Ninth Circuit Rules on Use of Statistics

Last week, the Ninth Circuit Court of Appeals discussed the use of statistics in proving employment discrimination. By statistics, I mean trying to prove discrimination by comparing the numbers employees: for example, men compared to women or people over age 40 compared to people under 40. Schechner & Lobertini v. KPIX-TV involved layoffs at the television station. William Schechner and John Lobertini alleged the employer favored younger employees and women when deciding who to keep and who to let go.

To understand the ruling, one must understand the three-part structure for proving employment discrimination. First, an employee must produce evidence that raises a sufficient stink of discrimination, which is called the prima facie case. For example, if a company kept only women and fired only men, then that smells fishy enough to require an explanation from the employer, which is part two. In part two, the employer must explain a legitimate, non-discriminatory reason for its decision. In part three, the employee must show that the employer’s stated reason is phony, a “pretext.”

The legal issue in last week’s case had to do with part one. The employees showed that the group of people laid off were older than the group of people who remained after the reduction in force by a statistically significant margin. The employer claimed that the statistical evidence was not valid, because it did not take into account its reasons for the decision. For example, the employer immunized high-profile news anchors from lay offs, and the statistics should have excluded them, too, claimed the employer.

The Ninth Circuit disagreed with the employer. The court held that employees may establish the prima facie case – the stink of discrimination – with the raw statistics, and employees’ statistics need not account for the employer’s stated reasons. “A plaintiff who relies on statistical evidence to establish a prima facie case… bears a relative low burden of proof.” The court cited previous cases where very weak statistics were enough to meet part one.

But that is only for part one. The employees still lost because they could not show that the stated reasons for the lay off decisions were mere pretexts or that age was not a motivating factor in the lay off decisions.

Jeff Merrick, Merrick Mediation

Copyright 2012