Antitrust Law Daily Wrap Up, ANTITRUST—N.D. Cal: Expert witness opinion admissible in part in employees’ antitrust action against high-tech companies, (Apr 7, 2014)

By Linda O’Brien, J.D., LL.M.

In a class action by former technical employees against major technology companies for allegedly engaging in a conspiracy to fix and suppress employee compensation, the testimony of the employees’ expert in relation to the use of certain variables in analysis was admissible in part, the federal district court in San Jose, California has decided. However, the expert’s opinion that raised a new theory in his rebuttal was excluded. Also, the defendant companies’ motion for summary judgment based on their motion to exclude the expert’s testimony was denied (In re High-Tech Employee Antitrust Litigation, April 4, 2014, Koh, L.).

Software engineers and other technical workers who were former employees of several major technology companies filed a class action suit against the companies, alleging that they engaged in a conspiracy to fix and suppress employee compensation and to restrict employee mobility through anti-solicitation agreements among the companies. In October 2013, certification of the proposed class was granted. The defendant companies moved to exclude the testimony of the employee’s expert and for summary judgment.

In denying the defendants’ motion to exclude, the court found that the plaintiffs’ expert’s opinion explaining the statistical significance of his use of the “total new hires” variable and real compensation metric in his original conduct regression analysis was proper rebuttal. Federal Rule of Civil Procedure 702 allows the admission of expert testimony if the expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine an issue of fact. The employee’s expert was asked to evaluate whether (1) classwide evidence could show that the anti-solicitation agreements artificially reduced the compensation for members of the class generally and all or most members of the class, and (2) there was a reliable classwide method capable of quantifying the amount of suppressed compensation suffered by the class.

The expert used regression analysis by contrasting compensation before and after when the agreements were in effect to generate estimates the effect on compensation. The detailed explanation in his reply to the motion to exclude his testimony was intended to clarify and contradict evidence presented by the defendants and did not provide any new analysis to justify his inclusion of the variable or the metric in his original analysis.

The court also denied the defendants’ motion to exclude the expert’s testimony under the Supreme Court decision in Daubert v. Merrell Dow Pharms, Inc., 509 U.S. 579 (1993). In null hypothesis testing, “standard errors” determine whether the model used provides statistically reliable evidence of the true value of the estimate. In rejecting the defendants’ argument that the plaintiffs’ expert’s regression analysis was unreliable, the court noted that the fact that two variables used were not significantly significant at the 1, 5, or 10 percent levels went to the weight and not admissibility of the expert’s model. The model’s failure to meet the three levels did not indicate that it was so methodologically flawed as to warrant exclusion, according to the court.

Additionally, the plaintiffs’ expert’s model was not excluded for failure to segregate any effects of unchallenged conduct, since it provided some basis to reasonably estimate the damages from the antitrust violations. Moreover, the defendants’ failed to present evidence on how the inclusion of the “total new hires” variable in the expert’s model was not inconsistent with the plaintiffs’ theory of liability.

However, the expert’s opinion that his original conduct regression with clustered standard errors should be evaluated using a 50 percent statistical theory was excluded, since it was first raised in his reply to the motion to exclude his testimony. In response to the employees’ motion for class certification, the defendants’ expert expressly criticized the plaintiff’s expert’s conduct regression as unreliable because it failed to meet the 1, 5 or 10 percent statistical significance levels and did not account for the fact that compensation for employees within the same firm was correlated. The plaintiffs’ expert had numerous opportunities to address the criticism and raise the new theory but failed to do so. Thus, the new theory was untimely disclosed and excluded as improper rebuttal.

The case is No. 5:11-cv-02509-LHK.

Attorneys: Anne Brackett Shaver (Lieff, Cabraser, Heimann & Bernstein LLP) for Siddharth Hariharan, Brandon Marshall, Michael Devine, Mark Fichtner, and Daniel Stover. Robert Addy Van Nest (Keker & Van Nest LLP) for Adobe Systems Inc. George A. Riley (O’Melveny & Myers LLP) for Apple Inc. Anne M. Selin (Mayer Brown LLP) for Google Inc. Gregory P. Stone (Munger Tolles & Olson LLP) for Intel Corp. Deborah A. Garza (Covington and Burling LLP) for Pixar.

Companies: Google Inc.; Apple Inc.; Intuit Inc.; Intel Corp.; Adobe Systems Inc.; Lucasfilm Ltd.; Pixar.

Cases: Antitrust CaliforniaNews