No Antitrust Injury in Libor Rate-Setting?—What Happened To Effects?
by Peter D. St. Phillip, Jr. (Shareholder, Lowey Dannenberg Cohen & Hart, P.C.) and Raymond P. Girnys (Associate, Lowey Dannenberg Cohen & Hart, P.C.)1
(this article also appears in the in the May 30, 2013 edition of Competition Policy International’s Cartel Column, and can be accessed here: https://www.competitionpolicyinternational.com/no-antitrust-injury-in-libor-rate-setting-what-happened-to-effects.)
“We recognize that it might be unexpected that we are dismissing a substantial portion of plaintiffs’ claims, given that several of the defendants here have already paid penalties to government regulatory agencies reaching into the billions of dollars.” In re LIBOR-Based Fin. Instruments Antitrust Litig., __ F. Supp. 2d __, 11 MD 2262 NRB, 2013 WL 1285338, at *62 (S.D.N.Y. Mar. 29, 2013) (“LIBOR”).
Unexpected, for some, may have been the wrong choice of word. Most lawyers, after reviewing the motion to dismiss hearing transcript, foresaw the outcome. What was surprising, however, was the legal reasoning the Court used to arrive at the conclusion that Plaintiffs, purchasers of financial instruments benchmarked to the allegedly collusively-set London Interbank Offered Rate, did not plausibly plead antitrust injury. This article examines the LIBOR Court’s rationale against the antitrust injury doctrine and suggests that the Court erred by failing to consider the anticompetitive effects of the collusive behavior instead of solely focusing on whether the conduct itself diminished competition. Additionally, the LIBOR Court concluded that plaintiffs failed to plead antitrust injury because plaintiffs could have been harmed in the same way absent collusion. In addition to coming close to begging a jury question, this rationale departs from that commonly undertaken by courts in evaluating “antitrust injury.”
LIBOR serves as the principal benchmark for short-term interest rates globally, affecting the pricing of trillions of dollars’ worth of financial instruments. LIBOR is calculated for ten currencies with fifteen maturities or tenors ranging from overnight through twelve months, including the U.S. dollar (“USD LIBOR”), the only LIBOR currency at issue in the LIBOR Court’s decision. The publication of LIBOR is overseen by the British Bankers’ Association (the “BBA”).
The BBA selects a Contributor Panel of banks for each LIBOR currency, whose interest rate submissions are considered in calculating the daily, composite LIBOR fix.2 Each business day LIBOR is fixed, the LIBOR Contributor Panel banks answer the following question: “At what rate could you borrow funds, were you to do so by asking for and then accepting inter-bank offers in a reasonable market size just prior to 11 AM?”
The Contributor Panel banks submit their rates in response to this question each business day by 11:10 AM London time to Thomson Reuters, the BBA’s agent tasked with calculating the daily LIBOR fix. Each bank allegedly “must submit its rate without reference to rate contributed by other Contributor Panel banks.” LIBOR, at *3. After receiving quotes from each bank on a given panel, Thomson Reuters determines the LIBOR for that day by ranking the quotes for a given tenor in descending order and calculating the arithmetic mean of the middle two quartiles, i.e., the middle eight quotes excluding the four highest and four lowest. Id. The arithmetic mean of these quotes is the resulting LIBOR fix for that business day. Id.
Plaintiffs, purchasers and sellers of myriad financial instruments pegged to USD LIBOR, allege that the Defendant Contributor Panel banks colluded to artificially depress their rate submissions during the period August 2007 to May 2010. For motive, they allege (1) “well aware that the interest rate a bank pays (or expects to pay) on its debt is widely, if not universally, viewed as embodying the market’s assessment of the risk associated with that bank, Defendants understated their borrowing costs (thereby suppressing LIBOR) to portray themselves as economically healthier than they actually were”; and (2) “artificially suppressing LIBOR allowed Defendants to pay lower interest rates on LIBOR-based financial instruments that Defendants sold to investors, including [plaintiffs], during the Class Period.” Id. at *4. Because the complaints under review by the LIBOR Court were filed before public disclosure of the volumes of instant messages culled by government regulators in support of their June 2012 settlement with Barclays, plaintiffs did not point to documentary evidence of collusion. Instead, they offered a host of statistical evidence to plead that USD LIBOR diverged from benchmarks that it would normally track. Plaintiffs contended that this empirical evidence of dislocation circumstantially suggested collusion.
