The global production pattern of the semiconductor industry: an empirical research based on trade network

Abstract Semiconductors are an important electronic component and play a central role in many industrial fields, whose production system is spread all over the world, involving trade in many raw materials, mechanical equipment, and finished products. Comprehending a broad global production picture of the semiconductor industry can be an arduous task. To tackle this complexity, the social network analysis method was used to capture the flows of products between economies in the semiconductor industrial chain. This study analyzed the evolution of the trade pattern of four key commodities in the semiconductor industrial chain, with a focus on identifying the status of major economies in the global semiconductor trade system. Additionally, the influencing factors of the formation of the trade networks were explored by using the Exponential Random Graph Model. The results showed that: (1) From 2001 to 2019, the import and export trade pattern of four key commodities in the semiconductor industrial chain has shown the characteristics of “rising in the east and decreasing in the west”, but the change range shows strong commodity heterogeneity, the change of the spatial pattern of integrated circuits trade is the most significant, and the integrated circuits trade proportion of Asian economies has increased hugely with more than 80%. (2) The trade connectivity of various commodities in the semiconductor industrial chain has increased, upstream support products with high technical requirements such as wafers and equipment have the lowest average degree and weak trade liquidity, and the average degree of trade network of packaging materials and integrated circuits is relatively high, with strong trade liquidity and closer and more prosperous trade connections. (3) From 2001 to 2019, the “core-periphery” system of global semiconductor industry trade has changed significantly, with the core circle undergoing obvious iterative reorganization. for example, China and Singapore have achieved status improvement to the core circle in the base materials trade network, while Malaysia has been squeezed into the semi-peripheral circle. (4) The formation of trade networks of various commodities in the semiconductor industrial chain was driven by three aspects: network self-organization, exogenous endowment attribute, and exogenous network embeddedness. Among them, the influence of the technological innovation level of economies on different commodity trade networks showed heterogeneous.