Group Polarization on Corporate Boards

Hongquan (David) Zhu

Ross School of Business, University of Michigan

umzhu@umich.edu

Although much research has been done to examine the external features of corporate boards and the networks created by interlocking directorates (Finkelstein, et al., 2008), limited attention has been given to the group processes within the board that may substantially influence directors’ collective actions and the functioning of directorial networks.

My research broadly explores 1) how group processes may influence strategic decisions made by boards, top management teams (TMTs), and other groups of key constituents; and 2) how local group dynamics may influence the functioning of social networks that connect key decision-making groups.

In the first essay in my dissertation, I examine how a fundamental group decision-making bias, referred to as group polarization, may influence board decisions and distort the diffusion of norms and practices through directorial networks. Group polarization is a systematic tendency for group members to amplify the group’s initial inclination in their collective decision following group interactions (Isenberg, 1986). Using and extending social psychological theories on group decisions, I explain how and why group polarization can occur on corporate boards and examine the consequences of polarization on board decisions about acquisitions premiums (i.e. percentage difference between the actual price paid to a target firm versus the market price of the target before the acquisition event). Specifically, the theory suggests that before board meetings directors tend to favor the average level of premium that they were exposed to when serving on other boards. Moreover, this predisposition of directors tends to bias information exchange and processing during board meetings, causing boards that are predisposed by their prior experience to favor relatively high (vs. low) level of premiums to approve an acquisition premium that is even higher (vs. lower) than the average premium they were exposed to in prior deals. In addition, I draw from behavioral theories to suggest that polarization biases tend to be stronger when directors 1) are more experienced in acquisition decisions; 2) were previously exposed to more homogeneous premiums; or 3) experienced previous acquisitions that are more similar to the focal acquisition. I also build on group process theories to suggest that demographic homogeneity among directors may exacerbate polarization biases.

I constructed a comprehensive annual directorial network dataset which includes all U.S. traded companies that connected to Fortune 500 firms between 1991 and 2006. This network dataset was utilized to track strategic decisions experienced by directors on the boards of Fortune 500 firms during these 16 years. Preliminary results from regressions and analysis of covariance support the group polarization hypothesis.

This study thus contributes to our understanding of boards and corporate governance by advancing a more social psychological and group process oriented perspective, which complements existing economic, social, and political approaches. This research also suggests that the diffusion of norms and practices through directorial networks may be subject to local group dynamics. Specifically, group polarization theory suggests that directors may approve more extreme decisions than what they were previously exposed to on other boards, thus reflecting a distortion of the network diffusion effect. As a central group decision-making bias, group polarization may also have implications for studies on top management teams. Group dynamics influences the way through which prior experience translates into subsequent decisions thus also has implications for research on experience effects. Finally, this study offers a novel behavioral mechanism to examine acquisitions and may have implications for other strategic decisions as well.