Case Information
UNITED STATES DISTRICT COURT
SOUTHERN DISTRICT OF NEW YORK
---------------------------------------------------------------------- X 10/25/2021
:
ISABEL LITOVICH, et al., :
:
Plaintiffs, :
: 20-cv-3154 (LJL) -v- :
: OPINION AND ORDER BANK OF AMERICA CORPORATION, et al., :
:
Defendants. :
:
---------------------------------------------------------------------- X
LEWIS J. LIMAN, United States District Judge:
Defendants, a group of investment banks and their affiliates [1] (collectively, “Defendants”), jointly move, pursuant to Federal Rule of Civil Procedure 12(b)(6), to dismiss Plaintiffs’ amended consolidated class action complaint at Dkt. No. 128 (the “Complaint” or “SAC”) for failure to state a claim.
For the following reasons, the motion to dismiss is granted.
BACKGROUND
The Court assumes the truth of the well-pled allegations of the Complaint and the documents incorporated therein for purposes of the motion to dismiss.
I. The Relevant Market
The Complaint relates to an alleged conspiracy in the secondary market for odd lots of corporate bonds (the “Relevant Market”). SAC ¶ 246.
Corporate bonds are debt instruments issued by corporations to raise funds for their operations; they are issued through individual offerings to one or more registered securities firms in the primary market. Id. ¶¶ 3, 59. The securities firms—the dealers—then trade these bonds with other dealers and investors in the secondary market. Id. ¶ 4. Corporate bonds are not traded on anonymous exchanges, but rather are generally traded individually with dealers in the secondary over-the-counter market. Id. ¶¶ 5, 65.
Within this market, corporate bonds are broken down into two categories based on the number of bonds included in the trade. “Round lots” consist of bond trades involving increments of 1,000 bonds, or that are greater than and divisible by $1 million in par value, while “odd lots” generally consist of bond trades involving fewer than 1,000 bonds, or that are less than $1 million in par value. Id. Because bonds from the same issue are fungible, odd lots of that issue can be combined into a round lot, and a round lot of that issue can be broken into odd lots. Id. Round-lot trades almost always involve institutional investors, whereas odd-lot trades are more likely to involve retail investors. Id. ¶¶ 81–85. Round-lot trades make up approximately 82% of total trading volume of corporate bonds in the United States. Id. ¶¶ 93, 116, 257 n.56.
In the secondary market, dealers trade in both round lots and odd lots, providing a bid price at which they are willing to purchase lots of a specific bond or an offer price at which they are willing to sell lots of a specific bond. Id. ¶ 6. Bid prices and offer prices are expressed as percentages of the bond’s par value. Id. ¶ 59. The difference between bid price and offer price—the bid-offer spread—becomes the profit dealers make on their trades. ¶ 7. For example, a bid-offer spread of 99/101 means that the dealer is willing to buy a bond at 99% of the bond’s par value, and is willing to sell the bond at 101% of the bond’s par value; the difference would be the dealer’s profit. ¶ 59.
Defendants are a group of ten investment banks—and their affiliates—who are dealers in the secondary corporate bond market, both in round lots and odd lots; Plaintiffs are three individuals, a trust, and a pension fund who traded in the Relevant Market.
II. The Alleged Conspiracy
Plaintiffs’ claim revolves around the allegation of “a conspiracy by Defendants from at least August 1, 2006 to the present . . . to restrain electronic advances in the marketplace that would have reduced transactional costs for investors in odd-lots of corporate bonds to the detriment of Defendants’ trading profits.” SAC ¶ 2; see also id. ¶ 279 (“Defendants have conspired and agreed with each other to engage in a group boycott as alleged above of certain odd-lot focused electronic trading platforms . . . that sought to increase pre-trade pricing transparency, allow all-to-all direct trading and/or anonymous trading, and/or otherwise promote pricing competition for odd-lot investors.”).
Although Plaintiffs do not bring a claim of price-fixing in the Complaint, see Dkt. No. 133 at 3, their anticompetitive theory is predicated on an allegation that Defendants trade odd lots at significantly higher bid-offer spreads than they do round lots: “Defendants display remarkable parallel pricing in terms of odd-lots versus round lots. Despite the high number of odd-lot trades executed by dealers and the fact that such trades are qualitatively identical to round lot trades, Defendants demand odd-lot investors, such as Plaintiffs . . . , pay spreads that are 25% to 300% higher than [those paid by] investors trading in round lots of the same issue.” SAC ¶ 10. Plaintiffs allege that “[n]o reasonable economic justification explains the magnitude of the pricing disparity between odd-lot and round lot trades of the same issue,” and that “[i]n a truly competitive market, multiple factors, such as advances in technology that improve pre-trade price transparency and dealers’ competitive desire to secure a greater share of the growing odd-lot market, suggest that Defendants should be narrowing their spreads on odd-lots toward parity with the already profitable round lot trades.” Id. ¶ 11. This central theory—that there is a huge discrepancy between the bid-offer spread for round-lot trades and odd-lot trades, and that this discrepancy has no reasonable economic justification and can only be explained by anticompetitive behavior—underlies Plaintiffs’ allegations of a conspiracy among Defendants.
A. The Group Boycott
Plaintiffs’ central claim is that Defendants “engaged in a joint scheme to boycott electronic trading platforms with increased pricing transparency, refusing to participate in them and provide them liquidity, even though gaining market share in the growing odd-lot market through such participation would be in each Defendants’ unilateral interest if they were not conspiring with one another.” Dkt. No. 133 at 2. Plaintiffs allege that “E-platforms have the ability to allow Plaintiffs and the Class to trade corporate bonds with greater transparency and significantly less cost, i.e. , with narrower bid-offer spreads. Therefore, in order to maintain wider spreads on odd-lot trades of corporate bonds, Defendants have engaged in a pattern of parallel conduct and anticompetitive collusion to restrict competition from those electronic platforms seeking to improve odd-lot pricing for bond investors and seeking to compete with Defendants in this market.” SAC ¶ 131. They allege that this collusive conduct included:
punishing those participants in the bond odd-lot markets that were engaging in trading activity that had the potential to narrow the bid ask spreads; investing in and acquiring control of various electronic platforms to ensure they did not improve pricing for old-lot investors (including one platform, TradeWeb, that was co-owned by all Defendants and used repeatedly to acquire and shut down platforms that threatened to provide pre-trade pricing transparency and increase pricing competition for retail odd-lot investors); engaging in a group boycott of other retail-focused (and therefore, odd-lot focused) electronic trading platforms; punishing others who attempted to offer support or liquidity to such retail-focused electronic trading platforms; denying liquidity to electronic platforms that might improve price competition for retail odd-lot investors despite the potential opportunity such platforms offer to increase each Defendant’s market share of odd- lot transactions; and using their market power to deny and/or delay access to essential facilities that competing retail-focused electronic platforms required to enter the secondary market for trading odd-lots of corporate bonds.
Id. ¶ 133. Plaintiffs’ allegations that Defendants punished those who threatened their conspiracy relate to two instances. First, Plaintiffs allege that “when odd-lot traders employed by InterVest engaged in trading that Salomon Smith Barney (later acquired by Citigroup) deemed to be ‘disruptive’ of the market, Salomon refused to do business with those traders. None of the other Defendants stepped in to do business with the traders, instead, as one would expect in a competitive market.” ¶ 137. Second, Plaintiffs allege that from 2015–2019, Blackrock used competitive regional banks and brokers such as First Tennessee for odd-lot trading, including on electronic platforms like MarketAxess, because they provided narrower spreads; Morgan Stanley and Citigroup threatened to limit their business with First Tennessee to penalize them for offering these narrower spreads. Id. ¶¶ 138–139.
The remainder of Plaintiffs’ broad group boycott allegations relate to six electronic trading platforms.
1. InterVest In the mid-1990s, InterVest Financial Services planned to debut a new electronic trading system for corporate bonds with anonymous, push-button trading on Bloomberg terminals, pursuant to an agreement with Bloomberg. SAC ¶ 141. However, once this was announced, bond dealers began to complain to Bloomberg that InterVest offered a service that competed with them. Id. ¶ 142. Bloomberg tried to withdraw from its agreement with InterVest, but after InterVest threatened legal action, the system launched on Bloomberg terminals in December 1996. Bloomberg terminated the service about a year later “due to pressure from Defendants.” Id. The Complaint does not include any specific allegations regarding this “pressure,” who specifically applied it, and how it was applied.
2. TradeWeb TradeWeb is an electronic platform that was founded in 1996, with initial funding by four of the defendants: Credit Suisse, Lehman Brothers (later acquired by Barclays), Saloman Smith Barney (later acquired by Citigroup), and Goldman Sachs. SAC ¶ 143. By 2004, Citigroup, Merrill Lynch, Morgan Stanley, JPMorgan, and Deutsche Bank held ownership interests in TradeWeb as well. Id. In 2004, Thompson Reuters purchased TradeWeb, but agreed to a joint ownership structure called “Project Fusion” that went into effect in January 2008, wherein Credit Suisse, Goldman Sachs, Lehman Brothers, Merrill Lynch, Morgan Stanley, JPMorgan, Deutsche Bank, and RBS would retain minority ownership stakes in TradeWeb; Citigroup joined this list in 2008 as well. Id. ¶¶ 150–151. According to at least one source, the trader defendants sold to Thomson Reuters because of “regulatory concerns over potential conflicts of interest and competition issues in dealer-owned networks”; the sale followed the issuance of antitrust civil investigative demands to similar electronic trading platforms. Id. ¶ 149.
Plaintiffs allege two separate kinds of anticompetitive behavior via TradeWeb. First, Defendants’ ownership of TradeWeb allowed them to use it as a “stalking horse” to “catch and kill” other electronic platforms that threatened to increase transparency and offer better pricing to retail odd-lot investors. Id. ¶ 152. Second, Defendants’ ownership of TradeWeb allowed them to block retail investors from the platform; “[t]o this day, in what can only be explained by Defendants’ refusal to support it as a competitive platform for odd-lot trades, TradeWeb does not allow access to retail investors to trade in odd-lot corporate bonds, and continues to maintain a dealer-to-dealer market structure rather than all-to-all trading.”
