Exchange rate movements' impact on international trade is a critical research topic, especially during the COVID-19 pandemic. Policymakers implemented measures to stabilize trade flows, making it crucial to understand exchange rate effects on trade and balance of payments. This study examines the impact of policy responses to the COVID-19 pandemic on the exchange rate elasticity of China's imports from Asian trading partners. China's role as Asia's largest importer and a central node in global value chains (GVCs) makes this analysis particularly relevant. The study utilizes a unique monthly Chinese provincial customs dataset covering the pandemic period, obtained using a Python crawler, offering a high level of disaggregation. This allows for examination of heterogeneous effects across provinces and products, capturing rapid dynamics during the crisis. The authors anticipate that the exchange rate elasticity of China's imports is significantly influenced by government support policies. More aggressive policies are expected to reduce export responsiveness to bilateral exchange rate changes, potentially reversing the sign of the elasticity. Currency depreciation by foreign competitors is expected to deter exports to China, an effect amplified by lower market concentration. Strong support policies should mitigate this foreign competition effect. The study further investigates a product quality channel, where support packages may lower bilateral exchange rate elasticity through quality upgrades in exported products. The results are robust across different trade modes, policy measures, and various controls, including regional and product heterogeneity analysis.

This paper connects to two main research areas. The first is the effect of the COVID-19 pandemic on international trade. Existing literature highlights the pandemic's severe impact on global supply chains and demand (Gruszczynski, 2020; Che et al., 2020; Vidya and Prabheesh, 2020; Hayakawa and Mukunoki, 2021; Zhao et al., 2021). Liu et al. (2021) emphasize the importance of third-country effects and heterogeneity in analyzing pandemic impacts on trade. This study complements this research by using detailed monthly provincial-level data to capture heterogeneous impacts across Chinese provinces and rapid dynamic responses. The second area focuses on the impact of competitor countries' exchange rate movements on domestic exchange rate elasticity. Previous studies demonstrated that competitor currency devaluations significantly affect domestic export performance (Feenstra et al., 2002; Cheung et al., 2016; Mattoo et al., 2017; Pennings, 2017). This paper uniquely analyzes exchange rate elasticity during the COVID-19 period with highly disaggregated data, integrating the effects of foreign competition and pandemic-induced policy responses to provide new insights into this important topic. The policy implications of the findings are significant for price and quantity adjustments to exchange rate movements during the COVID-19 crisis.

The empirical analysis utilizes monthly provincial-product-level Chinese customs data at the HS 2017 8-digit level from January 2019 to March 2021. The data's high level of disaggregation allows for detailed analyses across provinces, products, trading patterns, and trade modes. The import price is the CIF price in RMB. The study focuses on China's imports from Asian trading partners given China's central role in Asian GVCs and the high proportion of imports originating from Asia. The analysis distinguishes between ordinary trade and processing trade (pure assembly and processing with imports), which have distinct characteristics. The Oxford COVID-19 economic support index from the OxCGRT is used to quantify government policy responses, measuring support for unemployed people, debt relief, fiscal stimulus, and other public expenditures. The Herfindahl-Hirschman Index (HHI) measures market concentration of exporters in each provincial market. The main regression model examines the determinants of changes in import value, unit price, and quantity. Independent variables include changes in the real exchange rate of the RMB against Asian trading partners (RER_CNY), competitor real exchange rates (RER_Comp), the Oxford economic support index (Policy), the HHI, and the number of new COVID-19 cases in the exporting country. Lagged values of exchange rate and HHI variables are used to address potential endogeneity issues. High-dimensional fixed effects (province-time, province-product-destination, and monthly fixed effects) control for omitted variables. The model allows for the interaction of exchange rate variables with the policy index and HHI to capture the policy effect and market concentration effect, respectively. The analysis also includes robustness checks with alternative policy measures (containment and health index), alternative databases (UN Comtrade), the addition of dominant currency (USD and JPY) effects and lagged exchange rate effects, and further disaggregation into ordinary trade and processing trade (Pure Assembly and Processing with Imports). The analysis also explores product heterogeneity (classified by Lall's industrial classification) and regional heterogeneity (coastal versus inland provinces) and uses a Probit model to explore the extensive margin of adding a product-destination pair. Finally, a non-linear model is estimated to explore non-linear exchange rate effects.

The baseline regression results show that a 10% real depreciation of the RMB increases average provincial imports by about 17.58%, which seems counterintuitive but consistent with previous research on China's import behavior. This effect is primarily driven by quantity adjustments rather than price adjustments. The interaction term of the RMB exchange rate and the economic support policy index is significantly negative, suggesting that stronger economic support policies can change the sign of the exchange rate elasticity, making it more aligned with traditional economic intuition (RMB appreciation increases imports). This finding supports Proposition 1 from the theoretical model. The coefficient on competitors' real exchange rates is positive, confirming the 'beggar-thy-neighbor' effect (competitor currency depreciation reduces exports to China, supporting Proposition 2). The interaction term between competitors' exchange rate and policy is negative and significant, indicating that stronger economic support policies mitigate this foreign competition effect. The impact of the foreign competition effect is stronger when market concentration is lower. A more severe pandemic negatively impacts imports. The analysis of different trade modes shows that the effects of policy and exchange rates are larger for processing imports (PI) compared to ordinary imports. The product quality channel investigation using residuals from a quality inference model shows that the policy effect on exchange rate elasticity is significantly negative, indicating that policy support leads to quality improvements in exports, affecting the responsiveness to exchange rates. Robustness checks across different country income levels, provinces (coastal vs. inland), and industry classifications show consistency in the policy effect. Analyzing the different trade modes indicates that the exchange rate effects and policy impacts are more significant for processing imports, particularly PI, than ordinary trade. The extensive margin analysis (using Probit and Logit models) reveals that RMB depreciation increases the probability of adding new products from Asian exporters, but this effect is mitigated by stronger economic policies. The dominant currency effect is insignificant after controlling for other variables.

The findings address the research question by demonstrating how pandemic-induced economic support policies modify the relationship between exchange rate movements and China's import demand. The results highlight two key channels: the direct impact of policies on export responsiveness to exchange rates and the indirect impact through reduced foreign competition. These findings contribute significantly to the literature on exchange rate elasticity and international trade by integrating the unique context of the COVID-19 pandemic and using highly disaggregated data. The study's implications are particularly relevant for policymakers designing policies to boost economic recovery, maintain stable trade flows, and manage balance of payments. The findings suggest a more nuanced understanding of exchange rate effects on imports is needed, accounting for both bilateral and third-country effects, along with policy interventions.

This research provides novel empirical evidence on the exchange rate elasticity of China's imports during the COVID-19 pandemic. The study demonstrates the significant modifying role of government economic support policies on this relationship. The results highlight the importance of considering both bilateral and third-country exchange rate effects, as well as market concentration and the product quality channel. Future research could explore the long-term consequences of these effects, extend the analysis to trade in services, and further investigate the impact of GVC integration on exchange rate elasticity using more comprehensive input-output data. Further research into the interaction between the various policy channels would also be useful.

The analysis is limited by the availability of data, extending only to March 2021. While the use of the UN Comtrade database partially addresses this issue, a more complete and updated dataset would enable a more comprehensive analysis. The study focuses on China's imports, so generalizations to other economies should be made cautiously. The econometric model assumes a relatively simple functional form; including more complex non-linearities and interactions could improve the analysis. The study primarily focuses on the quantity adjustments, with limited emphasis on price adjustments. Finally, the causal inference could be further strengthened by utilizing more sophisticated econometric techniques that address endogeneity issues further.