Seite - (000300) - in Knowledge and Networks

297 SHINDIR GK d d c i n j n ij ind i n j n ij c h c h _ = = = = = ∑ ∑ ∑ ∑ 1 1 1 1 1 1 (14.6) Expression (14.5) is the share of the overall external reach that is due to direct external linkages made by gatekeepers, whereas expression (14.6) is the share of the overall external reach that is due to indirect external linkages mediated by gatekeep- ers. Such a distinction allows addressing two questions: (a) Whether indirect links mediated by gatekeepers are actually relevant for tech- nological recombination in cities as claimed in the literature; (b) whether direct relations held by gatekeepers are superior to direct relations held by other actors because of the higher inventiveness, creativity, and power char- acterizing gatekeepers as proposed in the literature. Fourth, we measured to what extent the indirect external reach mediated by gate- keepers (i.e., expression (14.2) above) is concentrated in the hands of few individu- als. To this purpose, we proceeded as follows. For each individual gatekeeper we computed how much the indirect external reach of the city would decrease by removing this gatekeeper. Then, we took the top 10 % of gatekeepers in terms of impact and computed how much the indirect external reach would decrease by removing these individuals (GKREMOV). It is worth noting that this exercise is somehow related to the notion of network redundancy. To the extent that there are multiple shortest paths and thus gatekeepers between inventor i (in city c) and inventor c, removing a specific gatekeeper does not have any effect on the overall external reach of the city. On the other hand, if there is only one shortest path and thus only one gatekeeper between them, then there is no redundancy, namely the gatekeeper has full control over any information flow between i and j. In terms of the measure discussed above, the greater the impact of the top 10 % of gatekeepers on the overall external reach, the lower the redundancy in information paths and the more the access to external sources of knowledge is concentrated in the few key individuals. A priori, the impact of this variable is uncertain. On the one hand, dis- persion of the gatekeeping function among many inventors, with a certain amount of redundancy resulting, could improve the reliability and continuity of access to external knowledge sources and mitigate the control power individual gatekeepers exercise on these sources of knowledge. On the other hand, as long as performing an effective gatekeeping function requires the possession of distinctive skills and attributes that are unlikely to be evenly distributed, concentrating that function among key individuals might result in a better outcome. Finally, the role played by gatekeepers could also vary according to their concen- tration or dispersion across firms within a city. To this purpose, we computed the share of the indirect external reach mediated by gatekeepers (i.e., expression (14.2) above), which is accounted for by the top four patenting firms (CONCGK). Once again, the effect of this variable is a priori uncertain. On the one hand, if one believes 14 Are Gatekeepers Important for the Renewal of the Local Knowledge Base?…