3. BondDesk BondDesk was a bond platform founded in 1999 that originally focused on smaller trades and investment advisors representing retail investors but was not directly open to retail investors. SAC ¶ 172. Over the next five years, BondDesk sold ownership stakes to fourteen major banks, including Goldman Sachs, Bank of America, JPMorgan, and Wells Fargo. Id. ¶ 173. In 2004, these fourteen banks held six out of eleven seats on the board of directors, including seats held by Brad Levy, who was affiliated with Goldman Sachs, and Matthew Frymier, who was affiliated with Bank of America. Id. ¶ 174. Plaintiffs allege that Defendants saw BondDesk’s innovations as threatening to their “supracompetitive profitability,” and therefore “conspired to use their positions on the BondDesk board to remove the existing management of BondDesk from their day-to-day leadership positions at the company in 2004.” Id. ¶¶ 175–176. Levy and Frymier led this effort—which entailed pressuring the management to leave by raising false concerns about existing stock option accounting treatment, something management would be held responsible for—by reaching out to BondDesk’s outside accounting firm and encouraging them to raise a red flag about the existing accounting treatment in exchange for referring additional clients to the firm. Id. ¶¶ 176–179. The firm agreed to do so, and the board, controlled by the board members affiliated with the bank owners, used this as an excuse to remove BondDesk’s management; subsequently, the accounting issue was resolved without any changes to existing procedure. ¶¶ 180–181.
In 2006, a private equity firm bought majority stake in BondDesk, which subsequently announced that it was extending its marketplace to institutional traders and portfolio managers, but not to retail investors directly. Id. ¶ 182. The platform became the primary bond-trading platform for retail odd-lot-sized trades for several major retail and institutional investment advisors, facilitating about a third of retail-sized trades. Id. ¶ 183. In 2011, BondDesk hired a new CEO who announced a plan to introduce a system for direct retail trading on BondDesk; he implemented BondWorks, which initially created workstations for advisors and brokers to access BondDesk and would “eventually” allow retail investors such access as well. Id. ¶ 184. BondDesk also partnered with another trading platform in 2011 and made this new combined service immediately available to its dealer clients but not retail investors. Id. ¶ 185.
Plaintiffs allege that Defendants were “threatened by these moves that would provide greater price transparency to retail odd-lots investors and allow retail investors the opportunity to trade outside of the Defendant-controlled and intermediated OTC market,” and responded by using TradeWeb—which, as described above, several Defendants had minority ownership interest in—to acquire BondDesk in 2013. Id. ¶ 186. Plaintiffs allege that Defendants, “(via TradeWeb) . . . closed off BondDesk access for retail investors,” except for indirect access. Id. ¶ 188. Plaintiffs never allege, however, that retail investors had direct access to BondDesk at any point prior to this acquisition. BondDesk became TradeWeb Direct, a platform open to institutional investors and dealers that allows trading in odd lots. Id. ¶ 187.
4. ABS/NYSE In 1976, the New York Stock Exchange (“NYSE”) introduced the Automated Bond System (“ABS”), which allowed trading in a variety of bonds, including corporate bonds. SAC ¶ 153. ABS failed to gain traction in the corporate bond market, however; by 2002, only 5% of corporate bonds were listed on ABS, and by 2006, only about 1% of corporate bond issues traded on ABS. Id. ¶ 154. In 2007, ABS was replaced by NYSE Bonds, which offered pre-trade transparency for investors on pricing. Id. ¶ 155. A 2014 study found that corporate bonds listed on NYSE Bonds had lower bid-offer spreads than similar bonds not listed on NYSE Bonds, and that these price effects were greatest for trades of less than $100,000. However, NYSE Bonds also failed to gain traction among dealers trading in corporate bonds. Id. ¶ 156.
Plaintiffs allege that this was due to “(a) a concerted boycott of the platforms by Defendants, and (b) collusive efforts by Defendants to deny or delay NYSE Bonds access to the Bloomberg TOMS facility, an essential venue necessary for any newcomer seeking to participate in and compete within the corporate bond market.” Id. ¶ 157. Although “[o]btaining access to Bloomberg TOMS should have taken a short period of time for NYSE Bonds, particularly given the significance and power of the New York Stock Exchange,” it took eighteen to nineteen months for NYSE Bonds to obtain this access. Id. ¶ 162. Defendants “used their market power and value to Bloomberg as Bloomberg Terminal customers to force Bloomberg to materially delay NYSE Bonds’ connection to the essential facility of Bloomberg TOMS,” and “forced Bloomberg to delay NYSE Bonds’ connectivity through Bloomberg TOMS by threatening to terminate or reduce their Bloomberg Terminal leases if Bloomberg failed to do so.” Id. ¶ 163.
5. Bonds.com Bonds.com is another company that sought to introduce an electronic trading platform. It launched a bond trading platform called BondStation in 2008 which was open to both retail and institutional investors; after three months, the company shifted its focus to the institutional segment “due to market conditions and other economic factors.” Id. ¶ 167 (internal quotation marks omitted). Plaintiffs allege that one of these market conditions was a group boycott of the platform by dealers. Id. In 2010, Bonds.com stopped using BondStation and offered a new platform called BondsPro, which aimed to offer institutional investors an alternative trading system for odd-lot fixed income securities. Id. ¶ 168. The platform continued to allow all-to-all, anonymous, exchange-style trading.
Plaintiffs allege that in 2012 and 2013, “Bonds.com sought order flow and participation on its platform from major corporate bonds dealers like Defendants,” but that “[n]one of the dealers would participate with Bonds.com and the Defendants monitored and policed their conspiracy to make sure there would be no defectors.” Id. ¶¶ 169–170. Specifically, Plaintiffs allege that Bank of America was interested in participating on BondsPro but was worried about “blowback it would suffer from other Defendants.” Id. ¶ 170. They allege that Bank of America said that it would only participate on the platform if at least one or two other large dealers did as well. Id. “As a result of this group boycott,” Bonds.com ran out of funding in 2013 and was sold in 2014. Id. ¶ 171.
6. Trading Edge Plaintiffs allege that most electronic platforms “failed within a few years,” and that even when they had success, “they were quickly acquired and shuttered by Defendant-backed platforms.” ¶¶ 191–192. As an example, Plaintiffs allege that Trading Edge, a startup trading platform that offered exchange-like electronic trading with anonymous matching on bond trades, had success in 1999–2000 and was subsequently acquired in 2001 by MarketAxess, an electronic trading platform that was open only to institutional investors and was founded by JPMorgan, among others. Id. ¶¶ 192–193. MarketAxess originally stated that it would integrate the anonymous trading with its current platform but decided to terminate the anonymous trading within seven months. Id. ¶ 194.
PROCEDURAL HISTORY
Plaintiffs filed a complaint against Defendants on April 21, 2020. Dkt. No. 1. On July 14, 2020, Plaintiffs filed an amended consolidated complaint. Dkt. No. 106. Defendants filed a joint motion to dismiss the amended consolidated complaint on September 10, 2020. Dkt. No. 116. Plaintiffs filed the second amended consolidated complaint on October 29, 2020. Dkt. No. 128. The Court denied Defendants’ pending motion to dismiss the amended consolidated complaint as moot. Dkt. No. 129. Defendants filed a new joint motion to dismiss the second amended consolidated complaint on December 15, 2020. Dkt. No. 130. Plaintiffs filed their response in opposition to the motion on January 28, 2021, Dkt. No. 133, and Defendants filed their reply in further support of the motion on March 15, 2021, Dkt. No. 136.
LEGAL STANDARD
To survive a motion to dismiss pursuant to Federal Rule of Civil Procedure 12(b)(6), a
complaint must include “sufficient factual matter, accepted as true, to ‘state a claim to relief that
is plausible on its face.’”
Ashcroft v. Iqbal
,
“A claim 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.”
“Determining whether a complaint states a plausible claim for relief will . . . be a
context-specific task that requires the reviewing court to draw on its judicial experience and
common sense.”
Id.
at 679. Put another way, the plausibility requirement “calls for enough fact
to raise a reasonable expectation that discovery will reveal evidence [supporting the claim].”
Twombly
,
Although the Court must accept all the factual allegations of a complaint as true, it is “not
bound to accept as true a legal conclusion couched as a factual allegation.”
Iqbal
, 556 U.S. at
678 (quoting
Twombly
,
“There is no heightened pleading standard in antitrust cases.”
In re Interest Rate Swaps
Antitrust Litigation
, (“
IRS I
”),
“A plaintiff’s job at the pleading stage, in order to overcome a motion to dismiss, is to
allege enough facts to support the inference that a conspiracy actually existed. . . . [T]here are
two ways to do this.”
Citigroup
,
“A horizontal agreement among competitors, the sort of pact alleged here, is commonly
based on claims of parallel conduct by the alleged co-conspirators.”
IRS I
,
DISCUSSION
Defendants move to dismiss Plaintiffs’ Complaint on four grounds. First, they argue that Plaintiffs have not pled a plausible boycott conspiracy. Second, they argue that the Complaint fails to connect any specific Defendant to the alleged conspiracy and instead engages in impermissible group pleading. Third, they argue that Plaintiffs’ claim is time-barred. Fourth, they argue that Plaintiffs fail to plead antitrust standing. The Court addresses each of these arguments in turn.
I. Plaintiffs Fail to Plead a Plausible Boycott Conspiracy
Defendants argue that Plaintiffs have failed to plead a plausible boycott conspiracy. They argue that Plaintiffs’ boycott theory is fundamentally implausible; that Plaintiffs have not alleged facts—either in the form of direct evidence of a conspiracy or parallel conduct and plus factors— that lead to the inference of a conspiracy; and that Plaintiffs’ allegations with regard to each of the electronic trading platforms that they argue Defendants either controlled or boycotted are individually defective.
The Complaint is ambiguous and unclear regarding the precise nature of the group boycott Plaintiffs allege. At some points, the Complaint appears to allege a conspiracy to boycott platforms that would allow access to retail investors and thereby increase transparency for such investors. SAC ¶¶ 195–200. That theory would require the existence, or at least the prospect of, platforms that would allow access to both retail and institutional investors. A group boycott of a platform that would allow access to retail investors could not occur absent the existence or the prospect of such a platform.
At other points, Plaintiffs appear to allege a group boycott of platforms that permit trading of both round lots and odd lots, or “a group boycott designed to maintain pricing opacity for odd-lot corporate bonds, whether they were bought to sold by retail or institutional investors.” Dkt. No. 133 at 31. That theory would not necessarily be undermined by the absence of a platform that permitted access to retail investors for the group to boycott; Plaintiffs allege that institutional investors, as well as retail investors, trade odd lots. It would, however, be undermined by evidence that members of the purported group did business with, and did not boycott, platforms that provided pricing for odd lots. A defendant cannot boycott one with whom it does business.
Plaintiffs’ briefing expressly disavows the former theory—a boycott of platforms that were open to retail investors. Dkt. No. 133 at 31. None of the platforms that Plaintiffs identify as part of Defendants’ group boycott allowed direct access by retail investors. The allegations of the Complaint undercut the latter theory. The Complaint alleges that the Defendants who were members of the conspiracy in fact supported TradeWeb and MarketAxess—platforms that “increase pre-trade pricing transparency, which results in better competition on pricing and lower transactional costs for institutional investors trading in corporate bonds.” SAC ¶ 198.
