaintiffs allege that—both before and after Tera and Javelin’s platforms emerged—the
Dealer Defendants insisted on maintaining a historical practice they called “name give-up,” in
which the names of each party to an IRS are identified to the other. The compulsory disclosure
of swap counterparties, plaintiffs claim, serves as a policing mechanism, allowing the Dealers to
retaliate against entities that attempt to trade on all-to-all platforms. In fact, buy-side investors
prefer not to disclose their trading strategies and needs. Although “name give-up” served a

32

purpose in an earlier era, as identification of the counterparty enabled assessment of its
creditworthiness, now that IRS trades are centrally cleared, there is no need for the
counterparty’s name to be disclosed. SAC ¶¶ 292–95 (quoting commentators as to lack of
justification today for “name give-up”); JTSAC ¶¶ 234–41 (same).
Plaintiffs allege that the Dealers enforce “name give-up” by various means. Dealer
officials—including at Goldman (Levy), Deutsche Bank (Wolf) and Barclays and Morgan
Stanley (Senft)—conditioned giving liquidity to a platform on its using “name give-up.” Other
Dealers joined this practice. Dealers also boycotted platforms, including Tera’s and Javelin’s,
that did not require “name give-up.” Third, the Dealers’ IDBs, at their insistence, use a service
called MarkitWire, a trade processing service, to deliver trades to clearinghouses for execution.
MarkitWire is operated by MarkitSERV, controlled by former Dealer officials. Before a trade
clears, MarkitWire discloses counterparties’ names to each other and gives each an opportunity
to terminate the transaction. Due to “collective pressure” from the Dealers, numerous SEFs—
including the largest IDB SEFs, seven of which plaintiffs name—maintain “name give-up,”
“thereby effectively preventing buy-side customers from trading on them.” The Dealers have
also pulled, and threatened to pull, liquidity from platforms that allow anonymous trading other
than among Dealers; one interdealer SEF received heated phone calls from executives at Credit
Suisse and J.P. Morgan over the prospect of introducing trade anonymity. Dealers have also
threatened buy-side customers with the “penalty box” if they attempt to trade on an all-to-all
trading platform, whether operated by an IDB or an independent SEF like Tera or Javelin.
Plaintiffs cite data indicating that platforms that do not use “name give-up” see virtually no IRS
activity. SAC ¶¶ 296–311; JTSAC ¶¶ 242–56.
J.

Impact of the Boycott

33

Plaintiffs claim that the boycott of Tera, Javelin, and TrueEx communicated that any
platform giving the buy-side access to anonymous all-to-all IRS trading would be “collectively
punished and strangled until it failed.” This “enforce[d] market discipline” and deterred others
from opening such platforms. Similarly, plaintiffs allege, the Dealers deterred customers from
using such platforms, by withholding clearing services from those who trade on them. SAC
¶¶ 288–91; JTSAC ¶¶ 226–33.
Plaintiffs claim that the boycott has resulted in the Dealers’ controlling 70% or more of
the IRS market and preventing structural changes in that market. As a result, investors pay more
money for IRSs, to which no other financial instrument is fully comparable. Plaintiffs allege that
a comparison of data with fixed income and equities products sold on exchanges reveals the
price benefits of such transparent all-to-all trading. SAC ¶¶ 312–47; JTSAC ¶¶ 267–72.
II.

Procedural History
The initial complaint in this action was filed on November 25, 2015. Dkt. 1 in No. 15

Civ. 9139. On June 2, 2016, the United States Judicial Panel on Multidistrict Litigation
(“JPML”) transferred all related cases to this Court for coordinated or consolidated pretrial
proceedings with the actions pending in this District. Dkt. 1.10 The cases encompassed by the
JPML’s order include cases brought on behalf of a putative class of IRS investors, and the
separate case brought by Javelin and Tera. On July 26, 2017, the Court held an initial
conference, and stayed formal discovery. Dkt. 10 (Order No. 3). On August 3, 2016, the Court
appointed Quinn Emanuel Urquhart & Sullivan, LLP, and Cohen Milstein Sellers & Toll, PLLC,

10

Unless otherwise specified, docket references herein refer to the docket in No. 16-MC-2704,
the “master case” docket which has been established for filings of significance in this case, Dkt.
2, which is intended to encompass motions to dismiss and associated filings. The Court cites to
No. 16-MD-2704 for filings not made in the master docket.
34

as interim co-lead counsel for the putative class. Dkt. 12 (Order No. 4). On September 9, 2016,
class plaintiffs, Dkt. 13, and the Javelin/Tera plaintiffs, Dkt. 14, each filed an Amended
Complaint. On November 4, 2016, defendants filed motions to dismiss. Dkts. 15–20; Dkts.
123–25, 127–29, 131–34 in No. 16-MD-2704. On December 9, 2016, class plaintiffs, Dkt. 23
(“SAC”), and the Javelin/Tera plaintiffs, Dkt. 145 in No. 16-MD-2704 (“JTSAC”), filed Second
Amended Complaints. On January 20, 2017, defendants moved to dismiss, through separate
motions filed by the Dealer Defendants, Dkts. 24–26, ICAP, Dkts. 28–30, HSBC, Dkt. 40 & Dkt.
163 in No. 16-MD-0274, and Tradeweb, Dkts. 169–71 in No. 16-MD-2704. On February 17,
2017, the class, Dkt. 32, and Javelin/Tera, plaintiffs, Dkt. 33, filed opposition briefs. On March
24, 2017, defendants filed reply briefs. Dkts. 36–40, & Dkts. 206–08, 210, 212 in No. 16-MD2704. On May 23, 2017, the Court heard argument on the motions. See Transcript, Dkt. 233 in
No. 16-MD-2704 (“Tr.”).
III.

The Motions to Dismiss: Overview and Standards
The Dealer Defendants move to dismiss plaintiffs’ Sherman Act § 1 claims on five

grounds: that (1) the SACs do not state a plausible claim of an antitrust conspiracy; (2) the class
plaintiffs lack antitrust standing; (3) the Commodities Exchange Act (CEA) and Dodd-Frank
impliedly preclude plaintiffs’ post-Dodd-Frank claims; (4) the SAC’s pre-2012 claims are timebarred; and (5) the SAC’s pre-2013 claims do not allege an injury-in-fact. Five individual
defendants—three Dealer Defendants (BNPP, HSBC, and UBS), and the two non-Dealer
Defendants (Tradeweb and ICAP)—move to dismiss for failure to state a § 1 claim for reasons
particular to them.11 Defendants also move to dismiss plaintiffs’ state-law claims.

11

BNPP and UBS did not file separate briefs but at argument pursued dismissal on individual
grounds.
35

To survive a motion to dismiss under Rule 12(b)(6), a complaint must plead “enough
facts to state a claim to relief that is plausible on its face.” Bell Atl. Corp. v. Twombly, 550 U.S.
544, 570 (2007). A claim only has “facial plausibility when the plaintiff pleads factual content
that allows the court to draw the reasonable inference that the defendant is liable for the
misconduct alleged.” Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009). A complaint is properly
dismissed where, as a matter of law, “the allegations in a complaint, however true, could not
raise a claim of entitlement to relief.” Twombly, 550 U.S. at 558. For the purpose of resolving
the motion to dismiss, the Court must assume all well-pled facts to be true, drawing all
reasonable inferences in favor of the plaintiff. Koch, 699 F.3d at 145. However, that tenet “is
inapplicable to legal conclusions.” Iqbal, 556 U.S. at 678. A pleading that offers only “labels
and conclusions” or “a formulaic recitation of the elements of a cause of action will not do.”
Twombly, 550 U.S. at 555.
There is no heightened pleading standard in antitrust cases. Concord Assocs., L.P. v.
Entm’t Props. Tr., 817 F.3d 46, 52 (2d Cir. 2016). Rather, at the pleading stage, plaintiffs need
only “raise a reasonable expectation that discovery will reveal evidence of illegality.” Mayor
and City Council of Baltimore v. Citigroup, Inc., 709 F.3d 129, 135 (2d Cir. 2013) (“Citigroup”).
IV.

Plaintiffs’ Sherman Act § 1 Claim
A.

Applicable Legal Principles

The Sherman Act bans “[e]very contract, combination in the form of trust or otherwise,
or conspiracy, in restraint of trade or commerce among the several States[.]” 15 U.S.C. § 1.
“The crucial question in a Section 1 case is therefore whether the challenged conduct ‘stems
from independent decision or from an agreement, tacit or express.’” Starr v. Sony BMG Music

36

Ent’m’t, 592 F.3d 314, 321 (2d Cir. 2010) (quoting Theatre Enters., Inc. v. Paramount Film
Distrib. Corp., 346 U.S. 537, 540 (1954) (alterations omitted)).
To plausibly allege a Sherman Act § 1 conspiracy, a complaint must allege “enough
factual matter (taken as true) to suggest that an [illegal] agreement was made,” that is, “enough
fact to raise a reasonable expectation that discovery will reveal evidence of illegal agreement.”
Twombly, 550 U.S. at 556. “The ultimate existence of an ‘agreement’ under antitrust law . . . is a
legal conclusion, not a factual allegation.” Citigroup, 709 F.3d at 135–36 (citing Starr, 592 F.3d
at 319 n.2 (“The allegation that defendants agreed to [a] price floor is obviously conclusory, and
is not accepted as true.”)).
As the Second Circuit has recognized, there are two ways, at the pleading stage, for a
plaintiff “to allege enough facts to support the inference that a conspiracy actually existed” so as
to overcome a motion to dismiss. Citigroup, 709 F.3d at 136.
First, a plaintiff may allege “direct evidence that the defendants entered into an
agreement in violation of the antitrust laws.” Id. (citing In re Ins. Brokerage Antitrust Litig., 618
F.3d 300, 323–24 (3d Cir. 2010)); see also Burtch v. Milberg Factors, Inc., 662 F.3d 212, 225
(3d Cir. 2011) (direct evidence is “evidence that is explicit and requires no inferences to establish
the proposition or conclusion being asserted”). “Such evidence would consist, for example, of a
recorded phone call in which two competitors agreed to fix prices at a certain level.” Citigroup,
709 F.3d at 136.
Direct evidence, however, is not required. And concrete “smoking gun” evidence of an
illegal conspiracy between sophisticated actors “can be hard to come by, especially at the
pleading stage.” Citigroup, 709 F.3d at 136; see also United States v. Snow, 462 F.3d 55, 68 (2d

37

Cir. 2006) (“[C]onspiracy by its very nature is a secretive operation, and it is a rare case where
all aspects of a conspiracy can be laid bare in court . . . with precision.”).
As a result, as an alternative to direct evidence, to prove a conspiracy a plaintiff may rely
on indirect or circumstantial evidence, that is, “inferences that may fairly be drawn from the
behavior of the alleged conspirators.” Anderson News, LLC v. American Media, Inc., 680 F.3d
162, 183 (2d Cir. 2012) (quoting Michelman v. Clark-Schwebel Fiber Glass Corp., 534 F.2d
1036, 1043 (2d Cir. 1976)); see also In re Elec. Books Antitrust Litig., 859 F. Supp. 2d 671, 681
(S.D.N.Y. 2012) (“eBooks”) (same).
A horizontal agreement among competitors, the sort of pact alleged here, is commonly
based on claims of parallel conduct by the alleged co-conspirators. However, as the Supreme
Court held in Twombly, “alleging parallel conduct alone is insufficient, even at the pleading
stage,” to survive a motion to dismiss. Citigroup, 709 F.3d at 136. Rather, to plausibly allege a
§ 1 violation, the parallel conduct must be “placed in a context that raises a suggestion of a
preceding agreement, not merely parallel conduct that could just as well be independent action.”
Twombly, 550 U.S. at 557.
Twombly illustrates these principles. After deregulation, regional telephone companies
had, in parallel, elected not to enter incumbent carriers’ local telephone and high-speed internet
markets. From this, the complaint inferred an agreement among these carriers to allocate
markets. Upholding dismissal, the Supreme Court held that the competitors’ parallel conduct, as
alleged, did not “render a § 1 conspiracy plausible.” 550 U.S. at 556. Although such conduct
was consistent with the existence of an agreement, it was also “just as much in line with a wide
swath of rational and competitive business strategy unilaterally prompted by common
perceptions of the market.” Id. at 554. Parallel conduct by competitors that would not state a § 1

38

claim, the Court observed, can be the result of “coincidence, independent responses to common
stimuli, or mere interdependence unaided by an advance understanding between the parties.” Id.
at 556 n.4 (quotation omitted). And there were sound reasons for each carrier independently to
have refrained from entering an incumbent’s market. Id. at 564–69. The Court cautioned: “Even
‘conscious parallelism,’ a common reaction of ‘firms in a concentrated market that recognize
their shared economic interests and their interdependence with respect to price and output
decisions,’ is ‘not in itself unlawful.’” Id. at 553–54 (quoting Brooke Grp. Ltd. v. Brown &
Williamson Tobacco Corp., 509 U.S. 209, 227 (1993)) (alterations omitted).
Following Twombly, for a complaint to state a § 1 claim based on parallel conduct by
competitors, it must plead facts sufficient to indicate that this conduct “flowed from a preceding
agreement rather than from [defendants’] own business priorities.” Citigroup, 709 F.3d at 137–
38; see also Twombly, 550 U.S. at 557 (“A statement of parallel conduct, even conduct
consciously undertaken, needs some setting suggesting the agreement necessary to make out a
§ 1 claim[, some] further circumstance pointing to the meeting of the minds[.]”). An inference
of conspiracy will not arise when the conspirators’ parallel conduct “made perfect business
sense,” Citigroup, 709 F.3d at 138, “there are obvious alternative explanations for the facts
alleged,” In re Ins. Brokerage Antitrust Litig., 618 F.3d at 322–23 (quotation and alterations
omitted), or the alleged facts “suggest competition at least as plausibly as [they] suggest
anticompetitive conspiracy,” In re Elevator Antitrust Litig., 502 F.3d 47, 51 (2d Cir. 2007).
However, the requirement of “‘plausible grounds to infer an agreement does not impose a
probability requirement at the pleading stage; it simply calls for enough fact to raise a
reasonable expectation that discovery will reveal evidence of illegal agreement,’” Anderson
News, 680 F.3d at 184 (quoting Twombly, 550 U.S. at 556) (emphasis added by Anderson News),

39

and “there may . . . be . . . more than one plausible interpretation of a defendant’s words, gestures
or conduct.” Id. at 189–90.
Post-Twombly courts have analyzed § 1 claims based on parallel conduct by horizontal
competitors by inquiring whether “plus factors” and/or other circumstantial evidence are present
that, along with the parallel conduct, make it plausible to infer an agreement among competitors.
See, e.g., Citigroup, 709 F.3d at 137; Gelboim v. Bank of America, Corp., 823 F.3d 759, 781 (2d
Cir. 2016). These factors describe “circumstances under which . . . the inference of rational
independent choice [is] less attractive than that of concerted action.” In re Ins. Brokerage
Antitrust Litig., 618 F.3d at 323 (citation omitted)). The Second Circuit has identified three
“plus factors” as ones that may support a plausible inference of conspiracy: “(1) ‘a common
motive to conspire’; (2) ‘evidence that shows that the parallel acts were against the apparent
individual economic self-interest of the alleged conspirators’; and (3) ‘evidence of a high level of
interfirm communications.’” Gelboim, 823 F.3d at 781 (quoting Citigroup, 709 F.3d at 136).
The Third Circuit has also identified as relevant circumstantial evidence facts “implying a
traditional conspiracy,” meaning “non-economic evidence ‘that there was an actual, manifest
agreement not to compete,’ which may include ‘proof that the defendants got together and
exchanged assurances of common action or otherwise adopted a common plan even though no
meetings, conversations, or exchanged documents are shown.’” In re Ins. Brokerage Antitrust
Litig., 618 F.3d at 322 (citations omitted).
B.

Discussion

The SAC alleges a nine-year—from 2007 to 2016—Sherman Act § 1 conspiracy among
11 Dealer Defendants (plus affiliates) and two other entities. The JTSAC, although focused on
the last four years of this period, alleges the same. Defendants argue that the Complaints do not

40

plausibly allege a conspiracy. They argue that, once conclusory allegations are stripped away,
the conduct pled largely consists of acts consistent with each Dealer’s independent self-interest.
And, they argue, there is insufficient circumstantial evidence or “plus factors” to suggest an
illegal agreement.
In assessing the motions to dismiss, it is useful to divide the period covered by the § 1
claim into two parts: 2007–2012 and 2013–2016 (ending December 9, 2016, when the SACs
were filed). That is because the structure and nature of the IRS market changed materially in or
around 2013, as a result of Dodd-Frank’s mandates; and because plaintiffs’ claims as to the
nature of the conspiracy, and the supporting allegations, differ in the two periods. During 2007–
2012, platforms offering anonymous all-to-all IRS trading accessible to the buy-side had not yet
developed; class plaintiffs’ claim is that the defendants conspired to inhibit such platforms from
coming into existence. During 2013–2016, the infrastructure for such platforms was in place and
several had emerged; plaintiffs’ claim is that the defendants conspired to quash these nascent
platforms, including those of Javelin and Tera.
1.

2007–2012

While plaintiffs’ theory over the entire class period is that the Dealers engaged in parallel
acts in circumstances suggesting collusion and not independent pursuit of self-interest, the SAC
is not explicit whether, as to 2007–2012, its claim of conspiracy is based on parallel acts, direct
proof of agreement, or a combination. The Court therefore examines this period from each
perspective to test whether a horizontal conspiracy has been plausibly pled.
a.

