I. Introduction
This paper examines how protectionist trade shocks propagate internationally through the uncertainty channel. Specifically, it investigates whether the 2018–2020 U.S. tariff escalations increased global economic policy uncertainty (GEPU) and whether such uncertainty depressed economic growth in major Asian economies. We hypothesize that tariff shocks originating from a dominant economy generate global policy uncertainty and that the macroeconomic effects of such uncertainty vary across countries according to their trade exposure to the shock-originating economy.
The theoretical mechanism follows real options models of irreversible investment under uncertainty (Bernanke, 1983; Bloom, 2009; Dixit & Pindyck, 1994). When firms face uncertainty regarding future trade policy, they delay investment and hiring decisions, reducing aggregate demand and output. Recent macroeconomic evidence shows that uncertainty shocks can operate as negative demand shocks capable of slowing economic activity (Bloom et al., 2018; Leduc & Liu, 2016). In a globally integrated environment, such shocks can propagate internationally through trade linkages, global value chains, and expectations, with the magnitude of spillovers depending on countries’ structural exposure to the source economy.
The resurgence of protectionism, particularly during the U.S.–China trade conflict, has renewed interest in the macroeconomic consequences of trade policy shocks. While existing studies primarily focus on welfare, price, and trade-flow effects (Amiti et al., 2019; Fajgelbaum et al., 2020), relatively little attention has been given to uncertainty-driven macroeconomic channels. Recent work shows that trade policy uncertainty can significantly affect global economic activity through expectations and investment decisions (Caldara et al., 2020; Handley & Limão, 2017). Moreover, global uncertainty indicators such as the Economic Policy Uncertainty index (Baker et al., 2016) and the World Uncertainty Index (Ahir et al., 2022) reveal strong cross-country spillovers.
Using a balanced panel of eleven Asian economies from 1997–2024, we interact GEPU with country-level U.S. trade exposure to identify heterogeneous effects. The results show that economies more exposed to the U.S. experience larger declines in economic growth when GEPU rises, highlighting the asymmetric macroeconomic consequences of protectionist trade policy.
This paper makes three contributions to the literature. First, it links protectionist trade shocks to macroeconomic outcomes through the GEPU channel, providing evidence that tariff escalations can generate internationally transmitted uncertainty shocks. Second, by exploiting cross-country variation in U.S. trade exposure, the paper identifies heterogeneous growth effects of GEPU, highlighting the role of structural trade linkages in shaping macroeconomic vulnerability. Third, the analysis contributes to the growing literature on international spillovers of policy uncertainty by showing that the macroeconomic consequences of uncertainty depend not only on global shocks but also on countries’ exposure to the policy-origin economy.
The remainder of the paper proceeds as follows. Section II describes the data sources, variable construction, and empirical strategy. Section III presents the baseline results and robustness checks. Section IV concludes with the main implications for trade-policy exposure, uncertainty transmission, and future research.
II. Data and Methodology
A. Data
The study uses an annual balanced panel of 11 Asian economies over 1997–2024, comprising 308 country-year observations. GDP per capita growth, sourced from the World Bank’s World Development Indicators (WDI), serves as the dependent variable. Global economic policy uncertainty (GEPU) is measured using the global index developed by Baker, Bloom, and Davis (2016). To capture the trade policy shock, a Trump Tariff (TT) dummy is constructed, equal to one for 2018–2020 and zero otherwise. World commodity prices are obtained from the World Bank’s Pink Sheet. Control variables, including trade openness, gross capital formation (GCF), foreign direct investment (FDI) inflows, inflation, and GDP, are drawn from the WDI. Crisis dummies for the Asian Financial Crisis (1997–1998), theGlobal Financial Crisis (2008–2009), and COVID-19 pandemic (2020–2021) are included to account for major global shocks.
B. Methodology
The empirical strategy proceeds in two sequential steps. In the first stage, the analysis tests whether the Trump tariff episode constitutes a significant shock to economic policy uncertainty. This is estimated using Newey-West standard errors to correct for heteroskedasticity and serial correlation:
\[GEPU_{t} = \alpha + \beta_{1}TTE_{t} + \beta_{2}X_{t} + \varepsilon_{t}\tag{1}\]
where denotes the global economic policy uncertainty index and is a dummy variable equal to one during the Trump tariff period (2018-2020) and zero otherwise, and includes global macroeconomic controls such as GDP, inflation, and major crisis dummies. This specification identifies whether U.S. tariffs functioned as exogenous shocks to global policy uncertainty.
In the second stage, the study examines the effect of global uncertainty on Asian economic growth by interacting with country-specific trade exposure to the U.S.
\[gdppc_{i,t} = \beta_{0} + \beta_{1}(GEPU_{t}*USExp_{i}) + \beta_{2}Z_{i,t} + \mu_{i} + \gamma_{t} + \varepsilon_{i,t}\tag{2}\]
where denotes the annual growth rate of GDP per capita in country at time measures the sum of exports to and imports from the U.S. relative to GDP, includes standard macroeconomic controls (inflation, trade openness, gross capital formation, FDI, and world Commodity price index), denotes country fixed effects, and captures time effects. Interaction with U.S. exposure introduces cross-sectional variation into the GEPU index, addressing concerns of inflated sample size and enabling identification of differential growth impacts across countries.