The Court’s discussion that follows is agnostic to the different theories. The analysis would apply to both theories. The Court identifies the allegations of parallel conduct, analyzes whether those allegations of parallel conduct plausibly support the existence of a preexisting agreement among competitors to engage in predetermined conduct, and then analyzes the relevant plus factors. Regardless whether the Complaint is understood to plead a conspiracy to boycott a platform that would permit retail investors or to boycott platforms that provide pricing for odd-lot transactions, it fails to state a plausible claim for relief.
1. Parallel Conduct
Plaintiffs acknowledge in their opposition that they do not plead “direct evidence of an
agreement among defendants.” Dkt. No. 133 at 16. They seek rather to support the existence of
a conspiracy by attempting to “present circumstantial facts supporting the
inference
that a
conspiracy existed.”
Citigroup
,
The first step in analyzing such a claim is to identify allegations of parallel conduct.
However, parallel conduct does not itself establish conspiracy or a plausible inference of
conspiracy. For one, “‘conscious parallelism’ [is] a common reaction of ‘firms in a concentrated
market [that] recogniz[e] their interests and their interdependence with respect to price and
output decisions.’”
Twombly
,
The Court first identifies the well-pled allegations in the Complaint that establish parallel conduct. They fall into three general categories: (1) parallel investment in certain trading platforms; (2) parallel refusal to support other trading platforms that Plaintiffs allege would improve pricing transparency, including two electronic trading platforms in particular; and (3) parallel application of punishment and pressure to boycott electronic platforms and specific dealers or traders that threatened to improve pricing. Dkt. No. 133 at 9–13, 27–29.
The Court rejects Defendants’ argument that conduct must be “unexpected or
idiosyncratic” in order to be parallel. Dkt. No. 136 at 4. That argument conflates and collapses
two separate inquiries—the identification of well-pled conduct alleged to be parallel and the
judgment whether that conduct, if accepted as true, supports a plausible inference of a
preexisting agreement among competitors.
See Citigroup
,
Defendants’ general point, however, is well-taken. In summary and as further elaborated below, the conduct alleged to be parallel is not suggestive of a preexisting agreement; rather it is suggestive of independent competitive conduct, whether analyzed in isolation or “holistically,” see Dkt. No. 133 at 30.
In essence, Plaintiffs assert that over a decades-long period, a group of ten Defendants invested in parallel in certain electronic trading platforms that allowed access only to institutional investors and did not provide transparent pricing for odd-lot transactions, and that they also did not support platforms that would allow access to retail investors or otherwise increase pre-trade pricing transparency for odd-lot transactions. They claim that such conduct must have been based on a preexisting agreement to restrain competition and to preserve what Plaintiffs claim was Defendants’ supracompetitive—i.e., not competitive—pricing.
Plaintiffs’ theory falters upon their own factual allegations. If, as Plaintiffs allege, each
Defendant had interests in (1) preserving a pricing structure that was profitable to that
Defendant, SAC ¶ 77, and (2) investing in platforms that otherwise threatened to take some of
their profits, which they could recoup by owning equity in those platforms,
id.
¶ 148, they would
not have needed to form a group boycott to pursue those interests. It would have been in the
interest of each Defendant individually to invest in platforms that they believed would be
successful and thus would immediately take some of their profits. It also would have been in the
interest of each Defendant individually not to invest in other platforms that did not have the
promise of success. There is no reason to assume the existence of a preexisting agreement from
each Defendant’s actions, even if all of them took similar actions. “[T]here is no reason to infer
that the [Defendants] had agreed among themselves to do what was only natural anyway.”
Twombly
at 566 (holding that no inference of conspiracy can be drawn by parallel conduct of
each defendant intended to maintain “its regional dominance”). An inference of conspiracy will
not arise when the alleged conspirators’ conduct “made perfect business sense,”
Citigroup
, 709
F.3d at 138, or where there are “obvious alternative explanations for the facts alleged,”
In re INS
Brokerage Antitrust Litig.
,
No inference of conspiracy can be drawn from the allegations that a Defendant would not participate in a platform without knowing whether others would do so as well. If, as Plaintiffs suggest, no platform could be successful without the participation of all or most of the dealers in the corporate bond market, it would be “only natural” for no individual dealer to commit to participate in a particular platform without knowing or believing that others would participate in that same platform. No preexisting agreement would be necessary or can be inferred. It would be in the independent business interest of each dealer not to make the investment of time and resources to trade on a platform unless it believed others would do so as well.
Finally, if, as Plaintiffs further allege, over the extended period of the purported conspiracy most of the start-up electronic trading platforms have quickly failed, SAC ¶ 132, it would be “natural” that each individual defendant acting unilaterally and in its own economic self-interest would be loath to make an investment of time and resources without a strong belief that a new platform would break the pattern and succeed.
No inference of preexisting agreement can be drawn from Plaintiffs’ allegations. At
various times during the lengthy period of the alleged conspiracy, some of the Defendants
invested in particular platforms, while others chose not to invest or to participate. There are no
allegations of parallel conduct that was so consistent in time,
see IRS I
,
In the end, then, Plaintiffs are left with the allegations that it is more expensive to purchase a bond in an odd-lot transaction than in a round-lot transaction and that a platform that provides pricing transparency for odd-lot transactions has been slow to develop. As further explained below, the first claim does not establish an antitrust conspiracy, much less a group boycott. Plaintiffs’ statistics do not show that Defendants each enjoyed the ability to impose supracompetitive, non-market prices they had an interest in preserving. Moreover, the Complaint shows why such pricing differentials would persist in a natural, competitive environment. Round lots of corporate bonds tend to be traded by institutional investors, who are “sophisticated, repeat participants in the market that maintain longstanding relationships with dealers and are willing and able to shop around for the best pricing,” and who “tend to be better informed than odd-lot or retail investors, who typically trade infrequently.” SAC ¶ 81. In other words, they have market power to drive better bargains. See id. ¶ 82. And since the “costs to Defendants for actual transmission and trading execution is . . . the same whether Defendants are dealing in [smaller] odd-lots or [larger] round lots of corporate bonds,” id. ¶ 232, there are obvious scale efficiencies which the round-lot purchasers enjoy. As to the second claim, even assuming that the platform the Plaintiffs desire has not yet emerged, there is nothing in the Complaint that suggests that this failure was the result of a preexisting agreement to restrain trade rather than the function of independent market forces.
The Court now turns to the three sets of allegations in detail.
a. Investment in and Control over Trading Platforms
The Complaint alleges that Defendants invested in parallel in certain trading platforms: TradeWeb, MarketAxess, and BondDesk. It also alleges conduct in connection with the direction of those platforms. First, it alleges that in 1996 two Defendants and two entities later acquired by two other Defendants provided initial funding for TradeWeb: Credit Suisse, Goldman Sachs, Lehman Brothers (later acquired by Barclays) and Salomon Smith Barney (later acquired by Citigroup). SAC ¶ 143. It further alleges that by 2004, Citigroup, Merrill Lynch, Morgan Stanley, JPMorgan, and Deutsche Bank had also obtained ownership interests in TradeWeb; at that point, Thompson Reuters purchased TradeWeb from its owners. Id. ¶ 143, 149. It then alleges that “Thompson Reuters proposed ‘Project Fusion,’ a joint ownership structure that went into effect in January 2008 that gave minority ownership stakes in TradeWeb to Credit Suisse, Goldman Sachs, Lehman Brothers (later acquired by Barclays), Merrill Lynch (later acquired by Bank of America), Morgan Stanley, JPMorgan, Deutsche Bank, and RBS,” and that Citigroup acquired ownership interest in TradeWeb in 2008 as well. ¶ 151. Second, the Complaint alleges that in 2000, MarketAxess was founded by JPMorgan, among others, id. ¶ 193, and it cites a 2000 Euromoney article saying that Credit Suisse and Lehman (now owned by Barclays) have also invested in MarketAxess, id. ¶ 147. Third, the Complaint alleges that by 2004, BondDesk, which was founded in 1999, “had sold ownership stakes to 14 major banks, including Defendants such as Goldman Sachs, Bank of America, JPMorgan, and Wells Fargo,” id. ¶ 173, before BondDesk was acquired by TradeWeb and folded into TradeWeb Direct, id. ¶¶ 186–188.
The Court treats this investment activity as parallel conduct. Although Defendants argue that “no court has treated this type of lawful joint-investment activity as a form of ‘parallel behavior’ that supports an inference of unlawful conspiracy,” Dkt. No. 136 at 6–7, there is no logical reason why parallel investment—like parallel failure to invest—cannot be considered parallel conduct. Conduct that would be lawful if taken independently is paradigmatic parallel conduct.
However, Defendants are correct that this conduct is “lawful parallel conduct” that “fails
to bespeak unlawful agreement,” and does not support an inference of unlawful conspiracy.
See
Twombly
,
Plaintiffs allege that, after some Defendants invested, the platforms either limited access
to institutional investors, did not provide transparent pricing for odd lots, or acquired other
platforms which (but for their acquisition) might have provided access to retail investors. These
allegations do not establish parallel conduct by the firms who invested in the platform and who
had representatives involved with the platform. Conduct that is engaged in by a company or
joint venture in which certain Defendants were investors or held board positions is not parallel
conduct. It is not simultaneous, similar conduct of competitors otherwise expected to conduct
business and compete with one another independently. It is the activity of a single joint venture.
“[V]iewing 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 standard imposes.”
IRS I
,
Moreover, even if decisions of the platforms could be considered parallel conduct by Defendants rather than unilateral activity by the joint venture, Plaintiffs have not adequately pled facts supporting the alleged use of the platforms to stifle transparent trading, including by blocking direct retail-investor access to the platforms. Indeed, Plaintiffs’ factual allegations undermine their own theory.
First, although Plaintiffs allege that “[t]to this day,” TradeWeb “continues to maintain a dealer-to-dealer market structure rather than all-to-all trading,” SAC ¶ 152, the Complaint also asserts that TradeWeb (as well as MarketAxess) “allow[s] investor-to-investor direct trading (without intermediary dealers), and increase[s] pre-trade pricing transparency, which results in better competition on pricing and lower transactional costs for institutional investors trading in corporate bonds,” id. ¶ 198. Second, although Plaintiffs allege that “Defendants’ ownership of TradeWeb (as well as other platforms, such as MarketAxess) gave them . . . the ability to shut out retail odd-lot investors from using these platforms,” id. ¶ 152, tellingly they do not allege that TradeWeb ever considered opening its platform to retail odd-lot investors whether before or after Defendants’ investment. Similarly, although Plaintiffs allege that BondDesk was seen “as a threat to the supracompetitive profitability [Defendants] enjoyed from wider bid-offer spreads on odd-lots of corporate bonds,” id. ¶ 175, they admit that “[f]rom its inception, BondDesk . . . was not directly open to retail clients,” id. ¶ 172, and do not allege that BondDesk ever considered any contrary business model. Thus, Plaintiffs do not allege the existence or prospect of any retail all-to-all platform for Defendants to “catch and kill” with the platforms they invested in.
