I am delighted to announce that our paper has been accepted for presentation at the International Conference on Trade Wars, Geopolitical Fragmentation and Repercussions for Financial and Monetary Stability. The conference will take place on November 4–5, 2026, at Brunel University of London, organized in collaboration with the University of Leeds.
I will present Bilateral Conflict Risk and Trade: Military Wars, Trade Wars, and Diplomatic Noise, joint research with Joshua Aizenman and Rodolphe Desbordes, available as NBER Working Paper No. 35077.
Our starting point is that different forms of geopolitical hostility should not be treated as interchangeable. Using GDELT event data and calibration against a human-curated reference dataset, we construct a monthly, directed bilateral conflict indicator. It separates kinetic fighting, military posture, trade-related hostility, and routine diplomatic tensions, allowing us to examine their distinct relationships with international trade.
In our gravity-model estimates, military conflict and trade-related hostility are associated with substantial reductions in bilateral trade. Routine diplomatic tensions, despite their prominence in aggregate measures, have no statistically distinguishable trade effect. The directed structure also reveals a retaliation channel that compounds trade losses over subsequent months.
Relative to a pre-escalation baseline, our estimates place approximately $334 billion of bilateral trade at risk, with the US–China relationship accounting for about half. This bilateral estimate should not be read as an equivalent net loss of world trade or global welfare.
The conference website announces a keynote by Beata Javorcik and panel contributions from Ricardo Reis, Gulcin Ozkan, and Jagjit Chadha.
I look forward to presenting this work and discussing the measurement and economic consequences of geopolitical fragmentation with colleagues in London.
Conference website · NBER working paper · Earlier EconMacro post on the paper