The Dynamics Of China's Electronic Information Industry Cluster Innovation Networks:Pattern,path And Mechanism

With the rapid development of science and technology,innovation has gradually become the primary driving force for regional economic development.With the change of ‘local space of factors' to ‘global space of flows',the knowledge flows and innovation linkages at different spatial scales have attracted the attention of regional economists.As an important carrier and effective way of the flowing and sharing of innovation elements,the coordination and cooperation of innovation actors,the transformation and utilization of innovation results,the innovation networks and multiscale innovation networks coupling have become commonplace within the regional economics.As the core strategy to seek regional development under the effect of globallocal tension,the study of cluster takes on the transformation from ‘local view' to ‘network view' in the context of open innovation.Some scholars propose a new survey of the relationships among clusters from the network perspective of multi-scale interaction between local and trans-boundary,and it is urgent to answer whether the cluster networks are universal and what are the differences between different types of cluster networks.In addition,according to the life-cycle theory,innovation networks are heterogeneous at different stages of development,which highlights the necessity of analyzing the characteristics of cluster networks in different stages of life-cycle based on the perspective of dynamic evolution.Following the research form of evolutionary economic geography ‘pattern-pathmechanism',an analytical framework of ‘actor-relationship-scale-structure' of innovation networks is set up to explore the dynamics of cluster innovation networks in different types and different stages of life-cycle.The electronic information industry integrated with the characteristics of technical knowledge intensive,industrial cluster pattern and globalization of innovation is selected as the object of study,and the location quotient and social network analysis are comprehensively applied to identify 32 electronic information industrial clusters in China.Based on the perspective of technological innovation of clusters,two-step cluster analysis and Logistic model are used to divide the cluster type and determine the life-cycle of the cluster.Based on the patent data issued by the State Intellectual Property Office of China from 1985 to 2015,chi-square test is used to verify the hypothesis of cluster networks.On this basis,the dynamic heterogeneity of cluster innovation networks is analyzed and the evolutionary trajectory of cluster innovation networks is absolved by using Ucinet,Pajek,ArcGIS and R language programming.Combined with the symposium of key enterprises and innovation organizations,as well as the field investigation of SEMICON China,IC China and other exhibitions between 2016 and 2017,the evolution mechanism of cluster innovation networks can be clarified.The main findings of this study are drawn as follows:(1)The innovation of China's electronic information industry shows a cluster networks.The innovation actors in cluster area have significant difference in spatial location from innovation actors in non-cluster area while choosing innovation partners due to a comprehensive consideration of local knowledge spillover and cross-regional ‘tacit knowledge pool'.As a result,they prefer to cooperate with the innovators in the local and other cluster areas and pay special attention to the formation of glocalization innovation networks so as to form a cluster networks model of industrial innovation.From the perspective of spatial pattern of innovation networks,industrial cluster is helpful for collective learning and cooperative innovation,therefore,the local innovation in cluster area is higher than that of the non-cluster area,the scope and intensity of the external innovation networks are closely related to the level of regional innovation,and the networks among clusters have gradually become the core sub-network of the trans-boundary innovation networks.(2)The evolution pattern of cluster innovation networks varies from different cluster types and different stages of life-cycle.From the perspective of innovation networks actors,at the germination stage of clusters,the innovation actors of endogenous innovation leading type(?),innovation following type(?)and innovation breeding type(?)are mainly universities and research institutes;and the innovation actors of exogenous innovation leading type(?)are mainly foreign enterprises,which are the main source of technical knowledge.The innovation cooperation is highly concentrated on a single scale.The clusters of class ?,class ? and class ? are mainly based on local scale cooperation among clusters,while the cluster of class ? is mainly based on the overseas scale cooperation among clusters.Influenced by unstable industrial development environment,incomplete regional innovation system and other factors,the distance of innovation networks space is in large fluctuation.With the gradual growth and maturity of clusters,prestigious science and engineering universities,large state-owned enterprises,and sino-foreign joint venture enterprises become the core actor of innovation networks.On the dimension of innovation linkages,a strong connection among headquarters-branches related to investment,joint ventures and institutions of industry-university-research with high level of innovation is formed.The innovation cooperation of the class I cluster is mainly among local enterprises,and between derivative enterprises from universities and enterprises and local research institutions;class II cluster forms a strong relation of global scale innovation with the Taiwanese enterprises in China based on organization and cognitive proximity;class III cluster forms a long distance innovation relation with national scale enterprises and university institutes based on organization and cognitive proximity;the strong relation of class ? cluster is mainly based on the innovation cooperation with local enterprises.Since then,the networks coupling at different spatial scales has become the new guidance of cluster innovation.The clusters of class ?and class ? have entered the mature stage of innovation in succession,the multi-scale innovation networks tends to be stable and the distance of innovation cooperation tends to decrease.The cluster of class ? is in the stage of rapid development,the crosscluster relation of non-local scale is increasing,and the distance of innovation cooperation is increasing.The development of class ? cluster is relatively lagging,the external connections of cross-cluster are relatively less,and the distance of innovation cooperation decreases.On the dimension of innovation networks structure,the features of ‘core-periphery' hierarchical networks are highlighted.The cluster innovation networks of China's electronic information industry feature ‘small-world' network properties,whereby dense clusters of network actors are linked to other clusters via a relatively small number of bridging links.The actor's abilities of innovation play a key role in the formation of innovation networks.The cluster innovation networks have significant differences in spatial scope and connection intensity.The external innovation linkages of clusters show a trend from ‘less core dominant' to ‘multi core interconnection'.(3)The ‘core-periphery' pattern of cluster networks based on the perspective of relationship can be optimized.The mechanism of path creation and path lock-in jointly plays a role in the location evolution trajectory of cluster networks.Most clusters have achieved networks location upgrading through interactive learning among innovators,new members in innovation cooperation,transformation of technological innovation mode and other ways based on the mechanism of path creation.Moreover,the location evolution of a few clusters shows a path locking effect.Constrained by low level of innovation in the cluster itself,weak external connections and other factors,the location of cluster edge is locked and even degraded.Throughout the life-cycle of technological innovation in China's electronic information industry clusters,the core motive force of forming innovation networks shows the transformation from geographic proximity to cognitive proximity and social proximity,the community structure of cluster innovation networks has gradually changed from the geographical boundary of clusters with a significant characteristic of localization to a pattern that local communities and crosscluster communities coexist,to a hierarchical structure with obvious difference in community level that surpasses the geographical boundary of clusters.(4)The dynamic of cluster networks is influenced by proximity,networks emmbeddedness and actor heterogeneity.Geographic proximity facilitates tacit knowledge communication and plays a fundamental role in the development of cluster innovation networks.Cognitive proximity improves the efficiency of knowledge exchange and absorption,which is the core driving force for the evolution of cluster networks.Social proximity can enhance mutual trust between two sides,which has become an ensuring factor in maintaining close innovation cooperations.Geographical proximity and cognitive proximity,geographical proximity and social proximity have an alternative effect on the evolution of cluster networks.Embeddedness plays an important role in shaping knowledge circulation and the dynamics of cluster innovation networks was clearly driven by embeddedness.Network status measures the costs of linkages which inhibit actors to be fully connected and the effect of network status is negative and significant.With respect to the individual characteristics,the positive and significant effects of innovation ability and collaborative innovation experience show that actors are more likely to partner with innovative and experienced actors.