The Court in
IRS I
rejected similar allegations. The plaintiffs there alleged that
defendants acquired ownership interests in TradeWeb and used that to ensure that it would not be
used for all-to-all trading. The court dismissed these allegations as conclusions “not supported
by well-pled facts.”
IRS I
,
b. Failure to Support Platforms that Would Increase Pricing Transparency Plaintiffs’ next set of allegations is a combination of inaction and action—the failure to support as well as the boycott of platforms that would admit retail investors or provide increased pricing transparency for odd-lot trades.
Much of what Plaintiffs allege is in generalities: Defendants engaged in conduct to “continu[e] to maintain a dealer-to-dealer market structure rather than all-to-all trading,” SAC ¶ 94, in that they individually did not support the development of a platform for all-to-all trading for both round-lot and odd-lot trades that if all of them collectively decided to support would have been successful. [2] Therefore, Plaintiffs conclude, the failure of any one Defendant to support the development of a platform that permitted all-to-all pricing for both round lots and odd lots must have been the product of a preexisting agreement between all Defendants.
The theory is flawed factually and logically. Factually, the Complaint acknowledges that TradeWeb and MarketAxess, which Defendants invested in and supported, “allow investor-to-investor direct trading (without intermediary dealers), and increase pre-trade pricing transparency, which results in better competition on pricing and lower transactional costs for institutional investors trading in corporate bonds.” SAC ¶ 198. Logically, the more plausible inference is that to the extent no such platform has yet developed, that is because no entrepreneur has developed a business model that would permit all-to-all pricing for round lots and odd lots that would compete with existing platforms and that has been attractive to any dealer individually. The Complaint shows the functioning of a competitive market and not the absence of one.
Plaintiffs’ more specific allegations are that Defendants “refus[ed] to participate in and provide order flow (supply of bonds for trading) to odd-lot trading platforms, NYSE Bonds and Bonds.com.” Dkt. No. 133 at 28; see SAC ¶¶ 156–158, 168. These allegations start with the failure of ABS and NYSE Bonds and reason backwards that such failure must have been the product of a group boycott. The Complaint alleges that:
• “ABS failed. By 2002, only 5% of all corporate bonds were listed on ABS for trading. By 2006, only 333 U.S. corporate bond issues (around 1% of the total number of unique TRACE-eligible corporate bond issues traded that year) traded on ABS.” SAC ¶ 154.
• “NYSE Bonds failed to gain traction in trading among dealers. As of November 2017, only 25 bond dealers continued to participate on NYSE Bonds.” ¶ 156.
• “ABS and NYSE Bonds failed to achieve larger-scale success among investors because of . . . a concerted boycott of the platforms by Defendants.” Id. ¶ 157.
• “Defendants engaged in a group boycott to not provide or allow order flow to ABS/NYSE Bonds, or to severely limit such order flow to a small number of corporate bond issues.” Id. ¶ 158.
These allegations fail to state a claim for group boycott. Plaintiffs “fail to . . . plead a
single time, place, or person involved in the alleged agreement to boycott ABS/NYSE Bonds”
and fail to “allege that any Defendant declined to participate on the platform.” Dkt. No. 136 at
11. “[N]o specific Boycott Defendant is identified,” and “the actions themselves are described in
generic terms.”
TSA
,
In the absence of any well-pled factual allegations, the Complaint asks the Court to accept the premise that it would be in the economic self-interest of each Defendant alone, and without the participation of others, to support ABS and NYSE. From that premise, Plaintiffs would draw the conclusion that the decision of every single one of Defendants not to participate in ABS or NYSE must have been a result of a preexisting agreement. The premise is not well-pled or supported, however, and thus the conclusion does not follow. From the Complaint’s allegations, no one Defendant would necessarily have had an interest in supporting ABS or NYSE without believing that others would do the same and that the platform was likely to succeed. The claim that none of the Defendants supported ABS and NYSE thus cannot support the conclusion that each of the Defendants is a party to a conspiracy.
The allegations regarding Bonds.com suffer from similar flaws. The Complaint alleges: • “Another example of Defendants’ collusive conduct designed to prevent competition from electronic platforms is their refusal to deal with Bonds.com.” SAC ¶ 164.
• “After just three months, . . . Bonds.com jettisoned BondStation’s retail odd-lot focus amidst pressure from dealers such as Defendants. In April 2008, the company ‘[r]efocused from the retail segment to the institutional segment due to market conditions and other economic factors.’ One of those ‘market conditions’ was a group boycott of the retail-focused BondStation by dealers.” Id. ¶ 167.
• “Between 2012 and 2013, Bonds.com sought order flow and participation on its BondsPro platform from major corporate bond dealers like Defendants, including Bank of America, JPMorgan, and Morgan Stanley, among others.” Id. ¶ 169.
• “None of the dealers would participate with Bonds.com and the Defendants monitored and policed their conspiracy to make sure there would be no defectors. Bank of America indicated that it had interest in participating on BondsPro, but that it could not do so due to the blowback it would suffer from other Defendants. Of course, threatening to punish cartel defectors is further evidence of the existence of the conspiracy. Bank of America stated that it would only be willing to participate on Bonds.com if at least one or two of the larger dealers (such as Morgan Stanley or JPMorgan) also participated and could provide it cover from retribution.” ¶ 170.
• “As a result of this group boycott by Defendants of Bonds.com’s all-to-all, anonymous odd-lot trading platform, Bonds.com ran out of money by late 2013 . . . .” Id. ¶ 171. The claim thus appears to be two-fold: (1) Bank of America, JPMorgan, and Morgan Stanley were solicited and failed to participate and thus they must have reached a collective agreement not to participate; and (2) Bank of America stated it would only be willing to participate if others participated as well.
The first claim does constitute parallel conduct. However, it is not parallel conduct
suggestive of conspiracy for the same reasons identified above with respect to ABS and NYSE.
An inference of conspiracy cannot be drawn from the fact that Bank of America, JPMorgan, and
Morgan Stanley each chose not to participate. The conduct suggests “rational and competitive
business strategy unilaterally prompted by common perceptions of the market.”
Twombly
, 550
U.S. at 554. The second claim illustrates the point. Not only is the unilateral expression of one
market participant insufficient to show the existence of a preexisting agreement not to compete,
but it is the expression of a market participant acting its own independent “competitive business
strategy.” It would “only be natural,”
id.
, for a market participant such as Bank of America
not to devote resources to a start-up platform such as BondsPro unless it thought that others
would participate as well and that the platform would enjoy the liquidity that would make trading
on it attractive. Thus, to the extent Plaintiffs allege any parallel conduct, it is the very same type
of parallel conduct that
IRS I
rejected, because there is “a ‘natural explanation,’ consistent with
unilateral action, for the Dealers’ decisions not to supply liquidity to” the platform without
knowing that at least some others would do the same.
IRS I
,
c. Enforcement of the Conspiracy via Punishment and Pressure Plaintiffs’ third set of allegations relates to what they claim was parallel conduct by Defendants in applying “punishment and pressure to boycott electronic platforms that threatened improved pricing.” Dkt. No. 133 at 10.
Plaintiffs allege that “when odd-lot traders employed by InterVest engaged in trading that Salomon Smith Barney (later acquired by Citigroup) deemed to be ‘disruptive’ of the market, Salomon refused to do business with those traders,” SAC ¶ 137, and that “none of the other Defendants stepped in to do business with the traders, as one would expect in a competitive market,” id. However, Plaintiffs plead no facts for their central premise that “one would expect” a different dealer to “step in” if Salomon chose not to do business with the traders. There is no allegation that the traders were particularly attractive or valuable or that any of the dealers lacked access to the traders before Salomon ceased doing business with them. Plaintiffs’ allegation rests on mere conclusion. Even if the individual decisions of each of the other Defendants not to “step in” could be considered parallel conduct (or non-conduct), it is not suggestive of conspiracy. It most plausibly suggests that no Defendant thought it could make more money transacting through the InterVest traders than they had been able to make transacting with other traders.
Plaintiffs also allege that “[f]ollowing the announcement of InterVest [which would debut a new electronic trading system for corporate bonds] on Bloomberg, bond dealers began to complain to Bloomberg that InterVest offered a service that competed with them.” ¶ 142. But Plaintiff does not name a single Defendant who complained or a single complaint that was made to Bloomberg. Moreover, even if the Complaint had contained a well-pled allegation that more than one dealer who paid money for services from Bloomberg complained when Bloomberg decided to do business with that dealer’s competitor, such allegation would not without more suggest that each complaint was the product of a conspiracy. Each dealer individually would have an interest in a competitor not taking market share from it. That is ordinary competitive activity. It does not require an agreement.
Next, Plaintiffs allege that Defendants pressured and threatened Bloomberg to stop it from quickly granting NYSE Bonds a connection to Bloomberg TOMS: “Defendants – which are large financial institutions with significant accounts with Bloomberg’s separate and profitable Bloomberg Terminal business – used their market power and value to Bloomberg as Bloomberg Terminal customers to force Bloomberg to materially delay NYSE Bonds’ connectivity through Bloomberg TOMS by threatening to terminate or reduce their Bloomberg Terminal leases if Bloomberg failed to do so.” Id. ¶ 163. Again Plaintiffs do not identify any complaint that was made to Bloomberg or any Defendant who made a complaint. The allegation rests on speculation: The fact that NYSE Bonds’ connectivity was delayed must have been a result of threats from “Defendants.” But, Plaintiffs do not allege what those threats were, that they were made by more than one Defendant, or how they impacted Bloomberg. There is nothing in that conduct that suggests a preexisting agreement rather than an individual decision of each alleged complainant as to its own economic interest.
Plaintiffs further allege that Defendants monitored and policed their boycott of Bonds.com through threats. The only specific factual allegation is that Bank of America was reluctant to participate on Bonds.com if other dealers did not participate and therefore it monitored whether others would participate:
None of the dealers would participate with Bonds.com and the Defendants monitored and policed their conspiracy to make sure there would be no defectors. Bank of America indicated that it had interest in participating on BondsPro, but that it could not do so due to the blowback it would suffer from other Defendants. Of course, threatening to punish cartel defectors is further evidence of the existence of the conspiracy. Bank of America stated that it would only be willing to participate on Bonds.com if at least one or two of the other larger dealers (such as Morgan Stanley or JPMorgan) also participated and could cover it from retribution. ¶ 170.
