I have news for Total Rewards and Compensation leaders: Eliminating statistical significant pay differences in pay equity audits is no longer enough. Why? It’s not what will really matter in the pay transparency era. Instead it’s:
- The EU Pay Transparency Directive
- Salary range disclosure laws in California, New York, and other U.S. states
- Shareholder proposals on pay from Arjuna Capital and others
These pressures demand a bigger goal: eliminating the pay gap. That requires you to eliminate statistical significance, but just as a starting point, not an end goal.
To close the pay gap and ensure pay fairness, a forward-thinking company needs an “all of the above approach.” With pay transparency, individual employees may learn that they are making less than their peers and they will not care if that difference is “significant” or not. Leaders must be able to tell stakeholders and employees that men and women are paid equally. To qualify this message with a note that existing gaps do not fall under the “p = .05” threshold may not be compelling enough.
In this era, companies need a strategy grounded in minimizing their pay gaps — both adjusted and unadjusted — and a solution flexible enough to help them get there. That’s why I work with and recommend Syndio. Syndio offers traditional pay equity analytics but also goes far beyond it to help you:
- Measure and remediate pay gaps to a specific number;
- Identify individual employee outliers and ensure you’re paying for what you say you pay for; and
- Make consistent pay decisions that your managers can explain.
In reflecting on the changes wrought by the pay transparency era, I’m frequently asked: how much does the methodology matter? It’s a complicated question, but one that more and more depends on your pay strategy and what it is you want to accomplish with your pay equity audit.
My advice: Analyze your data with a partner and a technology solution that gives you the flexibility to address the many reasons why you are running a pay equity analysis to begin with. Because more often than not, you need more than one methodology.
