President Obama recently announced new steps to advance equal pay including a proposal that the Equal Employment Opportunity Commission in partnership with the Department of Labor collect and report summary pay data by gender, race, and ethnicity from businesses with 100 or more employees. This is a giant leap toward identifying and ultimately rectifying wage inequities. But before we get into the implications these data will have for wage equality, let us first explain why these data are crucial for advancing our understanding of wage dynamics.
Social psychology research provides context for understanding why collection and reporting of these data represents real progress in identifying and rectifying potential sources of wage inequity. For simplicity, we will focus on gender-based inequity, though the same logics apply to race and ethnicity. We, as a society, hold implicit beliefs about gender that inherently conflate men with wealth. For example, our research shows that when a job description notes that the majority of job holders are female that job description is judged to have a lower average salary than the exact same job description that notes the majority of job holders are male. Because we implicitly assume men are associated with higher wealth, when we observe pay disparities among individuals, our minds often subconsciously accept any gender differences, and this is so whether or not other factors (education, experience, negotiation opportunity, etc.) actually support the disparity in pay. That is, we fall back on our buried associations of men with wealth so that our minds make excuses that allow us to mentally compensate for differences and in effect perceive disparities as equitable. This phenomenon whereby we rationalize the appropriateness of differential individual salaries for men and women should be neutralized when we have the opportunity to observe large-scale aggregate data. If more micro-level trends prove to be larger aggregated differences, it will be harder for our minds to mentally compensate.
As currently proposed, pay data would be reported across 10 job categories and by 12 pay bands, and will not include the reporting of individual salaries. This new reporting will not only shine light on hard numbers that document disparities in pay, but the evaluation of aggregate, not individual, data will not be as susceptible to implicit beliefs that explain away the gap.
Moreover, the current proposal of tracking across job categories and pay bands appears to recognize that the biases of which we are speaking are-- for the most part-- implicit, subconscious and automatic. Upon announcement of Obama's plan, many individuals and businesses were up in arms about the proposed new reporting claiming it would be an undue burden and wasted effort that would reveal no reliable information. We suspect that reaction can stem, in part, from some fear of prosecution. But, these new steps are not necessarily intended to trap companies, accusing them of overt discriminatory wage practices. The underlying assumption is that we have overcome overt discrimination. What is still at play are implicit biases associating levels of pay with gender, race and ethnicity. Until we have the opportunity to see large-scale hard numbers demonstrating the downstream effects of these latent biases it will be hard to acknowledge them, discover their magnitude, and correct for them. Understood in this context, the collection of summary pay data is being undertaken not to point the finger at anyone or any company but rather to allow us a society to take a hard look in the mirror and discover the extent to which all of us--men and women, executives and employees--might be complicit in perpetuating a system grounded in implicit bias.
Recognizing that the collection of the summary data will provide opportunity to reveal and analyze the outcome of hidden biases, what remains in question are the implications of the findings. Of course, we must wait until the numbers are in, but based on theory and research we can reasonably speculate on what the data might tell us.
• Implicit biases are likely to contribute to greater pay disparities when objective metrics of performance are less readily available. For example, pay disparities would be exacerbated in upper versus entry, or even mid-level, positions because subjective measures feature more prominently in evaluation at the managerial and executive levels.
• Though it will be a complicated task to match comparable jobs, aggregate data of this sort will provide opportunities to compare not only pay within companies and industries but as importantly across companies and industries. This will lead us to identify where wage equity is most lacking and bias more prevalent, as well as to highlight industries and organizations with the most successful pay practices and model recommendations based on their characteristics and experience.
The pay reporting requirement holds the promise to reveal and begin to erase implicit bias. The process of reporting will bring more transparency and greater accountability locally that likely will lead to efforts to correct inequities from within. And, at a much larger level, analysis of this aggregate pay data holds the promise to further societal-level understanding, accountability and correction. So, yes, show us the money! But let's not stop there, why not take the lead of Intel and disclose publicly diversity statistics to shine the light on not just pay inequities but hiring and promotion as well.
Emily Amantullah, research fellow at the Georgetown University Women's Leadership Institute at the McDonough School of Business, contributed to this article.