There are all kinds of issues with that. That kind of "sampling" is not going to result in an unbiased estimate of what the median CEO salary is. Also, I think it possible that it would overestimate the increase in CEO pay and perhaps even create the impression that there was an increase when there was a decline.Median CEO pay in 2010 was $9.0 million, based on 158 Standard & Poor’s 500 index companies with the same CEO serving all of 2009 and 2010 that have reported CEO pay, according to the USA TODAY analysis of data from GovernanceMetrics based on proxies that have already been filed
Since the sample only includes CEOs who served in 2009 and were still serving as CEOs in 2010, it is possible that the estimate is biased upward. CEOs that were successful and rewarded may be over-represented. In fact, if I had to bet I would bet that is the case.
To get an estimate of median CEO compensation of S&P 500 companies one would have to take a probability sample (like random) of CEO compensation levels. You'd do that for 2009 and again for 2010 in order to estimate the increase in CEO compensation. Or you could just look at every company and it would be reality rather than an estimate.
These guys took this hapazard, probably biased sample of CEO pay to derive an estimate they then compared to an estimate of "worker" pay dervied by the Bureau of Labor Statistics from probability sampling.
They should never have even gone through the exercise. Thanks again, media, for creating yet another false impression.