We estimate the model using fixed-effects panel regression with Driscoll-Kraay standard errors, which are robust to heteroskedasticity, serial correlation, and cross-sectional dependence across countries. This approach is widely applied in macroeconomic panel studies where cross-country spillovers and correlated shocks may bias standard errors (Driscoll & Kraay, 1998). We further conduct robustness checks, including alternative lag structures, lead specifications to test for reverse causality, and leave-one-out country analyses to ensure results are not driven by any single country.
III. Results
A. Trump tariffs and global economic policy uncertainty
Table 1 reports Newey-West estimates of the effect of Trump-era tariff escalations on GEPU. Results consistently show that tariffs were an independent and economically meaningful driver of GEPU. Across specifications, the coefficient on the tariff episode is positive, large, and statistically significant. In the baseline model (Column 1), a one-unit increase in tariffs is associated with a 130.68-point rise in GEPU. This magnitude remains stable when controlling for major global shocks, extending the lag structure, or applying robustness checks, ranging between 122 and 130, indicating that the results are not sensitive to dynamic specification or crisis controls.
Global GDP enters negatively and significantly (approximately -10 to -12), implying that stronger global economic performance dampens policy uncertainty. Inflation is statistically insignificant, suggesting that price movements do not independently explain GEPU once output and tariff shocks are accounted for. The GFC is insignificant, while the COVID-19 period positively and significantly elevated uncertainty, but the inclusion of these shocks does not reduce the tariff coefficient. The persistence of these results under alternative lags mitigates concerns about dynamic misspecification. These findings highlight the role of tariffs as an exogenous amplifier of GEPU.
B. Global economic policy uncertainty and economic growth in Asia
Table 2 reports country fixed-effects estimates of the impact of GEPU on growth, using country-level exposure to U.S. trade to introduce cross-sectional variation. Across all specifications, the interaction between GEPU and U.S. trade exposure remains negative and statistically significant, indicating that countries more exposed to the U.S. economy experience larger growth reductions when global policy uncertainty rises. In the baseline fixed-effects specification (Column 1), the coefficient suggests that a one-unit increase in GEPU exposure reduces growth by approximately 0.098 units. When crisis indicators are introduced (Column 3), the magnitude declines modestly to 0.078 but remains highly significant, implying that the estimated effect is not driven solely by major global downturns.
The preferred specification (Column 4), which includes crisis controls and employs Driscoll–Kraay standard errors to account for cross-sectional dependence, serial correlation, and heteroskedasticity, confirms the robustness of this finding. The stability of the GEPU coefficient across specifications strengthens the interpretation that GEPU affects domestic growth through exposure channels rather than merely reflecting global crisis episodes. The inclusion of crisis dummies increases the within R2 from 0.24 to 0.34, indicating that major global shocks account for a meaningful share of growth variation. However, the persistence of a statistically and economically significant GEPU exposure effect suggests that uncertainty operates beyond discrete crisis periods.
The control variables display expected signs. Inflation is negatively associated with GDP per capita, consistent with the distortionary effects of macroeconomic instability. Gross capital formation and trade openness exert positive effects on growth, highlighting the importance of real investment and integration. FDI appears positive in conventional FE estimates but loses significance under Driscoll-Kraay corrections, suggesting sensitivity to cross-sectional dependence. The crisis indicators (GFC, AFC, and COVID) enter with large and negative coefficients, confirming the severe contractionary effects of major global shocks. Their inclusion increases explanatory power substantially, yet the GEPU exposure coefficient remains stable, indicating that GEPU affects growth beyond discrete crisis episodes.
C. Robustness Checks
To further assess the stability of the results, we re-estimate the model after excluding China. The coefficient on GEPU exposure remains negative and statistically significant across all specifications. In the preferred Driscoll-Kraay fixed-effects model with crisis controls, the estimated effect is -0.074 (p < 0.01), which is economically and statistically comparable to the full-sample estimate. The crisis dummies (GFC, AFC, and COVID) also remain negative and significant. These findings confirm that China does not drive the baseline results and that they are robust to sample exclusion.
To assess potential reverse causality, we estimate a lead (placebo) specification including next-period GEPU (Column 5). The coefficient on future GEPU is negative but statistically insignificant, suggesting that anticipatory effects do not drive the baseline findings.
Additionally, Table 4 presents a leave-one-out robustness analysis, in which each country is sequentially excluded from the sample. The coefficient on global EPU exposure remains negative and statistically significant at the 1% level across all iterations, indicating that the results are not driven by any single country. Although excluding Vietnam increases the magnitude of the coefficient, the sign and statistical significance remain unchanged, confirming the stability of the findings.
IV. Conclusion
This study examines whether Trump-era tariff escalations heightened GEPU and whether such uncertainty constrained growth in major Asian economies. Using a two-step framework and conditioning GEPU on country-specific U.S. trade exposure, the results show that tariff shocks significantly increased GEPU and generated heterogeneous growth effects across Asia. Economies more intensively exposed to the United States experienced stronger contractionary effects, indicating that the transmission of external policy shocks depends critically on structural trade linkages rather than uniform regional dynamics.
The key contribution lies in advancing the uncertainty-growth literature by introducing exposure-based heterogeneity to sharpen the identification of global spillover effects. This approach moves beyond treating global uncertainty as uniformly exogenous and demonstrates how protectionist shocks propagate asymmetrically across integrated economies. The policy implications point to the need for strategic trade diversification, reduced overdependence on single partners, and stronger domestic policy credibility to mitigate vulnerability to future trade-policy disruptions. Future research may explore sectoral and firm-level adjustment mechanisms.