The crux of Plaintiffs’ alleged injury was that Defendants’ agreement to suppress USD LIBOR negatively impacted their financial instruments, either due to a formulaic increase in the costs or decrease in rates of return under the instruments’ terms.
THE LIBOR COURT’S ANTITRUST INJURY RATIONALE
The USD LIBOR Contributor Panel Bank Defendants moved to dismiss the antitrust damages claims on four grounds: plaintiffs (1) did not adequately plead a contract, combination, or conspiracy; (2) relatedly, failed to allege a restraint of trade; (3) lacked antitrust standing; and (4) were barred from recovery under the “indirect purchaser” rule enunciated in Illinois Brick Co. v. Illinois, 431 U.S. 720 (1977). Finding no “antitrust injury,” the LIBOR Court limited its discussion to this ground. LIBOR, at *10.
The LIBOR Court first observed that private antitrust plaintiffs seeking damages under the Clayton Act for a violation of the Sherman Act must have antitrust standing, comprising of both “antitrust injury” and “proper plaintiff” status under the four-part test articulated in Associated General Contractors v. California State Council of Carpenters, 459 U.S. 519 (1983). LIBOR, at *11 (citing In re DDAVP Direct Purchaser Antitrust Litig., 585 F.3d 677, 688 (2d Cir. 2009)). The LIBOR Court next canvassed “antitrust injury” precedent to determine the contours of that question. In doing so, it cited two main Supreme Court cases—Atlantic Richfield Co. v. USA Petroleum Co., 495 U.S. 328 (1990)(“ARCO”) and Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477 (1977)(“Brunswick”). Id. The LIBOR Court cited the holdings from both of these cases correctly:
- “a private plaintiff can recover [damages] for [an antitrust] violation only where ‘the loss stems from a competition-reducing aspect or effect of the defendant’s behavior.’” Id. (citing ARCO, 495 U.S. at 344 (emphasis in the original))
- “[p]laintiffs must prove antirust injury, which is to say injury of the type the antitrust laws were intended to prevent and that flows from that which makes defendants’ acts unlawful. The injury should reflect the anticompetitive effect either of the violation or of anticompetitive acts made possible by the violation.” Id. (citing Brunswick, 429 U.S. at 489)
From there, the LIBOR Court cited a decision from the Southern District of New York, Nichols v. Mahoney, 608 F. Supp. 2d 526, 543-44 (S.D.N.Y. 2009), for the proposition that “a plaintiff must demonstrate not only that it suffered injury and that the injury resulted from defendants’ conduct, but also that the injury resulted from the anticompetitive nature of defendant’s conduct.” Id. The LIBOR Court observed that a per se violation of the antitrust laws does not alone establish antitrust injury. Id. (citing ARCO, 495 U.S. at 344).
After setting out the antitrust injury standards, the LIBOR Court applied them to the allegations. First, the LIBOR Court concluded that “plaintiffs do not argue” that the rate-setting process itself violated the antitrust laws and cited allegations from the complaint that the anticompetitive conduct “had severe adverse consequences on competition” by rendering the prices of the LIBOR-based instruments they traded during the class period “artificial as a result of Defendants’ unlawful conduct.” Id. at *12. From this, the LIBOR Court reasoned that the harm plaintiffs identify did not stem from “any anticompetitive aspect of defendants’ conduct.” Id. Because, as plaintiffs conceded at oral argument, “the process of setting LIBOR was never meant to be competitive,” plaintiffs’ harm did not qualify as antitrust injury. Id.
While the LIBOR Court acknowledged that “there might have been antitrust injury if defendants had restrained competition in the market for LIBOR-based financial instruments or the underlying market for interbank loans,” it concluded Plaintiffs did not make either allegation. Id. at *13. The allegations of consequential harm to the financial instruments due to the rate rigging were not antitrust injury, the LIBOR Court reasoned, because “plaintiffs’ injury [did not] result from an anticompetitive aspect of defendants’ conduct.” Id. Critical to this reasoning was the LIBOR Court’s understanding that “the collusion must have been anticompetitive” by “involving a failure of defendants to compete where they otherwise would have.” Id. The effects of a collusive agreement in a non-competitive environment, according to the LIBOR Court, were irrelevant to the antitrust injury question.
