THE OLYMPIC-SIZED rewards for failure notched up in the last month by Merrill Lynch's Stan O'Neal ($8bn write-downs, $161m payoff) and Citigroup's Chuck Prince ($11bn write-downs, up to $100m payoff, with the use of office, car chauffeur and administrative assistant for five years, plus consultancy) have caused the usual bout of handwringing - and the usual resignation.
Even before the discovery of these performance black holes, US bosses had the grace to acknowledge they were being paid too much. Four in six of them, and 80 per cent of outside directors, admitted in a survey last month that CEOs were overpaid. But while the payoffs pocketed by the deposed bankers may have increased the embarrassment, there is little sign of a correction. On the contrary: instead of proving that the system is broken, each failure is used to justify still greater incentives to get the leaders facing in the same direction as shareholders.
But supposing it could be shown that there was a correlation between the composition of CEO pay and the size of the losses? We're not quite there yet, but some intriguing US research suggests there is an indirect link - and one that could have considerable implications for the way shareholders think about CEO compensation in the future.
In brief, a study in the influential US Academy of Management Journal finds that the way top managers are paid does influence a firm's performance - but not in the way the textbooks indicate. The article, by academics Gerard Sanders and Don Hambrick, focuses on the role of stock options, which still constitute the largest proportion of CEOs' ever-growing pay, and whose effects are surprisingly little researched.
Stock options, you recall, were supposed to stir top managers out of their assumed default mode of cautious self-serving and spur them to become bold, buccaneering entrepreneurs. So they do. CEOs loaded with options do indeed invest lots of shareholders' money in research and development, new ventures and acquisitions, and the result is extreme performance. What wasn't envisaged, however, 'was that the extreme performance delivered by option-loaded CEOs was more likely to be in the form of big losses than big gains'. Like Andrew Flintoff or Ian Botham, option-padded CEOs go for the big hits - but they usually get caught on the boundary rather than score sixes.
In a complicated formula, the professors looked at the performance of 950 US companies across the size range and related it to the stock-option content of the CEOs' pay. They found that the more executives were paid in options, the more the company invested and the more extreme its financial performance. It wasn't high investment that generated extreme performance, but, with stock options, it created a 'combustible combination' which encourages executives, the authors say, to take on big bets at long odds, with predictable consequences.
Why CEOs should do this is fairly obvious. If, as is their nature, options carry no downside, then CEOs will simply go for the projects with the largest potential payoff. 'If we accept the common-sense idea that the projects with the biggest possible upside are likely to also have the biggest possible downside, and then couple it with the assumption that option-loaded CEOs have little concern with the size or probabilities of downside outcomes, it is straightforward to expect that option-loaded CEOs have a relatively high likelihood of delivering big losses,' the authors say.
In the case of Merrill Lynch and Citigroup, O'Neal and Prince certainly had plenty of options. And although this is not part of the professors' case, if they don't have an option downside and, in addition, are certain to benefit from a lucrative exit payoff if they are fired, how much more likely are they to take on big, long-odds bets without fear of the consequences? If the numbers are life-changing enough - which they are - there is not even any danger to a future job, since neither of them will need ever to work again.
The Hambrick and Sanders findings are important because unlike other criticisms of stock options they strike at a feature that is at their heart: it is not that they fail to foster managerial aggressiveness, they do it only too well. They also take a swipe at agency theory, which has allotted options an important role in combating the supposed managerial tendency toward shirking, short-sightedness and risk aversion.
In the professors' perspective, the problem is fundamental and can't be fixed by tinkering with the way options are administered. The solution is simple: stop giving so many, or don't give any at all. 'These findings may help put the nail in the coffin of executive stock options,' remarks Hambrick. 'And even if not, they certainly ought to give the corporate world pause in using them nearly as extensively or heavily as they have been in the recent past.'
The Observer, 18 November 2007