Allegations of Parallel Conduct

41

The SAC’s main claim of parallel conduct during 2007–2012 is of parallel inaction: The
Dealers are each alleged to have disfavored as a matter of self-interest, and so not to have taken
steps to support, the emergence of platforms enabling all-to-all exchange-based IRS trading.
Otherwise, even treating generalized claims as cognizable, the SAC alleges limited
parallel conduct during this period. It alleges that the Dealers: (1) generally threatened to deny
liquidity (trading volume) to IDBs that invited the buy-side to trade on their platforms, see SAC
¶¶ 160–6112; (2) generally insisted on clearing IRS trades only through a Dealer-controlled
platform, SwapClear, rather than pivoting business to a new product and platform, “CME
Cleared Swaps,” that had longer-term potential to extend central clearing of IRS trades to the
buy-side, see SAC ¶¶ 184–8813; and (3) continued to practice “name give-up,” under which the
buy-side customer was identified to the counterparty, see, e.g., SAC ¶¶ 292–94.
These shards of parallel conduct do not give rise to an inference of an agreement to block
all-to-all trading. As in Twombly, the pleadings here supply good reason, as a matter of “rational
and competitive business strategy,” 550 U.S. at 554, for any individual Dealer independently to
have sought to maintain the status quo and to discourage, not facilitate, all-to-all IRS exchange

12

The SAC alleges one instance of a threat, during 2007–2012, to threaten to punish an IDB
platform for opening access to the buy side. In 2009, GFI Group (“GFI”), an operator of leading
IDB platforms, “attempted to implement anonymous trading protocols on its platform,” but this
bid was “immediately shut down by the Dealer Defendants,” who “responded in unison with
threats to pull their business off the platform,” leading GFI to “reverse course.” SAC ¶ 161. The
SAC does not elaborate on this claim. It does not name any Dealer or specify any threat. Id.
13

The SAC alleges that in 2008, the CME introduced plans to introduce a clearing product called
“CME Cleared Swaps” to all OTC swaps (including for IRSs) executed on its trading platform,
Swapstream. SAC ¶¶ 184–85. It alleges that “the Dealer Defendants viewed this as a serious
threat to lead to central clearing because CME had built successful clearinghouses and exchanges
for other asset classes.” Id. ¶ 186–88. The SAC alleges that “[t]he Dealer Defendants responded
to this threat by boycotting Swapstream” and “agreed to clear only interdealer IRS trades and
only on SwapClear (the entity they controlled),” and that as a result, “CME’s clearing solution
for IRS swiftly failed.” Id. (emphases in original).
42

trading platforms from taking root. The SAC’s theory, in fact, is that such platforms presented
an existential threat to the Dealers’ profit margins as market makers. See, e.g., SAC ¶ 28 (the
Dealers “‘want[ed] desperately to preserve the status quo’ of the OTC market” and to “stop[]
meaningful development of the market”); see also SAC ¶¶ 9, 74; JTSAC ¶ 6. It follows that
each Dealer had good reason to independently discourage (e.g., in its dealings with IDBs) and
not encourage (e.g., in not using a new clearing product) development of a new trading paradigm
that threatened, some day, to cannibalize their trading profits. “[C]ommon economic experience,
[and] the facts alleged in the complaint itself, [thus] show that independent self-interest is an
obvious alternative explanation for defendants’ common behavior.” In re Ins. Brokerage
Antitrust Litig., 618 F.3d at 326; see also id. at 349 (where defendants are “reaping enormous
profits,” it is natural for them to have “no desire to upset the apple cart”) (quotation omitted);
Twombly, 550 U.S. at 568 (no inference of conspiracy where each defendant “liked the world the
way it was” and was “sitting tight, expecting [its] neighbors to do the same thing”).
In three respects, the inference of collusion from the parallels pled, in fact, is far less
plausible than in Twombly.
First, the parallel activity in Twombly involved the ultimate act of refraining to compete:
The regional telephone carriers were alleged to have refrained from entering one another’s
markets as competitors. In contrast, the parallel behavior pled here implicates narrower business
practices (e.g., whether to use a new clearing product), not the ultimate decision whether to
compete or refrain.
Second, in Twombly, the decisions by the phone companies to forego competing in each
other’s markets were sharply pled. In contrast, here, as noted, plaintiffs’ claims here of parallel
practices during 2007–2012 are largely pled generally and collectively.

43

Third, in Twombly, the alleged agreement involved foregoing a live opportunity to
compete. In contrast, the alternative trading environment to which class plaintiffs claim to have
aspired not only did not exist in 2007–2012—based on the pleadings, the critical infrastructure
necessary for it to take root was not yet in place. The SACs acknowledge that central clearing
did not develop until 2013, when Dodd-Frank’s mandates made it compulsory for most IRS
trading. And without it, anonymous all-to-all exchange trading was impossible: For any trade, a
Dealer had to face its counterparty, so as to assess creditworthiness, manage risk, and consult any
ISDA that set the background terms for trades between the parties. See, e.g., SAC ¶¶ 71, 91, 95;
JTSAC ¶¶ 68, 88, 92. The advent of central clearing in 2013, however, eliminated the need for
these trade-specific inquiries, and thus enabled anonymous exchange trading: The clearinghouse
became each side’s counterparty; assured creditworthiness by requiring participants to have
posted advance collateral; and assured background terms and conditions for trades.14 But in the
preceding years when central clearing—the internal combustion engine of anonymous exchange
trading—was as-yet undeveloped, it is simply less plausible to infer a collusive agreement to
block such trading from the Dealers’ limited parallel actions. Before Dodd-Frank made this vital
infrastructure a looming reality, there would have no urgency for collective action to block allto-all exchange trade from emerging.15
b.

Alleged Direct Evidence of Conspiracy: Project Fusion

14

See, e.g., SAC ¶¶ 96, 212 (acknowledging that collateral must be posted to secure cleared
trades). The CFTC has recognized the substantial costs associated with payment of clearing fees
and posting of collateral to secure cleared trades. See Clearing Requirements Rule at 74,323.

15

Given the lack of central clearing, the Dealers’ common use of “name give-up” before 2013 is
an unpersuasive basis from which to infer collusion: Before central clearing, disclosure of the
buy-side name was a necessary incident to the trading process—a means of determining a
counterparty’s creditworthiness—as plaintiffs acknowledge. SAC ¶ 292.
44

Presumably given the sparse parallel acts in this period, the SAC makes the centerpiece
of its § 1 claims during 2007–2012 an episode of alleged actual collaboration. It involves
“Project Fusion,” in which most Dealers, in 2007, acquired a controlling stake in Tradeweb.
Plaintiffs claim that Project Fusion is direct evidence of a conspiracy.
As noted, the SAC alleges that the Dealers, which had founded Tradeweb in 1998,
repurchased it from Thomson in late 2007. Tradeweb had developed a dealer-to-client RFQ
platform in which buy-side customers traded consistent with the day’s norm: They received and
chose among non-binding quotes from Dealers. The SAC alleges that the Dealers feared that
Tradeweb might expand to introduce all-to-all trading to the IRS market. Therefore, the SAC
alleges, the Dealers devised a scheme, “Project Fusion,” using their collective leverage over
Thomson, to repurchase and take control of Tradeweb, so as to assure that its platform would
remain an RFQ platform and not be used for all-to-all trading. The SAC also alleges that the
Dealers, in the 2007 press release announcing their investment, concealed their majority control
of Tradeweb and their aim of steering the Tradeweb platform away from all-to-all trading.
In one obvious respect, the SAC’s allegations as to “Project Fusion” bolster the § 1 claim.
These allegations establish interfirm communications, and on the same subject area (IRS trading
platforms) as the alleged conspiracy. The SAC alleges collaboration among Dealers in acquiring
control of Tradeweb. See SAC ¶¶ 112–14. And it alleges in detail the longtime membership on
Tradeweb’s board of many Dealers, and that affiliates of Dealers were Tradeweb officers. See
SAC ¶¶ 126–42. Had there been meaningful parallel conduct by the competitors during this
period, these inter-firm communications would have been a “plus factor” supporting the
inference that this conduct resulted from an agreement.

45

As purported direct evidence of a § 1 conspiracy, however, the SAC’s pleadings as to
Project Fusion fall short, for two independent reasons.
First, the SAC’s factual allegations of an agreement to terminate a plan by Tradeweb to
open an all-to-all electronic platform for IRS trading are at best inferential, rather than explicit.
They are a far cry from the illustration of direct evidence of a § 1 conspiracy—a recorded phone
call in which competitors agreed to fix a price—that the Second Circuit has given. See
Citigroup, 709 F.3d at 136; see also Burtch, 662 F.3d at 225 (to qualify as direct, evidence must
be so explicit as to “require[] no inferences to establish the proposition or conclusion being
asserted;” plaintiff’s allegations, however, failed to “specify a time or place that any actual
agreement to fix credit terms occurred, nor do they indicate that any particular individuals . . .
made such an agreement”); In re Ins. Brokerage Antitrust Litig., 618 F.3d at 324 n.23 (“a
document or conversation explicitly manifesting the existence of the agreement in question” is
an example of direct evidence).
The SAC lacks any such allegations as to Project Fusion. It does not allege, for example,
a communication on a date in 2007 in which distinct persons agreed to put on ice Tradeweb’s
plan for an all-to-all IRS exchange. On the contrary, the SAC does not cite any evidence
supporting its critical background premise—that Tradeweb ever had such a plan. The SAC’s
claim that Tradeweb was “planning” in 2007 to introduce such a platform, which “plan” the
Dealers then sought to subvert, SAC ¶¶ 16, 111, is stated as a conclusion. It is not supported by
well-pled facts. Nor does the SAC allege anything specific that Tradeweb did—between
becoming majority-owned by the Dealers in 2007 and opening up a SEF for all-to-all IRS trading
in 2013 alongside Tera, Javelin, TrueEx and Bloomberg—that furthered the alleged conspiracy.

46

Far from qualifying as direct evidence of a § 1 conspiracy, the SAC’s lengthy allegations
as to Project Fusion largely consist of conclusory allegations and inferences. These include as to
why the Dealers invested alongside Thomson in Tradeweb, a financial technology company that
builds platforms for fixed-income and derivative products; why, in the transaction, the
reconstituted Tradeweb entities were given certain names; what defendants’ intentions were in
formulating the press release about the 2007 investment; and why the Dealers in 2007—a year
before the financial crisis and three years before Dodd-Frank mandated eventual central
clearing—viewed Tradeweb as a serious threat to morph into a platform for all-to-all trading.
See generally SAC ¶¶ 100–50.
Second, even assuming a well-pled agreement among Tradeweb’s Dealer owners to
terminate a plan to open an all-to-all IRS trading platform, the SAC does not plead facts under
which that agreement would be unlawful. A plaintiff alleging a § 1 violation may allege either a
per se unlawful agreement or one that is illegal under the “rule of reason.” Per se liability is
limited to “agreements whose nature and necessary effect are so plainly anticompetitive that no
elaborate study of the industry is needed to establish their illegality[.]” Nat’l Soc’y of Prof’l
Eng’rs v. United States, 435 U.S. 679, 692 (1978). As to these categories of restraints,
experience has enabled courts to “predict with confidence that the rule of reason will condemn
it,” and therefore permits to apply “a conclusive presumption that the restraint is unreasonable.”
Arizona v. Maricopa Cty. Med. Soc’y, 457 U.S. 332, 344 & n.15 (1982); see also Anderson
News, 680 F.3d at 182–83. “Paradigmatic examples are horizontal agreements among
competitors to fix prices or to divide markets.” Leegin Leather Prods., Inc. v. PSKS, Inc., 551
U.S. 877, 886 (2007) (quotation omitted); see also Klor’s Inc. v. Broadway-Hale Stores, Inc.,
359 U.S. 207, 211–12 (1959) (group boycotts are per se violations of the Sherman Act); Capital

47

Imaging Assocs., P.C. v. Mohawk Valley Med. Assocs., Inc., 996 F.2d 537, 542–43 (2d Cir.
1993). Agreements that do not fall under per se illegality are analyzed under the rule of reason
to determine whether they are an unreasonable restraint of trade. Under the rule of reason
analysis, “the plaintiff bears the burden of showing that the alleged [agreement] produced an
adverse, anti-competitive effect within a relevant geographic market.” In re Ins. Brokerage
Antitrust Litig., 618 F.3d at 315 (quotation omitted). Satisfying this burden includes a
demonstration of defendants’ market power. Id.; see also Concord Assocs., 817 F.3d at 52–53.
The SAC’s allegations as to Project Fusion do not fit into any category of agreement
recognized as per se illegal. It alleges a decision among participating Dealer Defendants16 as to
the strategic direction of a single financial technology company which they majority-owned
pursuant to a joint venture. But, viewing the operation of a legitimate joint venture as akin to
that of a single firm, modern antitrust law evaluates such joint conduct—including the creation
of the joint venture itself, its business focus, its product selection, and its pricing—under the rule
of reason, with the pleading requirements that that standard imposes. See, e.g., Texaco, Inc. v.
Dagher, 547 U.S. 1, 1 n.1 & 6–7 (2006) (“the pricing decisions of a legitimate joint venture do
not fall within the narrow category of activity that is per se unlawful under § 1”; “[a]s a single
entity, a joint venture, like any other firm, must have the discretion” to make decisions regarding
the conduct of the venture); see also American Needle, Inc. v. National Football League, 560
U.S. 185, 195 (2010); Major League Baseball Properties, Inc. v. Salvino, 542 F.3d 290, 316–18
(2008) (“MLB Properties”); cf. Copperweld Corp. v. Indep. Tube Corp., 467 U.S. 752, 772 n.18
(1984).

16

The SAC does not allege that two Dealer Defendants—BNPP and HSBC—participated in
Project Fusion or invested in Tradeweb. SAC ¶ 15.
48

Plaintiffs do not cite any case in which a decision by competitors to invest in or acquire
control over a business, or to direct the activities of the business, in the context of a legitimate
joint venture, has been evaluated under the per se standard. And the SAC does not plead facts
that remove the Project Fusion joint venture from this body of case law. It does not plead facts
indicating that, after its creation, Tradeweb, the subject of the joint venture, became a horizontal
competitor of the Dealer Defendants which thereafter conspired with them. And Tradeweb was
not such a competitor: As alleged, it was a provider of electronic trading platforms, not a market
maker. And the SAC does not adequately plead that the Project Fusion joint venture was an
illegitimate shell that offered no efficiency enhancements and served only to mask concerted
conduct. See, e.g., American Needle, 560 U.S. at 200 (“Agreements made within a firm can
constitute concerted action covered by § 1 when the parties to the agreement act on interests
separate from those of the firm itself, and the intrafirm agreements may simply be a formalistic
shell for ongoing concerted action.”). The SAC’s allegations as to Tradeweb—including as to
non-defendant Thomson’s minority-ownership of it, the Dealers’ infusion of $280 million in it,
and its existence and operation from 2007 forward—would not permit such an inference.17
That leaves the rule of reason. The SAC, however, fails to plead facts sufficient to
support the conclusion that, evaluated under rule-of-reason methodology, the Project Fusion joint

17

The SAC does not allege that Tradeweb was an unattractive investment opportunity, such that
only one with conspiratorial aims would have invested in it. Supporting the inference of
legitimacy, the SAC alleges that (1) the Dealers invested $280 million in Tradeweb; (2) they
owned Tradeweb alongside existing owner Thomson; (3) Tradeweb, post-acquisition, was
organized consistent with customary corporate formalities; (4) Tradeweb proceeded to expand
other aspects of its trading platform; and (5) its decision not offer an all-to-all IRS trading
platform until 2013 put it on par with all other trading platforms of the day, including nondefendants Tera, Javelin, TrueEx, and Bloomberg. See generally SAC ¶¶ 115–25, 142–50; see
also Twombly, 550 U.S. at 559 (declining to draw inference of conspiracy where company
decides not to “enter[] every market that an outside observer might regard as profitable.”).
49

venture—the centerpiece of plaintiffs’ pre-2013 § 1 claim—represented an unreasonable restraint
of trade. Under rule-of-reason analysis, “[t]he true test of legality is whether the restraint
imposed is such as merely regulates and perhaps thereby promotes competition or whether it is
such as may suppress or even destroy competition.” Chicago Bd. of Trade v. United States, 246
U.S. 231, 238 (1918); see also Nat’l Soc’y of Prof’l Eng’rs, 435 U.S. at 691 (rule-of-reason
analysis is of “whether the challenged agreement is one that promotes competition or one that
suppresses competition”). The factfinder applies this analysis to the restraint “under all the
circumstances of the case,” Maricopa Cty. Med. Soc’y, 457 U.S. at 343, and “must ordinarily
consider the facts peculiar to the business to which the restraint is applied, its condition before
and after the restraint was imposed; the nature of the restraint and its effect, actual or probable,”
as the “history of the restraint, the evil believed to exist, the reason for adopting the particular
remedy, [and] the purpose of the end sought to be attained, are all relevant facts,” Chicago Bd. of
Trade, 246 U.S. at 238.18

18

A burden-shifting framework drives the rule-of-reason analysis:
[T]he plaintiffs bear an initial burden to demonstrate the defendants’ challenged
behavior had an actual adverse effect on competition as a whole in the relevant
market. . . . Because the antitrust laws protect competition as a whole, evidence
that plaintiffs have been harmed as individual competitors will not suffice. . . . If
the plaintiffs satisfy their initial burden, the burden shifts to the defendants to offer
evidence of the pro-competitive effects of their agreement. . . . Assuming
defendants can provide such proof, the burden shifts back to the plaintiffs to prove
that any legitimate competitive benefits offered by defendants could have been
achieved through less restrictive means. . . . Ultimately, the factfinder must engage
in a careful weighing of the competitive effects of the agreement—both pro and
con—to determine if the effects of the challenged restraint tend to promote or
destroy competition.

