TY - JOUR
T1 - Being central is a double-edged sword
T2 - Knowledge network centrality and new product development in U.S. pharmaceutical industry
AU - Dong, John Qi
AU - Yang, Chia Han
N1 - Publisher Copyright:
© 2016 Elsevier Inc.
PY - 2016/12/1
Y1 - 2016/12/1
N2 - Today firms extensively use external knowledge from interfirm knowledge networks for their new product development (NPD). In light of this phenomenon, scholars and managers often believe that a higher centrality in interfirm knowledge networks is good for absorbing external knowledge and improving NPD performance. Since knowledge network centrality can be measured from different perspectives, however, we propose that some types of centrality might do more harm than good for NPD. Using a panel data set from the U.S. pharmaceutical industry, we empirically examine the impacts of three measures for knowledge network centrality (i.e., degree centrality, closeness centrality and eigenvector centrality) on NPD performance. We find that degree centrality in an interfirm knowledge network is positively associated with subsequent NPD performance. Counter-intuitively, closeness centrality and eigenvector centrality in an interfirm knowledge network have negative impacts on subsequent NPD performance. Taken together, our findings remind the danger of oversimplifying the complex impact of knowledge network centrality on innovation.
AB - Today firms extensively use external knowledge from interfirm knowledge networks for their new product development (NPD). In light of this phenomenon, scholars and managers often believe that a higher centrality in interfirm knowledge networks is good for absorbing external knowledge and improving NPD performance. Since knowledge network centrality can be measured from different perspectives, however, we propose that some types of centrality might do more harm than good for NPD. Using a panel data set from the U.S. pharmaceutical industry, we empirically examine the impacts of three measures for knowledge network centrality (i.e., degree centrality, closeness centrality and eigenvector centrality) on NPD performance. We find that degree centrality in an interfirm knowledge network is positively associated with subsequent NPD performance. Counter-intuitively, closeness centrality and eigenvector centrality in an interfirm knowledge network have negative impacts on subsequent NPD performance. Taken together, our findings remind the danger of oversimplifying the complex impact of knowledge network centrality on innovation.
UR - http://www.scopus.com/inward/record.url?scp=84979518006&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84979518006&partnerID=8YFLogxK
U2 - 10.1016/j.techfore.2016.07.011
DO - 10.1016/j.techfore.2016.07.011
M3 - Article
AN - SCOPUS:84979518006
SN - 0040-1625
VL - 113
SP - 379
EP - 385
JO - Technological Forecasting and Social Change
JF - Technological Forecasting and Social Change
ER -