Comments on: SEC proposes CEO pay ratio reporting

By: New guest column: "SEC Unveils Expensive Rule on CEO Pay Ratio" - Overlawyered

Mon, 21 Oct 2013 19:30:03 +0000

[…] a guest column at PointOfLaw.com on the Securities and Exchange Commission’s proposed rule (earlier) requiring public companies to calculate and make public the ratio between chief executive officer […]

By: Jonathan

Jonathan — Fri, 18 Oct 2013 12:48:40 +0000

The law uses MEDIAN not Average. The median is the middle value of all employee compensations. It is not an arithmetic calculation. So taking total labor costs and dividing by number of employees would not be a Median.
The discussion over using a regulatory prescriptive formula versus a discretionary formula for determining compensation comes up in a lot rulemakings. Providing discretion is great for business until the regulator questions the process. Then a business spends a lot of time and money defending its process. There’s more certainty when the regulator just tells you how to do the calculation. I think in this instance a prescriptive solution is best. First, it’s acknowledged that investors don’t need this information. The information is for political consumption only so the number doesn’t mean anything and any number that a business puts up will be used to beat the business over the head. The information systems required to calculate this ratio isn’t built yet so why have everyone go in different directions. There are way too many uncertainties of inputs to the equation for multinationals that would be best be prescribed, sort of like how annuity rates for pensions are mandated in GAAP.

By: William Nuesslein

William Nuesslein — Fri, 18 Oct 2013 12:38:54 +0000

Regulating has its own dynamic. To forestall gaming the system, calculations are defined in excruciating detail. But suppose the true ratio was 60 to 1 and the estimated ratio was 50 to 1. What difference would that make? What would be the correct ratio? Certainly it would depend on industry in some way.

I worked with executive pay data in the 1970’s. It was regressed to size of corporation, profit, etc. All correlations were small. Compensation was a random process.

By the way, Allen confuses an arithmetic average with a median.

By: Jim Collins

Jim Collins — Thu, 17 Oct 2013 17:50:43 +0000

“Other than that minor annoyance, how did you enjoy the theater Mrs. Lincoln?”

By: Jack

Jack — Thu, 17 Oct 2013 16:31:02 +0000

Why would it cost so much? I depends on how the regulation is written and enforced.
I worked for a multinational corporation that had 30,000+ employees in dozens of countries. A simple count of how many people worked for the company would take weeks, and the end result would only be an estimate as the actual number could vary by the hour. Differing laws and regulations regarding what constitutes a contractor vs employee or full time vs part time were a small part of the problem but could be a major factor in determining average salary. Add in currency fluctuations and differing benefit packages and it could be literally impossible to come up with an accurate average salary. Some how I doubt the government would be willing to take a back-of-the-envelope estimate.

By: Walter Olson

Walter Olson — Thu, 17 Oct 2013 15:36:51 +0000

Before Allan opens up his promising “Here, let me work it out for you on the back of an envelope” business consultancy, he might want to check out some relevant sections of the proposed SEC rule. Here are excerpts (footnotes omitted) from the commission’s explanation of its controversial decision to include overseas employees in the median wage pool:

According to [critical commenters], the international variation in compensation arrangements and benefits, in addition to cost-of-living differences and currency fluctuations, could distort the comparability of employee compensation to that of a PEO based in the United States. In addition, these commenters noted that the types of compensation that are recorded in payroll and benefits systems outside the United States may vary from those recorded as compensation in the United States due to local accounting standards and tax regulations. …

We acknowledge the concerns of commenters that the inclusion of non-U.S. employees raises compliance costs for multinational companies, introduces cross-border compliance issues, and could raise concerns about the impact of non-U.S. pay structures on the comparability of the data to companies without off-shore operations. …

In particular, we are cognizant that data privacy laws in various jurisdictions could have an impact on gathering and verifying the data needed to identify the median of the annual total compensation of all employees. Commenters have asserted that, in some cases, data privacy laws could prohibit a registrant’s collection and transfer of personally identifiable compensation data that would be needed to identify the median. We also understand that in many cases, the collection or transfer of the underlying data is made burdensome by local data privacy laws, but is not prohibited.

So remember, the SEC could throw the book at you if you use a currency conversion formula it doesn’t agree with, and foreign governments can throw the book at you if you export pay data on their citizens for purposes of doing the required calculation. Other than that, and maybe a few dozen other complications, Allan’s advice is perfectly safe to follow.

By: Allan

Allan — Thu, 17 Oct 2013 13:13:33 +0000

I do not understand why this would cost so much. Just take the amount paid to all employees and divide it by the number of employees. Walla – you have the average salary.

It is ridiculous to assert that it would cost more than a calculator to figure out the average employee’s pay or the CEO’s pay. If a company does not know how much it is paying its employees, it needs a new CFO and a new director of HR.

That said, there might be little purpose to this exercise. But to oppose it because of cost is simply ludicrous.