The LIBOR Court went on to conclude that there was no “harm to competition” in the interbank loan market as “LIBOR . . . does not necessarily correspond to the interest rate charged for any actual interbank loan,” instead the index was intended to convey information about prevailing rates. Id. at *13. As Plaintiffs did not allege a failure of competition in the interbank loan market, the Court did not consider the anticompetitive effects of defendants’ conduct in this market in rendering its decision. Id.
Confirming this analysis, the LIBOR Court determined it appropriate, under Brunswick, to evaluate whether plaintiff “could have suffered the same harm under normal circumstances of free competition.” Id. The LIBOR Court discussed Brunswick’s conclusion that the bowling center plaintiff’s injury resulting from the acquisition of failing competitors had no connection to the size of the acquirer or its competitors. Id. (citing Brunswick, 429 U.S. at 487). Because the loss in Brunswick did not occur due to anything that made the acquisitions unlawful—the allegedly improper concentration in the bowling center market—the harm was not antitrust injury.
The LIBOR Court also analyzed ARCO in connection with this confirmatory analysis. The LIBOR Court reads ARCO to hold that the lowering of prices to levels above the predatory was the “very essence of competition” and, therefore, the harm to the competitor plaintiff of missing out on the opportunity to charge higher prices was not antitrust injury. Id. at *14.
Analyzing the alleged harm against the backdrop of Brunswick and ARCO, the LIBOR Court concluded that it could have resulted from “normal competitive conduct.” Id. That was so, the LIBOR Court reasoned, because plaintiffs could have suffered the same injury if the Contributor Panel banks had not colluded, and had misreported their false quotes independently.
Contrasting what it viewed as traditional price-fixing conspiracies, the LIBOR Court concluded that supracompetitive pricing could only exist in those settings through collusion, whereas independent misreporting of LIBOR tenors was financially rational for the Contributor Panel banks. Because the independently misreporting banks would not have been concerned about losing business, “collusion … would not have allowed them to do anything that they could not have done otherwise.” Id.
Even accepting the proposition that defendants could not have “clustered” their submissions without collusion, the LIBOR Court considered this focus as answering the wrong question. Id. at *15. “[W]hether the quotes would have formed a ‘cluster’ or not is irrelevant; plaintiffs’ injury resulted not from the clustering of LIBOR quotes, but rather from the quotes’ alleged suppression.” Id.
Several factors led the LIBOR Court to conclude that less harm to competition resulted from plaintiffs’ allegations than from those in Brunswick and ARCO. Acknowledging that a change in LIBOR would have set the baseline from which market actors competed to set the price of LIBOR-based instruments, the Court did not view plaintiffs as complaining that this competition was impaired. Id. Moreover, plaintiffs did not contend that defendants’ conduct changed their marketplace positions relative to their competitors. There being no “structural effect” of collusive LIBOR rate-setting on the positions of the competitors, the LIBOR Court found more harm to competition alleged in Brunswick (the entry of the large competitor modified the competitive landscape for bowling centers) and ARCO (the maximum price-setting disrupted the competitive forces determining price for retail gasoline).
The LIBOR Court next turned to plaintiffs’ argument, advanced at oral argument, that because LIBOR is a proxy for competition in the interbank lending market, its collusive manipulation harmed competition. Id. at *16. The Court observed that this argument did accurately describe the conduct in the sense that a false LIBOR “no longer reflected competition in the market for interbank loans” so its value as a proxy was minimized. Notwithstanding the proxy argument’s intuitive appeal, however, the LIBOR Court concluded that this result derived from manipulation, not collusion.
Concluding its antitrust injury analysis, the LIBOR Court proceeded to distinguish plaintiffs’ cases on a variety of grounds: (1) several did not involve proxies similar to LIBOR; (2) some involved competitive harm not found for the reasons set forth earlier; and (3) the so-called “list price” and indices manipulation cases were factually distinguishable.
The Court’s Primary Misinterpretation of the Antitrust Injury Test Was Focusing Exclusively on Defendants’ Conduct, Rather Than The Effects of Defendants’ Conduct.
