Major stock exchanges challenge David Lauer’s opinions in HFT case
A lawsuit targeting some of the world’s biggest stock exchanges, including NYSE and Nasdaq, continues at the New York Southern District Court. The exchanges are now fighting the proposed plaintiffs’ class certification by challenging the opinions that support such a certification, including the detailed reports provided by David Lauer.
On November 19, 2021, Bats Global Markets, Inc., Chicago Stock Exchange, Inc., Direct Edge ECN, L.L.C., NYSE Arca, Inc., Nasdaq OMX BX, Inc., New York Stock Exchange, L.L.C., and The Nasdaq Stock Market L.L.C filed several documents at the Court, arguing that Lauer’s opinions have to be excluded.
For starters, let’s recall that this lawsuit was filed back in 2014 on behalf of investors that traded on a registered public stock exchange or a U.S.-based alternate trading venue, between April 18, 2009 and the present, and asserted claims against: (1) registered public stock exchanges located in the United States; (2) a class of brokerage firms; and (3) a class of HFT firms.
The plaintiffs claim, inter alia, that certain defendants allowed HFT firms to profit at the expense of the class and to manipulate securities markets in violation of federal securities laws.
The plaintiffs claim that the Exchanges violated Section 10(b) of the Exchange Act and Rule 10b-5 by offering proprietary data feeds, co-location services, and order types that the plaintiffs allege were used by HFT firms to gain an undisclosed advantage over other investors, including the plaintiffs.
On November 19, 2021, the Exchanges submitted a memorandum of law in further support of their motions to exclude the opinions and testimony of David Lauer.
The plaintiffs count on Lauer’s model for quantifying the harms specified in the complaint against the exchanges. Thus, they believe they can justify the proposed class certification. Proving damages is a basic part of justifying a complaint. So, David Lauer’s testimony and opinions may play a crucial role.
However, the exchanges argue that the model relied upon for class certification must actually measure damages that result from the class’s asserted theory of injury and that the plaintiffs’ assertion that Lauer’s excess markout analysis provides such a model is wrong.
One of the key arguments of the defendants is that Lauer’s methodology is unreliable because his classification of “HFT” and “Non-HFT” firms is arbitrary.
Lauer’s claim that by labeling the counterparties to a trade an HFT and a non-HFT he can establish both causation and impact requires him to reliably distinguish market participants who use the at-issue products to engage in allegedly illegal manipulative trading from those who do not. In his report, Lauer provides a list of 34 entities that he claims are “HFT firms”.
According to the exchanges, Lauer’s list of “HFT firms” is drawn from a headhunter’s website and a list of high volume traders created for a different purpose, and his own say-so. The exchanges claim the list rests entirely on Lauer’s asserted “familiarity with most of them from [his] time in the industry,” which ended in 2011.
The defendants add that Lauer offers no evidence that any entities on his list used the at-issue products, which he admits he did not even investigate.
On the other hand, many broker-dealers for institutional and retail investors which Lauer classifies as “non-HFT” are in fact co-located with Defendants’ exchanges. And Lauer offers no evidence of how any “HFT firms” actually traded during the class period or what services they used at any particular times, relying instead on unsupported generalizations about the trading advantages they supposedly enjoyed during unspecified times over “non-HFT firms.”
The defendants claim that this is inadmissible “expert” ipse dixit—the entities on Lauer’s list are “HFT firms” because he says so.
The exchanges state:
“Lauer just plucked his list off the Internet. He admitted that he conducted no analysis to establish that the firms on his list behaved like “HFT firms”—whatever that might mean—or even made inquiry about their behavior at all”.
Furthermore, according to the exchanges, Lauer has no method to determine that all orders entered by “HFT firms” were proprietary orders on behalf of those firms and that all orders entered by “non-HFT firms” were agency orders on behalf of putative class members.
According to the defendants, Lauer bases his opinions solely on the Market Participant Identifier (MPID) numbers of the executing brokers, which do not identify the investor on whose behalf a trade was made.
This is a critical flaw, the exchanges say, because when the executing broker executes agency trades on behalf of another party, the MPID does not identify who that other party is or whether that other party uses HFT strategies. Thus an institutional investor’s order executed by a firm such as Citadel (which Lauer classifies as an “HFT”) is treated by Lauer’s model as a proprietary “HFT” order, even if Citadel is acting as an agency broker for that institutional investor.
“And so how do you know that on orders submitted by Citadel isn’t an agency order?
I don’t.”
Similarly, Lauer’s model treats every order associated with the MPID of UBS (a broker Lauer classifies as a “non-HFT”) as a “non-HFT” order, even if the order is in fact placed by a proprietary trading firm, engaged in so-called HFT activity, which is using “sponsored access.”
The plaintiffs insist Lauer’s classifications are reliable because of “his years of experience in the industry.”Lauer asserts that certain firms belong on his list because “in my experience in the industry, it was well understood that each of these firms was engaged in high-frequency trading.”
Although he concedes that broker-dealers he deems “non-HFTs” used the at-issue products, he asserts without evidence that “in my experience, any broker-dealers that subscribed to co-location and direct feeds, only did so for their proprietary trading desks, and not their agency or algorithmic trading, despite marketing or sales claims suggesting otherwise.”
The Exchanges claim that Lauer makes all these assertions—based on nothing more than his “experience”–without having made any inquiries into what any particular firm was actually doing during the putative class period, and by ignoring without examination documentary evidence to the contrary.
The Exchanges state:
“Credentials—especially weak credentials like Lauer’s—do not substitute for analysis”.
Hence, according to the defendants, Lauer’s opinions should be excluded as unreliable.
Lauer’s opinions are also dubbed inadmissible for another reason: He presents no known error rate for his HFT classifications. Plaintiffs assert that error rates are “not a basis for exclusion” because it is a “granular factual topic … on which reasonable minds can disagree.” The Supreme Court disagrees: Known error rates “should” be considered in assessing admissibility of quantitative models like Lauer’s. Where the error rate is—as here— completely unknown and cannot be determined, the model is unreliable and inadmissible, the Exchanges claim.
They conclude by requesting that the Court exclude Lauer’s opinions.