To reserve a time to meet with the speaker, please register at the following form:
docs.google.com/spreadsheets/d/1bu8X1vUofi811JjwbVxP5o0ljbcedeX5oYVIo-MSLtQ/edit#gid=0

Abstract:
This paper empirically investigates the importance of network position for the productivity of scientists. I examine the role played by scientists who act as “bridges” on the productivity of their coauthors. I propose a pair-specific and asymmetric measure, called brokerage degree, based on the share of links of a scientist that provide exclusive access to new scientists to his coauthor. Using individual-level panel data for medical scientists covering 19 million publications between 1965 and 2009, I construct the coauthorship network. I exploit sudden and unexpected deaths of stars as a natural experiment which exogenously breaks network links and changes the local network structure. My results reveal heterogeneity in the impact of the death of a coauthor: depending on the brokerage degree between the deceased star scientist and his coauthor, the relative loss in annual publications can be up to 33% for surviving coauthors. Measuring brokerage in terms of new topics a star provides exclusive access to confirms that access to knowledge embodied in scientists further away plays a crucial role in scientific production.