An Optimal Transport approach to model the community structure of the International Trade Network

Abstract

This paper introduces a novel framework to analyze the community structure of the International Trade Network by integrating Optimal Transport theory with a gravity-based null model. Unlike traditional modularity approaches, our method accounts for socio-economic constraints and assesses the extent to which residual heterogeneity shapes the community structure. This allows for a more economically grounded and policy-relevant analysis, enabling scenario simulations of economic shocks, trade disruptions, and policy changes. Empirical results show that while our approach aligns with standard modularity methods in stable periods, it outperforms them during crises, capturing deeper economic and financial dynamics. Notably, our findings reveal that the degree of financial development of countries plays a critical role in shaping the emerging partitions.