Essays on the antecedents, outcomes and multiplexity of informal innovation networks in an industrial cluster
Item statusRestricted Access
Embargo end date25/11/2020
Golra, Owais Anwar
Innovation scholars have been studying social networks for a long time. The two major research concerns have been to understand the origin of social structures and their consequences on innovation. Considerable attention has been given to the analysis of network structures that favour innovation. This stream of research focuses on the structural properties of networks and their effects on innovation. On the flip side, a large number of studies have investigated the underlying mechanisms and driving forces behind these network structures. This stream of research focuses on the individual, dyadic and structural-level drivers of network formation. Despite these numerous contributions, there are at least three issues in innovation-related network studies that require further investigation. First, multiplexity has received little attention in innovation studies. Notably, scholars have overlooked the formation of multiplex innovation networks. Thus, there is a need to analyse the individual, dyadic and structural level drivers of the formation of multiplex innovation networks. Second, network research is dominated by studies conducted in the western context, and there is a lack of contributions from developing countries. Scholars have also highlighted this issue in recent studies. Third, innovation scholars have mainly focused on undirected networks and formal collaborations, and little attention has been paid to studying directed informal networks. Thus, this thesis aims to fill these research gaps and investigates the antecedents, outcomes and multiplexity of directed and informal innovation networks. The thesis constitutes three papers. The first paper, “Proximity and its impact on the formation of product and process innovation networks”, contributes to the stream of literature investigating the dyadic-level antecedents of the formation of multiple networks. It analyses the role of multi-dimensional proximity (a dyadic-level driver) in the formation of product and process innovation networks. Using multiple regression quadratic assignment procedure (MRQAP), a social network analysis technique, I study these networks among seventy-three firms in the Lahore textile cluster in Pakistan. I find a significant influence of four dimensions of proximity on the process of network formation. Notably, the impact of social, cognitive and organisational dimensions of proximity is found to be stronger for process innovation network than for product innovation network. Contrarily, geographic proximity plays a more critical role in network formation for product innovation than process innovation. The second paper, “Formation and dynamics of product and process innovation networks: evidence from a textile cluster in Pakistan”, also contributes to individual-level and network-level drivers of multiplex network formation. It investigates the influence of individual and relational attributes of actors, as well as endogenous network mechanisms on the formation of product and process innovation networks. Using exponential random graph models (ERGM), this study examines the effect of absorptive capacity and innovative capacity as individual-level attributes; business relations as a dyadic-level factor; and popularity, activity, reciprocity, multi-connectivity and transitivity as network-level characteristics, on the formation of product and process innovation networks. The study finds that individual attributes, relational attributes and endogenous network mechanisms show a significant influence on the formation of both innovations networks. The third paper, “Influence of a firm’s network position on its innovation outcome in a mature industrial cluster”, employs a social network perspective to investigate the influence of firms’ structural and relational embeddedness on their innovation outcome in a directed network in an industrial cluster. From the structural embeddedness perspective, the paper argues that a central position in an informal advice network does not bring equal innovation benefits to advice-seekers and advice-givers. Notably, in a mature industrial cluster, it is expected that the number of advice giving ties (popularity) positively influences the innovation outcome of firms, whereas the number of advice-seeking ties (activity) negatively affects the firms’ innovation. From the relational embeddedness perspective, the paper investigates the effect of strong and weak ties on the innovation outcome of firms in a mature industrial cluster. It expects a positive relationship between firms’ innovation output and strong ties, and a negative relationship between weak ties and the innovation output of firms. The findings suggest that activity has a significant negative impact on the innovation outcome of firms, while popularity shows a significant positive impact on the innovative outcome of firms. Strong ties show a positive and significant impact on innovation, while weak ties demonstrate a significant adverse effect on innovation. The study also finds that absorptive capacity fully mediates the relationship between advice-giving ties and innovation, and partially mediates the relationship between advice-seeking ties and innovation. This work has implications for cluster policymakers as well as research and development managers.