The Activist Investor Blog
The Activist Investor Blog
WSJ Abuses Statistics on CEO Pay
We used to think that the Wall Street Journal advocated for investors. After all, beginning with the name of the newspaper, it first and foremost reports on subjects of interest to the investment community. But, maybe not so much anymore.
Yesterday it ran an extensive feature on CEO pay, “CEO Pay Moves With Corporate Results”. Findings from their reporting include:
❖“Chief executives increasingly are being paid based on their companies’ financial results and share prices...”
❖“CEO pay during 2011 was more correlated to how well companies fared in the stock market...”
Some outside experts also endorsed the findings of the analysis:
❖Steve Kaplan, University of Chicago: “[pay is] highly correlated with performance”
❖Paul Hodgson, GMI Ratings: “There are nudges in the right direction”.
Yet, our analysis of the same WSJ data shows virtually no connection between CEO pay and company performance.
From whence does this good news come? “The analysis was based on a review of 300 proxy statements by consulting firm Hay Group”. Hay of course provides compensation consulting services to BoDs and CEOs.
The core data for their analysis looks like this:
We dug into this data, then. The WSJ provided the compensation and shareholder return data series that they used in their chart. We rebuilt the chart using their own data, and interpret it this way:
The analysis is also backwards. It seems to us that shareholder return should depend on CEO pay. Instead, the WSJ analysis makes CEO pay the dependent variable (on the y-axis). It doesn’t change the R-squared, but it does make more sense that way.
We also wonder about the definitions of the variables. In the published chart, “pay” is “Total Direct Compensation” for a single year (2011). That in turn includes salary and bonus paid, and equity grants. And, “return” is “Total Shareholder Return” for a single year (also 2011). TSR is simply share price appreciation plus reinvested dividends for the year. Confining either measure to a single year seems to short-term, though, and using equity grants rather than actual payout inflates the actual compensation that the company pays.
Oddly, the Hay data does include much better measures, but the WSJ didn’t publish them. Digging deep into the WSJ.com website, we found data for three-year realized compensation, and three-year TSR. We understand that these figures together show a much stronger link between CEO pay and investor returns. If so, we cannot figure out why the WSJ featured the flawed analysis that it did.
Looks like the WSJ will do anything to make CEOs look good at the expense of investors, even misrepresenting statistics on CEO pay.
Tuesday, May 22, 2012