The Wall Street Journal’s CEO pay survey includes a piece by Scott Thurm called Getting Bang for the Buck which looks at whether the best paid CEOs have delivered the best performance. The WSJ has produced a scatterplot that tries to map pay and performance, looking at three-year realized pay and three year total shareholder return (TSR). It looks like a target from a shooting gallery where someone used a shotgun at fifty paces. The stuff is all over the place. I happen to know that the r-squared is less than 0.1 (1 is a perfect fit, 0.1 is a long way from that). Yet we still see Steve Kaplan from Chicago University claiming that there is a strong correlation because for every 1 percent increase in TSR over the last three years we see $1 million extra in the CEO’s pocket. Thus the title of this blog.
Good grief, this is why I hate statistics. Give me a handful of companies to look at and I’ll tell you in every instance whether the CEO’s pay plans are performance-related or not. But you can prove just about anything you like as long as you keep trying tests until one comes back that gives you the answer you want.
What was my comment about the figures? More than half the CEOs delivered worse returns than a Treasury bond so they should really be just getting base salary and not be among the highest paid in the survey.
Look for The Corporate Library’s own preliminary CEO pay survey next week.
Paul Hodgson - Senior Research Associate
