One of the most important and challenging decisions faced by corporate directors is whether to promote a new CEO from within or to hire new talent when a CEO leaves the company. Both have advantages. Inside candidates, often groomed by their predecessors, bring firm-, market-, and industry-specific knowledge that outsiders might take years to acquire. In contrast, outsiders bring different experiential knowledge to the strategic decision-making position and are often seen as able to make dramatic changes that an insider might not consider.
So which CEOs are “better”? The empirical literature on CEO succession is mixed, finding either no differences in firm performance or that insiders have slightly better results than outsiders.
Prior studies have two common methodological limitations. First, analysis of firms’ inputs and outputs are not independent because purposeful decision making to change one measure based on the other causes problems when drawing statistical inferences. Second, the fit between theoretical descriptions of competitive advantage and commonly used organizational outcome measures is inadequate.
Our research, which focused on CEO succession in the American health care system, examined the impact of CEO succession on productivity and efficiency. We found that any type of leadership change will result in short-term adverse impacts on a firm’s operational efficiency, but outsider CEOs experienced a clear advantage in productivity gains. We reached these conclusions by comparing 490 CEO succession decisions with a control group of 1,150 firms in the health sector.
- Sponsored by Accenture StrategyHow to make your company more nimble and responsive.
The first challenge was to find a valid measure to compare firm performance. The ideal metric needed to simultaneously address: (1) the sector’s competitive dynamics, (2) the firm’s productivity relative to its competitors, and (3) what could be attributed to the new CEO. We turned to the classic Cobb-Douglas productivity measure, which assesses how firms transform human resource expenditures and capital investments into outputs such as sales or units produced. While Cobb-Douglas did give us a measure to meaningfully assess CEO performance, analyzing it in a way that is consistent with the idea of attaining a competitive advantage was not as simple.
Most statistical techniques focus on how firms deviate from the average. While being above average may qualify as a competitive advantage, it does not focus on firms that are truly superlative as the aim. Philosophically, firm performance should be measured as the distance to a best-practices frontier, where competitive advantage is most likely to be created. To that end, we employed a statistical technique called frontier analysis that identifies the firms that are pushing the performance envelope and measures everyone else relative to them. The result of frontier analysis is an index for each firm describing its efficiency at transforming inputs into outputs relative to the best performers operating at comparable scale.
Finally, we looked for a large enough set of organizations. Enter the U.S. hospital sector. There are thousands of U.S. hospitals, which gave us a useful sample size to work with. Because hospitals receive substantial governmental revenues, they make public their operating data, which we needed to calculate productivity. This ticked our second box. Lastly, they are large organizations with extensive executive cadres to draw potential leaders from as needed.
To conduct the analysis, we identified 490 hospital CEO succession events. Organizations that selected an internal C-level officer to become CEO (82 firms) were categorized as insider successions. The remaining CEOs were termed outsider successions (408 firms). Using a 3-to-1 matching of hospitals with no change as a comparison, we analyzed 1,640 firms in all. We then analyzed hospital performance data for a five-year window around the succession event.
The results indicated that hospitals that hired outsiders had a one-time decline in relative efficiency and, in turn, were less competitive in the short term. But in subsequent years, those hospitals started to close the efficiency gap, surpassing the performance of hospitals that hired insiders. When both groups were compared with hospitals that hadn’t changed CEOs, however, both groups fell behind over the five-year period we studied.
As U.S. corporations replace CEOs with increasing frequency, our findings indicate, they should consider that any change has a negative impact in the short term and that more patience might be warranted. But if a new CEO must be found, then recruiters may want to look outside rather than in.