Hiring an outside CEO: a board's best moves; The real issue isn't whether to consider an outsider--it's the need to set very different expectations for outsiders than for insiders.

• Author: Lucier, Chuck
• Position: Succession Planning

IF THERE'S ANYTHING that's guaranteed to spark intense debate among a company's directors, it's the suggestion that they look outside for the next CEO. For many boards, the very idea is heretical--a confession that the CEO succession plan was inadequate and that the company's senior management team is weak.

Of course, the board's reticence to select a CEO from outside is more than mere instinct or loyalty. There are good reasons to promote from within: The board has worked with the inside candidates and knows them far better than anyone hired from the outside; insiders understand the business, the company's culture, and the strengths and weaknesses of managers. In addition, advancing a CEO from within reinforces a company's self-confidence and reassures middle managers of a potential career path to the top.

It's possible that a board might set these benefits aside in the event that a superstar candidate comes along. But most often, boards look outside only as a last resort--when they feel they don't have a choice. Perhaps there aren't any good internal candidates; perhaps a major restructuring is required or new skills needed to set a new course. And even in these situations--perhaps especially so--the decision to go outside will trigger spirited debate.

In the debate, every director can cite some examples that support his or her position. Directors who laud the idea of an outside change agent capable of shaking up an organization may point to Lou Gerstner's success at IBM. Directors who believe an outsider is best suited to stabilize the organization and enhance operational excellence can cite Norm Payson at Oxford Health Plans. On the other hand, those opposed to looking outside might remind their colleagues that although Gary Wendt and Mike Armstrong seemed like great candidates, their tenures at Conseco and AT & T weren't successful.

A matter of expectations

Our systematic analysis of 392 CEOs of American corporations (see page 38 for description of methodology) suggests that the real issue isn't whether or not to consider an outsider--but rather the need to set very different expectations for outsiders than for insiders. Outsiders are a high-risk gamble: no more successful than insiders on average, but exhibiting a wide variation in performance. A successful outsider quickly leads a transformation of the company, delivering great returns to investors--for about five years. But performance during the second half of their tenure is poor--much worse than the performance of insiders.

We recommend that boards use insiders and outsiders to perform different roles--outsiders to drive a five-year change program; insiders to lead a decade-long evolution. Select them using different criteria, set different expectations for their tenure, and assess their performance differently.

Shaking things up

CEOs who came from the outside are the exception...