Contrary to popular belief, chief executives tend to stay too long, says Claudio Fernández-Aráoz. They – and their boards – have to plan succession

In the closing years of the last century there was a cult of the chief executive officer (CEO) as star. CEOs often appeared on magazine covers, and they published books dispensing their unique wisdom to aspirant followers. The collapse of Enron and WorldCom (both of whose leaders were fêted as visionary) put an abrupt end to this hero worship. For all that, the CEO is still seen as central to the success of the corporation, even if the qualities now sought (competence and a distaste for the limelight) are the reverse of what had at one time been thought essential to leadership. But for any chief executive and any employer, there are a couple of central questions. The CEO has to decide when to bow out. And – as this might not happen coincidentally – the board has to decide when and how to appoint a successor.

Why should a CEO retire? There are only two answers: for business reasons or for personal reasons. Obviously the ideal situation is when both of these are aligned.

Shorter, but not short enough
Let’s look first at business reasons. It is commonly felt that CEO turnover is out of control, given the shortening trend of CEO tenures. However, most evidence shows that CEOs stay too long, and can end up destroying value in a company. Perhaps the most comprehensive regular study of CEO succession is that conducted by consultants Booz Allen Hamilton. As chart 1 shows, returns to shareholders (adjusted by industries and regions) are significantly lower in the second half of CEO tenure regardless of whether the CEO was forced to leave or whether it was a regular transition, where the CEO retired or left for another job.

When segmenting by length of CEO tenure, the message becomes even clearer. It seems that CEOs who systematically perform poorly are properly prevented from staying on for too long. But it’s also true that for CEOs who stay for long periods (more than 10 years), the difference in performance between the first and second halves of their tenure is frankly dramatic (see chart 2). They contributed 5.3% of normalized median returns to shareholders in the first half of their tenure, but this fell to almost zero in the second half. There are many different reasons for this: lower motivation after too many years doing the one job; the lessening relative incentive of further remuneration; a natural loss of energy; and a lower level of competence in a changing situation.

It’s personal
People (even CEOs) seek meaning and happiness in life – and in work, since it takes up most of our waking hours. Happiness is driven by three factors: genetics (which we cannot alter), attitude and circumstances. The two most important preconditions of happiness are meaningful work and rich personal relationships.

How can chief executives be happy leaving a position where so many of their most meaningful relations have been developed? They cannot be, unless work on this transition has been started early. If the chief executive has cultivated other meaningful activities and rich relationships, the opportunity cost of staying might actually become larger than that of leaving.

Postponing the problem will only make things worse: the CEO is likely to be fired, relations will suffer more, and the range of choice for new activities will shrink.

Why is it then so hard to leave in a timely manner? The process is difficult because leaving is, in some way, confronting our mortality, realizing that neither our role, nor our strength, nor our impact lasts for ever. It is difficult to acknowledge the limits of the legacy we can leave, and to pass it on to someone else who will continue developing “our” organization to even higher levels of excellence, value and reputation.

If it’s so hard, how do you decide when it is time to retire? First, it should not necessarily be because of a short-term decline in performance. Responsible chief executives and boards should generally resist pressures from shareholders to change a chief executive just because performance has faltered. Performance can fall in the short term for many reasons, including investments in the future, or external cyclical or sectoral circumstances. Both the board and chief executive should act only on the basis of an in-depth analysis of the reasons for poor performance and after assessing whether a change at the top would improve matters.

A CEO should decide to retire either when competence falls well below that of the best alternatives, or when motivation fails, since performance is competence multiplied by motivation. It is usually easier to realize when our motivation is falling than to judge our relative competence.

Our relative competence can fall either because of a lack of energy and willingness to work hard enough, or because the environment has changed and our mix of skills is no longer appropriate. How can we best judge our relative competence? Usually we cannot. Research has demonstrated that we tend to overvalue ourselves, as a healthy mechanism for survival and happiness. Years of success as a chief executive, coupled with the unbalanced mix of praise typically received, make us quite blind to an accurate self-assessment.

Under these circumstances, the only way to find out when we should be leaving, so as to leave our company in better hands, is through taking advice. The right advisers are those who can provide an independent and insightful perspective. They might include board members, friends of a similar background and role, or professional advisers such as management consultants, auditors or executive search consultants. They should have the right combination of competence and trustworthiness, and be free of conflict of interest.

Board role
The key responsibility of the board is to select the chief executive who will, in turn, appoint and lead the management team. When making this decision, the board has four important questions to answer. What profile should we be looking for? Should we appoint an insider or an outsider? Should the roles of chairman and chief executive be split? And how should we go about choosing the successor?