This allegation begins with a conclusion and ends with a factual example that does not support the conclusion. Simply stating that “Defendants monitored and policed their conspiracy to make sure there would be no defectors” is a label and conclusion insufficient to get Plaintiffs over the Twombly/Iqbal line or subject Defendants to discovery. The allegation that Bank of America chose not to participate in BondsPro due its perception of “the blowback it would suffer from other Defendants” does not contain the factual content necessary to cross that line. It does not create an inference that any Defendant—much less a group of Defendants—in parallel threatened Bank of America for participation. The further allegation that Bank of America declined to participate in Bonds.com if at least one or two other large dealers did not also participate does not add support to the claim. The most plausible inference from the pleaded facts is that Bank of America arrived at that decision independently in the exercise of its own business judgment, and not collectively as the result of an agreement. No dealer would want to invest time or money in a trading platform absent some indication that other large dealers would also participate. Absent any allegations that other Defendants engaged in similar conduct, made similar excuses to avoid participating in new platforms, or similarly expressed reluctance to act alone, this allegation pertains just to one isolated comment by one isolated Defendant, and not to any parallel conduct.
Plaintiffs’ only specific allegation of parallel enforcement conduct is their claim that two of the eleven Defendants—Morgan Stanley and Citibank—each threatened to limit its business with First Tennessee, a regional bank, to punish it for offering narrow spreads for corporate bond transactions:
• “Defendants [sic] efforts to punish those who threaten to narrow odd-lot spreads continue to this day. . . . Blackrock began to use competitive regional banks and brokers such as First Tennessee, Piper Jaffrey, and McDonald & Co. for its odd-lots trading needs because of the narrower spreads they provided. When defendant Morgan Stanley learned that Blackrock was providing First Tennessee with axe sheets and allowing it to gain business because of the narrower spreads that it was providing, Morgan Stanley threatened to limit any business that it transacted with First Tennessee and blackball them. Defendant Morgan Stanley took such steps in order to punish First Tennessee for offering narrower spreads for such transactions.” SAC ¶ 138.
• “During the period from 2016 through 2018, Blackrock used electronic RFQ’s available on MarketAxess to execute directly with regional banks such as First Tennessee, Piper Jaffrey, and McDonald & Co. in order to benefit from the narrower spread that such dealers provided. Eventually, defendants Morgan Stanley and Citibank learned of these transactions and threatened to limit any business that they transacted with First Tennessee for offering narrower spreads for such transactions.” ¶ 139.
• “When individual employees of disciplining Defendant dealers tried to help the traders that were being penalized (by, for example, transacting business with them), they did so at risk of losing their jobs with the Defendant dealer. The substantial risks faced by employees ensured that the penalty box – the disciplining action – worked.” Id. ¶ 140. The claim that “defendants Morgan Stanley and Citibank learned of these transactions and threatened to limit any business that they transacted with First Tennessee for offering narrower spreads for such transactions,” id. ¶ 129, can be read to suggest parallel conduct, albeit barely. But the interest of each of them in not doing business with a bank that undercut its own individual pricing and the fact that all that Plaintiffs can allege is that two of the eleven Defendants made such a threat is also inconsistent with conspiracy. [3]
2. Plus Factors When stripped of conclusions and reduced to allegations of fact, the Complaint contains three well-pled allegations of parallel conduct: (1) Defendants’ parallel investments in TradeWeb, BondDesk, and MarketAxess; (2) two Defendants’ alleged retaliation against First Tennessee; and (3) the alleged group boycott of BondsPro. For reasons stated above, these allegations are not suggestive of a conspiracy whether viewed individually or collectively; rather, they are consistent with Defendants’ individual self-interest.
The Court next considers whether the Complaint pleads any “plus factors” which,
considered in conjunction with the parallel conduct, support the inference of a conspiracy. Plus
factors are “additional circumstances . . . which, when viewed in conjunction with the parallel
acts, can serve to allow a fact-finder to infer a conspiracy.”
Apex Oil Co. v. DiMaruo
, 822 F.2d
246, 253–54 (2d Cir. 1987). “These plus factors may include: a common motive to conspire,
evidence that shows that the parallel acts were against the apparent individual economic self-
interest of the alleged conspirators, and evidence of a high level of interfirm communications.”
Citigroup
,
Plaintiffs argue that the Complaint pleads four plus factors: (1) statistical evidence that shows that Defendants acted against their unilateral self-interest; (2) market concentration; (3) common motive to conspire; and (4) interfirm communication. Those factors do not support the alleged conspiracy.
a. Statistical Evidence of Actions Against Individual Self-Interest Plaintiffs argue that the Complaint’s “abundant statistical evidence of Defendants’ supracompetitive odd-lots pricing” is “market evidence that makes no rational, economic sense” and therefore “lends plausibility to the existence of a conspiracy.” Dkt. No. 133 at 44–45. Plaintiffs presume that each of the Defendants, and Defendants alone, enjoyed “supracompetitive” pricing and from that presumption seek to draw the inference that Defendants also enjoyed a common interest or motive to preserve a market where prices would not be competitive. They further assert that “if the relevant market were truly competitive, each Defendant would be competing on price and driving down odd-lot spreads to parity with already profitable round-lot spreads in order to gain a larger share of the market,” but that instead Defendants “ignored the logical competitive response of meeting customer demand and growing market share with better prices created by available technological innovations—which is in their unilateral interests—and instead have repeatedly and collectively settled for smaller market share by stifling any evolution in the market for odd lots of corporate bonds that would threaten their ability to change.” at 45. In other words, they suggest that their statistical allegations of a pricing disparity between odd-lot spreads and round-lot spreads demonstrate actions against each Defendant’s unilateral self-interest and indicate the existence of a conspiracy.
The Court analyzes Plaintiffs’ statistical allegations in detail. When carefully reviewed, they do not support the existence of a boycott conspiracy, nor do they support the claim that each Defendant enjoyed similar “supracompetitive odds-lots pricing,” or that any individual Defendant enjoyed wider spreads than any individual dealer in the non-Defendant group.
Plaintiffs’ first set of statistical allegations consists of evidence that the mean or average price per bond paid for the purchase of bonds in odd lots is higher and the spread is wider than for the purchase of bonds in round lots. SAC ¶¶ 86–94. Plaintiffs cite eleven studies that “show that odd-lot investors in corporate bonds pay average transaction costs (represented by the bid- offer spread) that are between 10% . . . to as much as 1,775% . . . greater than round lot investors.” Id. ¶ 90. Plaintiffs conclude from these allegations: “In sum, numerous peer- reviewed studies demonstrate that dealers, including Defendants, have engaged in a pattern of parallel conduct by charging odd-lot investors higher selling prices and paying them lower purchase prices than round lot investors for the same bond issue.” ¶ 94.
These allegations do not support the inference that Defendants enjoyed supracompetitive pricing or had an interest in engaging in a group boycott conspiracy. The quoted studies describe that, on average, purchasers of bonds in odd lots suffered a pricing disparity compared to purchasers of round lots. They do not reflect that each dealer imposed a pricing disparity on all purchasers of odd lots compared to round lots or that Defendants—and Defendants alone— imposed such a disparity. The Complaint alleges that “dealers, including Defendants,” charged odd-lot investors higher selling prices, not that all Defendants did so or that only Defendants did so. Indeed, Plaintiffs have abjured any allegation of price-fixing among Defendants.
The plausible inference to be drawn from Plaintiffs’ allegations is that any price disparity between the price paid for bonds purchased in an odd lot versus the price paid for bonds purchased in a round lot is the result of natural competitive forces. Plaintiffs themselves allege that “[r]ound lot transactions, given their size, almost always involve institutional investors that trade for numerous individuals and companies – sophisticated, repeat participants in the market that maintain longstanding relationships with dealers and are willing and able to shop around for the best pricing. As a result, they tend to be better informed than odd-lot or retail investors, who typically trade infrequently.” Id. ¶ 81. Those investors—by virtue of their size, sophistication, and market power—are able to demand better pricing. The Complaint further outlines why, in a competitive market, this translates to a pricing disparity between odd-lot and round-lot trades:
As a result, dealers responding to an RFQ for a round lot know that they are dealing with an institutional investor that is likely to be: (a) price sensitive; (b) willing and able to obtain multiple quotes from other dealers; (c) knowledgeable regarding the market and pricing due to their repeated role in trading; and (d) in control of large portfolios of bond that offer additional trading opportunities in the future if the dealer is competitive in its pricing. Responding to these incentives, dealers provide quotes for round lots at their best competitive prices, keeping their spreads narrow, in the hope of securing this (and other, future) trades from the round lot institutional investor – a process entirely consistent with economic and market microstructure theory.
Id. ¶ 82.
The Complaint also outlines the reasons why in a natural, competitive market dealers generally would be able to offer better prices to round-lot investors, regardless whether they were institutions. The “costs to Defendants for actual transmission and trading execution is . . . the same whether Defendants are dealing in [smaller] odd-lots or [larger] round lots of corporate bonds.” ¶ 232. The average price is lower and the bid-offer spread more narrow when those costs are able to be spread over a larger lot of bonds.
In other words, the Complaint itself provides specific and detailed allegations outlining
why each dealer and therefore also each individual Defendant would have both the ability to
offer better pricing to investors trading in round lots than investors trading in odd lots and the
interest in doing so, undercutting the theory that “[t]here is no explanation consistent with a
healthy, competitive market for why the differential in spreads between odd-lots and round lots
has persisted to the degree it has.”
Id.
¶ 85. Each individual Defendant has clear incentives to
offer lower prices to repeat institutional investors; although this may negatively impact prices for
Plaintiffs and other retail investors, “there is no reason to infer that the companies had agreed
among themselves to do what was only natural anyway.”
Twombly
,
That same conclusion follows from the analysis provided by Plaintiffs’ expert, set forth in the Complaint, who “analyzed the transactions costs for Riskless Principal Trades (“RPTs”) in the U.S. corporate bond market from January 2006 to December 2019.” SAC ¶ 95. The expert found “statistically and economically significant differences in transaction costs for round lot and odd lot trades over the period from 2006-2009.” Id. ¶ 99. The study reflects what one would expect in a market environment in which—as Plaintiffs allege—the institutional investors who command buy-side market power also transact primarily in round lots and in which there is obvious scale efficiency to purchasing in a round lot. Prices are lower for those who purchase through an institutional investor and do so in round lots. The study shows nothing distinctive about Defendants and nothing other than what one would expect from the natural operation of market forces.
Plaintiffs’ next set of statistical allegations sets out to remedy one flaw in the previous sets of allegations. However, it leaves the remaining flaws. Plaintiffs indicate that they “have gone a step further and analyzed a subset of actual trades by Defendants.” Id. ¶ 100. They allege that “[t]he analysis found that . . . Defendants charged an average of 86 basis points for smaller odd-lot trades (less than or equal to $50,000 in size) 94 basis points for odd-lot trades ranging from $50,000 to $100,000 in size and 22 basis points for odd-lot trades from $100,000 to $1 million. In contrast, average spreads on round lot RPT trades over $1 million executed by Defendants averaged 13 basis points.” ¶ 102. Plaintiffs then compare their data regarding “RPT spreads on odd-lot trades executed by Defendants to trades by non-Defendant dealers,” and allege that “transaction costs on odd-lot trades less than or equal to $50,000 in size executed by Defendant dealers average 62 basis points higher than non-Defendant trades of the same size,” and that “for odd-lot trade sizes ranging from $50,000 to $100,000 . . . Defendant trades average 17 basis points higher than same-sized trades by non-Defendant trades.” ¶ 103 (emphasis omitted).
