On March 2, 2022, the Pricing Conduct and Insurance and Financial Services Committees of the ABA Antitrust Section hosted a panel discussion titled “Update on Financial Benchmark Manipulation Litigation.” Deborah Elman (Garwin Gerstein & Fisher) and Robin van der Meulen (DiCello Levitt Gutzler) moderated the panel, which featured economists and attorneys from both prosecution and defense teams: economists Jamie McClave Baldwin (Infotech Consulting) and Kris Comeaux (Analysis Group), and attorneys Jefferson Bell (Gibson, Dunn & Crutcher) and Matthew Perez (DiCello Levitt Gutzler).
The panel addressed recent financial benchmark manipulation decisions and the treatment of economic evidence by courts, with legal and economic perspectives from counsel on both the plaintiff and defense sides. The discussion followed the framework of the “life of a case,” and focused primarily on the earlier stages of litigation.
For many financial benchmark manipulation matters, some sort of government investigation provides an early “tell” that something may be amiss. Since plaintiffs do not have the same investigative powers as regulators, plaintiffs’ counsel often look to public data sources to form the basis of a pleading that can meet plausibility standards. Although direct evidence – such as testimony from a confidential informant or market insider who has access to evidence such as chats among traders – exists in some cases, in many cases plaintiffs’ counsel rely primarily on indirect or circumstantial evidence (media reports, research of public data sources, etc.) to support an inference that something unusual and potentially collusive is happening that is not explained by normal market conditions.
To support their allegations, plaintiffs’ counsel may employ statistical analyses of publicly available and often aggregate-level data to identify abnormal patterns suggestive of collusive activity. Such analyses yield probabilities or odds ratios that can prove effective in demonstrating to a court the necessity of discovery and further investigation.
To identify unusual and potentially collusive patterns, counsel and any economists or statisticians they employ must first be able to explain how the markets and financial instruments in question work, including how they are priced, how they trade, and how they settle. Recognizing unusual trading patterns has become increasingly challenging in light of the rapid evolution of market manipulation tools, making collusive behavior progressively more difficult to detect. It can therefore be highly beneficial to engage economists or other experts before a complaint is filed to allow sufficient time to contend with complex markets and develop appropriate statistical analyses. The analyses presented need to be robust and explained in a way that is digestible and persuasive to a judge, many of whom do not have quantitative backgrounds; thus, it is critical to get these analyses “right.” To meet the pre-discovery plausibility standard, counsel, as well as their experts, must consider whether the data fit into a compelling narrative and whether they demonstrate causality – not just correlation. The nature of publicly available aggregate data may make it challenging to tie the identified atypical patterns to specific defendants; however, these data can be information-rich, and economists and statisticians can do a lot at this stage relying solely on public information. For example, demonstrating through odds ratios that unusual patterns are very low-probability events given what would be expected to occur under “normal” competitive forces can be effective in bolstering circumstantial evidence in the complaint.
On the defense side, the pre-complaint stage can be quiet, and the range of preparedness from the defense once a complaint is filed varies. As complaints have gotten more sophisticated, defense counsel have been charged with a burden that extends beyond simply dispelling allegations with straightforward explanations. Rather, they must evaluate the alleged causal link between the patterns in the data and the defendants’ behavior presented in the complaint. Thus, it may also be advantageous for defense counsel to involve economists or other experts in the early stages of the case. An expert might contribute by helping to explain the complexity of the market or a specific financial instrument, and demonstrating why aggregate statistics may not capture certain nuances. The defense may benefit from drawing on internal datasets that contain more granular data, such as time-stamped trading data, in addition to those that are publicly available, and may endeavor to demonstrate how an alleged pattern of unusual behavior identified by a plaintiff could be undercut by a closer look at the data. For example, an expert might counter a plaintiff’s injury claim by arguing that price fluctuations make it nearly impossible to show harm absent the plaintiff’s ability to identify a specific time when the harm occurred. Experts can also assist in defining the sample of transactions to test, and, by thinking ahead about the sample definition, they can suggest that defense counsel attempt to restrict the data produced in discovery to only the most relevant and responsive types. These approaches may require discovery of the data and analyses that underlie the statistics presented in the complaint and the motion to dismiss.
Engaging experts to analyze aggregate-level and publicly available data can also uncover defenses related to standing and the statute of limitations. Defense counsel may argue that a complaint that contains extensive analysis of publicly available data raises questions about data sources and how data were used. In addition, defense counsel may argue that aggregate statistics alone are insufficient to impute wrongful conduct to a particular defendant and that evidence uncovered in the discovery process will be necessary. This can be especially challenging in certain markets, such as foreign exchange markets, which are built on anonymity and in which only the exchange administrator has access to granular data. In these cases, the complaint may name the exchange administrator, such as the Chicago Board Options Exchange, and unnamed “Does” as defendants. The defense may also argue that the aggregate statistics are irrelevant given the complexity of the market.
However, the court may need – or even require – complex data analyses to be simplified and explained by the parties. Educating the court can be very important at this stage in a case; this may even involve judges requesting or counsel suggesting tutorials to educate judges on the markets and financial instruments at issue. Preventing the court from confusion and frustration at this stage is advantageous to both parties, but it is not uncommon for the court to decide to resolve data disputes by proceeding to litigation. Certain judges, such as those sitting in the United States District Court for the Southern District of New York, have greater familiarity with complex financial concepts than others by virtue of their relatively high exposure to these cases. Even there, however, educating the court on the complexities and nuances of these cases is important in the early stages of litigation.
Two threshold issues have emerged in financial manipulation cases: whether a plaintiff can serve as an “efficient enforcer” and how to establish “antitrust injury.” Recent court decisions have underscored the importance of being an efficient enforcer to be able to bring antitrust claims. Courts have historically been reluctant to use the antitrust laws to address alleged manipulation cases concerning financial benchmarks due to their pervasiveness in financial markets and the implications of allowing anyone to sue. When antitrust claims are proven, however, they can be disastrous for defendants in terms of damages and liability, and courts are increasingly sensitive to whether a particular plaintiff is an efficient enforcer through proximity to harm. Proximity may be defined in terms of relationships between the parties (e.g., direct trading among two individuals), or in terms of timing (e.g., trading at or near the time the benchmark is set). Courts have accepted other proxies to establish the plausibility of harm such as large market share (e.g., 90 percent) to demonstrate plausibility that a defendant caused harm such that the case may proceed beyond motion to dismiss.
Throughout the discussion, the panelists agreed that economists and experts can serve as valuable resources in financial benchmark manipulation litigation, particularly when they are involved early in the litigation process, as foundational arguments and analyses are presented to the court and opposing counsel. There was, however, some disagreement about the ability to veer away from the initial analyses underlying the allegations in complaints – that is, the extent to which arguments must retain fidelity to the analyses that form the basis of a complaint – throughout the life of the case. The panelists suggested this and other salient topics for a follow-up discussion.
Ms. Comeaux is a Vice President and Ms. O’Haver is a Manager at Analysis Group. The views expressed are those of the authors and do not necessarily reflect those of Analysis Group.
The statements in this summary are not attributed to any individual speaker and do not necessarily reflect the views of any other speaker(s). The statements are not made on behalf of any firm, client, or other party.
This article was prepared by the Antitrust Law Section's Insurance and Financial Services Committee.
