Rethinking the competition of export trade based on the bipartite network

As trade competition has regained attention in the current global context, this article seeks to develop a competition measurement grounded in a complex-network framework, thereby overcoming the limitations of conventional indicators that neglect “heterogeneity” and “interconnectedness”. We construct a country–commodity bipartite network based on the RCA index and apply the asymmetric reflection algorithm in combination with the Matrix-Estimation Exercise to calculate two indicators: generalized competition intensiveness of country (GCC) and of product (GCP), which quantifies the overall competition status at both the country and commodity levels. Subsequent algorithmic discussions and four case-based analyses support the validity and reliability of the proposed measures, further demonstrating that GCC and GCP are closely interrelated, mutually reinforcing, and determined by export structures rather than trade volumes. Robustness checks across different RCA thresholds, eigenvalue specifications, and trade-data statistical calibers consistently confirm the stability of the results. Cross-sectional comparisons with traditional indicators highlight the superiority of the proposed framework, showing that conventional measures tend to underestimate the competition intensiveness of underdeveloped economies while overestimating that of advanced ones. Finally, sensitivity analyses of intra-country and intra-sector shocks distinguish between two categories of countries, yielding policy insights that suggest divergent developmental pathways.