|Author Name||GOTO Hiromitsu (Kanazawa Gakuin University) / SOUMA Wataru (Rissho University)|
|Creation Date/NO.||September 2022 22-E-092|
|Research Project||Macro-Economy under COVID-19 influence: Data-intensive analysis and the road to recovery|
|Download / Links|
Understanding the industrial structure of inner and inter prefectures is crucial for policymakers to make economic policies according to evidence. To address this issue, using the dataset of financial statements and connections for one million firms in Japan collected by Tokyo Shoko Research Inc., we construct a multiplex supply network with 47 layers equivalent to prefectures. Applying clustering analysis based on the Jensen-Shannon distance between networks and the community detection techniques known as the Infomap method to this multiplex supply network, we clarify industrial structural similarities and differences for each prefecture. Finally, we compare the results for multiplex supply networks and the well-known facts for each prefecture's Input-Output table to evaluate our result's validity and complementarity. Our findings of this study are as follows. First, from 2011 to 2018, the industrial networks of 47 prefectures can be classified into three structural patterns by a degree of urbanization. Second, the hierarchical community structure can be observed using firm-level data. However, this hierarchical community structure cannot be seen in the conventional Input-Output table dataset. Therefore, our findings suggest a new classification approach for prefectures based on similarities in the industrial structure and contribute to a better insight into the geographical characteristics of each region's industrial structure.