As we speak to our clients and other consultants, we are finding that many companies have not started the process of calculating their CEO Pay Ratio for their 2018 disclosures.
Procrastination to date seems to stem from one of three factors:
- “I’m not spending any time on this because the resulting information is useless!” To this argument, we have no creative response (nor for that matter does the SEC, who itself admitted they could not find use for this information).
- “It’s going to be repealed or delayed.” Unfortunately, proxy season is only six months away and this no longer seems to be a reality.
- “It’s so complicated and cumbersome that we don’t even know where to start.” Fortunately, those clients who have gotten started are now finding that this may not be as difficult an exercise as many have suggested, as discussed below.
Interestingly, many of our clients have been able to rather accurately guesstimate who the median employee will be even before digging into their data, based on the demographics of their workforce. We are even seeing some friendly wagering about where the median employee will be found (which division, which country, etc.). Unfortunately, it’s a little like gallows humor as the teams all know they have to get this work done, but don’t like the fact they are spending time on something they don’t believe adds any value. It has the feel of mind numbing busywork.
Companies with enterprise-wide HRIS systems seem to be having an easier time of it. They generally believe they can get clean pay data from their systems to determine who the median employee is. These companies tend to have fewer hourly or seasonal workers whose data is often harder to clean up.
Companies are using a variety of methods to measure compensation for the median employee. Base salary or rates of pay are the most popular. Taxable earnings is also being used. Companies with multiple payroll systems (and a mostly domestic employee base) are using W-2 information to calculate their median employee. Those with international employees are using taxable income as a proxy for W-2s.
Some of our clients with more complex workforces (another way of saying messy data) are using statistical sampling, which is a time- and cost-saving approach. The term “statistical sampling” usually makes otherwise intelligent people cringe (flashback to required college courses we had hoped to forget after the final). But we have now found that the tools and techniques most of us have long forgotten (or never really understood) have a very practical application for determining the median employee. Sampling has proven to be most useful for our clients in multiple countries or for companies with many hourly and/or part time employees where cleaning up the data is difficult. The silver lining, if you go down this path, is that we are finding companies using sample sizes of 250 to 1,000 for employee populations of 5,000 to 15,000—tremendously cutting down on much of the mindless busywork in performing this calculation.
