International Trade Network: Statistical Analysis and Modeling

TL;DR

This paper analyzes the global trade network using statistical models to identify key determinants of trade relationships and assess structural stability amid recent disruptions.

Contribution

It applies ERGMs and SBMs to bilateral trade data, providing novel insights into the factors shaping international trade and the network's resilience to shocks.

Findings

Persistent nodal characteristics influence trade links

No major structural changes due to COVID-19 pandemic

Trade network remains stable over the studied years

Abstract

Globalization has rapidly advanced but exposed countries to supply chain disruptions, highlighted by the COVID-19 pandemic. This study exhaustively analyzes bilateral export data for 186 countries from 2018, 2020, and 2022, using Exponential Random Graph Models (ERGMs), to identify determinants of trade relationships, as well as Stochastic Block Models (SBMs), to characterize countries' roles in the trade network. Our findings show persistent, significant nodal characteristics driving bilateral trade and reveal no major structural changes in the trade network due to the pandemic.