Global Demand Triggering Local Cooperation: The Evolution of External and Internal Network within Bangalore IT Cluster

How do firm networks evolve? Nohria (1992:15) has first expressed the need for theories allowing for a deeper understanding of ‘how networks evolve and change over time’. His suggestion has been feeding an increasingly important research stream, which aims at extending further both network-related speculation and empirical research by breaking the classic static perspective and deploying a dynamic longitudinal approach. The speculative and empirical concern to investigate networks evolution over time can be considered as stemming from the well-known endogeneity issue impinging network-based research. Indeed, the causality pattern so far is still ambiguous, whether firm characteristics are antecedents to network structure or whether network structure is shaping firms. Beyond notable exceptions considering the impact of industry events (Madhavalan, Koka, Prescott, 1998) and environmental effects (Madhavan, Koka, Prescott, 2006) on network change over time, the main research efforts considering network evolution have been focusing on the network dynamics of young new ventures. Different types of networks are theorized to be differently leveraged over the firm early growth phases, according the evolution of the firm growth need (Lechner, Dowling, 1999; Lechner, Dowling, 2003; Schutjens, Stam, 2003; Hite and Hesterly, 2001; Renko, Autio, 1998). Though, two speculative limitations of these studies can be retrieved: a) they sharply focus on young entrepreneurial firms in their early stages, thus overlooking the dynamics of networks for established firms b) the different types of networks are considered as mutually independent,: any causal linkage between them has been strikingly overlooked so far. Given the presence of many sub-networks within the overall firm egocentric network, are some sub-networks’ characteristics likely to foster the evolution of other sub-networks, thus promoting an overall network evolution? Network speculation finds an ideal empirical setting within industrial clusters. Several studies have been arguing the effects of network embeddedness to be positive, because of the privileged access to novel and valuable knowledge flows (Porter, 2000). Though, a threshold is acknowledged, beyond which undesirable effects of overembeddedness can occur (Uzzi, 1997), which can prevent new knowledge flows from reaching the cluster and fuelling local innovation capability. Assuming the perspective internal/external, several studies argue the need for cluster firms to leverage channels to external environment, in order to benefit from new external knowledge and original knowledge re-elaboration and re-assembling (Owen-Smith, Powell, 2004; Giuliani, Bell, 2005; Keeble, Wilkinson, 2000). However, to our knowledge a causal link remains unexplored so far, between the type and nature of the relationships directed across-cluster, forming the external network, and the type and nature of the relationships directed within cluster, forming the internal network. More specifically, is external network likely to influence the evolution of internal network? Bridging the two perspectives, as well as the two gaps in the literature thus identified, this work aims at unfolding the causality between external relational structure and internal relational structure from a cluster-based pointof view, by deploying a longitudinal perspective. Overcoming the traditional endogeneity problems affecting the firmnetwork causal pattern, we originally propose the network to be self-feeding, thus finding a causal antecedent within the network itself. Following the widespread approach used to investigate network evolution, we deploy case-based inductive methodology, in order to pursue a theory-building attempt (Eisenhardt, Grabner, 2007). We observe the evolution of the egocentric networks of six companies within Bangalore IT cluster over three years, from 2005 to 2007. The sample encompasses both MNCs’ subsidiaries and ...