Performance and Pay: The Missing Links


Print | 1 Comment | Share/Save

Expert Perspective  by Grahall’s OmniMedia Editorial Board

In their May 6th, 2010 article published in Bloomberg (CBS Overpaid Moonves $28 Million in Study of Chief Executives) authors Jessica Silver-Greenberg and Alexis Leondis write:  “Pay expert Graef Crystal, a former adviser to Coca-Cola Co. and American Express Co., has concluded that pay for performance is a fiction.  Using formulas he developed over 30 years in the business, Crystal crunched the numbers to see whether higher shareholder returns, the gold standard of performance for investors, led to higher pay, and vice versa. No matter how he sliced the data, the answer was no.”

Time and time again we see overly simplified answers to highly complicated questions about CEO pay.  We have written in several occasions about the imprudence of broadly applying presumptions gleaned from ill-conceived studies of limited segments.  (See for example: Typical Examples Aren’t So Typical  and The Hurd Locker.

This “research” by Crystal again follows that well trodden but pointless path.  Let’s deconstruct what little we know of about Crystal’s “formulas” based on what is described in the article:

First the universe of CEOs is essentially the same size as the universe of companies.  In the US there are about 10,000 publically traded companies and perhaps twice that again in privately held companies.   Making the total number of companies in the US perhaps as large as 30,000.  They are companies of all sizes, in all industries, across all geographies, any they follow a myriad of business strategies.   A sample size of 1% used by Crystal might be sufficient if it is adequately diverse, the sample used by Crystal was not diverse. For his sample Crystal only “… included companies in the S&P 500 that had filed proxy statements for their 2009 fiscal years by April 16…”

Perhaps one thing that is common to most (if not all) of the 30,000 US companies is that the horizon for their business cycles and business strategies is more than one year.   It appears that Crystal is only considering performance and pay for the single year 2009.

We agree that there are some companies that do not pay for performance, but the right way to determine if CEOs are overpaid is to do the following, as we did in Grahall’s research on CEO and named officer pay :

1) Start with a large and representative sample of companies in all industries and of all size

2) Standardize and delineate the sample considering both revenues and market capitalization.

3) Consider a multitude of performance factors including return on sales, return of assets, return on equity over one and three year periods and consider total shareholder return over a five year period. 

4) And most importantly, eliminate the “beta” from the analysis. Think of this as the tide raising (or lowering) all boats. Subtract from the total shareholder return that portion of performance beyond the management team’s control. If the stock market goes down from 14,000 to 7,000 it is hardly the fault of the management team (and vice versa).

5)  Another consideration is to look at the total pay of the top 5 executives. There may be individual considerations with a specific CEO (founder, major intellectual contributor, early in career, etc.) so it is best to look at the top management team and let the Board and the CEO distribute the individual awards as is appropriate for the members of the team.

Interestingly when this article was published a month ago, the Dow Jones Industrial average was over 11,000, a number we hadn’t seen since September 2009, since then the market has lost 1,000 points. The idiom “a rising tide lifts all boats” is an important component of shareholder return, and has little to do with company performance.   Consumer confidence for example, can buoy the stock price of even the most poorly run company over the short term, but this works in both directions:  a lowering tide also lowers all boats.  Essentially, assessing CEO pay with shareholder return (that is with stock price) fails to recognize that broad market forces are playing a significant role in the performance of that stock. 

We believe that the right guide for determining relative pay and performance and reasonability of CEO and other named officer pay is to review an unbiased research study (such as our own Grahall research series on executive compensation.  With this information you can determine the performance percentile of the organization on one of OR all of the key performance statistics. Then determine the pay percentile that matches that relative performance. If the two percentiles generally match then the reward program is operating in a reasonable zone on a “relative pay for relative performance.” If the two percentiles don’t match (either the relative pay is too high or too low given the relative performance) then it would point to the need for a review of the executive compensation program.”

Contact us link to we can help you through a thoughtful and comprehensive review your CEO’s and other named officers’ pay packages.

Contact Grahall’s OmniMedia Editorial Board at

1 Comment

RSS Follow this discussion

  • Having read this I thought it was rather informative. I appreciate you taking the time and energy to put this informative article together. I once again find myself personally spending way too much time both reading and posting comments. But so what, it was still worth it!

Post a Comment