MLB Properties, 542 F.3d at 317 (quoting Geneva Pharm. Technology Corp. v. Barr Labs. Inc.,
386 F.3d 485, 506–07 (2d Cir. 2004)).
50

Here, however, there are no allegations in the SAC defining Tradeweb’s product or
geographic market, or, within that market, defining its market share or market power. There is,
in fact, no allegation that Tradeweb had any presence, let alone power, in any market. And,
vitally important, there are no allegations as to the pro-competitive benefits and anti-competitive
harms of Tradeweb after the joint venture, whether in general or as to Tradeweb’s specific
choices after 2007 as to which trading platforms and asset classes to pursue and which to forego.
Therefore, even assuming it adequately pled an agreement among Dealers to terminate a
Tradeweb plan to open an all-to-all trading platform, the SAC does not plead facts supporting,
under the rule of reason, the inference that that agreement was anti-competitive so as to violate
§ 1.19
In a final allegation regarding Project Fusion, the SAC claims that the Dealers and
Tradeweb concealed the Dealers’ majority interest in the entity (Tradeweb NewMarkets) that

19

With respect to the entire nine-year period covered by the SAC, plaintiffs appear to pursue a
per se theory of § 1 liability, see, e.g., Class Mem. at 5, although the SAC, in its charging
language, states summarily that defendants’ alleged agreement was unlawful per se and under
the rule of reason. SAC ¶ 396. As to the years 2013–2016, the allegation of a horizontal boycott
of three new trading platforms describes a recognized type of per se illegal conduct. But for
2007–2012, the SAC, whether as to Project Fusion or in its limited other allegations, does not
allege this type—or any other—of per se illegal conduct. In 2007, there was no all-to-all IRS
platform to boycott. There would not be such a platform for another six years. (Revealingly,
when questioned at argument as to what the Dealers were purportedly boycotting before 2013,
plaintiffs’ counsel replied: “The buy side.” Tr. 72–73.)
At argument, plaintiffs separately argued that Project Fusion might be subject to condemnation
under an intermediate “quick look” analysis. Tr. 65. The limited facts pled regarding
Tradeweb’s strategic direction following the Dealers’ acquisition of control, however, make it
impossible, even on a quick look, to balance the pro-competitive benefits and anti-competitive
harms, of that course. The Court is therefore unable to so condemn Project Fusion. See MLB
Properties, 542 F.3d at 317 (rejecting “quick-look” analysis as inappropriate for challenge to
joint venture marketing activity where arrangement “might plausibly be thought to have a net
pro-competitive effect, or possible no effect at all on competition”); see also In re Insurance
Brokerage Antitrust Litig., 616 F.3d at 318 (rejecting “quick look” argument).
51

housed Tradeweb’s IRS business. SAC ¶ 120. That claim proves inaccurate. As defendants
point out, securities filings of Thomson, Tradeweb’s parent, disclosed in November 2007 that the
Dealers who invested in Project Fusion had obtained (1) a 15% interest in the entity responsible
for Tradeweb’s “established markets” (Tradeweb Markets LLC) and (2) an 80% interest in a
separate entity that would pursue asset “asset class expansion” in additional markets (Tradeweb
NewMarkets LLC). See Thomson Corp. Form 6-K (Nov. 9, 2007) (quoted in Dealer Rep. Br. 17
& n.11); see also discussion infra, § V. In any event, even if well-pled, the Dealers’
concealment of their majority stake would not itself establish an agreement in violation of § 1. It
would instead be circumstantial evidence bearing on such a claim.
c.

Other Allegations Relating to 2007-2012

The Court next considers plaintiffs’ other allegations during 2007–2012 bearing on the
claim of a conspiracy to violate § 1.
OTCDerivNet: The SAC alleges that, in October 2000, eight Dealers formed a new entity
called OTCDerivNet; other named Dealers joined in 2001 and 2009. The Dealers’ goal in
forming the new entity, the SAC alleges, was to use it to secure control of SwapClear, an IRS
clearinghouse; soon after 2000, the Dealers did so. SAC ¶¶ 177–80. During 2007–2012, the
SAC alleges, the Dealer-affiliated board members of OTCDerivNet discussed how to “prevent or
delay the buy side’s ability to clear IRS trades.” Id. ¶ 183 & n.72. And in 2008, the SAC
alleges, “[m]eeting through OTCDerivNet, the Dealer Defendants agreed to clear only
interdealer IRS trades and only on SwapClear (the entity they controlled),” a “boycott” which
“starved” an “IRS product” created by CME called “CME Swaps on Swapstream.” Id. ¶ 188.
The allegations regarding OTCDerivNet, like those involving Project Fusion, plead the
existence of a forum for inter-Dealer communication during the alleged conspiracy. They also

52

adequately plead that the Dealers acknowledged a shared interest in influencing the process of
clearing IRS trades. But, for several reasons, the SAC’s allegations as to this one entity do not
more broadly support plaintiffs. They do not buttress the claim of an inter-Dealer agreement
beginning in 2007 to block all-to-all trading from emerging. And they supply an inadequate
basis for plaintiffs’ ambitious thesis that defendants—as opposed to a lack of buy-side interest—
are responsible for the pre-2013 absence of buy-side central clearing.
First, the Dealers founded OTCDerivNet in 2000, seven years before the start of the
alleged conspiracy. Its existence therefore is not evidence of the alleged conspiracy. Second, as
to the one action alleged to have occurred during 2007–2012, it is largely contradicted by class
plaintiffs’ original complaint: Contrary to the SAC’s claim that the members of OTCDerivNet
boycotted the CME product, plaintiffs initially admitted that the CME had never launched that
product. See Dkt. 13, ¶ 291. The SAC is notably silent as to that point, instead obliquely
alleging only that the product “swiftly failed” because of the alleged “boycott,” SAC ¶ 188.
These allegations, however, are conclusory. Second, the SAC’s allegations of a boycott of
clearing services accessible to the buy-side all take the form of conclusory group pleadings; the
SAC does not identify a single act by a single Dealer to boycott the CME product. See id. ¶ 186
(“The Dealer Defendants viewed this as a serious threat.”); id. 187 (“The Dealer Defendants
knew CME had the technical and commercial ability to offer integrated all-to-all trading and
central clearing for IRS to the buy side.”); id. ¶ 188 (“The Dealer Defendants responded to this
threat by boycotting Swapstream.”). Finally, the SAC does not allege that the Dealers boycotted
CME’s clearing services, just the one never-offered product. On the contrary, in an allegation
inconsistent with the claim that the Dealers conspired to “prevent . . . buy-side access to IRS

53

clearing,” id. ¶ 175, the SAC alleges that, after the 2008 financial crisis, the Dealer-controlled
SwapClear launched an IRS clearing product designed specifically for the buy-side. Id. ¶ 189.
Trade associations: The SAC alleges that the Dealers belonged to, and conspired
through, trade associations and related organizations. See id. ¶¶ 126, 151–57, 203. It alleges that
the Dealers entered into agreements at unidentified meetings of these organizations. Such bare
claims of “agreement,” however, are legal conclusions. They do not make the claim of a boycott
of all-to-all exchanges in this period plausible. See Citigroup, 709 F.3d at 135–36. And a “mere
opportunity to conspire” at legitimate meetings does not support an inference that “an illegal
combination actually occurred.” Capital Imaging Assocs., 996 F.2d at 545; see also Twombly,
550 U.S. at 567 n.12 (“belong[ing] to the same trade guild as one[’s] . . . competitors” does not
render conspiracy plausible); In re Musical Instruments & Equip. Antitrust Litig., 798 F.3d 1186,
1196 (9th Cir. 2015) (“mere participation in trade-organization meetings where information is
exchanged and strategies are advocated does not suggest an illegal agreement”).
d.

“Plus Factors” and Conclusion

Given the few well-pled allegations of parallel relevant activity among Dealers during
2007–2012, there is, arguably, no charter to inquire into the existence of the three “plus factors”
identified by the Second Circuit. The SAC, as noted, instead appears to anchor its claim of
conspiracy during this period on Project Fusion, which it presents, albeit incorrectly, as direct
evidence of a § 1 boycott. Nevertheless, to assure a careful review of the plausibility of
plaintiffs’ claim of conspiracy, the Court analyzes these factors.
One “plus factor” is clearly present. The SAC pleads “a high level of interfirm
communications,” Gelboim, 823 F.3d at 781, including via the Dealers common ownership of
TradeNet, their participation in OTCDerivNet, their participation on industry associations, and

54

the social and professional and social interactions among executives of Dealers in this market
niche. See, e.g., SAC ¶¶ 126–41, 151–58, 182–83.
But the other two “plus factors” are at best thinly pled. The SAC claims “a common
motive to conspire,” Gelboim, 823 F.3d at 781, during 2007–2012. But, while the SAC pleads
that many Dealers preferred RFQ trading and viewed all-to-all exchange trading as a long-term
threat to profit margins, see, e.g., SAC ¶¶ 102–105, 113, before 2013, there was no all-to-all
exchange trading to boycott. And to the extent the SAC’s theory is of a long-term plot to nip this
mode of trading in the bud, the SAC does not (other than conclusorily) allege that—before
Dodd-Frank’s mandates—central clearing, a necessary precondition for such trading, existed or
was imminent or seen as inevitable. Therefore, while defendants had a long-term interest in the
non-emergence of such trading, in the 2007–2012 period, on the facts pled, there was little
urgency to conspire against it.20
As to the remaining “plus factor,” the SAC does not plead activity by individual Dealers,
during this period, against their “apparent individual economic self-interest.” Gelboim, 823 F.3d
at 781. As to “non-economic evidence [of] an actual, manifest agreement not to compete,” In re
Ins. Brokerage Antitrust Litig., 618 F.3d at 322, the SAC does not plead it, either. The closest it
comes is the Dealers’ alleged coordination in connection with acquiring control of Tradeweb.
But, for the reasons addressed, the SAC’s allegations as to that joint venture do not adequately

20

Defendants cite extensively to buy-side firms’ submissions before the CFTC to the effect that
these firms often favored RFQ trading, were concerned about the costs and collateral outlays
associated with mandatory central clearing, and believed that RFQ trading and individualized
negotiations over trade terms—particularly as to customized IRS trades—would often yield
better pricing than exchange trading. See, e.g., Dealer Mem. at 7 nn.7–8, 8 n.10, 9 n.15, 11
nn.22–23, 12 nn.24–27, 13 n.28. The Court has not relied on these materials. Had the Court
done so, these would have reinforced that mandatory central clearing did not enjoy universal
buy-side support and that this trading infrastructure, before Dodd-Frank, was not viewed as
imminent.
55

plead an agreement to terminate a Tradeweb plan to initiate all-to-all exchange trading, let alone
that such an agreement was illegal.
All allegations considered, the SAC therefore has not pled a plausible conspiracy among
the Dealer Defendants, during 2007–2012, to block the emergence of all-to-all platforms for IRS
exchange trading. There are limited well-pled allegations of parallel activity among the Dealers.
There is no direct evidence of conspiracy, and the SAC’s main allegations of actual collaboration
among Dealers, the Project Fusion joint venture, as pled, do not describe an illegal agreement.
And the plus factors lend only light support to a conspiracy theory. As in other cases where the
factual allegations did not plausibly support the inference of an agreement in violation of § 1, the
SAC’s claims of such a conspiracy during 2007–2012 must therefore be dismissed as
implausible. See, e.g., Twombly, 550 U.S. at 570; Citigroup, 709 F.3d at 140; In re Insurance
Brokerage Antitrust Litig., 618 F.3d at 336 (dismissing all § 1 claims in claim of industry-wide
conspiracy except narrow sector-specific allegations of bid-rigging).
2.

2013–2016

In contrast to 2007–2012, plaintiffs’ pleadings for 2013–2016 allege a recognized type of
per se unlawful § 1 conspiracy: a group boycott. During this period, five platforms for all-to-all
exchange trading of IRSs emerged: Tera, Javelin, TrueEx, Tradeweb, and Bloomberg. The
SACs allege that the Dealer Defendants conspired to starve the first three of these platforms of
liquidity so as to destroy them—and that this boycott largely succeeded.21

21

As to the two remaining platforms: The SACs allege that Dealer-owned Tradeweb was a coconspirator. They do not allege that Bloomberg’s platform was an object of the Dealer boycott.
56

For this period, plaintiffs mainly argue that a conspiracy can be inferred from parallel
conduct by the Dealers coupled with circumstantial evidence and “plus” factors indicating
agreement.22 The SACs allege the following parallel conduct:


Parallel refusals to trade on Javelin, Tera, and TrueEX platforms: The three new
entities each sought liquidity from the Dealers for their new platforms permitting allto-all trading. The 11 Dealers (except, briefly, RBS) each refused to supply liquidity
to, or to trade on, these platforms. See, e.g., JTSAC ¶¶ 148, 151, 185, 189, 223–25.



Common excuses and vocabulary: The Dealers gave similar excuses for refusing to
provide liquidity to the new platforms. For example, at separate meetings at Javelin
between 2013 and 2015, senior employees at each Dealer cited “a need to conduct a
never-ending legal review of the Javelin SEF rulebook, a largely standardized
document already approved by the CFTC.” Multiple Dealers used the same
terminology in explaining to affiliated IDBs why they would not allow these IDBs to
trade on Tera. The Dealers called Tera’s platform a “Trojan Horse” and said they did
not wish to let Tera “off the mat.” See, e.g., id. ¶¶ 151, 154, 215.



Similar tactics at meetings with Javelin and Tera: The Dealers used meetings with
Javelin and Tera to explore ways to undermine the platforms and used similar tactics
at these meetings. Goldman Sachs executives used a September 2013 meeting to
“grill[] Javelin on whether its platform had to allow all-to-all trading and to fish for
the names of buy-side customers signed up for Javelin’s platform,” so as to head off
these customers from using the platform. A month earlier, Morgan Stanley’s Senft
had asked Javelin “for a list of buy-side customers that had expressed an interest in
Javelin, as well as the contact person for each firm.” As to Tera, Goldman Sachs,
Barclays, Bank of America, Credit Suisse, Citi, Deutsche Bank, and Morgan Stanley
arranged meetings with Tera, ostensibly to explore trading on its platform, and at
these meetings used “strikingly similar” “bait and switch tactics”: Tera “personnel
would walk into the meeting expecting to meet with the head of IRS trading to
discuss the bank signing up for the TeraExchange platform, but instead would be met
by personnel from the bank’s strategic investment group offering to take a stake in
TeraExchange itself,” allegedly with the goal of taking over and shutting down these
platforms. See, e.g., id. ¶¶ 144–45, 149, 151–54, 186–88.



Parallel withholdings by affiliated FCMs of clearing services: The Dealers’ affiliated
FCMs all withheld clearing services on the Javelin and Tera platforms, at the Dealers’
direction, effectively blocking market entry by these platforms. Withholding such

22

Plaintiffs also claim to have alleged direct evidence. They claim that the monthly meetings
among Dealers’ Strategic Investment Groups between 2013 and 2015, and other communications
among Dealers, were used to coordinate positions regarding the IRS market. See, e.g., JTSAC
¶¶ 127, 280, 327. The allegations as to the content of these meetings are too general and
conclusory, however, to qualify as direct evidence of an illegal agreement.
57

services cost the FCMs revenue, because FCMs earn fees for each trade submitted for
clearing. For example, the FCMs, citing the need for rulebook reviews, each refused
to conduct pre-trade credit checks for customers on both platforms, preventing trades,
despite routinely providing such checks to other platforms that did not offer
meaningful anonymous all-to-all trading, including Tradeweb and Bloomberg. As to
Tera: The Dealers’ affiliated FCMs—including Barclays, Citi, Credit Suisse,
Deutsche Bank, HSBC, and Morgan Stanley—refused to clear trades outright, or
quoted exorbitant fees. As to Javelin: The FCMs that refused to provide creditchecks to buy-side customers who sought to use its order book included Deutsche
Bank, which refused to allow its FCM to connect to the Javelin platform, and
Barclays, Goldman Sachs, and Morgan Stanley, which made “nearly identical claims
. . . as to the need to conduct reviews of Javelin.” See, e.g., id. ¶¶ 138–40, 162–70,
196–200.


Parallel action towards Tera in response to its first IRS trade: On June 13, 2014,
Tera conducted the first IRS trade on its platform. BNPP’s trading desk was notified
of the trade by its affiliated FCM, which, in a “transgression,” had cleared the
transaction. BNPP’s trading desk then contacted the parties to the transaction and
“threatened them with a loss of clearing and other banking services,” including
execution and general market research, if they continued to trade on Tera. On June
16, 2014, the next business day, four Dealers—BNPP, Citi, J.P. Morgan, and UBS—
separately notified Tera that they would not clear trades on Tera’s platform until they
had conducted a review of Tera’s rulebook, a largely standardized document that the
CFTC had already reviewed before giving Tera temporary SEF registration. Other
Dealers later similarly cited a need for (never-completed) rulebook reviews as reasons
not to trade on Tera. As to Javelin: Barclays, Goldman Sachs, and Morgan Stanley
and other Dealers each asserted a need to review its rulebook. See, e.g., id. ¶¶ 170,
204–05.



Similar statements to Tera personnel: Dealers’ representatives each “told personnel
at TeraExchange that its platform would never succeed.” See, e.g., id. ¶ 195.



Similar pressure applied to customers: Dealers, including Citi and Goldman Sachs,
pressured existing customers not to trade on Javelin or Tera, and penalized buy-side
entities caught trading on these platforms. See, e.g., id. ¶¶ 173–75, 214.



Similar clearing-fee differentials: Dealers, including Bank of America, Barclays,
BVPP, Credit Suisse, and J.P. Morgan, quoted much higher different clearing fees for
customers who sought to trade on the Javelin and Tera platforms than for clearing on
Dealer-friendly platforms. See, e.g., id. ¶¶ 201, 231–32.



Similar treatment of smaller IDBs: Dealers refused to consent to various smaller
IDBs’ use of Tera, and thereby blocked Tera from giving these IDBs access to SEF
services. See, e.g., id. ¶¶ 213–16.

58



Parallel adherence to “name give-up”: Dealers insisted in the practice of “name
give-up,” requiring the disclosure of each swap counterparty’s identity to the other,
and imposed this practice on IDBs, enabling Dealers to prevent buy-side customers
from trading on the IDBs’ electronic platforms. A trade-processing entity called
MarkitSERV, operated by former executives of Goldman Sachs and Deutsche Bank,
facilitated “name give-up”: The Dealers’ IDBs sent trades to its MarkitWire service
before they were cleared, and MarkitWire then shared the counterparties’ names with
each other. When an interdealer SEF operated by GFI stated in 2014 that it intended
to allow anonymous trading, GFI received heated phone calls from executives at
Credit Suisse and J.P. Morgan, causing GFI to reverse course. Another SEF,
Tradition, publicly attributed its similar decision to pressure from dealers. See, e.g.,
id. ¶¶ 19, 234–35, 242–47, 260–62, 332.



Common direction of Tradeweb: The Dealer Defendants caused Tradeweb, which
they jointly controlled and which was capable of launching an anonymous all-to-all
trading platform, not to do so. See, e.g., id. ¶¶ 23, 312–13.

