This story just won’t go away. Today the Economist published a survey on executive pay, which is balanced and objective sounding, but comes away with an erroneous conclusion: that the rise in CEO pay is, for the most part, warranted. Why is this conclusion wrong? Because it rests on a false premise: that CEOs are the cause of corporate performance.
Let’s go through the main arguments:
But over the past decade all markets have displayed the same pattern: widening gaps between managers and workers and greater emphasis on long-term incentives. It is hard to see how this could have been caused by universally poor governance across so many different systems.
“All markets” is sweeping and incorrect. It has not happened in Japan (in fact, there is greater pay compression between new hires and top managers than in the past), which is neck and neck with China as the second largest economy, and the rise in executive pay in Australia has been trivial compared to that of the US.
The chief mistake of the past 15 years was the granting of too many share options to too many people on terms that were too generous. That was costly and unwarranted, but it stemmed more from foolish accounting and tax policies supercharged by bull-market mania than from a sinister plot hatched in the executive suite.
First, now that share option grants that the Economist is willing to admit were “too generous” has become the norm, it’s well nigh impossible to roll them back. Second, who do they think was behind the “foolish accounting?” This didn’t come out of the air. Management is responisible for the financial records. The turn of the century saw an unprecedented level of fraud (not merely “foolish,” but criminal) at major US corporations.
OK, they relent, sort of:
This is not to deny the abuses and downright crookery that have marred executive pay. …Indeed, the case for reform is strong…
The lion’s share of the executives’ bonanza was deserved—in the sense that shareholders got value for the money they handed over…
Between 1993 and 2003 the total pay of the top five executives in the Standard & Poor’s 1,500, which accounts for roughly 80% of listed American companies by value, amounted to some $350 billion, according to Lucian Bebchuk and Yaniv Grinstein, of Harvard and Cornell Universities. The share of earnings consumed by those people’s pay rose from 5.2% in the first five years of that period to 8.1% in the second five. And this is without counting the value of pensions, which can boost the total by as much as a third.
That is a lot of money, to be sure—though not quite as much as it sometimes seems. The “average” pay that is often quoted, and which is used as the basis for comparison with the “average worker”, is the arithmetic mean. Chart 1 above, from a different study, shows the average earnings of the top three executives all the way back to the 1930s. Whereas mean pay at the peak was 320 times average earnings, the median pay was “only” 120 times. In 2000-03 their mean annual pay was $8.5m and the median $4.1m. The median is a better measure than the mean because the mean for those top three is skewed by a few huge payments, often to company founders or family managers who are not standard executives…
Shutting a subsidiary, sacrificing a pet project or forgoing a tempting acquisition is not much fun. Without the spur of high pay, managers tend to avoid such things.
Here is the core of the argument, and it fails any test of logic. The Bebchuk and Grinstein work shows that in a ten-year period, the proportion of earnings that went to the top five managers increased by 56% in the second half, and that could even be understated.
How is this justified? In fact, since it isn’t, the Economist moves along. In industries that want to motivate specific types of behavior, the tried and true method is either to pay a percentage (typically of revenues for sales staff) or a bonus or award, like a trip, for reaching a goal. The percentages are either fixed (the notorious 6% for real estate brokers) or if anything, they scale down as the size of the underlying transaction increases (as with M&A fees. The percentage may be set for a particular transaction, but the percentage chosen is smaller if it is a multi-billion, rather than multi-million, dollar deal). Increasing percentages for increasing base amounts is unheard of (except at very low levels of activity, to induce beginners or low performers to get to a desired level, which does not apply here).
It’s fair to point out that the use of median rather than mean CEO pay overstates the issue, but the flip side is that the use of mean data still does not change the basic pattern.
And the claim that CEOs need extra pay to do things that are hard, like closing down pet projects? The list is spurious, even laughable (and as we’ll address in our next post, the current regime encourages rather than discourages acquisitions). Doing any of those things requires that a leader swallow his ego. If he can’t do that, he shouldn’t be in charge in the first place. All corporate officers have a duty of care. That list suggests that the Economist believes that CEOs need extra rewards to do their job.
And the list conveniently omits cost-cutting, particularly headcount cuts, and reductions in compensation and benefits for the rank and file. A good deal of the current high level of corporate profitability is the result of expense reduction and underinvestment (see this article from Across the Board Magazine for some of the data). How much of the rise in CEO pay has been paid for by the issuance of pink slips? This is an important question, but oddly, only unions have raised it (for instance, the Allied Pilots Union is charging that executive bonuses at American Airlines will exceed the airline’s profits) is making it an issue with American Airlines) but unions are seen as not-credible parties these days. But the oft-repeated pattern, of CEOs laying off thousands of people and then paying themselves more, is one that many find galling. True leadership means sharing in the sacrifices, but too often what we have instead are imperial CEOs, prospering as their fellow employees suffer
Another article in the series …”start[s] with that moment in the 1980s when the long-established relationship between workers’ and managers’ pay began to break down.” This post is getting long, so we’ll parse it in our next edition.