In a subsection of the opinion describing its “antitrust injury” rationale, titled “Defendants’ Alleged Conduct Was Not Anticompetitive,” the LIBOR Court applied the Brunswick test to plaintiffs’ allegations. The Court ignored the effects of the collusive manipulation on plaintiffs’ financial instruments in favor of an exclusive inquiry into the conduct of the defendants. This was the primary error in the LIBOR Court’s rationale. In setting the antitrust injury standard, the LIBOR Court correctly quoted the holdings of Brunswick and ARCO, both of which directed the federal courts to look to the effect of the antitrust conduct in analyzing antitrust injury. Id. at *11 (citing ARCO, 495 U.S. at 344 (“the loss [must] stem from a competition-reducing aspect or effect of the defendants’ behavior”) (second emphasis added); citing Brunswick, 429 U.S. at 489 (“The injury should reflect the anticompetitive effect either of the violation or of anticompetitive acts made possible by the violation”)) (emphasis added). After citing these standards, the LIBOR Court veered off track in citing a trial court ruling requiring, for purposes of antitrust injury, that a plaintiff demonstrate that the conduct be anticompetitive.
There is ample precedential support in the case law that anticompetitive effects suffices to establish antitrust injury. Let’s start with the other leading Supreme Court case on antitrust injury that the LIBOR Court did not cite—Blue Shield of Virginia v. McCready, 457 U.S. 465 (1982).
In McCready, the Supreme Court renounced the very rationale used by the LIBOR Court in reaching its conclusion that the Complaint failed to adequately allege “antitrust injury.” There, plaintiff was a beneficiary of group health coverage offered by Defendant Blue Shield of Virginia to her employer. Blue Shield covered treatment by psychiatrists, but not psychologists, unless the treatment was supervised by a physician. McCready alleged that Blue Shield and Neuropsychiatric Society of Virginia agreed to engage in a group boycott of psychologist reimbursement in violation of the antitrust laws.
Blue Shield contended that McCready did not allege “antitrust injury” under Brunswick because she: (1) did not visit a psychiatrist whose bills were artificially inflated due to the conspiracy; (2) did not pay a physician to oversee her psychologist; and (3) did not claim that her psychologist’s bill was higher than it otherwise would have been without the conspiracy. Id. at 481. Citing Brunswick, Blue Shield argued that McCready’s injuries did not flow from the anticompetitive effect of the alleged violation. Id.
The McCready Court held that “Brunswick is not so limiting.” Id. at 482. The Supreme Court held “as we made clear in a footnote [in Brunswick], a § 4 plaintiff need not ‘prove an actual lessening of competition in order to recover. [C]ompetitors may be able to prove antitrust injury before they actually are driven out of the market and competition is thereby lessened . . . .’” Id. Concluding that McCready charges Blue Shield “with a purposefully anticompetitive scheme” that was “inextricably intertwined with the injury the coconspirators sought to inflict on psychologists,” the Supreme Court concluded that McCready’s injury “falls squarely within the area of congressional concern.” Id. at 484.
The Second Circuit recognizes the McCready Court’s limitations on Brunswick. The LIBOR Court cited its decision in In re DDAVP Direct Purchaser Antitrust Litig., 585 F.3d 677, 688 (2d Cir. 2009) for authority that a plaintiff must show antitrust injury. LIBOR, at *11. The DDAVP Court, however, rejected a similar line of reasoning offered by the USD LIBOR defendants—that the conduct of the defendants did not demonstrate antitrust injury to the plaintiffs. The Second Circuit held, “[a]lthough the defendants’ conduct at issue targeted their competitors, such as Barr, the plaintiffs’ claimed injury of higher prices was ‘inextricably intertwined’ with the conduct’s anti-competitive effects and thus ‘flow[ed] from that which makes defendants’ acts unlawful.’” DDAVP, 585 F.3d at 688 (citing Blue Shield of Va. v. McCready, 457 U.S. at 484) (emphasis added).
The LIBOR Court’s mistake in solely focusing on the nature of the conduct is plain from its reasoning. The LIBOR Court concluded that while Plaintiffs’ alleged injury “might suggest” that defendants fixed prices and thereby harmed Plaintiffs, “they do not suggest that the harm plaintiffs suffered resulted from any anticompetitive aspect of defendants’ conduct.” LIBOR, at *12. The Court highlighted that the process of setting LIBOR was never intended to be a competitive process; instead it was “a cooperative endeavor wherein otherwise competing banks agreed to submit estimates” of their borrowing rates.
