Page - (000193) - in Knowledge and Networks

187 collaboration network evolved from a neat core–periphery structure toward a new structural topology characterized by an increasing number of local and global bro- kers’ ties. This new topology closely resembles the buzz-and-pipeline model described by Bathelt et al. (2004). In analyzing this structural change, we have taken an evolutionary approach that takes both retention mechanisms and variations into account (Glückler, 2007). Among the retention mechanisms, the analysis identified the formation of a large cohesive core represennting key knowledge bases of IMAST. Examining the changing compo- sition of this cohesive core over time, we found an increasing integration of the sci- ence and technology bases. The core included not only private and public firms but also a growing number of university departments and research institutions. The con- vergence of knowledge bases through joint research activities undertaken at the core of the district remained a constant characteristic of the district’s development. In addition to this cohesive core, a bridging core also characterized the system at its initial phase. This bridging core can be thought of as IMAST’s institutional base constituted by the main local university and the national research institute. Over time, however, not only did the composition of the bridging core change to include firms, but new broker organizations also emerged. By 2013 the bridging function spread across clusters and became a characteristic behavior of broker organizations crossing structural holes rather than an institutional function. We note also that, although the original bridging core linked local cores to one another, new bridging ties linked the local cohesive core to external partners as in a buzz-and-pipeline model (Bathelt et al., 2004). From a policy perspective, the structural analysis undertaken in this study is important because it allows an assessment of how the network was governed over a 10-year period. The structural changes were the result of an active administration pursuing a coherent development strategy. A similar analysis can be replicated in other contexts to compare developmental trajectories across other policy-anchored districts in Italy and beyond. If under- taken, such study would probably open new questions about the variety of topolo- gies of innovation networks and the role local institutions play in managing the structural transitions from one topology to another. References Amin, A., & Thrift, N. (Eds.). (1995). Globalization, institutions, and regional development in Europe. Oxford: University Press. Ardovino, O., & Pennacchio, L. (2012). Le determinanti della cooperazione nei distretti tecno- logici italiani finanziati dal governo [The determinants of cooperation in government-spon- sored innovation networks: Empirical evidence from Italian technological districts]. Studi Economici, 108, 121–149. doi:10.3280/STE2012-108004 Asheim, B. T., Smith, H. L., & Oughton, C. (2011). Regional innovation systems: Theory, empirics and policy. Regional Studies, 45, 875–891. doi:10.1080/00343404.2011.596701 9 Topology and Evolution of Collaboration Networks