These allegations have the benefit of trying to isolate the Defendant group from the group of non-Defendants. But, because Plaintiffs make these allegations only in gross with respect to the Defendant group as a whole and the non-Defendant group as a whole, they do not say anything meaningful about the role of any individual Defendant and thus about the group of Defendants as a whole. Plaintiffs compare average spreads for odd lots amongst all Defendants to average spreads for odd lots amongst all non-Defendant dealers. Plaintiffs do not analyze whether each Defendant on average charges more for an odd-lot transaction than every non-Defendant or even whether a majority of Defendants charge more than the average non-Defendant for purchases done in an odd lot. It is impossible to know from Plaintiffs’ statistics whether the average for the Defendant group is driven by one particular dealer who charges a higher spread than the remainder and whose trades drive the average or whether the average for the non-Defendant group too is driven by one or a small group of dealers who charge a lower spread. From Plaintiffs’ allegations, it is equally if not more plausible that one or more than one Defendant charges lower prices on average for odd-lot transactions than one or more of the dealers in the non-Defendant group.
The allegation suffers from additional defects. Plaintiffs presumably proffer the statistics for the proposition that Defendants, and Defendants alone, charge higher prices for bonds purchased in odd lots than those purchased in round lots and thus that they have a pricing advantage with respect to odd lots over other participants in the market that they are trying to preserve. But the allegation does not compare the pricing of round-lot transactions by Defendants to those by non-Defendants. It compares only odd lots, see id. , and is entirely and conspicuously silent about the pricing for bonds purchased in round lots through non-Defendants. From the allegations, it is equally if not more plausible that for the non-Defendant group and for each dealer within the non-Defendant group there is a similar or greater differential as compared to that within the Defendant group between the price charged for a bond purchased in a round lot and one purchased in an odd lot.
The allegation thus shows only that bonds can be purchased cheaper if purchased as part of a round lot than if purchased as part of an odd lot. But, for reasons stated, those allegations do not show that Defendants had any particular pricing advantage they were trying to preserve, much less that each of them had such a pricing advantage. They show only that those who have market power and who can benefit from scale economies frequently do so in a natural competitive market environment.
Paragraph 112 of the Complaint offers the first—and only—statistical allegation that compares pricing differentials of individual Defendants with non-Defendant dealers. Plaintiffs compare “the differential between individual Defendants’ and non-Defendants’” non-competitive price markdowns for customer-initiated odd-lot sales. However, once again, Plaintiffs utilize an average of all non-Defendant dealers as a comparator; there is no way to infer from this that each Defendant’s pricing differentials were higher than those of every non-Defendant dealer, or even that each Defendant’s pricing differentials were higher than those of a majority of non-Defendant dealers. Moreover, the table Plaintiffs present—comparing each individual Defendant’s average customer-initiated round-lot sales price minus customer-initiated odd-lot sales price for the matching bond and date, which Plaintiffs refer to as the “markup,” with the average non- Defendant markup—highlights the problem with the earlier statistics that grouped together Defendants as a whole. The table omits two Defendants, Goldman Sachs and Credit Suisse, because their markups were lower than the average non-Defendant markup. Id. ¶ 112. And the significant disparity between the numbers presented for each of the eight individual Defendants included in the table undercuts any assumption that Defendants’ odd-lot bond pricing was supracompetitive and that Defendants were not competing amongst each other. Barclays’ average markdown was 83.7 basis points, which is significantly higher than Goldman Sachs’ average markdown of 27.3 basis points. This pricing disparity is consistent with a market in which Defendants “compete[d] for odd-lot business by improving execution prices,” contra id. ¶ 109, and as discussed above, the fact that pricing is still different for each Defendant between odd lots and round lots has a logical explanation also consistent with a competitive market. As such, Plaintiffs fail to present any statistical evidence or allegations plausibly suggesting that Defendants each, or all, benefitted from supracompetitive odd-lot pricing, that Defendants were not competing amongst each other, or that Defendants were acting against their unilateral self- interest.
b. Market Concentration
Plaintiffs argue that they have alleged a plus factor of market concentration. Dkt. No.
133 at 46. This factor looks at the feasibility of competitors reaching and enforcing an
agreement to restrain trade. The fewer the number of participants in the market the easier to
reach an agreement to restrain trade and the easier to enforce such an agreement, thus the more
plausible the inference of conspiracy.
See Todd v. Exxon Corp.
,
Plaintiffs allege that the U.S. corporate bond market is a concentrated market that Defendants control. The Complaint alleges that “OTC trading in the secondary markets for U.S. corporate bonds is highly concentrated, and is becoming even more so. This concentration makes it more likely that Defendants are engaging in collusion. Defendants have dominated the U.S. corporate bond market for well over a decade.” SAC ¶ 78. Plaintiffs allege that the ten Defendants “have controlled 65% or more of the bond underwriting market every year since at least 2014.” Id. ¶ 79. They allege that this market share means that Defendants “control the vast majority of trading, such that the market is highly concentrated,” and that this “dominance over the supply of U.S. corporate bonds has allowed them to effectively conspire to inflate the spread of odd-lot bonds in the secondary trading market.” Id. ¶ 229.
The Complaint only alleges specific facts about Defendants’ market share in the underwriting market, not in the Relevant Market—the secondary odd-lot corporate bond market. It alleges that a large share of the underwriting market translates directly to a large share of the secondary market: “The Defendants’ collectively large market share in the underwriting market for U.S. corporate bonds gives them power over the secondary market for trading in these bonds. That power flows from their control over the supply of bonds to be sold in the secondary market. From these leading positions as U.S. corporate bond underwriters, Defendants have secured a correspondingly larger aggregate share as the top dealers in the Relevant Market . . . for secondary trading in U.S. corporate bonds.” ¶ 80. The Complaint reiterates this assumption throughout: “In short, the Defendants control the largest inventory of bonds because they serve as underwriters and issuers of those bonds. That dominance over the supply of U.S. corporate bonds has allowed them to effectively conspire to inflate the spread of odd-lot bonds in the secondary trading market.” ¶ 229. Defendants argue that Plaintiffs’ premise—that a large market share in the underwriting market directly translates to a large market share in the secondary market—is false. They argue: “Bond underwriters purchase new bond from issuers and resell them to investors and other dealers in the primary market. After the underwriters resell the bonds, investors and other dealers trade them in the secondary market. The underwriters have no control over the bonds in the secondary market once they sell them in the primary market.” Dkt. No. 131 at 26 (citing SAC ¶¶ 3–4).
Plaintiffs’ allegations, taken as true, do not establish that the secondary corporate bond
market for odd lots—the Relevant Market—was concentrated. Both the number of entities
whom Plaintiffs contend “control” the market and the aggregate market share they enjoy are
inconsistent with the notion that the Relevant Market is concentrated. Defendants are ten entities
and their affiliates; even if they collectively control something around 65% of the secondary
corporate bond market, this does not constitute the “highly concentrated” market that Plaintiffs
argue enables a conspiracy. Ten market participants is a large number; there is no basis to
assume that an anticompetitive agreement among that large a number of banks would be easy to
reach or, once reached, easy to enforce, particularly when other market participants—not part of
the alleged conspiracy—already controlled the remaining 35% of the market.
See
Dkt. No. 133
at 14;
cf. Ross v. American Exp. Co.
,
Moreover, Plaintiffs’ allegations do not demonstrate that Defendants had control over the secondary corporate bond market for odd lots. Plaintiffs allege that “bonds being traded from the same issue are fungible,” so “odd-lots of that issue can be combined into a round lot, and, conversely, a round lot of a given issue can be broken into odd-lots of that issue.” SAC ¶ 5. They further allege that approximately 82% of trading volume by par value in the U.S. corporate bond market is estimated to be in round lots. ¶ 93. Given this, there is no way for any Defendant—or all Defendants—to control what happens with that 82% of corporate bonds once they are sold in round lots; any institutional investor who purchases them could easily break them down into odd lots and resell them at “competitive” rates. As such, even crediting Plaintiffs’ allegations that Defendants’ share of the underwriting market means that they have corresponding control of those bonds when they enter the secondary market, this does not mean that Defendants controlled the same share of the odd-lot secondary market, and furthermore this control does not extend to the level Plaintiffs suggest and does not support an inference that Defendants had control over pricing in the secondary corporate bond market.
But even crediting
arguendo
Plaintiffs’ assertion that they have pled that Defendants
control a large market share of the secondary corporate bond market, this does not make the
alleged parallel conduct suggestive of a conspiracy and therefore does not count as a plus factor;
rather, it is suggestive of mere interdependent conduct. “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 and of itself unlawful.”
Twombly
,
c. Common Motive to Conspire Third, Plaintiffs argue that Defendants have a common motive “to protect the pricing opacity that was jeopardized by the electronic platforms.” Dkt. No. 133 at 47.
This theory rests on two premises: first that Defendants enjoyed supracompetitive
pricing, which they were seeking to preserve and which increased pricing transparency would
jeopardize, and second that the platform Defendants allegedly boycotted—BondsPro—presented
an immediate threat to pricing opacity in a way that the platforms Defendants supported did not.
As
IRS I
held, a common motive to conspire is “at best thinly pled” where the allegedly boycott
platform did not present a threat to Defendants’ profit margins.
As to the former, as discussed above, despite its plethora of statistics, the Complaint fails to plausibly allege that Defendants enjoyed supracompetitive pricing for odd-lot bonds. The Complaint therefore does not allege that there was a pricing advantage that the Defendants as a group had a motive to preserve. Each Defendant had an interest in competing against the others as well as against those not named as defendants.
As to the latter, the Complaint fails to plausibly allege that BondsPro threatened pricing
opacity in a way that platforms like TradeWeb and MarketAxess, which Defendants supported,
did not. The Complaint also makes it clear that BondsPro, like TradeWeb and MarketAxess,
“focus[ed] on institutional odd-lot investors rather than retail investors.” SAC ¶ 168. It
therefore does not support a common motive to conspire to prevent the emergence of electronic
trading platforms that would improve transparency and pricing for
retail investors
. The
Complaint does allege that BondsPro “allow[ed] all-to-all, anonymous, exchange-style trading –
trading that would eliminate Defendants as middlemen, or force them through anonymous
pricing competition to lower odd-lot bid-offer spreads.”
Id.
