The Advisors' Blog

– Liz Dunshee

We’ve blogged that most companies are now conducting “equal pay audits” – and as compensation committees get more involved in human capital management, it’s likely the board will want to see the results of those assessments. But collecting the data is only the first step in the lengthier process of addressing pay inequities. This 8-page memo from Equity Methods points out that at first blush, the data might even give you more questions than answers, like:

– Should all employees paid below their expected pay range be adjusted?
– Should males and non-minorities flagged as having below-expected pay be adjusted, or only women and minorities with below-expected pay?
– Should we increase the pay for all or a large portion of women and/or minorities, or only for those with severe discrepancies?
– Should the messaging reference the occurrence of a pay equity study?
– What were the root causes that led to the situation and how can they be addressed?
– What non-pay-related programs would help drive sustainable improvement?

The memo explains how to use a regression analysis to pinpoint the root cause of discrepancies. Those causes might be surprising and appear unrelated to pay equity – e.g. differing turnover rates between men & women. Finding these causes allows the company to create a nuanced response that resolves the true issues. Usually, that requires some combination of short- and long-term strategies. For example:

Suppose the model estimates a 2% pay gap and indicates that anything up to 1% could be caused by random noise in the data. This means that aside from remediating outliers, an additional adjustment of 1% or so for all or most women could swing the modeled pay gap to zero.