The California legislature is poised to continue its trailblazing streak of equal pay legislation with a new pay gap reporting bill. If approved and signed by Governor Jerry Brown, AB 1209 would add Section 2810.7 to the California Labor Code and require certain large employers to report pay gap statistics on an annual basis beginning in 2019.
The proposed legislation follows on the heels of a new UK law imposing similar requirements on employers (see our previous coverage here). On May 31, 2017, the California Assembly passed a version of AB 1209 that would require employers with 250 or more employees to collect information including the difference in mean and median salary between male and female employees for each “job classification or title.” Employers would have to post the information on an employer-maintained website accessible by the public.
On July 5, 2017, the California Senate introduced amendments to AB 1209, making it applicable to employers with 500 or more employees. The Senate version of the bill would require covered employers to submit pay gap statistics to the California Secretary of State to be published on a public website, but eliminates the mandate that employers publish the information on their own websites.
However, as opponents of such reporting requirements have pointed out, without proper context and analysis, pay statistics can be misleading. To illustrate, consider this hypothetical pay data for sales consultants at imaginary Company XYZ:
Female Workers |
Salary |
Male Workers |
Salary |
Faith |
100,000 |
Marvin |
75,000 |
Farrah |
90,000 |
Matt |
75,000 |
Fatima |
75,000 |
Mel |
75,000 |
Fern |
40,000 |
Mike |
75,000 |
Francine |
30,000 |
Mohamed |
75,000 |
|
Mean: 67,000 |
|
Mean: 75,000 |
Median: 75,000 |
Median: 75,000 |
In this example, the mean salaries for female and male workers suggest a pay gap of nearly 10% in favor of men, while the median values say little about workforce pay differentials. The raw data tells a potentially different story altogether, with 60% of women earning as much or more than their male counterparts.
Without controlling for additional variables, the pay gap information above gives an imperfect description of gender pay equity at Company XYZ. The differences in mean salaries between female and male workers might be explained by an innumerable combination of lawful pay factors, such as tenure, education and performance. In theory, discrimination could also be at play, but discrimination might be based on factors other than (or in addition to) gender.
In some large workplaces, the problem of outliers—extreme data points that skew aggregate values in one direction or the other—may be diminished; more employees means more data points, and more data points means a reduction in the relative weight of individual extreme values in a statistical analysis. But other workforces covered by the proposed legislation are likely to have a small number of employees in certain positions. In such cases, employers may find it challenging to provide meaningful pay gap information to the government and the public.
Advocates and critics alike have proposed various solutions to pay gap reporting issues. One possible solution is the use of a multiple regression model that controls for non-gender related variables, but such analysis is costlier and more time-intensive than mean and median calculations. Also, many employers don’t presently collect and maintain in a central repository all of the demographic and pay factor information that would be needed for a reliable multiple regression model. Other proposals, such as supplementing the mean and median estimates with raw data or additional statistical values, could create privacy concerns. Simply put, the more comprehensive the data provided to the public, the more likely it can be traced to particular employees.
Revisiting our Company XYZ example, the mean and median figures fail to capture critical aspects of the employer’s compensation dynamics. Varying interpretations of the raw data could lead to alternate, and in some cases contradictory, conclusions about pay discrimination and whether female employees are favored or disfavored at Company XYZ. That women like Faith and Farrah are the employer’s most highly-compensated employees might matter in the context of a pay equity analysis, as Fern and Francine’s respective experience and aptitudes certainly would.
As AB 1209 continues to wind its way through the legislature, it remains to be seen whether and how such concerns will be resolved. In the meantime, stay tuned as we continue to monitor this and other developments in equal pay legislation.