These allegations of parallel conduct, viewed collectively and in conjunction with other
pled facts reviewed below, make plausible the inference of a § 1 conspiracy among Dealers to
boycott the three new platforms. That is so even discounting as worthy of less weight the SACs’
collective allegations about “Dealer Defendants” that lack allegations about a particular
Dealer. Importantly, the allegations concerning 2013–2016—unlike those for 2007–2012—are
not limited to isolated subsidiary acts. They include the core claim that the Dealers refused to do
business with the new all-to-all platforms. And the SACs’ claim that these refusals resulted from
a concerted plan to stop these platforms from gaining traction, as opposed to isolated decisions
by individual Dealers, is buttressed by the allegations of subsidiary parallel acts, practices, and
locutions.
At the threshold, the Dealers are correct that—standing alone—their collective refusal to
do business on the new platforms would not support inferring a conspiracy. There is a “natural
explanation,” Twombly, 550 U.S. at 568, consistent with unilateral action, for the Dealers’
decisions not to supply liquidity to Javelin, Tera, and TrueEx. As the SACs plead, the existing
RFQ mode of trading with the buy-side was highly profitable. All-to-all exchange trading,
59

however, threatened to slash the Dealers’ margins by “billions of dollars” by disintermediating
them. See SAC ¶ 5, 102–03, JTSAC ¶¶ 14, 275. Each Dealer’s decision to avoid the startup
platforms, like the decision by each phone company in Twombly not to compete in new markets,
is, in and of itself unremarkable. Considered alone, it is not—at all—suggestive of conspiracy.
See, e.g., Williams v. Citigroup, Inc., No. 08 Civ. 9208 (LAP), 2009 WL 3682536, at *4
(S.D.N.Y. Nov. 2, 2009) (alleged conspiracy among investment banks to boycott new financing
structure that “threatene[d] [banks’] positions” in derivative markets not plausible; conduct
suggests not “a wide-ranging conspiracy but rather unilateral action among the [banks], each of
whom wants to preserve its own market position”); In re Ins. Brokerage Antitrust Litig., 618
F.3d at 349 (natural for defendant who is “‘[r]eaping enormous profits” to have “no desire to
upset the apple cart’” (quoting Twombly, 550 U.S. at 568)).23
Problematic for the Dealers, however, are the allegations of other common behavior that
is less easily explained as unilateral action. Most telling are the events alleged on June 16, 2014,
the first business day after Tera’s initial IRS trade. That day, four Dealers (BNPP, Citi, J.P.
Morgan, and UBS) each separately contacted Tera. Each told Tera that it would not clear trades
on Tera’s platform until it had conducted a review of Tera’s rulebook. This suggests
coordination. Even assuming that the Dealers independently knew of Tera’s initial trade as
opposed to having been alerted to it by BNPP as the SACs allege,24 that four Dealers called Tera

23

Plaintiffs argue that refraining from trading on the new platforms was irrational conduct for a
Dealer, because each had an “individual interest” in participating in the startup markets to gain a
“first mover advantage.” SAC ¶¶ 342–44; JTSAC ¶¶ 376–77. The SACs, however, do not plead
concrete facts suggesting that a Dealer took any material risk by sitting tight and waiting to see if
the new platforms attracted sufficient support to survive. See SAC ¶ 273 n.102 (participation of
only “a smaller number of dealers” would not be enough to sustain an all-to-all platform).

24

The SACs do not allege any other means by which the four Dealers would have learned, by
June 16, 2014, of the June 13, 2014 trade. At argument, the Court inquired whether the trade had
60

the next business day, and made the identical demand of Tera (to audit its rulebook) as a
condition for clearing trades, is improbable enough to support an inference of collaboration.
That inference is strengthened by the SACs’ well-pled allegation that the Dealers’ interest in
Tera’s rulebook was pretextual. They allege that (1) the Tera rulebook had already been audited
by the CFTC, which had provisionally approved Tera’s platform, and (2) despite their
expressions of interest in the rulebook, no Dealer ever completed this supposedly important
audit.
Other parallel behavior supports an inference of coordinated conduct to put roadblocks in
the path of the new platforms. These include: causing affiliated FCMs to deny them pre-trade
credit checks and clearing services; scheduling meetings with Tera and Javelin with the
ostensible goal of exploring trading but using them to extract customer names and/or to probe
buying out the new entrants; using the bogus claim to require a rulebook audit to explain not
doing business with Javelin and Tera; pressuring existing customers not to trade on Javelin or
Tera and penalizing buy-side entities caught doing so; blocking smaller IDBs from using Javelin
and Tera; using a common invasion metaphor to describe Tera (a “Trojan Horse”) and a common
wrestling idiom to describe their goal towards it (not to let it “off the mat”); insisting on “name

been publicly reported by June 16 so as potentially to have become known to the Dealers “absent
communication among them.” See Tr. 18, 100, 130. In a post-argument letter, defendants
asserted that a subscription service called “SEFView” that took effect in February 2014 daily
reported SEF trading, including on Tera’s platform. Dkt. 223. In letter responses, plaintiffs
countered that the Dealers’ letter was tardy and improperly invited the Court to consider extrinsic
evidence. Dkts. 224–25.
The information reported about SEFView does not undermine the inference of coordination
among the four Dealers. Even assuming that BNPP, Citi, J.P. Morgan, and UBS each subscribed
to the SEFView service and accessed it and spotted the Tera trade before each contacted Tera on
June 16—and there is no cognizable evidence of any of this—the Dealers’ contemporaneous and
similar calls to Tera would still suggest coordination among them.
61

give-up” as an ostensible means to enforce the boycott, including as facilitated by MarkitWire;
and telling Tera it would never succeed.
To be sure, none of these practices, acts, or locutions are illegal, inherently irrational, or
self-destructive. Any one, if engaged in by multiple Dealers, could result from unilateral
decisions. But, viewing these areas of symmetry in combination, the inference of
communication and coordination among Dealers with the shared goal of grinding down the new
platforms is entirely plausible. To put the point differently, each Dealer’s independent interest in
maintaining the status quo would explain each’s decision not to supply liquidity to Javelin or
Tera. But it does not follow from that shared interest that Dealers would use similar stylized
stratagems to blunt the emergence of these newcomers. For a Dealer uninterested in doing
business with the new platforms, the obvious alternative would have been to “just say no.”
Of the three “plus factors” identified by the Second Circuit, two lend further support to
the inference of a conspiracy.
First, the SACs allege a common motive to conspire. It was to preserve the “profit
center” supplied by maximal use of non-anonymous RFQ trading with the buy-side. See, e.g.,
JTSAC ¶ 14. A conspiracy, if successful, would assure that the new platforms were starved of
liquidity; there was otherwise a risk that enough Dealers would participate to give the new
platforms enough sufficient liquidity to be viable. See, e.g., Gelboim, 823 F.3d at 781–82; In re
Credit Default Swaps Antitrust Litigation, No. 13-MD-2476 (DLC), 2014 WL 4379112, at *10
(S.D.N.Y. Sep. 4, 2014) (“In re CDS”) (“no single [defendant] could prevent exchanges from
emerging, but all [defendants] would profit from such prevention”).
Second, the SACs allege a high degree of interfirm communications. See Apex Oil Co. v.
DiMauro, 822 F.2d 246, 254 (2d Cir. 1987). They allege extensive communications among

62

high-level officials at Dealers with responsibilities for IRS and/or overall swaps trading,
including among members of strategic investment groups; monthly meetings among the heads of
the Dealers’ trading desks; communications among their clearing personnel; common
participation in trade associations; and informal meals and other gatherings. The SACs also
reflect that officials involved in IRS trading sometimes switched employment from one Dealer to
another, tying the firms closer. Dealer executives’ common service on Tradeweb’s boards,
finally, supplied another opportunity for communications among relevant Dealer personnel—and
in a forum in which discussion of competing platforms like Javelin and Tera might naturally
arise.
The third “plus factor”—actions against defendants’ self-interest—only marginally, at
best, enhances the claim of a conspiracy. The SACs do allege that the Dealers caused their FCM
affiliates to forego fees when they refused to clear trades on the new platforms. See, e.g., JTSAC
¶ 137. But the focus on FCMs’ profitability is myopic. Plaintiffs’ overarching allegation is that
preventing the all-to-all IRS trading platforms from taking root enriched the Dealers by
preserving their wider spreads and profit margins on IRS trading. Plaintiffs do not claim that
foregone FCM fees approached, let alone exceeded, the profit margins preserved by squelching
the new platforms. Each Dealer’s overall interest therefore lay in aborting the new platforms. A
Dealer was not acting against its individual self-interest in directing the FCM not to act to fortify
the new platforms.
Finally, other alleged incidents and statements by Dealer personnel circumstantially
support the conspiracy claim. Four examples are illustrative.

63

First, Morgan Stanley’s Senft asked that Javelin’s non-disclosure agreement be modified
to allow one Dealer to discuss Javelin with the others. See id. ¶ 149. Senft’s request, to which
Javelin agreed, supports the inference that the Dealers discussed Javelin among themselves.
Second, on August 6, 2013, Javelin personnel gave Goldman’s Smith, at his request, a list
of 80 buy-side firms in various stages of discussion with Javelin. Senft stated that “[t]his is
exactly what I needed,” and then shared this list with other Dealers, to alert them which buy-side
customers were using or considering trading on the Javelin platform. Id. ¶ 145. Smith’s request
and statement support the inferences that Dealers communicated about the new platforms and
coordinated putting pressure on buy-side customers to shun these platforms.
Third, the Citadel Fixed Income Master Fund, a buy-side entity, executed a series of IRS
trades on Javelin between January and April 2014 but then abruptly abandoned Javelin.
Citadel’s chief operating officer is quoted as explaining that Citadel had received “calls from
representatives of Dealer Defendants pressing him to do so.” Id. ¶ 175. Mitsubishi, another buyside entity, is similarly reported as stating that pressure for J.P. Morgan’s FCM had caused it to
stop trading on Javelin. Id. ¶ 178. Citadel’s and Mitsubishi’s experiences support the claim of a
plan among Dealers to pressure the buy-side to boycott the new platforms.
And fourth, the chief executive of a small IDB told Tera’s chief executive that the IDB
could not do business with Tera because the Dealers would not allow it. Id. ¶ 216. Although the
SACs do not specify the basis for the IDB executive’s belief, this allegation supports the claim of
coordinated Dealer pressure on the buy-side.
The allegations reviewed above, viewed as a whole, plausibly allege a group-boycott
conspiracy among Dealer Defendants between 2013 and 2016 aimed at the three new platforms,

64

in violation of Sherman Act § 1. Plaintiffs have pled enough facts to permit them to test this
claim in discovery.25
In so holding, the Court has carefully considered defendants’ various agreements why the
SACs fall short. The Court addresses here three not considered above.
Group pleading: Defendants argue that the many references to “Dealer Defendants” are
improper “group pleading.” The SACs do indeed often cluster these 11 defendants—each a
corporate family—together. And some collective allegations are so sweeping and conclusory as
to prevent them from being credited, except as topic sentences summing up more specific
allegations. See, e.g., SAC ¶ 6 (“[T]he Dealer Defendants conspired to stop the IRS market from
developing in ways that would help the buy side.”); ¶ 13 (“To ensure the IRS market would
remain in an OTC-like environment, the Dealer Defendants conspired to implement roadblocks
and monitoring and enforcement mechanisms at every step of the trading pipeline.”); ¶ 160
(“The Dealer Defendants agreed to punish any IDB that dared to consider opening their
platforms to the buy side.”).26

25

The SACs allege that a putative class action was filed against many of the same defendants
alleging § 1 violations in connection with the credit default swaps (CDS) market; that these
claims largely survived a motion to dismiss, see In re CDS, 2014 WL 4379112, at *9–13; and
that there were regulatory investigations of the CDS market. See SAC ¶¶ 348–56; JTSAC
¶¶ 380–89. In sustaining plaintiffs’ claims as to 2013–2016, the Court has disregarded these
allegations. The civil claims did not result in any findings or admissions of liability; in any
event, the civil claims involving CDSs are materially different from those here involving IRSs;
and the regulatory investigations were closed without adverse findings. See In re Commodity
Exch., Inc., Gold Futures and Options Trading Litig., 213 F. Supp. 3d 631, 661–62 (S.D.N.Y.
2016) (“Commodity Exch.”); see also In re Elevator Antitrust Litig., 502 F.3d at 51–52
(discounting allegations of anti-competitive conduct by same defendants in Europe as irrelevant;
court rejects claim that “if it happened there, it could have happened here”).
26

The Court similarly disregards the complaints’ characterizations of defendants’ behavior as a
“boycott,” “conspiracy,” “cartel,” and “illegal agreement.” See, e.g., SAC ¶¶ 4, 5, 13, 19, 20, 22,
28, 30, 81, 163, 202, 206, 208–09, 244, 278, 282, 286. These labels are legal conclusions. See,
e.g., Twombly, 550 U.S. at 561; Citigroup, 709 F.3d at 135–36.
65

But the lengthy complaints27 are also rife with specifics. As illustrated above, they
contain many allegations against individual defendants that are quite particular, both as to the
actor and the act. The SACs often name the executive or business unit at a particular Dealer who
took a concrete action with respect to a specific target, including Javelin and Tera. And many
references in the SACs as to the “Dealer Defendants” are proper. Given plaintiffs’ thesis that the
Dealers participated in a collective boycott, it is natural that the SACs express some factual
allegations collectively.
In the end, the dispositive issues are whether, viewing the complaints as a whole, they
satisfactorily allege the existence of the § 1 conspiracy, and, as to individual defendants, whether
it is adequately pled that they “in their individual capacities, consciously committed themselves
to [this] common scheme designed to achieve an unlawful objective.” AD/SAT, a Div. of
Skylight, Inc. v. Associated Press, 181 F.3d 216, 234 (2d Cir. 1999). The SACs adequately plead
the existence of the group boycott between 2013 and 2016. And—save for Dealer HSBC, as
discussed infra—the SACs adequately plead that each Dealer Defendant participated in that
group boycott. They give each such defendant “fair notice of what the claim is and the grounds
on which it rests,” Anderson News, 680 F.3d at 182, including the factual connection of that
defendant to the scheme and the identity of its alleged co-conspirators. See In re Foreign Exch.
Benchmark Rates Antitrust Litig., 74 F. Supp. 3d 581, 594 (S.D.N.Y. 2015).
Non-uniformity: Defendants note that as to certain practices alleged to further the
boycott, the SACs do not plead that the Dealers acted uniformly. For example, unlike the other
Dealers, RBS initially made markets with Javelin before reversing course. And, as to TrueEx,
the SACs allege only that six Dealers “declined to use” that platform. SAC ¶ 283; JTSAC

27

The SAC is 144 pages. The JTSAC is 134 pages.
66

¶¶ 222–23. Relatedly, while plaintiffs’ claim that Dealers refused to allow affiliated FCMs to
clear trades on the new platforms, as to Javelin, the SACs do not allege specific instances of
clearing difficulties with respect to eight such FCMs; and as to Tera, they do not allege specific
instances as to six.
Unanimity of action, however, is not required. At the pleading stage, the issue is whether
the inference of conspiracy, viewing the well-pled allegations holistically, is plausible. The
Court has so found here. See, e.g., SD3, LLC. v. Black & Decker (U.S.) Inc., 801 F.3d 412, 428–
29 (4th Cir. 2015) (rejecting defendants’ argument that alleged parallel conduct must be
“simultaneous” or “identical” or that, for conspiracy to be plausible, defendants must move in
“lockstep”); cf. United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 222 (1940) (“Nor is it
important that the prices paid by the combination were not fixed in the sense that they were
uniform and inflexible.”). That RBS initially made markets with Javelin—but soon backed away
so as to align with other Dealers—does not undermine the claim of a group boycott. And the
absence of specific allegations as to every Dealer Defendant with respect to each means or
method of hobbling the new platforms is not fatal, either. Conspiracies, particularly when
alleged among a large group, are not always tidy and symmetric. Conspirators may aid the
common venture via techniques and stratagems that are consistent and reinforcing but not
entirely overlapping. And at the pleading stage, before discovery, the failure to allege that a
given defendant engaged in a given practice may simply reflect that plaintiffs’ information is asyet incomplete, a scenario that is particularly realistic as to defendants’ conduct towards nonparty TrueEx. Discovery will test the extent to which there was dissonant conduct among
Dealers, and what such asymmetry says about the existence of a conspiracy and its membership.

