Loading [MathJax]/jax/output/SVG/jax.js
Badmus, S. K. (2026). A Global Analysis of Trade Tension, Energy Security and Sustainable Transition. Energy RESEARCH LETTERS, 7(Early View). https://doi.org/10.46557/001c.162652

Abstract

We present global evidence on the impact of trade tensions (TTs) on energy security and the transition to renewable energy for a panel of 36 countries. In the short run, trade tensions increase energy security and carbon emissions, while in the long run, they reduce energy security and carbon emissions. The effects of trade tensions on energy security are more pronounced for the U.S. trade tensions than for the U.S.-China trade tensions.

I. Introduction

This study investigates the effects of trade tensions on energy security and the transition to renewable energy across a panel of 36 countries. Rising energy prices, driven by shifting trade relationships and geopolitical risks, have heightened the significance of trade tensions for energy security (Zhao et al., 2025). The pursuit of energy security is closely tied to renewable energy transitions, as nations invest to mitigate climate shocks and enhance energy availability, affordability, and accessibility. Notably, the recent U.S.–China tariffs on solar panels and related components, such as batteries and transition metals, directly link trade tensions to renewable energy investments. At the same time, research indicates that trade disruptions can stimulate domestic development by encouraging innovation and solutions that may offset trade deficits in the long term (Freund, 2023; Stepanov et al., 2024). Therefore, we hypothesize that trade tensions yield long-term benefits for energy security and emissions reduction, while they may reduce energy security and emissions reduction in the short term.

Theoretical perspectives suggest that trade tensions can disrupt energy supply chains in the short run, undermining energy security and emission reduction goals. However, these disruptions can also drive innovation, diversification, and investment in greener energy systems, leading to improved energy security and emissions reduction over the long term (Dunford & Han, 2025; Romer, 1990). Using annual data from 1995 to 2024 for 36 countries, the study finds that U.S. trade policy uncertainty influences energy security and sustainability. For robustness, the analysis is also conducted using the U.S.–China trade tension index. Six distinct subsamples, namely Asia-Pacific (includes China, India, Japan, South Korea, Indonesia, Singapore, Australia, and New Zealand), America (includes United States, Canada, Brazil, Mexico, Argentina, and Chile), Europe (includes United Kingdom, France, Germany, Italy, Ukraine, Austria, Belgium, Portugal, Spain, Hungary, Switzerland, Finland, Denmark, Turkey, Norway, Poland, and Iceland), Africa (South Africa, Egypt, and Morocco), advanced economies (includes G7 countries), and emerging economies (non-G7 countries) are included in the analysis. Prior studies by Gao et al. (2024) and Zuo & Majeed (2024) have focused on trade uncertainty and renewable energy consumption in the U.S. and China. This study addresses a gap in the literature by offering a global perspective on the relationship between trade tensions, energy security, and sustainability. It utilizes diverse measures of trade tensions and accounts for heterogeneity across multiple subsamples.

The rest of the paper is organized as follows. Section II describes the methodology used in this study followed by main findings presented in Section III. Final section concludes the study.

II. Methodology

The study employs a sample of 36 countries that account for more than 85% of the world’s gross domestic product. Following the specified theoretical background, the econometric formulation of a bivariate model linking trade tensions to energy transition and sustainable development is defined as follows:

Sit=α+βKi,t1+μi+εit

i=1,2,3,,N; t =1,2,,T

Sit is defined as the percentage change in total energy supply per capita or combustion from CO2 emissions, which is specific to each country; Kit is the percentage change in either the U.S. trade policy uncertainty index or the U.S.-China trade index, which is general for all sampled nations; μi stands for country-specific effects, while εit is the stochastic error term.

