Trade Networks and Natural Disasters: Diversion, not Destruction

Job Market Paper - With Timothée Gigout.

Preliminary draft

We study how international trade networks react to natural disasters. We combine exhaustive firm-to-firm trade credit and disaster data and use a dynamic differences-in-differences identification strategy. We establish the causal effect of natural disasters abroad on the size, shape and quality of the French exporters’ international trade networks. We find strong and permanent negative effects on French suppliers’ exports and trade credit sales to affected destinations. This effect operates exclusively through a reduction in the number of buyers, particularly among those with good credit ratings. This induces a negative shift in the distribution of the quality of firms in the destination affected by the natural disaster. On the supplier side, we find that large multinationals divert trade toward unaffected destinations, leaving their overall export level to be mostly unaffected. Trade diversion is higher for large multinationals trading more homogeneous products.

Cross-Sector Interactions in Western Europe: Lessons From Trade Credit Data

Preliminary draft

Large-scale analyses to map existing interactions between financial constraints at the sectoral level are still scarce. To fill the gap, this paper identifies short–term predictive relationships between sectors’ financial health thanks to the construction of an original micro-based indicator. Using defaults on trade credit as a sign of tighter financial constraint and a high-dimensional VAR model, I test for Granger non-causality and identify predictive relationships across sectors. Past tightening in a sector’s financial constraint is shown to be predictive of a narrower budget constraint in other sectors after controlling for macroeconomic determinants. These interactions, both in their existence and magnitude, follow the pattern of input–output network. Such structure of interactions is in line with shock propagation mechanisms described in the literature, both for direct propagation through supplier-buyer input-output relationship but also for indirect one among two suppliers of substituable goods. A few sectors - among which construction, wholesale and retail, and motor vehicles - tend to display a wider set of predictive relationships, making them key to track in monitoring processes.

Trade Credits and Sudden Stops

with Maéva Silvestrini