On July 25, 2015, Judge Barbara Lynn of the Northern District of Texas issued a formative opinion in the class actions securities arena. The case,The Erica P. John Fund, Inc., et al. v. Halliburton Co., et al., No. 3:02-CV-1152-M, is viewed as a bellwether among securities class actions due to its treatment of novel issues regarding, among other things, a defendant’s ability to disprove reliance—i.e., a causal link between alleged misrepresentations and an eventual drop in stock prices upon correction—for purposes of class certification.

Rather than requiring plaintiffs to prove reliance for each individual shareholder, securities class action cases have long permitted a more efficient approach to establish the necessary causal link. This approach, set forth in Basic v. Levinson, 485 U.S. 224 (1988), invokes a rebuttable presumption in favor of reliance if certain elements are met. Recently, in connection with the Halliburton case, the Supreme Court held this presumption can be rebutted if a defendant shows an alleged misrepresentation did not affect the market price of a security. If the presumption is rebutted, the class cannot be certified.

In the July 25 opinion, the Halliburton court addressed one of the key questions left open after the Supreme Court’s ruling: What level of proof is necessary to rebut this reliance presumption? After considering plaintiffs’ motion for class certification for claims arising from six corrective disclosures of Halliburton, the court granted class certification with respect to one of the alleged disclosures, and denied certification with respect to the remaining five.

In arriving at this decision, the court first concluded that Halliburton, as defendant, bore the burden of both production and persuasion. This meant Halliburton was required to “persuade the Court that its expert’s event studies are more probative of price impact than [plaintiffs’] expert’s event studies.” The court also declined to consider Halliburton’s argument that the alleged disclosures were not “corrective,” because such an argument went to the underlying merits of the claim and was more properly considered at a later stage of the litigation.

Regarding evidence of price impact (i.e., impact of alleged disclosures on the market price of Halliburton shares), the court conducted a careful analysis of the evidence and expert testimony presented by both sides. One noteworthy aspect of the court’s analysis is its endorsement of a “multiple comparison adjustment” advocated by defendants’ expert, despite acknowledging that such adjustments are rarely utilized in event studies for securities litigation. The court indicated its decision to apply these adjustments, which help mitigate the risk of relying on “statistical flukes,” was due to the substantial number of comparisons being tested for statistical significance in plaintiffs’ expert’s analysis. The court’s decision was also at least partially influenced by allegations from Halliburton that plaintiffs reverse engineered their claims by initially running a statistical comparison to uncover thirty-five dates found to have a statistically significant price movement and, from these dates, selecting events to allege misstatement based on their coinciding with news releases.

Other determinations by the court included utilizing an additional “Analyst Index” advocated by plaintiffs to measure the statistical significance of stock price movement. The court also determined it would consider the effects of events on the market price of stock with reference to a one-day, rather than a two-day, window following the announcement or event in question.

Utilizing these methods, the court determined only one of the six alleged disclosures had an impact on the market price of Halliburton stock. The disclosure in question was Halliburton’s announcement on December 7, 2001, that it had lost a substantial verdict as the defendant in an asbestos litigation case. Plaintiffs argued this disclosure “corrected” Halliburton’s previous disclosures that it was not exposed to significant liability in asbestos cases and was efficiently handling such litigation. Despite Halliburton’s argument that this drop was caused by events other than the disclosure plaintiffs alleged, the court found that Halliburton failed to meet “its burden of showing lack of a price impact” with respect to the December 7 announcement. This finding led the court to grant plaintiffs’ motion to certify with respect to only the December 7 disclosure.

The most recent Halliburton class certification decision is not altogether unsurprising: it likely reflects the decision of a judge whose class certification decisions were twice reversed by the U.S. Supreme Court. The decision is a win for plaintiffs, who finally obtained class certification after a seven-year battle. Defendants, however, are well positioned going forward since they successfully defeated five of the six corrective disclosures and are now left with a single corrective disclosure to attack on summary judgment. Judge Lynn’s analysis will likely be a roadmap for both securities class action plaintiffs and defendants in evaluating class certification in the future.