Abstract: A study that won Best Paper prize at the 2010 SMS Meeting argued that prior estimates of the CEO effect are conflated with events outside the CEO’s control, largely the result of random chance, and that the true CEO effect is substantially smaller than has been previously estimated. We suggest that the empirical methodology employed in this study substantially overstates the “random chance” element of the CEO effect and that its conclusions are flawed because of a failure to consider overall model fit statistics as well as its use of an improper modeling technique. We replicate these findings, highlight methodological issues, offer alternative conclusions, and, using multi-level modeling, ultimately show that this prior study significantly under-estimates the CEO effect.

Abstract: Research connecting CEO characteristics to firm innovation has not yielded a consistent set of findings. We set out to address this situation by examining more complex relationship forms and also the role of managerial discretion in the more complex dynamics. Both of these tactics proved fruitful in developing greater insights into the effects of CEOs and a deeper understanding of past inconsistent findings. Specifically, using a longitudinal sample from the pharmaceutical industry, our findings suggest that: 1) simple longevity expressed in terms of age and tenure outweighs other CEO factors in innovation outcomes; 2) age and tenure exhibit interesting non-linear relationships with overall innovation productivity; and 3) age and tenure affect the locus of emphasis in the innovation domain contingent on managerial discretion.

Abstract: Although many firms now separate the CEO and board chair roles, scholars know little about the extent to which board chairs affect firm performance, and even less about how the effect of board chairs differs due to macro-institutional context. Using multilevel analysis, we empirically decompose the variance of the performance of a large sample of publicly listed firms in the U.S. and China from 2002 to 2009. Results provide strong, robust evidence that while the effect of board chairs is statistically significant in explaining the variation of firm performance in both countries, the impact of the board chairs is greater in China than in the U.S.

Abstract: Separate literatures have explored the effects of CEOs alone as well as the effects of top management teams (TMTs), CEOs included, on firm performance. However, relatively little research has investigated the relative contributions of CEOs and non-CEO members of the TMT. In this paper, we draw on the method of variance decomposition to isolate the effects of non-CEO members of top management teams and compare those effects to the effects of the CEO. Initial results suggest that both CEOs and non-CEO members of the TMT have significant effects on firm performance, but that the effects of non-CEO members are relatively greater. Our preliminary findings highlight the importance of CEOs’ indirect effects on firm performance through influencing other top managers (e.g., selection, evaluation, rewarding, and coaching).