techVer la versión en español aquí

In January, the U.S. Department of Labor (“DOL”) brought suit in a San Francisco administrative court against Google, Inc. to require Google, as a federal contractor, to allow the government to inspect Google’s pay records.  While the DOL was demanding various documents regarding employees’ compensation history for compliance purposes, it was clear the underlying reason was the DOL’s suspicion that Google was committing gender pay discrimination.

Ironically, in 2016, Google published advice to help other employers analyze their pay structures in order to close the gap in pay between genders. On Equal Pay Day earlier this month, Google also announced to the world that it had proudly “closed the gender pay gap globally.”

google pay

In the administrative court case, the DOL responded: not so fast! Janette Wipper, a DOL regional director, testified last week that the DOL found “systemic compensation disparities against women pretty much across the entire [Google] workforce.” Janet Herold, regional solicitor for the DOL, added “The investigation is not complete, but at this point the department has received compelling evidence of very significant discrimination against women in the most common positions at Google headquarters.”

The internet giant doubled down on its claim that it had closed the gender gap by posting a blog about its pay methodology. Google explained that each year, an amount for each employee’s compensation is proposed based on role, job level, job location, and current and recent performance ratings.  Google stated that the analysts who calculate the suggested compensation have no access to employees’ gender –in essence, it is a “blind” analysis.

Google’s blind analysis uses the following methodology:

google pay2

How can both Google and the DOL be so sure about their respective, completely opposing opinions? Could the disagreement be rooted in the way the DOL and Google define the “roles”? For example, if Google and the DOL compared compensation between similar roles instead of exact roles, would the gender pay gap widen or narrow? If there is a pay gap, could it be that the information used by the Google analysts to “blindly” assign pay is tainted by gender bias—for example, are women not being hired for certain “roles” with higher rates of pay because only men traditionally held those roles? Or are women being scrutinized more harshly in their evaluations, which are then used in the “blind” assessment as a factor in determining pay? The answer is unclear.

Ultimately, the DOL has yet to provide its methodology for determining that Google is discriminating against women. It will be interesting to see how this case unfolds over the next few months. Regardless of the outcome, the case serves as a good reminder to employers that they need to  identify and address any potential inconsistencies or biases that could unlawfully affect compensation and benefits.

*An earlier version of this post stated that the DOL lawsuit was brought in federal court. The blog post has since been revised to clarify that the suit was brought in administrative federal court.