Identifying influential directors in US corporate governance network

 

IRENA VODENSKA-CHITKUSHEV

BOSTON UNIVERSITY

 

Corporate governance is important for developing company policies and assuring business growth and innovation. For successful industry positioning and increased competitiveness, companies tend to elect directors that are influential and well known in the business world as well as in the community. We analyze the structure of US corporate governance network for the period between 1996 and 2006. We examine the Investor Responsibility Research Center (IRRC) directors’ database containing approximately 1,600 companies and 10,000 directors per year. More specifically, we study the influence and power of directors in the network by developing Influence Factor, a new centrality measure based on information flow in the directors’ network. Influence Factor measures the level of information that is shared among corporate directors. If there is a direct link between directors, the information that is shared between these directors contributes to a higher Influence Factor score compared to information shared between directors who are connected through intermediate directors. We find that the normalized Influence Factors of directors for different years follow the same distribution, which enables us to compare the influence of directors over time. We also study Influence Factors for different industries and find that the Influence Factor is efficient measure for identifying powerful directors within specific industries. We compare our results with the results obtained by using other centrality measures such as degree, betweenness centrality, and K-shell structure of the nodes, and find that Influence Factor is superior to other centrality measures in identifying influential directors. We also compare our findings with popular lists of prominent people in business such as Fortune magazine and Networking magazine lists and find overlapping directors between our quantitatively generated top lists and the popular magazine lists based on qualitative survey data.