Modelling Global Trade with Optimal Transport

2024-09-10

arXiv.org

AI Summary

Global trade is jointly shaped by explicit costs (e.g., freight, tariffs) and implicit factors (e.g., geopolitical tensions, institutional ties), yet conventional gravity models—relying on pre-specified covariates and fixed functional forms—fail to capture its dynamic complexity. This paper introduces the first end-to-end dynamic trade modeling framework that integrates optimal transport theory with deep neural networks, enabling data-driven learning of time-varying, interpretable implicit transport cost functions without structural priors and naturally supporting uncertainty quantification. Methodologically, it unifies time-series modeling with counterfactual causal inference. Empirically, the model significantly outperforms classical gravity specifications in agricultural trade forecasting; uncovers asymmetric disruptions to Global South wheat trade induced by the Ukraine war; and detects structural breaks associated with FTA deepening, U.S.–China trade tensions, and Brexit.

Technology Category

• Search and Optimization: Learning to Search
• Reasoning under Uncertainty: Relational Probabilistic Models
• Machine Learning: Deep Neural Architectures and Foundation Models

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Global trade is shaped by a complex mix of factors beyond supply and demand, including tangible variables like transport costs and tariffs, as well as less quantifiable influences such as political and economic relations. Traditionally, economists model trade using gravity models, which rely on explicit covariates but often struggle to capture these subtler drivers of trade. In this work, we employ optimal transport and a deep neural network to learn a time-dependent cost function from data, without imposing a specific functional form. This approach consistently outperforms traditional gravity models in accuracy while providing natural uncertainty quantification. Applying our framework to global food and agricultural trade, we show that the global South suffered disproportionately from the war in Ukraine's impact on wheat markets. We also analyze the effects of free-trade agreements and trade disputes with China, as well as Brexit's impact on British trade with Europe, uncovering hidden patterns that trade volumes alone cannot reveal.

Problem

• Modeling global trade dynamics beyond traditional supply-demand factors
• Capturing subtle trade drivers like political relations and transport costs
• Analyzing trade impacts of geopolitical events and policy changes

Innovation

• Uses optimal transport for trade modeling
• Learns time-dependent cost via neural network
• Outperforms traditional gravity models accuracy