Spatiotemporal analysis of the dynamic evolution and driving factors of trade networks in the Belt and Road countries

As the Belt and Road Initiative (BRI) continues to advance, trade networks among BRI countries have evolved significantly. Understanding development patterns within these trade networks is crucial for promoting further growth. This study adopts a spatiotemporal perspective to analyze the dynamic evolution and driving factors of trade networks among BRI countries, utilizing the Separable Temporal Exponential Random Graph Model (STERGM) and a change point detection model. These methods assess the impact of endogenous structural variables, exogenous edge-level covariates, and exogenous nodal variables on the formation and dissolution of trade networks, as well as on stage-specific changes within these networks. The findings reveal that: (1) around 2017, the trade networks underwent a significant shift, with high-trade-value relationships growing faster than low-trade-value ones, and the networks have a small-world character. (2) China, Turkey, India, and Russia hold central positions in the trade networks, functioning as “bridges” and “hubs”; the prominence of Poland, the Czech Republic, and Ukraine has increased, while Thailand and United Arab Emirates have seen a relative decline; (3) geographical proximity, bilateral investment treaties, and shared legal origins foster trade network development, whereas exchange rate volatility and political distance have a negative impact. Countries with high urbanization, large populations, and strong economies are more likely to form trade relations. And these effects on the formation and maintenance of trade relations changed significantly before and after 2017. Therefore, while enhancing their own economic and social development, BRI countries should work to strengthen trade relations by bridging political differences and establishing trade agreements.