On the seventh anniversary of the federal Lilly Ledbetter Fair Pay Act, the Equal Employment Opportunity Commission (“EEOC”), in partnership with the U.S. Department of Labor, Office of Federal Contract Compliance Programs (“OFCCP”), announced a proposed rule to collect data from employers that will be used to identify discriminatory pay practices. Under the proposed rule, companies with 100 or more employees, both private employers and federal contractors, would be required to report wages from W-2 earnings and total hours worked for all employees by sex, race, and ethnicity within a 12-month period. It is projected that these new proposed requirements will affect over 63 million employees.

This proposed rule is now in the comment period until April 1, 2016. The EEOC also plans to conduct a public hearing regarding the new rule at some point. If things progress as expected, this rule becomes effective for the September 30, 2017 reporting period.

While the EEOC highlights that the proposed rule also is a benefit to employers because it assists employers “in evaluating their pay practices to prevent pay discrimination” and to avoid enforcement actions, there are legitimate concerns regarding how such data will be interpreted and used by government agencies. Some concerns include the strong likelihood of this data producing false positives and the ability to keep this information confidential.

Pay Data

Currently, the EEO-1 form collects data regarding the number of employees, along with their sex, race and ethnicity, in 10 specifically designated job categories. Under the proposed rule, an employer also would be required to report the number of employees by their sex, race, and ethnicity, within 12 specified pay bands in each of the 10 job categories. These pay bands track the 12 pay bands used by the Bureau of Labor Statistics in the Occupation Employment Statistics survey. The specific pay bands are:

Click here to view table.

The proposed new section of the EEO-1 form is available on the EEOC’s website (click here).

The pay data will be taken from employees’ total W-2 earnings for a 12-month period looking back from a pay period between July 1st and September 30th. The EEOC believes the benefit to using W-2 earnings is that it includes total earnings, including wages, salaries, and other compensation such as commissions, tips, taxable fringe benefits, overtime pay, shift differentials and bonuses. Also, the EEOC insists that using W-2s places the least amount of burden on an employer because this information is already gathered and most human resources information systems allow for calculations for any 12-month period, not just the calendar year.

A concern for employers is that there is no way to indicate on the EEO-1 form neutral factors, such as experience, education, or performance that might account for or explain any pay differentials. Accordingly, it is anticipated that this data may produce many false positives which will force employers to exert additional time and resources to defend their pay practices.

Total Number of Hours Worked

Under the proposed rule, an employer would also have to record the total number of hours worked by employees, broken down by sex, race, and ethnicity, in each pay band. The EEOC states that the reason for providing the number of hours worked is to take into account part-time or partial-year employees. Specifically, data on number of hours worked “will allow analysis of pay differences while considering aggregate variations in hours.”

As the rule is currently drafted, it is unclear how this information will achieve that purpose when it does not take into account factors which could skew results such as overtime hours, or other supplemental earnings like bonuses or commissions, which may be less due to part-time work. Another issue not addressed by the EEOC is how hours for salaried employees would be calculated. In fact, the EEOC acknowledges that it is not certain how to report hours worked for salaried employees and is requesting employer input on that issue.

Data Analysis

The EEOC states that it plans to use the pay data to: (1) assess complaints of discrimination; (2) focus agency investigations; and (3) identify existing pay disparities that may warrant further examination. The agency claims the information from the pay bands will be used to “compute within-job-category variation, across-job-category variation, and overall variation” to discern potential discrimination. The EEOC plans to develop statistical tools for staff to use on their computers so that they can conduct this type of analysis. The EEOC will also publish aggregate data so that employers can conduct their own analysis of their pay practices

Yet, the EEOC has not identified what statistical methodology it plans to use. Thus, it is not possible to assess whether the EEOC’s statistical analysis would hold up under judicial scrutiny or would be rejected by the courts.

Confidentiality Concerns

The EEOC does not guarantee that the pay data will be kept confidential and not subject to FOIA requests through both the EEOC and the DOL. Specifically, the EEOC states that Title VII forbids it from making public the EEO-1 data before a Title VII proceeding is instituted. As for OFCCP, it promises to keep the EEO-1 data confidential “to the maximum extent permitted by law, in accordance with the Freedom of Information Act Exemption 4 and the Trade Secret Act.”

The EEOC attempts to counter any confidentiality concerns by claiming that since the data is provided in the aggregate and not on based on individual employees, there is no confidentiality issue. Aside from the fact that it’s really more of a privacy issue, that response underplays the importance that the compensation data could provide to competitors and ignores the free discovery that it would provide to the plaintiff’s bar by allowing access to this pay data.

Next Steps

Given the complexity of defending discriminatory pay claims, in preparation of the enactment of this new rule employers should conduct pay equity analysis to assess any issues prior to submitting any pay data. Additionally, companies affected by the proposed rule may wish to consider submitting comments.