Artigo muito bom da revista The Economist tratando do comportamento de gerentes de empresas, particularmente dos gerentes de bancos nesta crise financeira!
Aug 11th 2009
Is it good if bosses feel strongly for the firm?
GETTING bosses to act in the best interests of a company’s shareholders has long been one of capitalism’s trickiest problems, identified early on by Adam Smith. In “The Wealth of Nations” he worried that “Being the managers of other people’s money than of their own, it cannot well be expected that they should watch over it with the same anxious vigilance with which the partners in a private co-partnery frequently watch over their own.” In the 1920s, Adolph Berle and Gardiner Means followed up on the problems of separating ownership from control in their classic study, “The Modern Corporation and Private Property”. From the 1970s, this dilemma acquired its own branch of economics—agency theory, which studied the problems that can arise when “principals” (ie, shareholders) hire “agents” (executives) to run their firm.
Michael Jensen, one of the theory’s fathers, argued that the key to minimising “agency costs” is to set the right incentives for executives, so that in their decisions they identify as much as possible with the interests of shareholders. Others argued for investment by shareholders in closer monitoring of bosses—though this could be costly and impractical. Mr Jensen thus favoured private ownership over public ownership, defending buyout firms that undertook controversial debt-financed leveraged buyouts in the 1980s, and their later incarnation as “private equity”. He also argued that executives should have a significant proportion of their wealth in shares of the firm they run, and that much of their remuneration should be in company stock.
The bursting of the dotcom bubble sowed doubt in Mr Jensen’s mind, especially about one hugely popular form of equity-based executive pay, share options, which he dubbed “managerial heroin”, because it encouraged a focus on short-term highs, regardless of the potentially destructive long-term consequences. Yet the notion of linking executive pay to share-price performance—or, rather, to long-term share price performance—continues to have widespread support.
Indeed, this has been a recurring theme of those trying to reform pay on Wall Street, where, they say, short-term incentives were among the main causes of the financial meltdown last year. The recent bill on compensation passed by America’s House of Representatives even gives financial regulators a duty to vet the pay structures of financial firms to ensure that they do not encourage short-term behaviour that could lead to the sort of systemic crisis that we have just lived through.
Two new papers challenge this assumption. The first, “Bank CEO Incentives and the Credit Crisis”, by Rüdiger Fahlenbrach and Rene Stulz, looks at the long-term incentive pay of the bosses of banks wrecked during the financial meltdown, compared with those of bosses of banks that were less badly damaged. The study found that in 2006—arguably the year that the decisions were made that made or broke banks when things started to go wrong in 2007—bank bosses typically had very large long-term equity incentives. The median value of a chief executive’s equity stake (taking into account options) in his bank was $36m, worth, on average, over ten times his compensation.
Moreover, the more significant the long-term equity incentives, the worse a bank performed during the crisis. A bank’s return on equity in 2008 was negatively related to its chief executive’s holdings of shares in 2006. An increase of one standard deviation in dollar ownership by the boss was associated with approximately 10% lower return on equity. As for managerial heroin, different mixtures of shares and share options made no discernable difference to the bank’s performance. Using data on insider trading to estimate sales of shares, the two economists found that bank bosses made extremely large losses on their holdings of shares in their banks because they typically did not sell shares, and they also made large losses on their options.
What to conclude from this? First, that bank bosses had very high incentives to maximise shareholder wealth. Second, say Messrs Fahlenbrach and Stulz, “this evidence makes it implausible that the credit crisis can be blamed on a misalignment of incentives between chief executives and shareholders.” It is worth pointing out, as Richard Posner does in his excellent book on the crisis, “A Failure of Capitalism: The Crisis of ’08 and the Descent into Depression”, that bank shareholders may rationally support actions by their managers that involve some risk of bankruptcy, in the hope of achieving higher returns—which is why banking needs an outside regulator to constrain activities that pose a systemic risk.
The second study, presented at the annual meeting of the American Academy of Management on August 10th, takes a non-monetary approach. In “Me or We: The Effects of CEO Organisational Identification on Agency Costs”, a team of management professors led by Jim Westphal of the University of Michigan posed a series of questions to the bosses of 793 large American companies, to determine the extent to which each boss’s “self identity is intertwined with the fate of the organisation, or the degree to which the CEO defines him/herself in terms of the organisation”. The statements each boss had to respond to included “when I talk about [the organisation] I often say ‘we’ rather than ‘they’,” and “when someone criticises [the organisation], it feels like a personal insult.”
Mr Westphal’s team found that bosses with a high degree of identification with their firms did far fewer of the things for which bosses are notorious: they were much less likely to diversify into completely new businesses, and were far less likely to indulge in perks such as corporate jets. For instance, a boss who strongly agreed with the statement “being a member of [the organisation] is a major part of who I am” is 68% less likely on average to use a company aircraft than one who only somewhat agreed or somewhat disagreed with the statement.
APWas Fuld hooked?
Strong social self-identification with the firm greatly reduced the need for monitoring through an independent board of directors or incentivising with lots of equity, the study found. It argues that firms could impose less burdensome monitoring in cases where the boss’s self identity with the company is high (although surely this would provide a strong incentive for bosses to cheat to get lighter oversight by pretending to have a strong identification with their firm). Unfortunately, the study has nothing to say about how to encourage such feelings.
Nor, in common with the first study, does it offer a solution to the Dick Fuld question. The boss of Lehman Brothers (pictured) when it went bust was heavily incentivised with shares in the investment bank. It is also hard to think of a company boss, on Wall Street at least, whose self-identity was more wrapped up in the fate of his firm than Mr Fuld. Yet none of this stopped him from bankrupting Lehman. Indeed, so dedicated was he to keeping Lehman alive that he may actually have been unwilling to accept offers of help which could have saved it from bankruptcy, and him from a great fall. By the time Mr Fuld was emotionally ready to sell, it was too late.