67

“Market realities”: Most substantially, defendants argue that facts contained in
secondary material cited in the SACs (the “Secondary Material”) establish “important market
realities,” Dealer Mem. at 2, inconsistent with the alleged group boycott. The Secondary
Material consists largely of (1) articles regarding the swaps industry; (2) submissions to the
CFTC by industry participants, including the buy side, when the CFTC weighed swap-trading
rules implementing Dodd-Frank; and (3) the CFTC’s findings and commentary issued along with
these rules. Drawing heavily on these materials, defendants argue that at least four “market
realities” make the § 1 claim implausible.
The first is that buy-side support for all-to-all trading was more limited than the SACs
portray. The Secondary Material suggests that many IRS trades were “bespoke” (i.e.,
customized) and did not lend themselves to trading based on a preexisting order book. A 2012
Federal Reserve study, for example, stated that “the vast majority of IRS contracts traded only
once during the three-month period studied” and that “no single instrument in the IRS data set
traded more than 150 times per day,” whereas stocks and other exchange-traded securities trade
several thousand times a day on multiple venues. See Dealer Mem. at 7–9 & nn.8–15 (exhibit
citations omitted). Similarly, the CFTC reported that buy-side entities had strongly supported
maintaining RFQ trading alongside order-book trading. RFQ trading, these buy-side entities told
the CFTC, facilitated customized trades and saved costs. See Dealer Mem. at 7 & nn.7–8
(exhibit citations omitted). Such input, the CFTC stated, persuaded it to authorize RFQ trading
for IRSs on SEFs and to reduce the minimum number of RFQ recipients (originally proposed as
five) to three. See Dealer Mem. at 1–12 & nn.22–27 (exhibit citations omitted). Other
Secondary Material, defendants argue, shows that since SEF trading began, RFQ trading has

68

proven much more popular than order-book trading because the buy-side views it as yielding
better prices. Id. at 12–13 & n.28 (exhibit citations omitted).
A second “market reality” is that, among the three platforms (Javelin, Tera, and TrueEx)
the Dealers are alleged to have boycotted, there was more Dealer support for TrueEx than the
SACs portray. The Secondary Material indicates that more than 17 IRS dealers have been
trading on TrueEx, that J.P. Morgan supported TrueEx from its launch, that UBS has been in
talks with TrueEx “to deliver client volume” to TrueEx’s SEF, and that TrueEx today enjoys
“record volumes” and “on good days it handles as much as 38 percent of trading” in its markets.
See Dealer Mem. at 18–19 & n.37 (exhibit citations omitted).
A third “market reality” is that Javelin and Tera’s failure to gain traction is due, at least in
part, to factors other than a boycott. The Dealers depict Tera as having had lackluster interest in
all-to-all trading of IRSs: The JTSAC, they note, does not allege that Tera affirmatively
recruited the Dealers to its site, while it does plead that, “in the first instance,” Tera “focused on
attracting non-traditional liquidity providers to make markets on its platform.” JTSAC ¶ 190.
As for Javelin, the Dealers note, not only did RBS initially made markets on its platform, but
Secondary Material indicates that UBS aided Javelin’s by making available its UBS’s “Neo
price-aggregation service.” See Dealer Mem. at 19–20 & n.38 (exhibit citations omitted).
A fourth “market reality” that defendants derive is that Javelin and Tera encountered
challenges that had nothing to do with a boycott. For example, one of the Secondary Materials, a
press release from a “credit hub,” Traiana, reports that Javelin and Tera first connected to that
hub, which facilitates credit checks, in January 2014. If in fact Javelin and Tera had not
previously been connected to a credit hub, defendants argue, then, up to that point, the Dealers’

69

FCM affiliates had every reason not to clear trades on these platforms. See Dealer Mem. at 21–
22 & nn.39–40 (exhibit citations omitted)
The parties, invoking competing lines of case law, differ on whether it is proper to draw
to the extent defendants do upon the Secondary Materials to develop “market realities”
undermining the SACs’ group boycott claim. In general, “it is proper to take judicial notice of
the fact that press coverage, prior lawsuits, or regulatory filings contained certain information,
without regard to the truth of their contents.” Staehr v. Hartford Fin. Servs. Grp., Inc., 547 F.3d
406, 425 (2d Cir. 2008) (emphasis added). Plaintiffs rely on cases holding that, although a court
can take notice of the fact of public filings and documents attached to or cited in a complaint, it
may not consider these materials for “’the truth of the matters asserted.’” See, e.g., Int’l Star
Class Yacht Racing Ass’n. v. Tommy Hilfiger U.S.A., Inc., 146 F.3d 66, 70 (2d Cir. 1998) (noting
that such materials ordinarily do not meet the test of indisputability required for judicial notice
under Fed. R. Evid. 201); see also Global Network Commc’ns, Inc. v. City of N.Y., 458 F.3d 150,
156 (2d Cir. 2006) (judicial notice improper where court “relied on those materials to make a
finding of fact that controverted the plaintiff’s own factual assertions”); eBooks, 859 F. Supp. 2d
at 686 (a “[c]ourt accepts as true for purposes of the motion to dismiss only those factual
allegations that actually appear in the Complaint” and will not “take judicial notice of the
complete contents of articles cited in the Complaints” or “draw inferences against the plaintiffs
from those facts that are in the Complaint”) (internal citation omitted); Sira v. Morton, 380 F.3d
57, 67 (2d Cir. 2004) (“Limited quotation from or reference to documents that may constitute
relevant evidence in a case is not enough to incorporate those documents, wholesale, into the
complaint.”) (citation omitted). They argue that defendants may not draw upon other aspects of
the sources cited in the SACs to develop a “counter-narrative.” Defendants counter by invoking

70

caselaw that a court may disregard a complaint’s factual allegations where these are
“contradicted by the complaint itself, by documents upon which the pleadings rely, or by facts of
which the court may take judicial notice,” Perry v. NYSARC, Inc., 424 F. App’x 23, 25 (2d Cir.
2011) (summary order).
On this methodological dispute, plaintiffs are more nearly correct. For the most part,
although by no means entirely, the facts and data that defendants extract from the Secondary
Material do not contradict specific factual allegations in the SACs. The embedded facts and data
instead are instead used as building blocks for an alternative non-boycott narrative. Crediting,
for example, that much of the buy-side preferred RFQ trading, there would be less impetus for a
boycott; and crediting that Javelin and Tera had little demand and did limited Dealer outreach,
there would be less reason to blame Javelin and Tera’s failure to gain traction in IRS trading on a
boycott. On a motion to dismiss, however, as plaintiffs’ case authority reflects, defendants may
not use plaintiffs’ sources as an archive from which to build what amounts to a factual rebuttal.
Defendants may argue implausibility on other grounds, but not by stacking facts extracted from
secondary sources to construct an alternative narrative, at odds with the SACs’ well-pled
allegations.
In any event, even if the Court could draw upon this secondary material for this purpose,
a Dealer conspiracy to impede the new all-to-all trading platforms would not be implausible
(although such a conspiracy would be substantially less consequential than the SACs portray).
Plaintiffs’ view would remain one reasonable way to view events. See Gelboim, 823 F.3d at 781
(“‘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’”; an antitrust plaintiff “need not
show that its allegations suggesting an agreement are more likely than not true or that they rule

71

out the possibility of independent action”) (quoting Anderson News, 680 F.3d at 184); see also
Iqbal, 556 U.S. at 678 (claim is facially plausible so long as plaintiff “pleads factual content that
allows the court to draw the reasonable inference that the defendant is liable for the misconduct
alleged”).
To be sure, the “market realities” which defendants describe, were they established,
would be in tension with plaintiffs’ theories of liability, causation, and damages. The Court
expects counsel, in discovery, to probe these issues. But on 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.
Plaintiffs therefore have stated a plausible § 1 claim of a group boycott among Dealer
Defendants between 2013 and 2016.28 The Court denies the motion to dismiss that claim as to
that period.
C.

Arguments Particular to Individual Defendants

In briefs or argument, defendants BNPP, UBS, and HSBC each argue that, even if the
SACs adequately pled a conspiracy among Dealers in general, the facts pled do not connect them
to it. See Dkts. 39–40 (HSBC briefs); Tr. 54–62 (arguments by HSBC, BNPP, and UBS).29 The
two defendants who are not Dealers, Tradeweb and ICAP, also pursue dismissal on grounds
specific to themselves.
1.

BNPP

28

The Court has used “2013” as shorthand to capture the start of the plausibly-pled conspiracy.
Today’s decision does not, of course, pinpoint when in 2013 the alleged conspiracy began.
Conceivably, too, discovery may show that, if there was a conspiracy, it began later than 2013.
29

As to the remaining Dealer Defendants, the Court, following careful review of the SACs, has
found sufficient factual allegations linking each to the group boycott conspiracy.
72

BNPP argues that the SACs have not adequately pled its participation in the conspiracy.
It notes that the complaints do not allege that it was part of Tradeweb or Project Fusion. It also
notes that there are no allegations that BNPP did anything to boycott TrueEx.
Those specific allegations were not required. A conspiracy may be proven by
circumstantial evidence, and to show that a defendant joined, the plaintiff “need not show that
the defendant knew all of the details of the conspiracy, so long as he [or she] knew of its general
nature and extent.” United States v. Huezo, 546 F.3d 174, 180 (2d Cir. 2008) (quotation
omitted). The plaintiff need not show that “the defendant knew the identities of all the other
conspirators”; a single act may support an inference of involvement in a conspiracy if “of a
nature justifying an inference of knowledge of the broader conspiracy.” Id.; see also eBooks,
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”; “the 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”).30
Here, BNPP is alleged to have been one of the four Dealers who, in parallel, called Tera
on June 16, 2014, the first business day after Tera executed its first IRS trade, for the pretextual
purpose of auditing its rulebook. BNPP’s FCM affiliate also allegedly notified BNPP’s trading
desk of the trade, through which word spread to other Dealers. See JTSAC ¶¶ 203–05.
Although BNPP argues that its motives in calling Tera and asking to audit the rulebook were
benign, the Court cannot so assume on a motion to dismiss. BNPP also allegedly was among the

30

Plaintiffs urge the Court to apply an incorrect standard to this issue—that “once a conspiracy is
shown, only slight evidence is needed to link another defendant with it.” Apex Oil, 822 F.2d at
257. The Second Circuit has repudiated that formulation as inaccurately describing the burden
of proof in conspiracy cases. Huezo, 546 F.3d at 186 & n.2.
73

firms that threatened to raise clearing fees for buy-side firms in retaliation for their attempts to
trade on Javelin and Tera. See id. ¶ 231. These allegations, in conjunction with the core
allegation that BNPP refused to make markets on the new Javelin and Tera platforms, are
sufficient to link BNPP to the conspiracy.
2.

UBS

UBS notes that, in industry publications and conferences and before the CFTC, it
opposed, as a matter of policy, the practice of “name give-up,” which plaintiffs claim was a
means by which the Dealers policed the conspiracy. See JTSAC ¶ 240 (quoting UBS official as
stating, in October 2015 panel discussion, that “name give-up on order books is pointless for
cleared swaps”); cf. id. ¶¶ 250, 252.31 UBS argues that this iconoclastic policy stance makes its
participation in that conspiracy implausible.
For much the same reasons as with BNPP, the Court sustains the claims against UBS.
Like BNPP, UBS is alleged to have been among the four Dealers who, the day after Tera’s first
successful IRS trade, in concert told Tera “that they would not clear any trades executed on
TeraExchange’s platform until they conducted an audit of TeraExchange’s rulebook[.]” JTSAC
¶ 205. That act, and the claim that UBS refused to make markets on the new all-to-all platforms,
id. ¶ 56, is sufficient to link UBS to the alleged conspiracy. UBS’s opposition to “name give-up”
would certainly supply a basis not to infer its participation, but it does not make the claim of
such participation implausible.
3.

HSBC

31

For purposes of this discussion, the Court assumes arguendo, with UBS, that the fact that it
opposed “name give-up”—a fact based on materials attached to the SACs—is cognizable on this
motion.
74

HSBC argues that the allegations specific to it are too sparse to sustain the § 1 claim
against it. HSBC is correct. A close review of the SACs reveals no allegations that HSBC took
any action against any trading platform that proposed all-to-all trading or against any customer
that sought to so trade. The SACs do allege that HSBC or HSBC personnel occupied posts in
industry trade associations, like the ISDA and FIA and on OTC DerivNet. But they do not claim
that HSBC, in those posts, did anything specific to further the conspiracy.32
The one specific claim made against HSBC is that its FCM “gave TeraExchange the
runaround for over a year before refusing to clear for the platform,” in that an HSBC official,
despite receiving approximately a half-dozen “successful demonstrations” of Tera’s trading
platform, “refused to give a direct answer on whether HSBC would clear customers’ trades on
the platform.” Id. ¶ 198; see also SAC ¶ 230. That lone claim is insufficient to link HSBC to the
alleged conspiracy among Dealers.
The Court therefore dismisses the § 1 claim against HSBC.
4.

ICAP

ICAP is a London-based IDB that brokers trades, including of IRSs, between dealers.
SAC ¶ 11. Plaintiffs’ claims against ICAP center on events in 2009 and in 2013.
In 2009, plaintiffs allege, ICAP “joined [the Dealer] Defendants’ conspiracy,” when
ICAP agreed not to expand its European swap platform, i-Swap, into the United States. Class
Mem. at 15 (citing SAC ¶ 166). Plaintiffs allege that, in April 2009, Tradeweb had launched
Dealerweb, a platform for trading mortgage bonds, which cut heavily into ICAP’s competing
mortgage-bond trading platform. SAC ¶ 167. In retaliation, ICAP “threatened to launch an all-

32

HSBC is also not alleged to have had a stake in Tradeweb, on which plaintiffs’ (nowdismissed) claims of a pre-2013 conspiracy centered.
75

to-all trading platform for IRS that would be open to the buy side,” a threat plaintiffs claim was
credible because ICAP was then developing an electronic-trading platform for the European
inter-dealer IRS market called “i-Swap,” “which ICAP could have launched as an all-to-all,
anonymous trading platform in the United States.” Id. ¶ 168. “Alternatively,” plaintiffs claim,
“ICAP could have chosen to allow buy-side investors to trade on its existing IDB platform.” Id.
However, plaintiffs allege, later in 2009, “ICAP and the Dealer Defendants agreed to a détente
whereby the Dealer Defendants would not further expand Dealerweb into the IDB space in
exchange for ICAP not establishing an all-to-all anonymous IRS trading platform.” Id. ¶ 169.
Plaintiffs’ allegations with respect to ICAP in 2009 do not state a plausible claim of
participation in a conspiracy. As held above, the SAC does not state a claim of a § 1 conspiracy
among Dealer Defendants during 2007–2012. There was, therefore, no conspiracy in 2009 for
ICAP to join. Even if there had been, the SAC’s claim of a 2009 “détente” between ICAP and
the Dealers is conclusory. The SAC does not allege direct evidence of such an agreement. It
merely declares, without specifics, that ICAP “communicated with representatives of the Dealer
Defendants in the late spring and early summer of 2009” and that “[a]s a result of their
discussions” the parties “agreed to a détente.” Id. ¶ 169.
The SAC’s basis for claiming entry into a conspiracy in 2009 instead is circumstantial: It
asserts that ICAP’s decision not to abandon its IDB business to launch an untested electronic IRS
exchange in the United States was economically irrational. That allegation does not sustain a
conspiracy theory. That ICAP maintained its existing business model in Europe and forewent a
new venture in the United States does not, without more, give rise to an inference of agreement
with the Dealers. The SAC does not allege that any of ICAP’s brokerage rivals then offered an
anonymous all-to-all electronic exchange, that the clearing infrastructure to support such an

76

exchange in the United States was then in place, or that there was market demand at the time for
such an exchange. And the SAC does not allege any other act by ICAP to further a conspiracy to
block all-to-all IRS trading by the buy side. ICAP is not alleged, for example, to have done
anything to “punish” the buy-side or to restrict its access to clearing services.
As to 2013, the SAC alleges that, in February 2013, ICAP, effectively expanding iSwap,
launched an electronic trading platform, ICAP SEF, in the United States that, among other
things, facilitates IRS trades. But, plaintiffs claim, ICAP “limited participation to the Dealers.”
Id. ¶ 171. This, plaintiffs claim, was against ICAP’s self-interest, which lay in opening its
platform to the buy-side. Id. ¶ 172. Plaintiffs also fault ICAP for maintaining “name give-up”
on its U.S. platform, “thereby effectively preventing buy-side entities from trading on [it].” Id.
¶ 301. Plaintiffs claim that in exchange for ICAP’s not creating a U.S. platform for anonymous
all-to-all IRS trading, the Dealers, “to fulfill their end of the bargain,” have supported ICAP by
directing business to it and have supplied Dealerweb with low trading volume, while “reminding
ICAP and other IDBs that Dealerweb is an ever-present threat.” Id. ¶ 173.
These allegations, although occurring during the period (2013–2016) as to which the
Court has found sufficient allegations of a Dealer Defendant conspiracy, do not support an
inference that ICAP joined it. The SACs do not plead direct evidence of such an agreement.
Indeed, the SACs do not allege any communications between the Dealers and ICAP in or after
2013, let alone communications substantiating the quid pro quo or continuing “détente” with
Tradeweb that the SACs imagine.
And the complaints’ limited circumstantial allegations do not fill that vacuum. That
ICAP first launched an electronic trading platform, a SEF, in 2013, is unremarkable. That is
when Dodd-Frank took effect; numerous non-defendants launched U.S. trading platforms then.

77

As for the claim that ICAP “limited participation to the Dealers,” id. ¶ 171, it is elsewhere
repudiated by the SAC, which implicitly acknowledges that ICAP SEF, consistent with the
CFTC’s impartial access requirement, is open for IRS trading by the buy side, see id. ¶ 301
(asserting that the SEF’s condition of “name give-up” is what “effectively prevent[s] buy-side
entities from trading” on it).33 And in maintaining “name give-up,” ICAP’s SEF has company
from numerous interdealer SEFs, including six non-defendant entities identified in the SAC. Id.
And the complaints reveal an obvious non-conspiratorial reason for such a SEF unilaterally to
require name give-up, as the CFTC permits34: They allege that the Dealers, who allegedly
control IRS liquidity, prefer platforms with this protocol and are apt to move liquidity from
platforms they disfavor. See, e.g., SAC ¶¶ 113, 224, 276; JTSAC ¶¶ 127, 210, 238, 285; see
Citigroup, 709 F.3d at 138 (declining to infer a conspiracy from conduct that “made perfect
business sense”). Finally, as in 2009, there are no allegations that ICAP otherwise acted to
disadvantage the buy-side. Plaintiffs’ thesis that ICAP joined the conspiracy alleged among
Dealers and designed its business model to further that conspiracy is, therefore, conclusory.
The Court, therefore, dismisses the § 1 claims against ICAP.
5.

Tradeweb

33

Although not necessary to this ruling, ICAP has submitted documentation that in fact its SEFs,
consistent with their impartial access rules reviewed and approved by the CFTC, are all-to-all
platforms open to buy-side participation. See ICAP Mem. at 3 (exhibit citations omitted).
34

In its rule-making process, the CFTC expressly allowed platforms to use post-trade name giveup. See 78 Fed. Reg. 33,476, 33,499 (June 4, 2013); id. at 33,567 (giving SEFs operational
“flexibility,” including name give-up, to “improve the efficiency, competitiveness, and financial
integrity of the swaps market”). The CFTC approved ICAP’s SEF, including its post-trade name
give-up protocol. See ICAP Mem. at 9 (exhibit citation omitted).
78

As noted, Tradeweb builds and operates fixed-income and derivatives trading platforms,
including for IRSs. Since 2007, it has been majority-owned by Dealers, whose representatives
sit on its board. The parties’ discussions of Tradeweb center on the 2007–2012 period, during
which the complaints depict Tradeweb as the “principal forum,” SAC ¶ 151, of the alleged
conspiracy. The Court, however, has dismissed plaintiffs’ § 1 claims for that period. The issue
then is whether the complaints adequately plead conduct by Tradeweb during 2013–2016 that
link it to the conspiracy which the Court has found viable—among Dealers to boycott the all-toall IRS trading platforms of Javelin, Tera, and TrueEx whose emergence in 2013–2014 was
facilitated by Dodd-Frank’s clearing mandate.
Once pre-2013 allegations are stripped away, the SACs allege essentially three facts in
support of the claim that Tradeweb joined and furthered the conspiracy to boycott these new
platforms. First, Tradeweb served as a forum which the Dealers used “to hold secret
conspiratorial discussions” “under the cover of a supposedly lawful and independent enterprise”
so as “to coordinate their conduct.” Id. ¶¶ 17, 138–40. Second, Tradeweb, to “the present day,”
has not launched an anonymous all-to-all platform for IRS trading accessible to the buy-side. Id.
¶¶ 16, 144–46. Third, after Dodd-Frank took effect, Tradeweb operated two SEFs: Dealerweb
SEF, which was “exclusively for Dealers” and allows “anonymous, competitive trading,” and
Tradeweb SEF, which was designed for non-dealer market participants and “always discloses
counterparty identities.” Id. ¶ 147; JTSAC ¶ 316. To trade on Dealerweb SEF, Tradeweb
charged $50,000 per month, whereas it charged only about $100 a month to trade on Tradeweb
SEF. This cost differential was intended to “entrench a bifurcated dealer-to-dealer and dealer-tocustomer marketplace.” SAC ¶ 148; JTSAC ¶ 317.

