Deciphering the global production network from cross-border firm transactions

2025-08-17

AI Summary

Global supply chain research has long been hindered by the unavailability of firm-level cross-border transaction data. Method: Leveraging 1 billion real-world cross-border transactions across 20 million firms, we construct a directed production network to characterize global intermediate-good flows at the micro level for the first time. We innovatively integrate an LLM-generated product semantic network with NAFTA rules of origin for dual structural validation, identifying three major intermediate-good clusters: textiles; chemical & food; and machinery & metals. We propose a country–product diversification prediction metric based on forward and backward linkage intensities. Contribution/Results: Empirical analysis demonstrates high predictive accuracy for national industrial expansion paths. Findings reveal that countries with higher industrial complexity and deeper integration into dense supply chains exhibit stronger product diversification capacity. Moreover, European industrial nations and China serve as core hubs for critical intermediate goods.

Technology Category

• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP
• Application Domains: Internet of Things, Sensor Networks & Smart Cities
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Web Mining and Content Analysis: Bridging structured and unstructured data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Social Networks and Social Media: Social media analysis through the lenses of networks

Abstract

Critical for policy-making and business operations, the study of global supply chains has been severely hampered by a lack of detailed data. Here we harness global firm-level transaction data covering 20m global firms, and 1 billion cross-border transactions, to infer key inputs for over 1200 products. Transforming this data to a directed network, we find that products are clustered into three large groups including textiles, chemicals and food, and machinery and metals. European industrial nations and China dominate critical intermediate products in the network such as metals, common components and tools, while industrial complexity is correlated with embeddedness in densely connected supply chains. To validate the network, we find structural similarities with two alternative product networks, one generated via LLM queries and the other derived by NAFTA to track product origins. We further detect linkages between products identified in manually mapped single sector supply chains, including electric vehicle batteries and semi-conductors. Finally, metrics derived from network structure capturing both forward and backward linkages are able to predict country-product diversification patterns with high accuracy.

Problem

• Lack of detailed data on global supply chains
• Infer key inputs for 1200+ products from transactions
• Validate and predict supply chain diversification patterns

Innovation

• Utilizes global firm-level transaction data
• Transforms data into directed network clusters
• Validates network with alternative product networks