Working Paper: CEPR ID: DP17395
Authors: Shekhar Aiyar; Davide Malacrino; Adil Mohommad; Andrea Presbitero
Abstract: While standard demand factors perform well in predicting historical trade patterns, they fail conspicuously in 2020, when pandemic-specific factors played a key role above and beyond demand. Prediction errors from a multilateral import demand model in 2020 vary systematically with the health preparedness of trade partners, suggesting that pandemic-response policies have international spillovers. Bilateral product-level data covering about 95 percent of global goods trade reveals sizable negative international spillovers to trade from supply disruptions due to domestic lockdowns. These international spillovers accounted for up to 60 percent of the observed decline in trade in the early phase of the pandemic, but their effect was short-lived, concentrated among goods produced in key global value chains, and mitigated by the availability of remote working and the size of the fiscal response to the pandemic.
Keywords: Trade; Spillover Effects; Lockdown; COVID-19
JEL Codes: F14; F60; I18
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Stringency of lockdown measures (C54) | Bilateral imports (F10) |
| Lockdown spillovers (F69) | Decline in imports (F14) |
| Fiscal responses (E62) | Spillover effects (F69) |
| Teleworkability of jobs (J62) | Spillover effects (F69) |
| Lockdown stringency (H77) | Supply disruptions (L99) |
| Lockdown stringency (H77) | International repercussions (F69) |