Because of recent high-profile cases claiming gender-based pay discrimination, the Equal Pay Act has taken on a new life. Newspapers continue to tout the controversial statistic that women earn only 77 cents for each dollar men earn. That statistic does not distinguish among jobs and is actually a comparison of apples to oranges, argue critics. Nevertheless, there are certainly situations where women on average are paid less than similarly-situated men. The fact that retail giant Wal-Mart is the defendant in the first major case of this kind in many years has placed all retailers in the crosshairs of the wage and hour plaintiffs' bar.

What Makes These Claims Different?

Unlike run-of-the-mill discrimination claims, Equal Pay Act claims rely heavily on the use of statistical analysis of disparities in pay as evidence of discrimination. Statistical analyses can reveal abnormalities among pay in various groups. But statistics are not arithmetic where there is only one correct answer to the problem. The inclusion or exclusion of factors other than the challenged one can reveal that nondiscriminatory characteristics, such as education and experience, have more bearing on the disparity than gender.

One of the most important, but hardest to test for characteristics, is the willingness to negotiate. In general, women are more likely than men to accept the employer's first offer. For this reason, pending legislation designed to address the pay gap includes provisions for training women in negotiating skills. Social scientists have numerous other theories about other gender differences that play a role.

Men's greater willingness to relocate, to seek physically demanding or dangerous jobs, and to sacrifice home life and quality time with their children, are all additional reasons men on average appear to earn more. But regardless of the strength or weakness of these factors, the possibility that a statistical analysis will reveal a significant pay gap between genders in any job is serious enough that retailers need to perform some self-analysis.

The problem in performing such a self-analysis is that there is no predicting what it will reveal; no company wants to create Exhibit A for the plaintiffs in an equal pay case. In a majority of jurisdictions, there is no "self-critical-analysis" privilege to protect this type of report against disclosure in discovery.

If performed with no threat of litigation on the horizon, the attorney work-product doctrine will not apply. Since the underlying data will never be protected, the possibility of being forced to turn over a statistical analysis that cost tens of thousands of dollars and countless hours of time is not pleasant.

Our Advice

In light of this, here are some steps that employers can take to provide the greatest amount of protection against their statistical analysis being discoverable. First, it is better if the analysis is performed in response to some claim of pay discrimination. Even if it is only a demand letter or an EEOC charge, the existence of a legal concern creates the ability to claim that the analysis is protected work-product.

The company should also hire outside counsel to provide it with a legal analysis of potential liability. That firm should then engage the statistical experts on behalf of the employer. The employer should not receive the actual underlying report of the statistician. This provides far too much of an opportunity for the report to be widely disseminated and lose the aura of attorney-client-privilege that helps protect documents from disclosure. While the statistician's report can be some evidence regarding the impact of gender on pay, the report standing alone is not in and of itself the answer to the question. Instead, outside counsel should draft a legal analysis presented to the in-house counsel or senior human resources professional. These steps provide the best protection against being ordered to disclose the report in litigation.

The Man In The Mirror

Before setting out on an analysis of pay disparities, the most important decision an employer must make is a commitment to address the findings. If you are not committed to correcting gender-based disparities revealed by the testing, then it is better not to do the testing at all. In the words of the Michael Jackson song, "make that change." Knowing about possible discrimination and not addressing it is the type of conduct that opens employers to claims for punitive damages.

Pay disparities, while often small in the individual case, can quickly become significant in the aggregate. The fix can be millions of dollars.