Acyclic Depth Partition of a Complex Subcontracting Network

Abstract

The present research unveiled inter-industry dynamics deeply embedded in the extremely complex subcontracting networks in a large-scale industrial district, where over 7,000 small- and medium-size enterprises (SME) functioned as suppliers for leading Japanese manufacturing firms, based on relational data from 1994-95. It analyzed the regional production mechanisms embedded in the large-scale industrial district, applying the analytical concepts of cycle and network centrality in combination. Applying the concept of acyclic network, the analysis converted the complex regional subcontracting networks into a simpler form. Relative positions of the firms were mapped out across seven hierarchical clusters, by reorganizing the flows of goods and services into linked hierarchical stages of manufacturing processes. Based on degree centrality, relative positions of most central prime buyers were scrutinized, in order to identify inter-industry linkages among the linked, but different, industries. Two key findings were as follows. First, SMEs in the relatively low clusters collectively filled a role of supporting industries for the leading prime buyers located at relatively high layers, offering a variety of specialized manufacturing processes and services. The extended networks were spread out far beyond the geographical boundary. Second, there was a hierarchical order among the linked, but different, embedded industries, as the relative positions of the most central prime buyers indicated. In contrast to conventional claims that assumed the existence of random networks in large-scale industrial districts, the present research articulated the underlying social structure based on quantifiable relational data, not only between suppliers and buyers, but also among and across the embedded industries in an industrial district, for the first time in academia.