However, the Complaint
acknowledges that TradeWeb and MarketAxess, which Defendants supported, “allow investor-
to-investor direct trading (without intermediary dealers), and increase pre-trade pricing
transparency, which results in better competition on pricing and lower transactional costs for
institutional investors trading in corporate bonds.” ¶ 198. Plaintiffs offer no plausible reason
why Defendants would have a common motive to support some platforms that increased pricing
transparency and competition, removed Defendants as intermediaries, and did not offer access to
retail investors yet also boycott a platform because it threatened to increase pricing transparency
and competition, threatened to remove Defendants as intermediaries, and did not offer access to
retail investors. They similarly do not offer any reason to believe a platform like BondsPro,
which was similar to platforms Defendants supported but also similar to the many start-up
trading platforms that quickly failed, presented any
immediate
threat to Defendants’ profits—as
such, “there was little urgency to conspire against it.”
IRS I
,
Thus, Plaintiffs have not plausibly pled a common motive to conspire to boycott platforms that presented an immediate threat to their supracompetitive pricing, both because Plaintiffs have not demonstrated supracompetitive pricing and because Plaintiffs have not demonstrated any distinct threat posed by BondsPro.
d. Interfirm Communication
Plaintiffs further argue that the Complaint alleges “high levels of interfirm
communications among Defendants.” Dkt. No. 133 at 14. This plus factor is relevant because
the existence of communications among firms could permit an anticompetitive conspiracy to
form and to flourish. Thus, where there are “extensive communications among high-level
officials at Dealers with responsibilities” for the products at issue and where communications
occur in fora where discussion of competitive threats “might naturally arise,” those
communications tend to be supportive of the existence of a conspiracy.
IRS I
, 261 F. Supp. 3d at
476–77. This is particularly true where the communications “‘represent[] a departure from the
ordinary pattern’ of communications between defendants” or “where there is evidence that
defendants exchanged confidential information or sought to conceal their communications.”
Anderson News, L.L.C. v. American Media, Inc.
,
The Complaint alleges no such communications.
Cf. Gelboim v. Bank of America Corp.
,
Instead, the Complaint focuses exclusively on the fact that traders from different dealers communicate about pricing in the course of trading. It alleges that “there is a constant communication loop among a small group of bond trading insiders” and that “Defendants’ traders see live quotes from their competitors and likewise coordinate their pricing.” SAC ¶¶ 237–241. It is conceivable that those fora could provide a venue for traders to fix prices. See id. (alleging that the high levels of interfirm communication “makes fixing prices in odd-lots of corporate bonds easier to accomplish,” and that “it appears that Defendants use these channels of interfirm communication to collude, rather than to find ways to compete that would improve prices for odd-lot investors”); see also City of Philadelphia v. Bank of America Corp. , 498 F. Supp. 3d 516, 529 (S.D.N.Y. 2020) (holding that a high level of interfirm communications was established and supports the inference of a price-fixing conspiracy where those communications were alleged to include “Banks routinely . . . shar[ing] information about their base rates, inventory levels, and planned rate changes with each other over the telephone and through electronic communications . . . [and] Banks communicat[ing] their future rates to each other”). The Court need not reach that issue as Plaintiffs have abandoned any Section 1 price-fixing claim.
The allegations do not support a plus factor of high levels of interfirm communication relevant to the alleged boycott . There is no reason to believe that the traders who communicate with one another about pricing in the course of the day are communicating confidential information much less that they are discussing the platforms they will support and those that they will boycott. The conduct Plaintiffs challenge involves decisions regarding investment in and management of alternate platforms for the trading of bonds. The Complaint offers no reason to believe that the traders who transacted on individual trades would have had the authority to make decisions about such matters or the interest in communicating about them on behalf of the firms for which they worked, nor does it offer any reason to believe that any of those who had authority or an interest in such strategic matters engaged in interfirm communications.
In contrast, in
IRS I
, the court found that this plus fact was “clearly present” for a boycott
claim when the complaint alleged communication “via the Dealers common ownership of
TradeNet, their participation in OTCDerivNet, their participation on industry associations, and
the social and professional . . . interactions among executives of Dealers in this market niche.”
II. Plaintiffs’ Complaint Fails to Connect Any Specific Defendant to the Alleged
Conspiracy
It is fundamental to pleading in the post-
Twombly
era that before a defendant is forced to
be held to account for conduct as serious as a conspiracy in restraint of trade in violation of the
Sherman Act that the pleader inform the defendant what it is alleged to have done. It is not
sufficient for the plaintiff to allege that a group has done something wrong unless the allegations
give reason to believe that the defendant, as a member of the group, also has committed the
wrong. “[A]uthorities overwhelmingly hold that a complaint that provides no basis to infer the
culpability of the specific defendants named in the complaint fails to state a claim.”
In re
Mexican Government Bonds Antitrust Litig.
(“
MGB
”),
For example, in
TSA
, the court held that as to allegations which referred to only one
defendant by name and “refer[red] to the other Boycott Defendants collectively – as in ‘among
other Boycott Defendants’ and ‘gave in to the Boycott Defendants’ threat’ – the Complaint
engages in impermissible group pleading.”
The Complaint fails those precepts. It does not contain allegations as to any individual Defendant that would establish that such Defendant engaged in a group boycott. And, in the absence of such allegations as to any Defendant, the Complaint must be dismissed against all Defendants.
The thrust of the Complaint is that Defendants were all among the largest underwriters of bonds and that as such each individual Defendant had an interest in preserving the market structure. In essence, because a Defendant ranked as number 1, or number 3, or number 7 in the ranks of bond underwriters, the Complaint alleges that such Defendant must have been a participant in what is claimed to be a group boycott. The deficiencies of that approach have already been discussed. The fact that a bank is large and active does not mean that it is an antitrust conspirator.
At oral argument, when asked to identify a particular Defendant against whom they claimed the allegations were well-pled and to identify the facts that supported the claim, Plaintiffs directed the Court to their allegations against Goldman Sachs. Oral Argument Transcript at 26:2–5. Examination of those allegations, however, demonstrates the flaws in Plaintiffs’ pleading. The Complaint alleges that Goldman Sachs held between a 7% and 8% share of the U.S. corporate bond underwriting market between 2014 and 2018, SAC ¶ 79; that Goldman Sachs’ prices obtained for their customers selling odd lots as compared to round lots of the same bond on the same date were 27.3 basis points lower, id. ¶ 111; that Goldman Sachs, along with three other banks, provided initial funding for TradeWeb, regained a minority ownership stake in TradeWeb as part of “Project Fusion” in 2008, and was also invested in BondBook, id. ¶¶ 143, 147, 151; that Goldman Sachs had ownership interests in BondDesk— which was later acquired by TradeWeb—“by 2004,” and that two “individuals affiliated with” Goldman held seats on the board of directors of BondDesk in 2004, which they used “to remove the existing management of BondDesk from their day-to-day leadership positions at the company in 2004,” id. ¶¶ 173–181, 186. At oral argument, Plaintiffs summarized their allegations of parallel conduct as to Goldman Sachs specifically as: “So our allegations of the parallel conduct are defendants’ use of these companies to catch and kill potential competitors that threaten their market. We also make other allegations about refusal to provide liquidity and delaying access.” Oral Argument Transcript at 28:21–25.
The allegations amount, at most, to the claim that Goldman Sachs had a less than 10%
market share of the U.S. corporate bond underwriting market and that it invested in TradeWeb
and in BondDesk, which was later acquired by TradeWeb. That conduct, however, is as
consistent with the rational and competitive decisions of a business which wants to at least
preserve its market share and, if possible, gain market share, and believes that it can do so by
investing in a platform that offers traders a further venue through which to obtain a price. The
Complaint also goes into significant detail about actions of board members of BondDesk who
were affiliated with Goldman Sachs, but activity by board members within a lawful joint venture
cannot be imputed against Goldman as parallel conduct relevant to a per se Section 1 group
boycott claim. Absent from the Complaint is any allegation of concerted action by Goldman
Sachs with any of the other Defendants alleged to be a member of the group boycott. Indeed, as
discussed in detail above, the only statistical allegation proffered by Plaintiffs cuts against the
inference Plaintiffs would draw of conspiracy. The Complaint compares pricing differentials of
certain
individual
Defendants with non-Defendant dealers for customer-initiated odd-lot sales but
omits Goldman Sachs entirely from the analysis because their markups were
lower
than the
average non-Defendant markup. SAC ¶ 112. In short, assuming it is even “conceivable” that the
allegations against Goldman Sachs would be consistent with anticompetitive conduct, those
allegations fail to push the inference that Goldman Sachs participated in a group boycott over the
line from “possible” to “plausible,” and thus fail to “state a claim to relief that is plausible on its
face” as against Goldman Sachs.
Iqbal
,
Other than Goldman Sachs, the Defendant whose name is most frequently mentioned in the Complaint is JPMorgan. The allegations against JPMorgan are that it held between an 11% and 12% share of the U.S. corporate bond underwriting market between 2014 and 2018, SAC ¶ 79; that its average markup for odd-lot sales as compared to round-lot sales was 55.7 basis points, as compared to a 41.7 basis point average markup among non-Defendants, id. ¶ 112; that it held an ownership interest in TradeWeb by 2004, id. ¶ 143; that it regained a minority ownership stake in TradeWeb as part of “Project Fusion” in 2008, id. ¶ 151; that Bonds.com “sought order flow and participation on its BondsPro platform from major corporate bond dealers like Defendants, including . . . JPMorgan,” but that “[n]one of the dealers would participate with Bonds.com,” id. ¶ 169; that it had ownership interests in BondDesk—which was later acquired by TradeWeb—“by 2004,” and that one “individual[] affiliated with” Bear Stearns & Co., later acquired by JPMorgan, held a seat on the board of directors of BondDesk in 2004, which it used “to remove the existing management of BondDesk from their day-to-day leadership positions at the company in 2004,” id. ¶¶ 173–181, 186; and that JPMorgan was one of the founders of MarketAxess, which acquired Trading Edge in 2001 but did not carry over its anonymous trading feature, id. ¶ 193.
Those allegations fail to state an antitrust claim against JPMorgan. As with Goldman Sachs, activity within a lawful joint venture that JPMorgan had ownership stakes in cannot be imputed against JPMorgan as parallel conduct relevant to a per se Section 1 group boycott claim. What is left, then, is the allegation that JPMorgan’s pricing disparity that disadvantaged odd-lot sales was slightly worse than the average among all non-Defendants (absent any basis from which to conclude that it was worse than that of all, or even most, non-Defendants), the allegations that JPMorgan participated in TradeWeb and MarketAxess, and the allegation that BondsPro “sought order flow and participation” from JPMorgan but that JPMorgan did not participate. Viewed holistically, the allegations do not make out a plausible inference that JPMorgan was part of a group boycott conspiracy. The Complaint does not establish that JPMorgan enjoyed supracompetitive pricing as compared to any non-Defendant dealer; JPMorgan’s lawful investment activity has a rational explanation and thus does not give rise to a plausible inference of conspiracy; and the allegation that JPMorgan chose not to participate on BondsPro is similarly consistent with its rational and competitive business interests.