79

These factual allegations do not connect Tradeweb to the discrete conspiracy among
Dealers that the Court has sustained: to boycott and thereby cripple Javelin, Tera, and TrueEx’s
platforms. Indeed, the portions of the SACs that make these allegations, strikingly, do not
reference any of those new platforms. Plaintiffs’ first allegation is that the Dealers’ ownership
and membership on Tradeweb’s board gave them an opportunity to conspire. But the complaints
do not allege, except in conclusory fashion, that the Dealers actually used gatherings at
Tradeweb to arrange this boycott. See Capital Imaging Assocs., 996 F.2d at 545 (“The mere
opportunity to conspire does not by itself support the inference that such an illegal combination
actually existed.”). And, even if there had been a well-pled allegation that the Dealers used a
Tradeweb board meeting to plot an aspect of the boycott, that would not implicate Tradeweb.
See In re Processed Egg Producers Antitrust Litig., 821 F. Supp. 2d 709, 753 (E.D. Pa. 2011)
(dismissing § 1 claim against trade association because it was improper to “impute the activities
of [an] organization’s members to the organization itself”); see also Sky Angel U.S., LLC v. Nat’l
Cable Satellite Corp., 947 F. Supp. 2d 88, 102 (D.D.C. 2013) (dismissing complaint because
“merely pleading that multiple entities hold positions on a board of directors” without pleading
any factual content for alleged agreement does not establish § 1 conspiracy).
Plaintiffs’ second allegation is that Tradeweb did not launch, and to “the present day”
does not have, an anonymous all-to-all platform for IRS trading accessible to the buy side.35 But
the CFTC did not require that IRSs be traded via an anonymous, all-to-all trading platform.

35

Because plaintiffs acknowledge that Tradeweb SEF was accessible to the buy side, plaintiffs’
complaint that a Tradeweb platform was functionally inaccessible to the buy side appears to
relate to its Dealerweb SEF, which allowed anonymous trading but which plaintiffs claim was
functionally inaccessible to the buy side due to its cost. Compare JTSAC ¶ 316 with id. ¶ 318.
In contrast, plaintiffs’ complaint that a platform did not allow non-anonymous IRS trading
appears to relate to the Tradeweb SEF.
80

Provided that it complied with operative regulations, the Tradeweb joint venture—regardless
who owned it—was under no obligation to open a platform to plaintiffs’ preferred specifications.
It was free unilaterally to maintain a business model that entailed non-anonymous trading.
More importantly, Tradeweb’s not having adopted anonymous all-to-all IRS trading
would not, without more, give rise to an inference of participation in a conspiracy. And
plaintiffs do not explain why Tradeweb’s decision to forego anonymous all-to-all trading stood
to further the particular conspiracy found plausible here—to injure the three new entrants that
made the opposite decision. The SACs do not, for example, claim that Tradeweb had market
power, or structured its business, or otherwise acted to siphon customers or support services from
Javelin, Tera, or TrueEx, so as to starve these infant platforms. Nor do the SACs allege that the
Dealers, abetted by Tradeweb, steered all their IRS market-making to Tradeweb’s SEFs, so as to
functionally to deny the new platforms of Dealer liquidity. On the contrary, plaintiffs depict
both Dealerweb SEF and Tradeweb SEF as close to irrelevant. The SAC alleges that the Dealers
have maintained Dealerweb’s IRS market share “at a very low level” (2%), ostensibly to
maintain their end of the alleged “détente” bargain with ICAP and to reward other IDBs “for
keeping the buy side off their platforms.” SAC ¶¶ 173, 306 n.124. The SAC alleges that
Tradeweb SEF is “inactive.” Id. ¶ 149.
Third, plaintiffs note the different pricing structures for Tradeweb’s SEFs. Dealerweb
SEF, plaintiffs claim, was prohibitively expensive to join as a member; whereas Tradeweb SEF
was reasonably priced but did not permit anonymous trading. Plaintiffs infer that Dealerweb
SEF was priced to be off-limits to the buy-side, such that a buy-side party that wished to trade
through Tradeweb was, realistically, compelled to disclose its identity. But even crediting that
inference, plaintiffs fail to explain why Tradeweb’s pricing practices held down competing SEFs

81

like Javelin, Tera, and TrueEx, so as to support an inference that Tradeweb was participating in a
Dealer boycott of them. The rational inference is the opposite: The less appealing to the buy
side that Tradeweb made its SEFs as platforms for IRS trading, the more relatively attractive
alternative platforms would be. And—as with ICAP—Tradeweb itself is not a Dealer positioned
to deny business to Javelin, Tera, or TrueEx. And plaintiffs do not allege any other actions by
Tradeweb tending to inhibit buy-side access to those new platforms.
The Court accordingly dismisses plaintiffs’ § 1 claims against Tradeweb.
V.

Statute of Limitations
A Sherman Act § 1 claim is subject to a four-year statute of limitations that runs from the

date of injury. The class plaintiffs filed an initial complaint on November 25, 2015. Defendants
therefore argue that their claims, to the extent based on injuries incurred before November 25,
2011 (hereinafter, “pre-2012 claims”), are time-barred.36 The class plaintiffs counter that the
statute of limitations should be equitably tolled due to fraudulent concealment. Although the
Court’s ruling that the pre-2013 claims of a § 1 conspiracy have not been plausibly pled has
independently eliminated the class’s pre-2012 claims, in the interest of completeness, the Court
addresses this argument.
A claim of fraudulent concealment must be pled with particularity. In re Commodity
Exch., 213 F. Supp. 2d at 675 (citations omitted). To show fraudulent concealment, a plaintiff
must prove: “‘(1) that the defendant concealed from him the existence of his cause of action, (2)
that he remained in ignorance of that cause of action until some point within four years of the
commencement of his action, and (3) that his continuing ignorance was not attributable to a lack

36

All defendants but HSBC and Morgan Stanley were named in the original complaint. Those
defendants were added on February 25, 2016; they argue that claims predating February 25, 2012
are time-barred.
82

of diligence on his part.’” In re CDS, 2014 WL 4379112, at *15 (quoting New York v.
Hendrickson Bros, Inc., 840 F.2d 1065, 1083 (2d Cir. 1988)); see also Veltri v. Bldg. Serv. 32B-J
Pension Fund, 393 F.3d 318, 323 (2d Cir. 2004) (“Where defendant is responsible for concealing
the existence of plaintiff’s cause of action, this Court has held equitable tolling arises.”). The
fraudulent concealment doctrine “ensures that wrongdoers are not permitted, or encouraged, to
take advantage of the limitations period to commit secret illegal conduct without penalty.”
Supermarket of Marlinton, Inc. v. Meadow Gold Dairies, Inc., 71 F.3d 119, 125 (4th Cir. 1995).
As to the first element of the fraudulent concealment test, concealment of the cause of
action, the Court focuses on the well-pled facts that plaintiffs have argued—albeit incorrectly—
support inferring a § 1 conspiracy during the pre-2012 period. As described above, plaintiffs’
claims during that period overwhelmingly center on Tradeweb, which defendants are claimed to
have used then as their vehicle for blocking the emergence of all-to-all IRS trading platforms.
Plaintiffs argue that there was fraudulent concealment as to Tradeweb, because the Dealers
allegedly hid their majority ownership, inhibiting the class from learning that the Dealers, not
others, set the direction of Tradeweb’s platform.
The Tradeweb allegations do not support a finding of fraudulent concealment. Plaintiffs
do not claim that the nature of Tradeweb’s platform and offerings was ever concealed. And, as
defendants demonstrate, contrary to the SACs’ claim, the Dealers’ acquisition of a majority stake
in Tradeweb NewMarkets, the entity which housed Tradeweb’s IRS business, was disclosed
close to contemporaneously. The Dealers’ investment in Tradeweb NewMarkets was announced
in an October 11, 2007 press release. It was entitled “Nine Global Dealers and Thomson
Financial Form Premier Electronic Trading Venture Using Tradeweb.” SAC ¶ 118 n.33 (exhibit
citation omitted). And the press release named Thomson Financial as Tradeweb’s owner and

83

referred readers to Thomson’s SEC filings for further information. These, beginning in
November 2007, disclosed the Dealers’ majority (80%) stake in Tradeweb NewMarkets, and
identified Tradeweb NewMarkets as responsible for “asset class expansion,” which in turn were
defined to include “interest rate and credit default swaps.” See Dealer Mem. at 17–18 & nn.11&
14 (exhibit citations omitted). Later securities filings by Thomson, including in each of 2008,
2009, and 2010, again disclosed the Dealer investors’ majority stake. Id. at 18 & n.12.
In a separate argument as to the concealment element, plaintiffs allege that the Dealers’
pre-2012 scheme was “self-concealing,” because it was effectuated “through private meetings
under the guise” of the Dealers’ presence on boards such as those of Tradeweb and trade
associations. Certain conspiracies, by the nature of the wrong itself, are “self-concealing.”
Hendrickson Bros, Inc., 840 F.2d at 1083. Depending on the facts, a group boycott of exchange
trading, like a covert price fixing scheme, could be of such a nature, where the perpetrators’
collective venture “must remain concealed to be successful.” See In re CDS, 2014 WL 4379112,
at *15.
However, the pre-2012 scheme, given its unusual nature as alleged, is not of this nature.
There was, as yet, no all-to-all anonymous IRS trading exchange to boycott. Plaintiffs’ pre-2012
claim instead is, at heart, that the Dealers agreed to block Tradeweb from evolving into such an
exchange. Given the visible nature of Tradeweb’s trading platform and of the Dealers’ majority
ownership, the direction they gave to Tradeweb was not “self-concealed.” Moreover, plaintiffs’
logic is that, prior to 2012, the benefits of all-to-all trading were already “widely recognized by
market participants, regulators, and economists” such that “absent a conspiracy . . . [e]volution to
all-to-all trading that would have been open to the entire market would have been inevitable[.]”
SAC ¶¶ 87, 342; see also id. ¶ 99 (“There were . . . no natural or technological reasons why the

84

IRS market did not evolve by 2008, at the latest, to allow the buy side to conduct all-to-all
trading of IRS on electronic platforms.”). On plaintiffs’ theory, the failure of Tradeweb to
evolve into an all-to-all IRS trading exchange occurred in plain sight and in contrast to the
market’s expectations. This, if anything would have invited, rather than lulled, skeptical
attention. See In re Publ’n Paper Antitrust Litig., No. 304-MD-1631 (SRU), 2005 WL 2175139,
at *4 (D. Conn. Sept. 7, 2005) (“[T]he circumstances that are alleged . . . point to the opposite
conclusion. The plaintiffs allege that the price increases were not explainable by ordinary
market forces[.]”).37
Class plaintiffs’ allegation that the pre-2012 conspiracy was accomplished under the
“cover” of Dealers’ memberships in trade associations does not change this result. The SAC’s
claims of secret meetings under the aegis of such groups to further the pre-2012 plot are general
and conclusory. These ill-pled allegations cannot discharge plaintiffs’ burden to show
concealment. Cf. Precisions Assocs., Inc. v. Panalpina World Transp. (Holding) Ltd., No. 08
Civ. 42 (JG) (VVP), 2011 WL 7053807, at *50 (E.D.N.Y. Jan. 4, 2011) (rejecting, as basis for
fraudulent concealment, claim that defendants “formulated the conspiracy during secret
meetings”).
For similar reasons, class plaintiffs fail to plead ignorance of their pre-2012 claims—the
second element of fraudulent concealment. For better or worse, plaintiffs’ express claim as to
this period is that the failure of the IRS market to move towards all-to-all anonymous exchange
trading in this period was unnatural and contrary to expectations, see, e.g., SAC ¶¶ 87, 99,
suggesting conspiratorial manipulation. In fact, as the Court has held, that claim is implausibly

37

To the extent plaintiffs’ claims for the pre-2012 period separately involve OTCDerivNet, the
Dealers’ control of that joint venture, dating to the early 2000s, was also public knowledge. See,
e.g., Dealer Mem at. 74 n.88 (exhibit citation omitted).
85

pled. The absence of such platforms until 2013 instead is plausibly ascribed to the lack of
critical infrastructure such as central clearing. But, accepting class plaintiffs’ premise that only a
plot can explain the missing platforms, class plaintiffs had every basis, in real time, to smell a
rat. At a minimum, they were on inquiry notice. “[A]ll that is necessary to cause the tolling
period to cease is for there to be reason to suspect the probability of any manner of wrongdoing.”
131 Maine St. Assocs. v. Manko, 179 F. Supp. 2d 339, 348 (S.D.N.Y. 2002) (courts “equat[e]
suspicion with knowledge in the context of fraudulent concealment” and “only inquiry notice is
necessary”); see also In re Processed Egg Prods. Antitrust Litig., 2012 WL 6645533, at *3 (“[I]n
pleading fraudulent concealment, plaintiffs must plausibly suggest[] that they did not have
knowledge or did not possess any information about the conspiracy that would have given rise to
inquiry notice.”).38
As to the third required element of fraudulent concealment, due diligence, defendants
argue that class plaintiffs fail to allege making any inquiries in response to the publicly available
information suggestive of a pre-2012 claim, and instead make only generalized “boilerplate
recitals” about an “investigation.” Defendants note that plaintiffs do not allege inquiring of
anyone about the pre-2012 period or following up on the public domain materials from that
period cited in the SAC. See In re Merrill Lynch Ltd. P’ships Litig., 154 F.3d 56, 60 (2d Cir.
1998) (upholding dismissal where plaintiffs “make no allegation of any specific inquiries of

38

Class plaintiffs based their initial Complaint, like the SAC, on voluminous news articles and
other public sources, many published well before November 25, 2011. This too supports a
finding that plaintiffs were on inquiry notice, as a plaintiff may not “simultaneously claim that
the generalized evidence cited as the basis of its complaint—the vast majority of which involves
factual allegations published prior to the [limitations period]—is insufficiently detailed to state a
cognizable claim for relief and that, nevertheless, these facts were insufficiently particular to
cause the statute of limitations to run.” Woori Bank v. Merrill Lynch, 923 F.Supp. 2d 491, 497
(S.D.N.Y. 2013) (dismissing complaint as time-barred where it relied on “extensive publicly
available information” to support its claims), aff’d, 542 F. App’x 81 (2d Cir. 2013).
86

Merrill Lynch, let alone detail when such inquiries were made, to whom, regarding what, and
with what response”). Class plaintiffs counter generally that they monitored their investments,
reviewed news reports, and consulted with sophisticated investment managers who monitored
their investments, and that none of these sources alerted them of a possible conspiracy. They
claim that they first realized in June 2014 that a conspiracy had long been afoot. Class Mem. at
103–04.
Although defendants raise substantial questions about the adequacy of class plaintiffs’
pre-2014 diligence, the Court elects not to resolve this third element at this time. There is no
reason to do so, with the Court’s having resolved the first two elements of fraudulent
concealment in defendants’ favor, and with the pre-2012 § 1 claims having been separately
dismissed for failure to state a plausible claim. The Court is, moreover, mindful of the
complexity of this data-rich case, as exemplified by the source-heavy 145-page SAC. To resolve
reliably the adequacy of the class’s diligence would entail a sustained inquiry that is unnecessary
and unproductive to undertake at this time.
The Court, therefore, based on its findings as to the first two elements, holds that class
plaintiffs have not established fraudulent concealment. Accordingly, class plaintiffs’ claims
based on injuries incurred before November 25, 2012 are time-barred.39
VI.

Antitrust Standing
Section 4 of the Clayton Act establishes a private right of action to enforce § 1 of the

Sherman Act, and entitles “[a]ny person who [is] injured in his business or property by reason of
anything forbidden in the antitrust laws” to treble damages for those injuries. Gatt Commc’ns,

39

In light of this holding and the dismissal of the pre-2013 claims for failure to state a claim, the
Court has no occasion to reach defendants’ alternative argument that class plaintiffs have not
alleged an injury-in-fact with respect to the pre-2013 period.
87

Inc. v. PMC Assocs., LLC, 711 F.3d 68, 75 (2d Cir. 2013). Applying the decision in Associated
General Contractors of Cal., Inc. v. Cal. State Council of Carpenters, 459 U.S. 519 (1983)
(“AGC”), the Second Circuit has held that “a private antitrust plaintiff [must] plausibly [] allege
(a) that it suffered a special kind of antitrust injury, and (b) that it is a suitable plaintiff to pursue
the alleged antitrust violations and thus is an ‘efficient enforcer’ of the antitrust laws.” Gatt, 711
F.3d 68, 76 (2d Cir. 2013) (citations and internal quotations omitted). Such “antitrust standing is
a threshold, pleading-stage inquiry.” Id. at 75–76 (citation omitted).
Defendants do not dispute that Javelin and Tera have antitrust standing. Defendants also
do not dispute that the injury the class plaintiffs claim to have suffered—higher prices on IRSs as
a result of an anti-competitive boycott of the nascent all-to-all trading platforms—qualifies as an
antitrust injury.40 Defendants dispute, however, that the class plaintiffs meet the second
requirement for antitrust standing: that they are “efficient enforcers” of the antitrust laws.
To determine whether a putative antitrust plaintiff is an “efficient enforcer” of the
antitrust laws, courts examine the following factors:
(1) the directness or indirectness of the asserted injury; (2) the existence of an
identifiable class of persons whose self-interest would normally motivate them to
vindicate the public interest in antitrust enforcement; (3) the speculativeness of the
alleged injury; and (4) the difficulty of identifying damages and apportioning them
among direct and indirect victims so as to avoid duplicative recoveries.
Gatt, 711 F.3d at 78 (quoting Paycom Billing Servs., Inc v. Mastercard Int’l, Inc., 467 F.3d 283,
290–91 (2d Cir. 2006)).

