Yesterday’s announcement by Microsoft that insider Satya Nadella would take over as CEO hardly seemed to inspire investors. The company’s stock price barely budged, no doubt in part because Nadella had been the rumored frontrunner, but also perhaps because he was widely seen as the “safe” choice. But does that imply that an outsider CEO would be better for Microsoft? The evidence says otherwise.
Instead, the appointment of a new CEO from within Microsoft’s ranks is evidence of a healthy succession process, says Joseph Bower, professor at Harvard Business School. “The board should be developing insiders,” he argues, and the fact that Microsoft has done so is a signal that it is succeeding at the succession process.
Insider CEOs have an intimate knowledge of the firm’s capabilities, customers, and employees. Outsiders, by contrast, can take years to acquire the necessary familiarity with a company’s structure and operations. “There’s a tendency to look at the job of CEO like driving a car,” said Bower. “When you rent a car, you adjust the mirror, you turn the key and there you go. You never look under the hood. Companies aren’t like that.”
Research by Bower and others suggests that insider CEOs outperform outsiders, as Bower wrote for HBR in 2007:
Strong evidence supports the notion that a well-groomed insider is a key to sustained company performance. In my analysis of 1,800 successions, for instance, I found that company performance was significantly better when insiders succeeded to the job of CEO. Other researchers, including Jim Collins in Good to Great, have come to similar conclusions working from different data sets.
More recent research published in 2010 by HBR found more mixed results, with insiders and outsiders performing roughly the same. But notably, firms tend to pay more for outside CEOs than for insiders.
If outside CEOs are more costly and offer, at best, comparable performance, why do many firms pass over internal candidates? “When an outsider comes in, you get a pop in the stock,” Bower told me, citing evidence from the hire of GE executives by other firms. Yet research has shown that market reaction to a new CEO does not predict subsequent performance.
If Nadella’s twenty-plus years of experience at Microsoft will be an advantage, the challenge for insiders like him is maintaining enough outside perspective to know which things must change. Here again Bower is optimistic, pointing to some evidence that Nadella took contrarian positions with respect to open source and developer tools in previous roles.
Although industry and macroeconomic trends likely play a larger role in firm performance than does any single executive, data suggests that the CEO does have a measurable impact, albeit one that varies across industries. The impact of any Microsoft CEO likely falls in the middle of this spectrum. Somewhat counterintuitively, the fact that Microsoft is competing in a dynamic, high growth, and opportunity-rich industry suggests the CEO will have less impact. (When opportunities are hard to come by, the importance of CEOs taking advantage is magnified.) At the same time, the fact that the firm does have the resources to invest in R&D and acquisitions suggests the opposite.
One potential pitfall in what otherwise seems to have been a well thought-out succession process will be the relationship between Nadella as CEO, John Thompson as chairman, and Bill Gates in his newly expanded product role. “Conventional wisdom,” said Bower, “is that this is very, very hard to pull off.”