Do Chief Executives Get Fired When They Should?

The Wharton School of the University of Pennsylvania has recently reported on the firing of under-performing chief executives. (1) This prompted some reflections of my own based on the hundreds of board/chief executive relationships my colleagues and I have observed over the years. Three conclusions came quickly to mind.

• Boards are often slow to decide to replace an under-performing chief executive; they almost always lament that it took them so long to do so.
• While some under-performing chief executives, therefore, stay in the job longer than they might, quite capable chief executives often get prematurely ejected from their positions because of what, ultimately, are the shortcomings of their boards.(2)
• The cost of ‘changing horses’ (or is it jockeys?) is far greater than is fully understood. If it were better understood boards would likely put far more effort into ensuring that their chief executives are successful and remain in the position for as long as possible.

It is in relation to this third conclusion that the Wharton article is most useful. Commentators in the public domain often suggest that boards of directors are far too slow to replace chief executives perceived to be under-performing. However, as Wharton finance professor Luke Taylor (3) points out, it is difficult to either agree or disagree with such assertions.

In this part of the world the number of chief executive who leave their jobs unwillingly is unclear because these situations are almost always subject to PR smokescreens. In the US according to Taylor an average 2% of large U.S. corporations’ chief executives are fired each year. While noting that this rate seems low Taylor observes that it is not clear what rate of forced chief executive turnover should be expected from a well functioning board.  Consequently, it is difficult to judge whether the observed 2% rate is low or high.

To try and resolve that question Taylor set out to model the decision to fire a chief executive and to quantify the various forces at work. He identified two costs of firing a chief executive: direct costs and ‘entrenchment’ costs.  Direct costs include severance and retirement packages, fees to HR firms to find replacements - not just for the chief executive but for other senior executives who become collateral damage - and any other costs that affect profitability.  These will include the time it will take for the new team to build relationships inside and outside the company and to set a new strategy. 

Taylor defined entrenchment costs as the intangible costs that relate, for example, to severing personal relationships between directors and the chief executive.  There is a personal cost to board members who terminate the company leader: the time and stress of making a management change and the loss that directors face in the departure of a business ally or personal friend.  The degree of stress and distraction to a board contemplating whether or not to sack its chief executive cannot be underestimated.  Another contributory factor suggested by Taylor is that the board may not care all that much about maximising shareholder value at least not as much as keeping a chief executive with whom they feel comfortable. 

From his model Taylor concluded that the combined cost of these two factors was in the order of $1.3 billion per business. Of these the intangible entrenchment cost was by far the greater.

Much of the external pressure on a commercial board to fire its chief executive comes from concerns about profitability. Not surprisingly, however, Taylor found that a company's profitability is not a good predictor of whether or not its chief executive will be fired.  A board looks at a number of factors when it evaluates its chief executive's performance.  This includes everything from how the chief executive is spending his or her time, to changes in market share, to new projects that may not yet be contributing to earnings.  Taylor suggests that paying more attention to these factors makes complete sense because there is a lot of ‘noise’ in profit numbers including accounting charges and other forces outside the chief executive’s control.

Taylor's analysis hints at a number of other considerations as to whether enough chief executives are being fired.

• Increased board and shareholder activism already means that conspicuously poor chief executives do not last long.  An increase in the replacement rate is therefore likely to take its toll on chief executives who are, comparatively, more effective.  The ‘upside’ to the replacement of this next cadre of chief executives is therefore much lower.
• Because a board cannot directly observe its chief executive's ability but instead learns about it progressively over time, there is a lag in the opportunity to decide whether or not to replace the chief executive. Each time that opportunity occurs, the board must decide whether to replace the chief executive with a new one of uncertain ability.

Taylor concludes that the results of his model should provide ammunition for boards that are under pressure from irate shareholders to replace their chief executives. They could argue that the reason they don't fire their chief executives more often is it takes time to learn about his or her ability and because the firing comes with very real costs to shareholders.

(1) ‘The Cost of Retrenchment: Why CEOs Are Rarely Fired’: In Knowledge@Wharton, January 19, 2011
(2) Common reasons for this - worthy of an article on their own - include: poor original selection of the chief executive, inadequate and ineffective performance management, board dysfunctionality, etc.
(3) Lucian Taylor (2101) Why Are CEOs Rarely Fired? Evidence from Structural Estimation. Journal of Finance (forthcoming) (Source: http://www.afajof.org/journal/forth_abstract.asp?ref=628)

Top of the Article
Back to the Periodical