The Effects of US Sectoral Shocks through the World Input-Output Network

In this paper, I estimate the global effects of hypothetical 1% changes in US sectoral productivity. To do that, I formulate a multi-sector Armington trade model with import tariffs, trade in intermediate goods, sectoral heterogeneity, and input-output linkages. Because a closed form for changes in welfare is not available in the model, as opposed to Ossa (2015), I solve for equilibrium to conduct the counterfactual exercises. In particular, knowing that the gravity equations in this model are identical to Caliendo and Parro (2015) once we calibrate the trade elasticity and industry level productivity to the corresponding data, I modify their computation approach to estimate the counterfactual productivity changes. The model predicts that the primary channel through which the sectoral shocks affect welfare is terms of trade. I also find that both US productivity’s direct effect and effects through export prices are substantial for countries such as the US, Canada, Chile, and Mexico. On the other hand, changes in volume of trade are small and their directions of changes are also mixed across countries.