Japan-US bilateral commodity-level trade and trade policy-related uncertainty under the COVID-19 pandemic: the nonlinear ARDL model

Abstract

This study examines whether trade policy-related uncertainties, scaled by TPU index, and the COVID-19 pandemic affect Japan’s bilateral commodity trade balance with the U.S concerning 60 industries. To this end, the nonlinear ARDL (autoregressive distributed lag) model is applied. Empirical findings of the disaggregated data model indicate that while changes in Japan’s TPU index have significant, either improving or worsening, impacts on this balance concerning 23 industries, changes in US TPU index have impact on only 32 industries. Additionally, based on the behaviors of Japanese and US consumers through their import demands, it can be interpreted that neither of them is uncertainty-sensitive to each other’s products and they continue to buy these products even if both countries’ TPU indexes keep rising. Another expected empirical finding is that the COVID-19 pandemic worsens the trade balances of Japan for different industries shown in the tables.