Author's School

Olin Business School

Author's Department/Program

Business Administration


English (en)

Date of Award

Spring 4-27-2014

Degree Type


Degree Name

Doctor of Philosophy (PhD)

Chair and Committee

Jackson A. Nickerson


CEO succession is a vital organizational decision because organizational strategy is a reflection of the experiences of a firm's top leader. The dissertation studies CEO succession by exploring its pattern, which reveals a dynamic nature of many succession decisions. One specific succession pattern, leadership vacillation, is observed in a grounded case of PepsiCo. Leadership vacillation occurs when organizations sequentially select their CEOs between "output" and "throughput" functions. To explain the existence of leadership vacillation, the study combines organizational vacillation theory (Nickerson and Zenger, 2002) with insights from upper echelons research (Hambrick and Mason, 1984) and studies on expert performance (Bereiter and Scardamalia, 1993). The study argues that some organizations dynamically balance exploration and exploitation by switching leaders between output and throughput functions, thus exhibiting the phenomenon of leadership vacillation. An empirical examination of 200 large, public companies indicates that the phenomenon is common and non-random. In addition, a probit model accounting for sample selection and a modified treatment effect model are employed to correct for two endogeneity problems when assessing antecedents, boundary conditions, and performance implications of leadership vacillation. The empirical analyses provide initial evidence that leadership vacillation is a product of a series of endogenous successor choices, meaning that a departing leader's functional background has an impact on that of a successor. Knowledge of the phenomenon of leadership vacillation, which has not been explored in previous research, contributes to CEO succession studies in specific and to organization and strategy research in general. Finally, the dissertation discusses its limitations and suggests possible future research on leadership vacillation.


Permanent URL: