Board interlocking network and financial decisions

Previous literature demonstrates that interlocking director networks are significant in various financial areas. However, the distinction between network selection and behavioral influence effects is not considered. Few studies in the finance discipline have sufficiently considered social networks in a statistically robust way. This inattention may cause bias similar to omitted variable bias, resulting in incorrect conclusions being reached from the reported results. We employ recent advances in statistical modeling to investigate the effect of the interlocking director network on executive compensation, capital structure and mergers and acquisitions. Results show selection and influence effects are strongly significant. Firms select directors from firms with a similar proportion of fixed executive compensation (such as salary) and a similar level of acquisitiveness. Firms are influenced via the interlocking director network to become more similar with respect to executive compensation practices and capital structure choice, and more dissimilar for the number of acquisitions undertaken. That firms become more dissimilar in level of acquisitiveness may be attributed to negative feedback on acquiring because prior research shows that acquisitions are generally value destructive for the acquirer. The results suggest that the interlocking director network is a potent mechanism for the spread of firm practices. Common dependent variables from prior literature are less significant in explaining observed outcomes when the network is taken into account. For executive compensation, the profitability of a firm is no longer significant for the model including the interlocking director network but is significant when the same data are used in a standard linear regression. A similar occurrence happens for asset tangibility in the capital structure models. Tobin’s q and free cash flow are significant for the same data in a Poisson regression model for count data, and not significant for interlocking director network mergers and acquisitions models