Disembodied knowledge flows among industrial clusters: A patent analysis of the Korean manufacturing sector

Understanding inter-industrial knowledge flows lays the foundation for building knowledge clusters and designing a national innovation system (NIS). The types of knowledge measured can be distinguished as embodied knowledge flow and disembodied knowledge flow. Embodied knowledge flow takes place through the purchase of machinery, equipment, and components that incorporate new technologies; disembodied knowledge flow is the process whereby knowledge is disseminated through human mobility and research spillovers. Previous attempts to identify industrial clusters have mainly focused on the embodied knowledge flow. This study seeks to identify industrial clusters in the Korean manufacturing sector using patent citation relationships as a proxy for disembodied knowledge flow. The six identified clusters are: supplier-dominated, material-intensive, machinery-intensive, chemicals-based, information/precision-based, and electrical/electronics-based. The technological characteristics of these six clusters are examined through various patent indicators. An inter-cluster knowledge network is constructed to identify the roles of each cluster in the overall knowledge flows. The findings should provide valuable implications for innovation policy making and design of NIS in Korea.