Fallible Executives, Centralization of Decision-Making and Corporate Performance

In this paper we explore some possible consequences of fallibility in managerial decisionmaking for Þrm performance. Based on Sah and Stiglitz (1991), we develop the hypothesis that if managers are fallible, Þrm performance will be more variable as the number of managers participating in decision-making decreases, i.e. as the Þrm becomes more centralized. We use characteristics of the Executive Office to develop a proxy for the number of executives participating in top decision-making. For example, we argue that if the Chairman of the Board is not the CEO, decision-making in the Þrm will be more decentralized because the Chairman will also participate in decision-making. We test our hypothesis using this proxy (which we call the centralization index), and Þnd that the evidence is consistent with our hypothesis. Firm performance (measured by Tobin’s Q, stock returns and ROA) is significantly more variable for Þrms with greater values of our centralization index. The results are consistent across various tests designed to detect differences in variability.