The words chosen by the Court to state the test critically matter to the outcome. The Court focused on “any anticompetitive aspect of defendants’ conduct” not the effects of such conduct. Id. This phrasing mimics that of the Supreme Court in ARCO when characterizing “antitrust injury” as harm “attributable to an anti-competitive aspect of the practice under scrutiny” 495 U.S. at 334 (citing Cargill, Inc. v. Monfort of Colorado, Inc., 479 U.S. 104, 109-110 (1986)). In applying the “antitrust injury” test, the LIBOR Court interpreted ARCO to focus on defendants’ conduct—the practice of rate-setting—and not the effects of that conduct.
But the LIBOR Court did not cite McCready. The McCready Court stated bluntly that an antitrust plaintiff need not prove a lessening of competition in order to show antitrust injury. The LIBOR plaintiffs’ injuries—having to pay more for their financial instruments that were tied to the LIBOR rate—flowed from the same conduct that violated the antitrust laws, the collusion that allegedly moved the rate from what it otherwise would have been. The effect was consumers paying higher prices—the bull’s eye upon which the Sherman and Clayton Acts both train. Plaintiffs were not competitors complaining about lost profit opportunities like those in Brunswick, Cargill, and ARCO, they were the victim consumers of collusively-rigged financial products. The holders of these LIBOR-based financial instruments do not complain that the conspiracy prevented them from making more profits on these products. Just the opposite, they complain that they lost money as a result of the conspiracy.
The “antitrust injury” hurdle was easily surmounted by the LIBOR pleadings.
The LIBOR Court’s Second Erroneous Focus on What Would Have Happened Absent the Collusion.
The LIBOR Court made a second mistake in analyzing antitrust injury—it concluded that because plaintiffs could have suffered the same injury “could have resulted from normal competitive conduct,” LIBOR, at *14, antitrust injury was absent. This rationale, taken to its logical conclusion, would eliminate all damages recovery for antitrust violations. Many courts have rejected this same argument. See, e.g., Virginia Vermiculite, Ltd. v. W.R. Grace & Co.- Connecticut, 156 F.3d 535, 540 (4th Cir. 1998) (recognizing that Defendants are “foreclosed from challenging causation [on a Sherman Act Section 1 claim] simply on the basis that it could have achieved the same result through lawful means”); In re Cardizem CD Antitrust Litig., 105 F. Supp. 2d 618, 648, n. 16 (E.D. Mich. 2000) (“To accept Defendants’ argument, the Court must also accept the argument that there can never be an antitrust violation if the antitrust defendant can posit an argument that it could have lawfully done the same thing it is accused of doing collusively. . . . There are many things a defendant can do unilaterally without offending the antitrust laws that it cannot do collusively. For example, consider two gas stations that have control over a large geographic market and independently price their gas in such a way that they are within a penny or so of each other. That is not an antitrust violation. However, if the two gas stations, which have a monopoly over gas in the geographic market area, agree to fix the price of gasoline, then there is an antitrust violation. The violation would meet the Brunswick criteria for antitrust injury because the claimed injury is of the type the antitrust laws were meant to discourage; agreements to fix prices. Also, the plaintiff's injury (having to pay higher, deliberately set prices) is causally related to the defendant’s anticompetitive acts. The same analysis applies here.”) (citing Virginia Vermiculite, Ltd. v. W.R. Grace & Co., 156 F.3d 535, 539-40 (4th Cir. 1998)). See also, Andrx Pharm., Inc. v. Biovail Corp. Int’l, 256 F.3d 799, 813 (D.C. Cir. 2001) (same).
As these decisions demonstrate, positing that the same injury could have happened without collusion does nothing to advance whether a complaint adequately pleads “antitrust injury.” The LIBOR Court’s reliance on a counterfactual hypothetical (the harm could have occurred if the defendants acted independently), instead of crediting the LIBOR Complaint’s assertion that defendants’ colluded, erroneously discounted the pleaded facts.
The LIBOR Court’s adoption of defendants’ “antitrust injury” argument rested upon two errors of law. It should not withstand appellate scrutiny.
1 The authors, Peter D. St. Phillip, Jr., Esq. and Raymond P. Girnys, Esq., are a shareholder and associate, respectively, with the firm Lowey Dannenberg Cohen & Hart, P.C. The firm is prosecuting similar rate-setting cases involving other financial instruments.