40

It clearly does. “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.’” Gelboim, 823 F.3d at 772 (quoting Brunswick Corp. v. Pueblo Bowl-OMat, Inc., 429 U.S. 477, 489 (1977)).
88

The importance assigned to these factors “will necessarily vary with the circumstances of
particular cases.” Id. (quoting Daniel v. American Bd. of Emergency Med., 428 F.3d 408, 443
(2d Cir. 2005). However, in other contexts, the Supreme Court has noted that the first factor,
requiring proximate causation, “must be met in every case.” Lexmark Int’l, Inc. v. Static Control
Components, Inc., 134 S. Ct. 1377, 1392 (2014). In contrast, the third and fourth factors are
“problematic,” and the “‘potential difficulty in ascertaining and apportioning damages is not . . .
an independent basis for denying standing where it is adequately alleged that a defendant’s
conduct has proximately injured an interest of the plaintiff’s that the statute protects and other
relief might be available.” DNAML PTY, Ltd. v. Apple, Inc., 25 F. Supp. 3d 422, 430 (S.D.N.Y.
2014) (quoting Lexmark, 134 S. Ct. at 1392) (emphasis in original); Commodity Exch., 213 F.
Supp. 3d at 653–54; In re CDS, 2014 WL 4379112, at *8.
Defendants here raise challenges as to each factor.
A.

Indirect Injury

“Directness in the antitrust context means close in the chain of causation.” Gatt, 711
F.3d at 78 (citation omitted). It is essentially a proximate cause analysis. Lotes Co. v. Hon Hai
Precision Indus. Co., 753 F.3d 395, 412 (2d Cir. 2014) (inquiring whether plaintiffs’ injury is
“within the scope of the risk created by” defendants’ alleged conduct so as to be a natural or
probable consequence of that conduct); Blue Shield of Va. v. McCready, 457 U.S. 465, 480–84
(1982) (inquiring whether plaintiffs’ injury is a “type of loss” the alleged conspiracy “would be
likely to cause”).
Class plaintiffs meet that standard. They allege that the conspiracy’s prevention of all-toall exchange trading on the nascent platforms injured IRS buy-side investors by leaving them
without an alternative to the wider bid/ask spreads characteristic of the historical RFQ trading

89

mode. See, e.g., SAC ¶¶ 323–24 (alleging that “[c]ompetition lowers costs” because “the more
competition for business, the better quotes the dealer tend to give and the narrower the spreads);
id. ¶¶ 21, 319–39, 401. As alleged, this scheme proximately and predictably harmed buy-side
investors who were denied the superior prices of an allegedly tighter-priced trading platform.
See McCready, 457 U.S. at 480–84 (class of healthcare consumers had standing to challenge
group health plan’s exclusion of psychologists, not psychiatrists; consumers “within that area of
the economy” were endangered by the “breakdown of competitive conditions” and alleged an
injury that “was inextricably intertwined with the injury the conspirators sought to inflict” on the
excluded competitors); In re DDAVP Direct Purchaser Antitrust Litig., 585 F.3d 677, 688–89
(2d Cir. 2009) (although defendants’ conduct targeted their competitors, direct purchasers of
drug had standing because their “claimed injury of higher prices was ‘inextricably intertwined’
with the conduct’s anti-competitive effects and thus ‘flow[ed] from that which makes
defendants’ acts unlawful’”) (quoting McCready, 457 U.S. at 484); In re CDS, 2014 WL
4379112, at *9 (conspiracy aimed at impeding exchange trading of CDS market alleged direct
injury to investors); cf. New York Citizens Comm. On Cable TV v. Manhattan Cable TV, Inc.,
651 F. Supp. 802, 811 (S.D.N.Y. 1986) (“Consumers have standing when they are injured as a
result of a defendant’s improper exclusion from the market.”).
As in In re CDS, the defendants rely on Paycom. There, a merchant, Paycom, alleged it
had been injured by MasterCard’s competitive programs policy (“CPP”), which prohibited
MasterCard member banks from acting as issuers or acquirers for any other payment-card
network, with the exception of Visa. Paycom claimed that, absent the CPP, American Express
and Discover would have expanded their networks, and that increased competition from
American Express and Discovery would have caused MasterCard to adopt policies more

90

favorable to merchants like Paycom. Paycom, 467 F.3d at 288, 293. Rejecting the claim of
antitrust standing, the Second Circuit noted that the CPP did not prevent Paycom from accepting
Discover or American Express cards as payment options, and elimination of the CPP would have
benefitted Paycom only through the increased use of Discover and American Express cards.
“Consequently, any injury suffered by Paycom was indirect and flowed from the injuries
suffered by Discover and American Express.” Id. at 293.
As Judge Cote recognized in In re CDS, Paycom presents a more attenuated causal chain
than a claim of a dealer conspiracy to block a more price-competitive exchange from emerging:
There, absent the CPP, MasterCard member banks might have decided to act as
issuers or acquirers for American Express and Discovery; American Express and
Discovery night have increased their networks; MasterCard might have felt
pressure to compete for merchants; and MasterCard might have changed some of
its policies to be more merchant-friendly. Here, by contrast, plaintiffs allege that,
absent defendants’ agreement to prevent a CDS exchange, [a new exchange for
CDS] would have entered the market, immediately allowing plaintiffs to avoid
trading directly with Dealer Defendants and paying their inflated bid/ask spreads.
In re CDS, 2014 WL 4379112, at *8.
The other cases on which defendants rely also involve distended causal chains between
the alleged misconduct and the asserted injury. Apart from other complexities, the theories of
injury in each of these cases “depend[ed] upon a complicated series of market interactions”
across multiple markets. Reading Indus., Inc. v. Kennecott Copper Corp., 631 F.2d 10, 13 (2d
Cir. 1980); see also id. (misconduct occurred in refined copper market; injury alleged to
participant in copper scrap market; and “[t]o establish a causal chain, the actions of innumerable
individual decision-makers must be reconstructed”); In re Aluminum Warehousing Antitrust
Litig., 13-MD-2481 (KBF), 2016 WL 5818585 (S.D.N.Y. Oct. 5, 2016) (market for aluminum
warehouse services versus market for physical aluminum); Boyd v. AWB Ltd., 544 F. Supp. 2d
236, 245 (S.D.N.Y. 2008) (Iraqi wheat market versus U.S. wheat market; plaintiff’s claim
91

implicated a “multitude of intervening developments” between asserted antitrust violation and
injury); de Atucha v. Commodity Exch., Inc., 608 F. Supp. 510 (S.D.N.Y. 1985). This case, in
contrast, involves a single market.
B.

Other Plaintiffs

This factor “simply looks for a class of persons naturally motivated to enforce the
antitrust laws,” In re DDAVP Direct Purchaser Antitrust Litig., 585 F.3d at 689, and asks
whether the plaintiff group’s self-interest would “motivate [it] to vindicate the public interest in
antitrust enforcement.” Paycom, 467 F.3d at 294 (quoting AGC, 459 U.S. at 542). Although
there is no categorical rule to this effect, buyers and sellers of securities are often held to be
proper plaintiffs to vindicate § 1 claims of collusion affecting the securities markets. See, e.g.,
Alaska Electrical Pension Fund v. Bank of America Corp., 175 F. Supp. 3d 44 (S.D.N.Y. 2016)
(purchasers and sellers of derivatives affected by manipulated LIBOR rates); Commodity Exch.,
213 F. Supp. 3d at 656–57 (purchasers of futures and gold on exchange market); In re CDS, 2014
WL 4379112, at *7–9 (purchasers of credit-default swaps).
Defendants here do not contend that the class plaintiffs, were they the only plaintiffs in
this consolidated action, would fail this standard. They clearly meet it: The class plaintiffs claim
to have suffered a direct and consequential financial injury from defendants’ alleged boycott of
the nascent IRS trading platforms. As in the above cases, this will “adequately motivate them to
vindicate the public interest in antitrust enforcement.” Alaska Electrical Pension Fund, 175 F.
Supp. 3d at 61.
Defendants instead argue that because Javelin and Tera have also brought suit, and
because their injuries, as the allegedly boycotted platforms, were more direct than those of the
class, that Javelin and Tera are “the more efficient enforcers.” Therefore, defendants argue, the

92

class plaintiffs are “not necessary” and dropping them as superfluous “will not “leave an alleged
antitrust violation . . . unremedied.” Dealer Mem. at 54 (quoting In re CDS, 2014 WL 4379112,
at *9).
This argument is not persuasive. Courts have indeed often inquired whether there is a
more direct victim of the asserted violation than the plaintiff. Where the “superior” victim has
not brought suit, an inference may arise that the plaintiff’s claim is dubious. See, e.g., AGC, 459
U.S. at 542 n.47 (“[I]f there is substance to [the plaintiffs] claim, it is difficult to understand why
the[] direct victims of the conspiracy have not asserted any claim in their own right.”); cf. Phillip
Areeda & Herbert Hovenkamp, Fundamentals of Antitrust Law, § 3.01c, at 3-9 to 3-10 (4th ed.
2011) (“If the ‘superior’ plaintiff has not sued, one may doubt the existence of any antitrust
violation at all.”) (both quoted in Gatt, 711 F.3d at 79). The existence of a superior victim does
not, however, require dismissal. “Inferiority to other potential plaintiffs can be relevant, but it is
not dispositive.” In re DDAVP Direct Purchaser Antitrust Litig., 585 F.3d at 689. And
defendants cite no case dismissing an otherwise qualified plaintiff for lack of antitrust standing
on the ground that a co-plaintiff was a superior enforcer; or dismissing investor or purchaser
plaintiffs on the ground that a fellow plaintiff had standing as a competitor.
In any event, here, effective enforcement of the antitrust laws is enhanced, not inhibited,
by continued collaboration of the class and the Javelin/Tera plaintiffs. Class plaintiffs, through
counsel, were responsible for much of the investigation on which this action is based and filed
the initial complaint. Class counsel has played a lead role in advocacy and appears to have
coordinated well with counsel for Javelin and Tera, which have, as to a number of issues,
adopted class plaintiffs’ agreements. And, with the case now moving to discovery, it is possible
that these SEF plaintiffs, proceeding alone, might prove unable to fund this litigation at a level

93

equal to the heavy demands that may lie ahead. Partnership with the class, and the resources that
come with it, will help assure that the public interest in vigorous antitrust enforcement is
vindicated.
C.

Speculative Injury

Defendants argue that it is unduly speculative to posit that, absent collusion among the
Dealers, the class plaintiffs would have experienced injury. Two aspects of that argument have
merit.
First, class plaintiffs’ theory of pre-2013 injury is extraordinarily conjectural, such that,
had the § 1 claim as to that period not otherwise have been dismissed as implausible (and had the
pre-2012 aspects of that claim not been dismissed as time-barred), the claim would not satisfy
the requirements of antitrust standing. Among other infirmities, it is “entirely uncertain,” Gatt,
711 F.3d at 79, that, absent the scheme, the necessary infrastructural preconditions for
anonymous all-to-all trading, such as central clearing of IRS trades, would have developed
before Dodd-Frank willed them into being in 2013. Plaintiffs’ alternative history of IRS trading
for the first five years of the class period (2008–2012) requires too many leaps of imagination
and guesswork for a claim of class injury to be viable. See Reading, 631 F.2d at 10 (no antitrust
standing where indirect purchaser at end of vertical distribution line “predicate[d] its claim of
injury on a basis too tenuous and conjectural for a valid causal finding of anticompetitive effect
and damages”); see also Paycom, 467 F.3d at 293 (no antitrust standing where chain of causation
was “highly speculative” and was built on “conclusory allegation[s]”).
Second, as to the 2013–2016 period, class plaintiffs’ theory of injury must be limited to
those IRS trades that were suited by nature to anonymous all-to-all exchange trading permitting
the buy-side to draw upon an order book. On plaintiffs’ theory, “bespoke” trades—trades with

94

customized or idiosyncratic features requiring individualized negotiation—or trades otherwise
defying commoditization would not have fit on or benefitted from an all-to-all anonymous
exchange platform. Consider, for example, a municipality seeking to enter into an IRS to contain
the risk presented by 30-year bonds it had issued with floating interest rates. Depending on the
bonds’ terms and the municipality’s specifications for its IRSs, it might have made no sense to
turn to an anonymous exchange like Javelin, Tera, or TrueEx to fill the municipality’s bid, even
assuming such exchanges had flourished. The order books of such platforms might have lacked
counterparty offers aligning with the municipality’s idiosyncratic needs. To fill its bid, the
municipality might have no choice but to turn to the more dynamic, negotiation-based, RFQ
process, in which the buy-side or a broker acting on its behalf sought out multiple offers tailored
to align with the required specifications.
The SAC, however, does not exhibit such self-restraint. It sweepingly defines the class
as consisting of “[a]ll persons or entities who, from January 1, 2008 to the present, directly
entered into fixed-for-floating, floating-for-fixed, or floating-for-floating interest rate swaps with
the Dealer Defendants” or their affiliates. SAC ¶ 385. At argument, class counsel, while broadly
asserting that the presence of of all-to-all trading platforms would have benefitted the entire buyside by sparking “transparency and competition throughout the interest rate swaps market,”
stated that the class was intended only to be limited to “plain vanilla swaps.” Tr. 88. The Court
will hold class plaintiffs to that necessary commitment. As to any IRS trades other than “plain
vanilla” trades, any claim that defendants’ alleged boycott of the Javelin, Tera, and TrueEx
platforms caused price injury would require the finder of fact unacceptably to “pile inference
upon inference,” Lopez v. United States, 514 U.S. 549, 567 (1995), starting with the counter-

95

intuitive premise that a bespoke trade would have been filled by means of drawing,
anonymously, on an order book.41
With that vital limitation, class plaintiffs’ theory is not unduly speculative. Plaintiffs cite
empirical research, academic analyses, and analyst reports of the effects of migrating securities
trading (including of fixed-income products) to exchanges. These, plaintiffs claim, reveal that
“significant spread compression,” and price savings for the buy-side, has followed from these
transitions. SAC ¶¶ 319–39. See In re CDS, 2014 WL 4379112, at *9 (finding claim of injury in
CDS market non-speculative based on similar source materials).
D.

Potential for Duplicative Recoveries

The final factor inquires whether class plaintiffs’ claims invite duplicative recoveries or
problems of apportionment.
There are no problems of duplicative recovery. Class plaintiffs seek damages for price
overcharges (based on the difference between the bid/ask spreads to which they were subject and
the tighter bid-ask spreads that they posit would have inhered). The SEF plaintiffs, Javelin and
Tera, in contrast, seek lost profits. Compare In re CDS, 2014 WL 4379112, at *9. These
measures are different.
Defendants argue that apportioning damages “would be prohibitively complex.” That is
not necessarily so. The two types of damages are different. Whether at trial or in settlement
discussions, the parties presumably would tabulate each type of damages based on the salient
metrics. Defendants claim that apportioning damages would be challenging because class and

41

“Plain vanilla,” of course, is not a defined term (at least in the context in which it is used here).
The Court expects that in discovery, the parties will give due attention to the characteristics of
IRS trades (e.g., their terms and the necessary number of like-termed bids or offers pending
during a given time period) to make them suitable for exchange trading.
96

Javelin/Tera damages might prove “negatively correlated,” on the theory that, had the SEFs
succeeded and the Dealers’ profit margins shrunk, benefitting the buy-side, the Dealers might
have stepped back from some IRS business. That is possible. It is also possible that the Dealers
would have maintained their level of IRS trading, even at lower profits. The possibility of this
modest degree of tension between the damages theories of the two groups of plaintiffs does not
justify eliminating the claims of the class plaintiffs.
E.

Overall Assessment

For the reasons above, the Court holds that class plaintiffs—subject to the limitations
addressed above—are “efficient enforcers.” Like Javelin and Tera, they have antitrust standing
to pursue the § 1 claims.
VII.

