Beginning in 2018, most publicly held companies will have to include in their proxy statements a ratio that compares the compensation of their chief executive with the compensation of their ‘median employee’. The disclosure must be based on the existing proxy statement rules for calculating compensation in the summary compensation table and must also include a brief explanation of the ratio and any assumptions that went into it. Certain emerging growth companies and smaller issuers will be exempt.
On September 21, 2017 the SEC released additional guidance on the pay ratio disclosure requirements, providing relief that should reduce the overall time burden on companies as well as provide greater flexibility in how they calculate and disclose their pay ratio. There had been hope the new administration would delay or significantly ease the rule, but it appears that this is not going to happen.
By this point, you’ve had the opportunity to read thousands of pages of law firm and consulting firm reports describing the complexities of the required pay ratio disclosure and the many different alternatives for determining the pay ratio. Here, we present a framework and proposed approach to pay ratio disclosure that might greatly simplify your analysis and work plan. We would encourage you to conduct a trial run soon to increase your chance of success for next proxy season. One benefit of a trial run is that you can see how results vary by exercising discretion with respect to the varying methods.
There is a key concept driving the whole analysis: determining the median employee. To select the median employee, the issuer must determine both the population of employees to consider and the appropriate compensation measure. Importantly, using reasonable estimates, assumptions or methodologies as permitted by the regulations will not be the basis for an SEC enforcement action – unless the disclosure is made or reaffirmed without a reasonable basis or was provided other than in good faith. We recommend that companies state in their proxy statements that the pay ratio disclosure was made based on reasonable estimates. The SEC has said that such statements are permissible.
In determining the median employee, it is important to consider the following two questions: who is likely to review the disclosure, and how will he/she react to it? On the one hand, shareholders and analysts will examine pay ratios and likely compare them among companies. For this reason, having a higher-paid median employee that drives a lower pay ratio may be preferred.
On the other hand, any company employees who read the proxy statement may quickly hone in on what the median employee is paid (the CEO’s pay has been in the proxy for years). Employees will be looking to see whether they are in the bottom half of employees by pay. For this reason, having a lower-paid median employee may be useful. HR and management should be prepared for complaints and questions from employees who believe they should be paid more than the median employee.
DETERMINATION 1: WHAT EMPLOYEE POPULATION IS USED?
The rules require that you determine, as of a date within three months of the end of the relevant fiscal year, the population of employees of the company and its subsidiaries that you will use to determine the median employee.
‘Employee’ for this purpose can be determined under any widely recognized test, so if you have in your records classification of individuals as employees or contractors based on relevant local legal or tax requirements, you can use those classifications. You are permitted to exclude non-US employees if data privacy rules don’t allow you to share data to the US, and/or if your non-US employees constitute less than 5 percent of your workforce.
If all of your non-US workers are relatively low paid, you might want to include them even if they constitute less than 5 percent of your workforce so that your median employee pay is lower. If you have a workforce with a lot of seasonal employees, you may wish to pick a date before they come on board to avoid dragging the pay ratio down. If your goal is to decrease the reported median employee pay, it may be helpful to pick a date in the middle of the holiday season. At a minimum, it would be helpful to know the extent to which methodology choices like this impact the result.
DETERMINATION 2: WHAT COMPENSATION IS USED TO DETERMINE THE MEDIAN EMPLOYEE?
Although the rule requires using the summary compensation table rules to determine the pay ratio, you aren’t required to use those rules to determine the median employee. The median employee can be selected using any consistently applied compensation measure that reasonably reflects how the workforce is generally compensated. This can be done using human resources or payroll records.
One efficient approach might be to combine total cash compensation paid during the relevant period with the grant date value of equity compensation granted during the period to figure out your median employee. This has the following benefits:
- Cash compensation is already tracked in payroll records and conversion of those amounts to a single currency for comparison purposes shouldn’t be too difficult
- The grant date value of equity awards should be readily accessible, stored in a clean and auditable manner and already reflect the company’s functional currency.
FIVE DATA ADJUSTMENTS OF UNEQUAL USEFULNESS
There are five primary adjustments or exclusions you can apply to the raw data you pull together as described above. The usefulness of these adjustments or exclusions varies substantially:
- Data privacy – The rule exempts companies from sending data that is prohibited from being sent due to local data privacy rules. Relying on this exemption requires an opinion of counsel, and the use of anonymized data would likely be permissible and just as useful as data with a name attached. This exemption therefore may not be very useful
- Annualization – Employees who are hired mid-year or who take leaves of absence can have their compensation annualized for the year. After a trial run without annualized data, you could consider whether annualization is worth the effort
- Cost of living adjustments (Colas) – Although the rule allows you to adjust non-US data to reflect cost of living differences when determining your median employee, you must still disclose the pay ratio without any Colas. We understand this has deterred many companies from adjusting for the cost of living
- De minimis exemption – The rule allows you to exclude jurisdictions until the total of all employees in all excluded jurisdictions makes up 5 percent of your total workforce. This exemption may be important because it can reduce the number of payroll systems you interact with from a very large number to a much smaller one
- New acquisitions – Employees in acquired organizations can be excluded in the first year of acquisition. Many companies will rely on this adjustment given system-integration speed bumps and employee changes that may occur post-M&A.
STATISTICAL SAMPLING IS NOT A PANACEA
There are two methods for calculating the pay ratio – one involving use of the full data population (subject to the exemptions already mentioned), and one in which you sample from a subset of the population. Statistical sampling may or may not be the answer for you; it’s a tool, nothing more. The problem with statistical sampling is that it doesn’t address the fundamental challenge of data access – it simply allows you to work with less data volume.
Here’s what that means: the rules require you to sample from each unique jurisdiction – you cannot use your easily obtainable US data as a sample for how people in Asia are paid. Instead, you need to obtain representative samples from each jurisdiction. The way modern databases work is that if you can obtain 10 records from a system, you can probably obtain 10,000 at virtually no additional effort or cost. Nevertheless, best-of-breed sampling methods have their benefits in isolated cases involving missing and erroneous data, so do as you would with any problem: define the problem, then identify the right tool to fix it.
Before engaging in a lengthy exercise of pulling the various levers available for a pay ratio calculation, it might make sense to use the methodology described above based on your 2016 data to see the results and think about potential disclosure before you start optimizing for a different result.
John Aguirre, Lisa Stimmell and David Thomas are partners based in the Palo Alto, California office of Wilson Sonsini Goodrich & Rosati
This article originally appeared in the Winter issue of Corporate Secretary.
