In a recent decision, In re Xcelera.com Securities Litigation, No. 00-11649 (D. Mass. Apr. 25, 2008), Judge D.J. Zobel of the United States District Court for the District of Massachusetts was called upon to assess the testimony of a putative expert witness, Dr. Scott Hakala, tendered by the plaintiffs in a pending securities fraud class action. In Xcelera.com, the plaintiffs had argued that Xcelera misled investors by failing to disclose that a corporate transaction involving Mirror Image could cause their interests in Xcelera to be significantly diluted. Plaintiffs sought to rely on Hakala’s opinions for both their proposed damages theory and their need to prove loss causation, which is a mandatory element of any claim under Section 10(b) of the Securities Exchange Act of 1934. Defendants sought to exclude Hakala’s testimony pursuant to Fed. R. Evid. 702 and the Supreme Court’s decision in Daubert v. Merrell Dow Pharmeceuticals, Inc., 509 U.S. 579 (1993).
Only a few days before the case was to go to trial, the District Court postponed the trial and scheduled a preliminary hearing on defendants’ Daubert motion concerning Hakala. After oral argument on the motion, the District Court ruled that Hakala would not be allowed to testify, which meant that plaintiffs no longer had any means to attempt to establish through expert testimony that defendants’ activities caused any injury to Xcelera shareholders. Based on this ruling, the Court also granted summary judgment to the defendants.
Specifically, the Court found that Hakala’s “event studies and his application of the methodology to the facts in [the] case [were] flawed in several respects,” and not the “product of reliable principles and methods.” Xcelera.com, No. 00-11649, slip op. at 1 (quoting Fed. R. Evid. 702). The Court noted four areas where Hakala’s work was unsound, all of which led the Court to exclude him from testifying in the case: (1) his use of “dummy variables;” (2) his failure to use relevant event dates in his event studies; (3) his failure to square his methodology with the theory of market efficiency; and (4) his failure to consider other factors that could have affected Xcelera’s stock price during the relevant time period.
Misapplication of Dummy Variables: Dummy variables are used in event studies to capture and control for the effects of company-specific events by assigning a value for the days under study and zero elsewhere. Hakala used dummy variables in his event studies for “every date on which he claims there was any news at all about Xcelera that might have affected the stock price.” Id. at 1. The Court held that, “[a]lthough the academic literature supports the use of dummy variables for events in which significant company-specific news is released, no peer-reviewed journal supports the view that dummy variables may be used on all dates on which any company news appears.” Id. (emphasis in original).
Failure To Acknowledge Other Relevant Dates: The Court also discussed the fact that a key date in the chronology of relevant facts was August 1, 2000 – the day after Xcelera issued its Form 20-F, containing the purported corrective disclosure, and the same day an analyst report was released that discussed the dilution issue. The Court was troubled by the fact that Hakala failed to include August 1 in his event studies. Id. at 2. From Hakala’s testimony during the hearing, it appeared that he chose to exclude August 1 (the first trading day after the announcement) because he concluded the stock price fluctuated on August 1, and the stock did not have an overnight drop until August 2. (Hakala Apr. 25, 2008 Hr’g Tr. 59:6-13.) The Court ultimately rejected these attempts to explain away the omission of August 1. Hakala had not included either July 31 or August 1 as a relevant date in his original event study, but added July 31 only after defendants filed their summary judgment motion. Hakala claimed that he had added July 31 because the stock price declined on that day due to “negative anticipation” of the dilution issue. Id. The Court observed, however, that no evidence existed in support of such a theory. Id. (noting that “[q]uite simply, his theory does not match the facts”).
Opinions Inconsistent With Market Efficiency: The Court further faulted Hakala’s decision to include August 9, 2000 as a relevant date where he claimed the stock price went down due to additional information related to dilution. Id. at 2-3. The Court could not square the theory of market efficiency with Hakala’s conclusion that August 1 should be excluded (the day after the relevant announcement), but that events starting on August 2 through August 9 should be considered as having a possible effect on the stock price associated with this same issue. Id. at 3 (noting Hakala’s contradictory testimony that, “in an efficient market the release of previously-known information will not effect [sic] the stock price”).
Ignoring Potentially Confounding Factors: Last, the Court was troubled by Hakala’s failure to consider other factors that could have had an effect on Xcelera’s stock price in August 2000, including tax liability issues, the inclusion of a one-time gain that impacted net income per share, the acquiree company’s potential inability to generate revenue, and insider stock sales. Id. at 4-5.
For all of the above reasons, the Court granted the defendants’ motion to exclude Hakala’s testimony.