Preclusion by Dodd-Frank
Defendants next argue that the Commodities Exchange Act (CEA) and the 2010 Dodd-

Frank Act impliedly preclude plaintiffs’ post-Dodd-Frank claims.
In Credit Suisse Securities (USA) LLC v. Billing, 551 U.S. 264 (2007), the Supreme
Court recognized that application of the antitrust laws may be implicitly precluded by another
federal regulatory regime. At issue in Billing was an antitrust action against underwriting firms
that marketed and distributed shares in initial public offerings (“IPOs”). The plaintiffs alleged
that the underwriters had unlawfully agreed together not to sell shares of a popular new issue
unless the buyer committed to buy additional shares of that security later at escalating prices, to
pay unusually high commissions on later purchases, or to purchase from the underwriters other
less desirable securities. Id. at 267. The Supreme Court ruled that federal securities law
precluded applying the antitrust laws, including Sherman Act § 1, to the underwriters’ alleged
anticompetitive conduct. The Court identified four considerations that bear upon whether “the

97

securities laws are ‘clearly incompatible’ with the application of the antitrust laws” in a
particular context: (1) location within the heartland of securities regulation; (2) SEC authority to
regulate, (3) ongoing SEC regulation, and (4) conflict between the two regimes. Id. at 285.
Two years later, the Second Circuit reached a similar result in Electronic Trading Group
v. Banc of America Securities LLC, 588 F.3d 128 (2d Cir. 2009). Applying the four “Billing
factors,” the Circuit held precluded § 1 and other antitrust claims brought by a class of short
sellers against financial institutions that served as “prime brokers” in short-sale transactions. Id.
at 138.
Drawing on these decisions, defendants argue that, if the Billing factors are considered
here, they reveal Congress’s implicit intent to preclude application of plaintiffs’ § 1 claims. The
Billing framework, however, is inapposite. As the Supreme Court explained in Billing, the
Billing test, inquiring “whether, and in what respects, they implicitly preclude application of the
antitrust laws,” is reserved for cases “[w]here regulatory statutes are silent with respect to
antitrust.” 551 U.S. at 271 (emphasis added). In such circumstances, the Billing factors guide a
court’s inquiry as to whether “there is a ‘plain repugnancy between the antitrust and regulatory
provisions,’” id. at 272 (quoting Gordon v. New York Stock Exch., 422 U.S. 659, 682 (1975)
(internal citations omitted), that is, “whether the two are ‘clearly incompatible.’” Id. at 275
(citation omitted).
In contrast, the Supreme Court explained in Billing, “[s]ometimes regulatory statutes
state whether they preclude application of the antitrust laws.” Id. at 270. The Court gave an
example both of a statute that expressly provided for antitrust immunity (the Webb-Pomerene
Act, 15 U.S.C. § 62) and of one that expressly stated that the antitrust laws remain applicable
(the Telecommunications Act of 1996, 47 U.S.C. § 152). Id. When Congress has spoken, the

98

Supreme Court stated, a court is to apply Congress’s command as to the extent, if any, to which
antitrust laws are abrogated. Id. (citing Verizon Communications, Inc. v. Law Offices of Curtis
V. Trinko, LLP, 540 U.S. 398, 406–07 (2004) (“Trinko”) (analyzing the antitrust saving clause of
the Telecommunications Act).
Unlike the statutes at issue in Billing and Electronic Trading Group but like that at issue
in Trinko, Dodd-Frank expressly addresses whether it precludes operation of the antitrust laws.
As defendants acknowledge, Dodd-Frank includes an “antitrust savings clause”:
Nothing in this act . . . shall be construed to modify, impair, or supersede the
operation of any of the antitrust laws, unless otherwise specified. For purposes of
this section [with some exceptions not relevant here], the term “antitrust laws” . . .
mean[s] [the Sherman Act, parts of the Wilson Tariff Act, the Act amending the
Wilson Tariff Act, and the Clayton Act].
12 U.S.C. § 5303; see also 15 U.S.C. § 12(a). There is, therefore, no charter for undertaking a
Billing inquiry into implied preclusion. See Trinko, 540 U.S. at 406 (antitrust savings clause
“bars a finding of implied immunity”). The question instead is one of statutory interpretation:
specifically, given the text of the antitrust savings clause, “whether, and where, Dodd-Frank has
‘otherwise specified’ that it is modifying sections of those antitrust statutes.” In re CDS, 2014
WL 4379112, at *16 (analyzing similar claim that Dodd-Frank precludes § 1 boycott claim).
In In re CDS, which involved a similar defense claim that a § 1 claim of collusion among
swap dealers was precluded by Dodd-Frank, Judge Cote addressed this question. Dodd-Frank,
she noted, does not elsewhere mention the Sherman Act (or the Wilson Tariff). And it explicitly
modifies the Clayton Act in four provisions, none relevant there. Id. at *16 (statutory citations
omitted). Because these four provisions were the only parts of Dodd-Frank to refer to the
antitrust laws, Judge Cote held, Dodd-Frank did not displace operation of the antitrust laws,
which therefore “apply with full force to the conduct at issue here.” Id.

99

On its independent review, the Court adopts this analysis. No other part of Dodd-Frank
“specifie[s]”—or references, either generally or by name—the existing antitrust laws. And the
four references to the Clayton Act are equally irrelevant here as in In re CDS. As a matter of
plain language, the exception to Dodd-Frank’s clause preserving plaintiffs’ right to bring
antitrust claims is not implicated here.
In response, defendants, as in In re CDS, focus on a separate pair of provisions in DoddFrank. These, in parallel, address the duties of swap dealers, on the one hand, and security-based
swap dealers, on the other. They provide:
Antitrust considerations: Unless necessary or appropriate to achieve the purposes
of this chapter, a swap dealer [or security-based swap dealer] . . . shall not— (A)
adopt any process or take any action that results in any unreasonable restraint of
trade; or (B) impose any material anti-competitive burden on trading or clearing.
7 U.S.C. § 6s(j)(6); 15 U.S.C. § 78o-10(j)(6) (emphasis added). Defendants argue that these
provisions fall within the “unless otherwise specified” exception to the antitrust savings clause.
They argue that the introductory clause in these provisions—“[u]nless necessary or appropriate
to achieve the purposes of this chapter”—invites the same inquiry used in Billing to resolve
claims of implied preemption. And, defendants argue, the Billing factors require dismissal of
plaintiffs’ § 1 claims.
This argument is unpersuasive for two reasons.
First, the swap-dealer provisions do not “specif[y]” or say anything about—they simply
do not mention—existing antitrust laws, as required to override the antitrust savings clause. Nor
do they function to eclipse existing antitrust law. Rather, as Judge Cote recognized in In re CDS,
the thrust of these provisions is to impose additional duties on swap dealers “above and beyond
what the antitrust laws themselves require”:
[W]hereas the antitrust laws criminalize ‘contract[s], combination[s] . . . or
conspiracy[ies], in restraint of trade . . . among the several States, or with foreign
100

nations,’ 15 U.S.C. § 1, the antitrust-considerations provisions more broadly forbid
‘adopt[ing] any process or tak[ing] any action that results in any unreasonable
restraint of trade.’ 7 U.S.C. § 6s(j)(6); 15 U.S.C. § 78o-10(j)(6). It follows that the
carve-outs from the antitrust considerations provisions permit neglect of the
heightened antitrust considerations when necessary or appropriate to achieve the
purposes of Dodd-Frank, but do not permit neglect of the baseline antitrust laws.
2014 WL 4379112, at *17 (provisions impose on swap dealer “a duty to avoid taking actions that
could have antitrust implications, even if those actions fall short of actually violating the antitrust
laws”). In particular, the swap-dealer provisions go beyond existing antitrust law by forbidding
the swap dealer from engaging—even unilaterally—in conduct that “results in an unreasonable
restraint of trade” or “impose[s] any material anti-competitive burden on trading or clearing.” In
contrast, Sherman Act § 1—prohibiting conspiracies in “restraint of trade”—does not proscribe
unilateral conduct, but only joint action. It follows that the introductory clause merely negates
these additional duties where “necessary or appropriate to achieve the purposes of [DoddFrank].” It does not implicate the swap’s dealer’s duties under existing antitrust law.
Second, even if the introductory clause to the swap-dealer provisions had the potential to
override existing antitrust law, it would not invite the broad inquiry that defendants propose into
whether the Billing factors point towards implied preclusion. It would instead entail applying the
text of that clause to the allegations at issue: It would ask whether defendants’ alleged conduct
was “necessary or appropriate to achieve the purposes” of Dodd-Frank. That inquiry does not
avail defendants here. It cannot credibly be claimed that an inter-Dealer conspiracy to boycott
three new all-to-all trading platforms was “necessary or appropriate” to achieve Dodd-Frank’s
purposes.
In short, the antitrust savings clause applies here. Dodd-Frank preserves, rather than
precludes, plaintiffs’ § 1 claim.

101

The Court, finally, addresses a point pursued by the defense at argument. Defendants
expressed concern that plaintiffs have advanced and will advance propositions in support of their
§ 1 claim that are inconsistent with the CFTC’s implementation of Dodd-Frank. For example,
defendants asserted, plaintiffs have treated Dealers’ post-2013 retention of RFQ trading as proof
of a boycott, whereas the CFTC has expressly preserved RFQ trading as a permissible mode of
IRS trading that can promote efficiency and competitiveness. See, e.g., Tr. 38–39. The point is
well-taken. As this litigation progresses, the Court will not permit the parties to pursue
arguments and inferences inconsistent with the statutory and regulatory design, and will entertain
motions—e.g., in limine—to exclude such improper advocacy.
VIII. State-Law Claims
Defendants, finally, move to dismiss plaintiffs’ state-law claims.
A.

Donnelly Act

Javelin and Tera bring a claim under New York’s Donnelly Act, New York’s antitrust
statute, N.Y. Gen. Bus. Law § 340 et seq. JTSAC ¶¶ 405–10. “The Donnelly Act is patterned
after the Sherman Anti-Trust Act and is generally construed in light of federal precedent.” Great
Atl. & Pac. Tea Co. v. Town of E. Hampton, 997 F. Supp. 340, 352 (E.D.N.Y. 1998).
Defendants first pursue dismissal of the Donnelly Act claim on the ground that Javelin
and Tera have not stated a § 1 claim. The Court has, however, sustained the § 1 claim to the
extent that it covers the period 2013–2016, except as against defendants HSBC, ICAP, and
Tradeweb, as to whom the claim has been dismissed.
Defendants separately challenge the Donnelly Act claim on the ground that it does not
apply where “‘the conduct complained of principally affects interstate commerce, with little or
no impact on local or intrastate commerce.’” H-Quotient, Inc. v. Knight Trading Grp., Inc., No.

102

03 Civ. 5889 (DAB), 2005 WL 323750, at *4 (S.D.N.Y. Feb. 9, 2005) (quoting Two Queens v.
Scoza, 296 A.D.2d 302, 303–04 (N.Y. App. Div. 1st Dep’t 2002)). In fact, the JTSAC alleges
that the conspiracy affected interstate commerce, JTSAC ¶¶ 64, 402, and intrastate commerce,
id. ¶ 410. And the latter claim is made plausible by the allegations that plaintiffs’ principal
places of business are in New York, id. ¶¶ 29–32, that most defendants are based in New York,
id. ¶¶ 34–66, and that activities relating to the conspiracy occurred in New York. Id. ¶¶ 181,
310, 324, 327, 395; see also In re Digital Music Antitrust Litig., 812 F. Supp. 2d 390, 417
(S.D.N.Y. 2011) (impact on intrastate commerce adequately alleged where “many of the
Defendants are headquartered in New York City and/or incorporated in New York, and they
clearly conduct significant business in New York”).
The Court therefore grants the motion to dismiss Javelin and Tera’s Donnelly Act claims
against HSBC, ICAP, and Tradeweb in their entirety, but denies the motion to dismiss the
Donnelly Act claims against other defendants, while limiting these claims to the period 2013–
2016.
B.

Tortious Interference

Javelin and Tera bring a claim of tortious interference with business relations under New
York law. JTSAC ¶¶ 414–16. To establish such a claim, a plaintiff must show that “(1) there is
a business relationship between the plaintiff and a third party; (2) the defendant, knowing of that
relationship, intentionally interferes with it; (3) the defendant acts with the sole purpose of
harming the plaintiff, or, failing that level of malice, uses dishonest, unfair, or improper means;
and (4) the relationship is injured.” E.g., Goldhirsh Grp. v. Alpert, 107 F.3d 105, 108–09 (2d
Cir. 1997); see also, e.g., Innovative Biodefense, Inc. v. VSP Techs., Inc., 176 F. Supp. 3d 305,
328 (S.D.N.Y. 2016) (same). “This tort is a difficult one to sustain, with requirements more

103

demanding than those for interference with [the] performance of an existing contract.” PKG
Grp., LLC v. Gamma Croma, S.p.A., 446 F. Supp. 2d 249, 251 (S.D.N.Y. 2006).
This Court dismisses this claim as inadequately pled. As defendants note, the JTSAC’s
articulation of this claim consists of three boilerplate sentences. The JTSAC does not identify
any customer with whose prospective business relations with Javelin and Tera the Dealer
Defendants sought to interfere or the particular nature of the prospective business relations.
Although the JTSAC—to the extent that it alleges a viable § 1 claim—adequately pleads the
wrongful purpose of impairing Javelin and Tera’s trading platforms and hence generally
harming their businesses, the JTSAC does not allege any particular customer relationship of
those plaintiffs with which the Dealer Defendants’ actions interfered. See, e.g., Commercial
Data Servers, Inc. v. Int’l Bus. Mach. Corp., 166 F. Supp. 2d 891, 898 (S.D.N.Y. 2001)
(“Plaintiff must specify some particular, existing relationship through which plaintiff would have
done business but for the allegedly tortuous behavior. A general allegation of interference with
customers without any sufficiently particular allegation of interference with a specific contract or
business relationship does not state a claim.”); see also Diario El Pais, S.L. v. The Nielson Co.,
(US), Inc., No. 07 Civ. 11295 (HB), 2008 WL 4833012, at *7 (S.D.N.Y. Nov. 6, 2008) (“To
render Plaintiffs’ tortuous interference claim ‘plausible,’ Plaintiffs must provide some factual
allegations that but-for Defendant’s alleged acts, Plaintiffs would have entered into contracts
with specific prospective advertisers.”); Stephens v. Trump Org. LLC, 205 F. Supp. 3d 305, 311
(E.D.N.Y. 2016) (“[t]he failure to identify a specific business relationship with a third party is
fatal”).
C.

Unjust Enrichment

104

Both complaints bring a claim for unjust enrichment under New York law. SAC ¶¶ 402–
05; JTSAC ¶¶ 411–13. Under New York law, a claim for unjust enrichment requires that “(1)
[the] defendant was enriched, (2) at plaintiff’s expense, and (3) equity and good conscience
militate against permitting defendant to retain what plaintiff is seeking to recover.” Diesel Props
S.r.l. v. Greystone Bus. Credit II LLC, 631 F.3d 42, 55 (2d Cir. 2011) (quotation omitted). The
“‘essence’ of such a claim ‘is that one party has received money or a benefit at the expense of
another.’” Kaye v. Grossman, 202 F.3d 611, 616 (2d Cir. 2000) (quoting City of Syracuse v.
R.A.C. Holding, Inc., 285 A.D.2d 905, 905 (N.Y. App. Div. 4th Dep’t 1999)).
Defendants first argue that, to the extent the § 1 claims have been found deficient, the
unjust enrichment claims must fall. That is correct, because, without a viable underlying claim
of illegality, an unjust enrichment claim “must be dismissed.” See, e.g., Kramer v. PollockKrasner Found., 890 F. Supp. 250, 257 (S.D.N.Y. 1995); see also In re Aluminum Warehousing
Antitrust Litig., 2014 WL 4743425, at *4 (S.D.N.Y. 2014) (dismissing unjust enrichment claim
because it was “predicated on defendants’ alleged violations of antitrust or consumer protection
laws . . . which the Court has dismissed”), aff’d, 833 F.3d 151 (2d Cir. 2016).
Defendants next argue that the complaints do not allege that plaintiffs conferred a benefit
on defendants. The analysis of this claim differs by plaintiff group. The SAC alleges that the
named plaintiffs and every class member “directly traded IRS[s] with one more of the Dealer
Defendants,” SAC ¶ 403, thereby enabling these defendants to “obtain[] profits from the inflated
bid/ask spreads on IRS transactions with [the class plaintiffs],” id. ¶ 404. That allegation is
sufficient to state an unjust enrichment claim. As alleged, the Dealers benefitted directly at the
expense of the class—the wider the bid/ask spread enjoyed by the Dealers as a result of
squashing the more buyer-friendly trading platform, the more money kept by the Dealers. See In

105

re CDS, 2014 WL 4379112, at *18 (sustaining unjust enrichment claim by class of purchasers of
CDSs allegedly harmed by conspiracy to block CDS exchange trading).
The JTSAC’s boilerplate claim of unjust enrichment of the Dealers at Javelin and Tera’s
expense, presupposes a different mode of enrichment. With narrow exceptions not relevant here
(e.g., RBS’s brief provision of liquidity to Javelin), the JTSAC does not allege that the Dealer
Defendants traded on the Javelin or Tera platforms. Instead, fairly read, it alleges an indirect
theory of unjust enrichment, in that, by destroying Javelin and Tera, the Dealer Defendants cost
these new entrants per-transaction usage revenue, while enabling themselves to benefit from the
wider bid-ask spreads characteristic of other modes of IRS trading.
That claim is too indirect to state an unjust enrichment claim. As Judge Caproni recently
observed, courts “require proof that the defendant received a ‘specific and direct benefit’ from
the property sought to be recovered, rather than an ‘indirect benefit.’” In re Commodity Exch.,
213 F. Supp. 3d at 676 (quoting Kaye, 202 F.3d at 616). Although the plaintiff “need not be in
privity with the defendant to state a claim for unjust enrichment,” neither can the relationship
between the parties be “‘too attenuated to support such a claim.’” Id. (quoting Sperry v.
Crompton Corp., 863 N.E.2d 1012, 1018 (N.Y. 2007)); see also Reading Int’l, Inc. v. Oaktree
Capital Mgmt., 317 F. Supp. 2d 301, 334 (S.D.N.Y. 2003) (an unjust enrichment claim “requires
some type of direct dealing or actual, substantive relationship with a defendant”) (quoting
Redtail Leasing, Inc. v. Bellezza, No. 95 Civ. 5191 (JFK), 1997 WL 6034996, at *8 (S.D.N.Y.
Sept. 30, 1997)). On the basis of this line of authority, Judge Caproni dismissed a unjust
enrichment claim that was pendant to viable price-manipulation claims brought under the
Sherman Act and Commodities Exchange Act, on the grounds that that “[p]laintiffs have failed

106

to allege that they had any relevant relationship with the [d]efendants.” In re Commodity Exch.,
213 F. Supp. 3d at 677–78 (numerous case citations omitted).
The same line of authority requires dismissal here. There is no relationship, direct or
indirect, between the Dealers and Javelin and Tera. And the usage revenue of which Javelin and
Tera were allegedly deprived is different from the money the Dealers ostensibly gained (via wide
trading spreads).
The Court therefore grants the motion to dismiss the unjust enrichment claims brought by
Javelin and Tera; grants the motion to dismiss the unjust enrichment claims against defendants
HSBC, ICAP, and Tradeweb in their entirety; and denies the motion to dismiss the unjust
enrichment claims brought by the class plaintiffs, while limiting these claims to the period 2013–
2016.
CONCLUSION
For the reasons set out above:
(1)

The Court grants defendants’ motion to dismiss class plaintiffs’ claims of a

Sherman Act § 1 conspiracy to the extent covering the period 2008–2012. The Court denies
defendants’ motion to dismiss plaintiffs’ § 1 claims to the extent covering the period 2013–2016.
(2)

The Court grants defendants’ motion to dismiss the claims of plaintiffs Javelin

and Tera of tortious interference with business relations.
(3)

The Court denies defendants’ motion to dismiss the claims of plaintiffs Javelin

and Tera under the Donnelly Act, but limits these claims to the period 2013–2016.
(4)

The Court grants defendants’ motion to dismiss the claims of plaintiffs Javelin

and Tera of unjust enrichment. The Court denies defendants’ motion to dismiss the class
plaintiffs’ claims of unjust enrichment, but limits these claims to the period 2013–2016.

107