Page - (000176) - in Knowledge and Networks

170 The initial enthusiasm for clusters gradually thus dampened, and thought turned to the question of what type of connections were really to be encouraged, given that “being connected” is just not enough. Moreover, the promised gains of regional competitiveness were difficult to assess, given that the institutional complexity involved in regional innovation systems could not be easily translated into competi- tiveness as though clusters were firms (Martin & Sunley, 2003). The evaluation of regional innovation systems is even more problematic when clusters’ members have key public targets, as in the case of policy-anchored districts, whose major drivers can be government tenants such as capital cities, military or defense facilities, pub- lic universities, or national research centers (Markusen, 1996). The specific issues inherent in these policy-anchored districts still need to be properly addressed. Researchers contributing to a growing body of studies are investigating social network analysis as an effective means with which to examine the structure and dynamics of regional innovation systems. Collaboration networks represent the backbone of systemic learning. According to Powell et al., (1996), the wider the industrial knowledge base is, the more collaboration networks become essential for exploiting and exploring a firm’s capabilities. Such networks are not one-time dyadic interactions aimed at filling in a firm’s gaps in knowledge. On the contrary, networks are the loci of innovation representing the means through which collective learning unfolds. Studying clusters as networks has therefore become increasingly popular in the literature. Network analysis has been used to explain how the innova- tive capacity and performance of firms vary with network attributes such as central- ity or density (Boschma & Ter Wal, 2007; Giuliani, 2007; Tsai, 2001). Alternatively, network properties and configurations were explained as resulting from firms’ char- acteristics, such as knowledge bases (Baum, Shipilov, & Rowley, 2003). However, to the best of our knowledge, there are no studies formally analyzing networks as coordination mechanisms administered by a formal local institution devoted to orchestrating the network’s developmental trajectory (Provan, Fish, & Sydow, 2007). Our study contributes to this literature on regional innovation systems through a formal network analysis of a network-administered, policy-anchored district, Ingegneria dei Materiali polimerici e compositi e Strutture (IMAST), located in the region of Campania in southwestern Italy. Policy-anchored districts are often criti- cized as suffering from technological and political lock-in, where member organiza- tions are well-established, large corporations primarily interested in consolidating and reproducing a given structure through multiple embedding mechanisms (Tödtling & Trippl, 2005). We want to determine the extent to which a process of consolidation of existing structures applies to IMAST or whether behavioral changes are emerging. IMAST is a high-tech district focusing on polymeric and composite materials, engineering, and structures. This modern corporate research center was built in 2004 on the initiative of a public university, a national research center, and a set of public and private firms in strategic industries such as aeronautics, transport, and defense. A dedicated administration was established with the explicit mission of orchestrating members’ relations and facilitating the integration of the different L. Prota et al.