The allegations against the other Defendants are even more sparse.
Absent allegations that would support a claim against an individual defendant, there is no
basis for the Court to sustain the complaint against any defendant.
MGB
,
III. Plaintiffs’ Claim is Time-Barred
Defendants also move to dismiss on the separate ground that Plaintiffs’ group boycott
claim is time-barred. “Although the statute of limitations is ordinarily an affirmative defense
that must be raised in the answer, a statute of limitations defense may be decided on a Rule
12(b)(6) motion if the defense appears on the face of the complaint.”
Ellul v. Cong. of Christian
Brothers
,
A claim under Section 1 of the Sherman Act is subject to a four-year statute of limitations that runs from the date of injury. Plaintiffs filed their initial complaint on April 21, 2020; Defendants argue that any claim based on conduct that occurred before April 21, 2016 is therefore time-barred, and that because the Complaint does not allege any anticompetitive conduct that occurred after this date, the entire Complaint must be dismissed. Plaintiffs argue that the Complaint should not be dismissed under the statute of limitations because “each sale of an odd-lot bond is a continuing violation” and because Defendants fraudulently concealed their conspiracy.
A. Continuing Violation
Plaintiffs argue that “each sale of an odd-lot bond is a continuing violation, starting the
running of the four-year statute of limitations period from that time.” Dkt. No. 133 at 49. The
Supreme Court has held that “[a]ntitrust law provides that, in the case of a ‘continuing violation,’
say, a price-fixing conspiracy that brings about a series of unlawfully high priced sales over a
period of years, ‘each overt act that is part of the violation and injures the plaintiff,’
e.g.
, each
sale to the plaintiff, ‘starts the statutory period running again.’”
Klehr v. A.O. Smith Corp.
, 521
U.S. 179, 189 (1984). However, this rule only applies to overt acts that are
part of the violation
;
“an overt act committed more than four years prior to the filing of the complaint whose effects
were first felt outside the limitations, therefore, usually will not support a cause of action even
the
effects
persist into the limitations period.”
In re Nine West Shoes Antitrust Litigation
, 80 F.
Supp. 2d 181, 191 (S.D.N.Y. 2000);
see also US Airways v. Sabre Holdings Corp.
,
Plaintiffs also argue that “[o]ther acts also continued after April 2016. Morgan Stanley
took steps to punish First Tennessee (a competitive regional bank) that was offering Blackrock
odd-lot trades at narrower spreads than those offered by the Defendant and (¶¶ 138-139) that
Defendants shut down retail investor access to TradeWeb and MarketAxess. (¶¶15, 135, 147,
152, 193.)” Dkt. No. 133 at 49. As to the former, paragraphs 138 and 139 of the Complaint do
not identify with specificity when “Morgan Stanley [took] steps to punish First Tennessee.” But even assuming that this conduct occurred during the statutory period—and assuming, for the
purposes of the statute of limitations analysis that this allegation is well-pled and relates to a
broader well-pled conspiracy—it does not extend the conspiracy as a whole into the statutory
period. As Plaintiffs point out in their briefing, “In a continuing antitrust conspiracy, . . . the
general limitations rule ‘has usually been understood to mean that each time a plaintiff
is injured
by an act of the defendants a cause of action accrues to him to recover
the damages caused by
that act
and that, as to those damages, the statute of limitations runs from the commission of the
act.’” (quoting
Zenith Radio Corp. v. Hazeltine Research, Inc.
,
B. Fraudulent Concealment
“A claim of fraudulent concealment must be pled with particularity.”
IRS I
, 261 F. Supp.
3d at 487. “[A]n antitrust plaintiff may prove fraudulent concealment sufficient to toll the
running of the statute of limitations if he establishes: (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 lack of diligence on his part.”
New York v. Hendrickson Bros.,
Inc.
,
For the first element, “the plaintiff may prove the concealment element by showing either that the defendant took affirmative steps to prevent the plaintiff’s discovery of his claim or injury or that the wrong itself was of such a nature as to be self-concealing.” Id. at 1083–84. Plaintiffs argue both that the alleged conspiracy is “inherently self-concealing” and that “Plaintiffs also pleaded active concealment.” Dkt. No. 133 at 51. However, the group boycott allegations do not support a finding of active concealment or of a self-concealing conspiracy.
First, Plaintiffs argue that “Defendants used secret Bloomberg messages, dealer-to-sales- desk-to dealer channels and online platforms closed to retail investors to collude. They regularly communicated pricing information via these channels that they withheld from odd-lot investors.” (citing SAC ¶¶ 16, 237–241, 261). Plaintiffs fail to address, however, that the alleged secret communications relate only to “pricing information,” id. , not to any group boycott conspiracy. As such, the allegations about private Bloomberg messages and communications among individual dealers and traders on platforms closed to retail investors are entirely irrelevant to concealment of a group boycott conspiracy.
Plaintiffs also argue that Defendants actively concealed their conspiracy by taking
“affirmative steps to throw Plaintiffs off of their scent,” by “hold[ing] out to the public that their
activities are in good faith through detailed codes of conduct promising the highest ethical
standards.” (citing SAC ¶ 263). The Complaint alleges that “each Defendant’s code of
conduct represented that their operations were above-board, providing a false sense of security to
corporate bond investors.” SAC ¶ 263. These statements “are not sufficient to invoke fraudulent
concealment,” because “communications to the community at large will not generally support a
finding of fraudulent concealment.”
In re Merrill, BofA, and Morgan Stanley Spoofing
Litigation. (“Spoofing”)
,
Nor is the alleged conspiracy self-concealing. In
IRS I
, the court emphasized “the visible
nature of Tradeweb’s trading platform and of Defendants’ majority ownership,” and therefore
reasoned that the conspiracy was not self-concealing because “[o]n 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,” which “if anything would have invited, rather than lulled,
skeptical attention.”
In addition, even if Plaintiffs could establish either that Defendants actively concealed
their alleged conspiracy or that the alleged conspiracy was self-concealing, the Complaint’s
allegations establish that Plaintiffs were on inquiry notice of the conspiracy. “[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
,
IV. Antitrust Standing
Defendants also argue that the Complaint should be dismissed because no Plaintiff has antitrust standing to assert a group boycott claim.
“Section 4 of the Clayton Act establishes a private right of action for violations of the
federal antitrust laws, and entitles ‘any 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
Comms., Inc. v. PMC Assocs, L.L.C.
,
Plaintiffs’ Complaint fails to plead antitrust standing at the first inquiry. Defendants argue that Plaintiffs lack antitrust standing because “they do not allege that they or a broker or investment advisor acting on their behalf ever sought to trade corporate bonds on an electronic platform but were blocked from doing so because of the supposed boycott.” Dkt. No. 131 at 59. Plaintiffs respond in their briefing that their injury is not that they were prevented from trading on electronic platforms, but rather that “the Complaint alleges harm through the inflation of prices resulting from continued opacity in a market that, in the absence of Defendants’ collusive conduct, would have seen the development of robust electronic trading platforms.” Dkt. No. 133 at 58. They argue that “[t]he conspiracy affected pricing throughout the market, not just for those that tried to purchase their bonds through the platforms that Defendants conspired to eliminate.” at 58–59 (citing SAC ¶¶ 2, 10). The factual allegations of the Complaint do not support this conclusory statement. The Court therefore does not need to reach the question whether the link is too attenuated to constitute antitrust injury or whether Plaintiffs are efficient enforcers of the antitrust laws.
The Complaint’s introductory section alleges that the case “involves a conspiracy by
defendants to restraint electronic advances in the marketplace that would have reduced
transactional costs for investors in odd-lots of corporate bonds” plus the boilerplate claim “[a]s a
result of Defendants’ conspiracy, Plaintiffs and the Class paid more when buying, and received
less when selling, their corporate bonds, suffering antitrust injury under Section 1 of the Sherman
Act.” SAC ¶ 2. This allegation is merely a conclusion and a recitation of the required elements.
See Rosner v. Bank of China
,
Notably, the section of the Complaint that directly addresses antitrust injury contains a single paragraph—paragraph 258. It alleges that “Plaintiffs and the Class have suffered the quintessential antitrust injury – purchasing a price-fixed product directly from horizontal competitors.” SAC ¶ 258. It makes no reference to a group boycott. This allegation, and the Complaint as a whole, fails to establish that Plaintiffs suffered any antitrust injury stemming from the alleged group boycott conspiracy, rather than from the price-fixing conspiracy that Plaintiffs originally claimed but have now abandoned.
CONCLUSION
The motion to dismiss is GRANTED. Because the Court concludes that any amendment would be futile and also because the statute of limitations on Plaintiffs’ claims has run and Plaintiffs have not proffered a reason for tolling the statute of limitations, and also because Plaintiffs have not requested leave to amend the Complaint if the motion to dismiss is granted, the Complaint is dismissed with prejudice. [4]
The Clerk of Court is respectfully directed to close the motion at Dkt. No. 130 and to prepare a judgment and close the case.
SO ORDERED. Dated: October 25, 2021 __________________________________
New York, New York LEWIS J. LIMAN United States District Judge
Notes
[1] Defendants are Bank of America Corporation, Merrill Lynch, Pierce, Fenner & Smith, Inc., and BofA Securities, Inc. (collectively, “Bank of America”); Barclays Capital Inc. (“Barclays”); Citigroup Inc., and Citigroup Global Markets Inc. (collectively, “Citigroup”); Credit Suisse Securities (USA) LCC (“Credit Suisse”); Deutsche Bank Securities Inc. (“Deutsche Bank”); The Goldman Sachs Group, Inc. and Goldman, Sachs & Co., LLC (collectively, “Goldman Sachs”); JPMorgan Chase & Co. and J.P. Morgan Securities LLC (collectively, “JPMorgan”); Morgan Stanley, Morgan Stanley & Co., LLC, and Morgan Stanley Smith Barney LLC (collectively, “Morgan Stanley”); NatWest Markets Securities Inc. (“NatWest” or “RBS”); and Wells Fargo & Co., Wells Fargo Securities LLC, and Wells Fargo Clearing Services, LLC (collectively, “Wells Fargo”).
[2] Or drawn them all down collectively.
[3] Although Plaintiffs seem to allege that this was part of a course of parallel conduct along with the earlier-referenced InterVest “penalty” by Salomon, two isolated actions that are somewhat similar but separated by about two decades do not constitute parallel conduct. See TSA , 2021 WL 1226670, at *19 (holding with regard to events that “took place ten years or more before most of Plaintiffs’ remaining allegations” that “Plaintiffs have not pled sufficient intervening facts to tie together the alleged 2003 conduct with conduct that took place in 2013 or later”).
[4] Moreover, Plaintiffs have already amended their complaint once in response to Defendants’ first motion to dismiss, at Dkt. No. 117, which raised virtually identical arguments to those considered here.