The Making of Brokers: When Second-Order Brokerage Leads to Formation of Neighborhood-Spanning Ties

Is the formation of bridging ties by an individual (first-order brokerage) affected by the presence of bridging ties of his partners (second-order brokerage)? Despite the recent interest in the consequences of second-order brokerage, little is known about its impact on network dynamics. We propose that individuals will have more bridging ties when they are connected to partners who have many bridging ties of their own. This effect will be positively moderated by the second-order broker’s performance and negatively moderated by its status. We find support for these relationships using data on collaboration networks in the French television game-show industry between 1998 and 2012. Our results also show that the performance consequences of bridging ties depend on the roles which individuals play in their organizations. Bridging ties bring benefits to people whose role requires access to diverse information, but are irrelevant to the performance of other individuals. Our findings suggest that individuals do not always understand the benefits of brokerage: some of them build bridging ties to imitate high-performing second-order brokers even when these ties have no impact on their own performance.