The Small World Network Structure of Boards of Directors

In this paper we present a random graph model to explain the network structure of boards of directors. We investigate the conditions under which corporate boards can be said to be a "small-world". Our empirical results show that the random graph model is remarkably good at explaining board structure and connectedness in the United States, the United Kingdom and Germany. Although there are small-world traits such as "clustering" and "short-paths" in the corporate world, they are no more pronounced than would be expected by chance in a statistically similar, but randomly assembled corporate universe. Finally, our results show the existence of positive degree correlation: directors who sit on many boards do so in the company of other directors who sit on many boards. This result helps explain the distribution of board interlocks.