WSJ's Carol Hymowitz on the CEO Pay Gap

The Wall Street Journal has a most thoughtful article called Pay Gap Fuels Worker Woes.  It talks about how when there is a huge gap between what the CEO makes and what everyone else makes, it can undermine morale.  And it quotes some accomplished researchers like Wharton's Pater Cappelli and Harvard's Rakesh Khurana.  I want to go a bit beyond this article to show that these concerns about excessive CEO pay aren't just political rhetoric. I went through and reviewed the research on CEO pay (something I have been meaning to do for awhile), and although there are one or two studies that show paying the CEO a lot more than others might increase performance under some conditions, the weight of the evidence looks negative to me (although in the spirit of evidence-based management, this is a tentative conclusion and I need to  read the research I gathered at least once more and search for other published studies might have missed).  Here are some major findings:

1. A 2006 study in the journal Organization Science by James Wade and his colleagues suggests that overpaying the CEO leads to a dilemma: You can overpay other senior executives too and thus entice them to stay; our you can create a big gap between the overpaid CEO and everyone else, which leads other senior executives to jump-ship.  Either way, overpaying the CEO has costs beyond the extra dollars the CEO gets.

2. A 2001 study by Donald Hambrick and Phyllis Siegel in the Academy of Management Proceedings of 64 firms showed that when there are bigger pay differences between the CEO and other members of the top management team, organizational  performance tends to suffer -- and the negative effects of such pay dispersion is most pronounced in high-technology firms.

3.  A 2002 study of 199 firms in the S&P 500  by  Mason Carpenter and WM. Gerard Sanders in the Strategic Management Journal suggested that firms that had a bigger gap between CEO and the other members of the top management team performed less well in subsequent years.

4. Finally, to preview a study I will likely be talking about more in the future, Arijit Chatterjee and Donald Hambrick published a paper on CEO narcissism in the Administrative Science Quarterly called "It is all about me."  They develop an inspired six-item narcissism measure: 1. CEO cash compensation relative to the second highest paid executive; 2. CEO non-cash compensation relative to the second highest paid executive; 3. length of the CEOs Who's Who entry;  4. CEO prominence in company press releases;  5. CEOs use of personal pronouns during press interviews ("I" and "me" versus "us" and "we"); and (I love this one) 6. prominence of the CEO's photograph in the annual report (maximum score if there was a picture of the CEO alone in the annual report and it took up more than half a page in the annual report).  The intercorrelation among these six items was .76, which is well-within the range for an acceptable index.  So these index suggests that -- performance issues aside -- when the CEO is getting a lot more money than the next executive,  he or she will likely be afflicted with other signs of narcissism.  The firm performance effects in this study are complicated, and I will talk about them at later point.  But here is a post with a link to the whole paper -- in unpublished form -- if you want to read it.   

I want to end by saying again that, although the message in this research so far seems to be that paying the CEO a lot more than others isn't a good thing for the company, there are some studies that suggest this isn't always the case.  I also find that one of the most interesting studies is the one by James Wade and his colleagues, which suggests that if the CEO is overpaid, the decision to overpay the rest of the top team isn't a purely good thing -- reducing pay dispersion when the CEO is overpaid can cause a company to waste even more money.

Comments

[Greg]

Bob,

What's your take on Elliott Jaques' "Felt-fair pay"?

[Derek Scruggs]

Good to see Harold Geneen, do that, but I've seen the same situation and the risk was effectively removed ex post facto. I worked for a company who's new CEO borrowed money to buy shares. When the company cratered and was forced to sell for pennies on the dollar, he negotiated a bailout of his loan as part of the acquisition price. Some risk.

[dblwyo]

Bob - beyond the studies I think we need to look for mechanism. My belief (& prior argument: http://tinyurl.com/2xqwtg)is that enterprises are social organisms which require commitment and legitimacy to function well. Excess CEO pay reduces the legitimacy, excess at the expense of the health of the company more so, especially while people are being laid off. Pretty soon you have a vicious cycle of degeneration where everybody spends all their time watching their backs and looking for the next opportunity. Effort levels drop enormously to the detriment of performance because everyone is looking for their next opportunity. And this behavior is perfectly rational on the part of the employees.

[Jan]

There is even a more serious side of this. During the last 10 to 15 years we have seen redistribution of wealth in a scale not seen before, from middle income families to less than 1% of the US population. Accumulating this amount of capital has no point, unless you put it into work somewhere. When people are not credit worthy, things like sub-prime mortgages are invented. Thus the credit crisis we have now. In a sense, this redistribution of wealth destroys the market. And this is taught in Business Schools...

[Wally Bock]

Bob, I agree that the Journal article was thoughtful, but I also thought it was muddled.

Ratios for the pay of CEOs (no data about company size) to all workers are mixed with statements about the relation of CEO compensation to the other top 25 executives.

Big and small company examples are mixed.

There's no discussion of the performance of the companies used as examples and no discussion of how people feel about the ratios between CEO and other compensation in profitable versus unprofitable companies.

There is a statement about what jobs MBA graduates are pursuing today versus 1970 without a shred of information about how that fits into the discussion.

I think this is an important issues, but I think the Journal article to help us sort it out.

[Bill Burnett]

High pay often means 'no risk'. Pay is not really tied to company or individual performance. If the CEO is shielded from risk, it's got to be hard to really 'lead the troops'. Leadership is showing that "risks can be surmounted, and surmounting these risks is what we are going to do!"

When Harold Geneen was given a large block of options by his board, he immediately went out and borrowed a lot of money to exercise the options and put back in the risk. That's leadership.