What qualities should the board be looking for? A few years ago we conducted a major study of the profiles of successful and failed executives in several different cultures. We looked at three general characteristics: previous relevant experience (situational, functional or sectoral), intelligence and emotional intelligence (the ability to manage oneself and one’s relationships). Emotional intelligence was measured in this study as a combination of several personal and social competencies, including self-confidence, self-control, trustworthiness, conscientiousness, adaptability, achievement drive, commitment, initiative, empathy, political awareness, influence, communication, leadership, collaboration and team capabilities.

As chart 3 shows, the apparently appropriate profile of someone who is highly intelligent and who has relevant experience is much more likely to be a failure than a success, unless he or she also has a very high level of emotional intelligence.

Conversely, someone who is not so bright, but who has a very high level of emotional intelligence and relevant experience, is much more likely to succeed than to fail. This obviously does not imply that chief executives do not need to be intelligent. What happens is that typically candidates for these positions have already been filtered in terms of intelligence, both through their education and career path. At the chief executive level, they need to be carefully appraised in terms of emotional intelligence as well as of experience.

When replacing a CEO, is it better to employ an insider or an outsider? On average, promoting an insider maintains the previous level of results and performance. Hiring an outsider, however, can have either strong negative or positive results. As chart 4 shows, Rakesh Khurana and Nitin Nohria of Harvard Business School have found that when the departure of the predecessor was unforced, bringing in an outsider reduced performance by almost 6%. However, when the predecessor was forced to retire, bringing in an outsider was clearly the best solution, adding on average 4.4% to operating return. So in general, it adds significant value to bring an outsider when things have not been going well. Having said this, these are average findings, and the spread in performance is so large that even when things go well, a top outsider can create – and a poor insider destroy – huge economic value.

When choosing the governance model, it is important to remember that more value cannot be generated simply by deciding to split the roles of chairman and chief executive. As chart 5 shows, the median returns to shareholders seem to have been higher when the same person was appointed simultaneously chairman and chief executive from the start, rather than when the chief executive was never appointed chairman. This finding applies globally.

A note of caution is appropriate: the fact that a well-chosen and powerful chief executive might be the best solution to company leadership does not imply that split roles should be avoided. As recent scandals show, too much power in the hands of one individual makes it too easy for corruption to arise. Aristotle ranked enlightened one-man rule as the best, followed by oligarchy, with democracy third best. However, he also cautioned that, in a less than ideal situation, the order is reversed, with democracy being the least bad, and perverted dictatorship the worst. Given human nature, he preferred democracy.

How should we do it?
Lastly, how should the board go about actually implementing a succession? There are four key lessons. The generation and assessment of internal candidates should be a continuing process.

Second, given the huge emotional and social pressures that affect this process, the board should go out of its way to ensure independence and objectivity. This is only possible when there are board members of high personal integrity and selfless independence, complemented by competent and independent external advisers.

Third, the board should remember that what matters for value is performance against the competition. The difference between an average chief executive and the best is extremely large. Several studies have shown that the spread in performance grows exponentially with the complexity of the job. This implies that top chief executives will produce results that are several times better than the average, while less qualified ones will definitely generate results in the red. For this reason, the board should make sure that internal candidates have been benchmarked against the best possible external alternatives.

Finally, given the critical importance of emotional intelligence-based competencies for success at the top, the board should make sure that these soft competencies are made hard by means of a thorough and reliable evaluation. This is only possible by checking the candidates’ previous history and behaviour in depth, rather than basing decisions on résumés, or superficial interviews, or recommendations.

In the same way that deciding when to leave is not easy, choosing a chief executive is also extremely difficult and can have huge negative consequences. Choosing people is never easy, otherwise there would be no divorces, no broken friendships, and no lawyers would ever be needed. The difficulty of the task grows exponentially with the complexity of the job, since it becomes much more difficult to understand what profile is needed, and to evaluate accurately the soft skills so critical for success. Likewise, since there is so much at stake, strong emotional and political pressures build up. The process requires courage and a significant level of investment of time and energy. The consequences for organizations are, however, so huge that they clearly justify the effort. For companies, it may mean the difference between lasting success and extinction in a scandal. For society as a whole, the choice can have dramatic consequences in terms of employment, happiness and prosperity

Claudio Fernández-Aráoz
Claudio Fernández-Aráoz is a partner at Egon Zehnder International, a global executive search firm, and a member of its executive committee. His publications include the best-selling Harvard Business Review article “Hiring without firing”.