We use the panel ARDL model proposed by Pesaran & Smith (1995) and Pesaran et al. (1999), with extensive procedural and empirical illustrations provided by Salisu & Ndako (2018). The model addresses the issue of dynamic heterogeneity that is often associated with panels with large N and large T. In line with the objectives of the study, Equation (1) is re-specified in a linear Panel ARDL (p,q) form as follows:

ΔSit=ϑi(Si,t1ψ0iψ1iKi,t1)+pj=1λijΔSi,tj+qj=0γijΔKi,tj+νit

i =1, 2,3,,N; t =1,2,,T

where Δ is the first difference operator; γij measures the short-run relationship between the two variables, while ϑi captures the long-run equilibrium in the model and represents the speed at which deviations of short-run dynamics adjust to the long-run equilibrium. The study refers to Salisu & Ndako (2018) for additional derivations on the mean group[1] (MG) and pooled mean group[2] (PMG) estimators.

Annual data on total energy supply per capita and CO2 emissions from combustion are obtained from The Energy Institute (https://www.energyinst.org/statistical-review/resources-and-data-downloads). Monthly data on the U.S. trade policy uncertainty and the U.S.-China Trade index are sourced from https://www.policyuncertainty.com/trade_cimpr.html, credited to Baker et al. (2016) and Rogers et al. (2024). Aggregation to annual data was performed by averaging all monthly observations.

Our sole focus on energy supply as an index of energy security is justified by its centrality to various perspectives of energy security, such as availability, affordability (i.e., prices), and stability (Kim et al., 2025). Thus, the energy security index adopted in this study captures the behavior of energy-vulnerable economies in the face of supply disruption. In addition, the intensity of CO2 emissions has been shown to significantly decrease as countries adopt more renewable and efficient energy systems (Lau et al., 2023; Pakrooh et al., 2025).

The current study acknowledges the limitations of the empirical inquiry in addressing other dimensions of energy security, such as import dependency and energy diversification. Additionally, transforming monthly indices into annual averages may obscure the effects of volatility in energy supply and trade tensions.

III. Results

The section begins with the presentation of results with a preliminary analysis of trade tensions, energy security, and sustainable transition. Table 1 shows that advanced (196.21), Asia-Pacific (177.13), and European (172.65) economies record a greater supply of energy per capita than other blocs and regions, highlighting higher levels of energy security. The Asia-Pacific (1447.67), advanced (1295.27), and American (1151.48 ± 133.91) economies had higher CO2 emissions from combustion than other regions (mean = 640.28 ± 1497.92).

Table 1.Descriptive statistics
Full Sample Asia America Europe Africa Advanced Emerging
Total Energy Supply
Mean 158.818 177.1333 146.3271 172.6468 47.37653 196.2056 149.7934
Max 903.532 649.1912 355.6465 903.532 102.9409 355.6465 903.532
Min 10.55273 10.55273 31.512 37.8222 13.80795 93.02262 10.55273
Std. Dev. 133.9146 154.7662 121.9204 129.4788 31.79054 80.49267 142.4402
No. obs. 1080 240 180 570 90 210 870
CO2 Emissions
Mean 640.2837 1447.666 1151.482 206.0242 215.1781 1295.271 482.1833
Max 11172.85 11172.85 5885.034 914.8445 475.9517 5885.034 11172.85
Min 1.813847 27.98577 42.28093 1.813847 27.14705 254.2081 1.813847
Std. Dev. 1497.922 2450.287 1880.498 203.1834 163.1448 1674.629 1408.127
No. obs. 1080 240 180 570 90 210 870
The U.S.-China Trade Tension Index
Mean 103.9192
Max 199.5386
Min 56.54682
Std. Dev. 36.16974
The U.S. Trade Policy Uncertainty
Mean 118.3419
Max 797.1224
Min 28.73999
Std. Dev. 156.2814

Note: This table reports descriptive statistics of all variables used in this study.

The main results of the empirical analysis conducted in this study are presented in Tables 2–3 and Tables A and B in the appendix. The following outcomes are deduced: (i) In the short run, the total energy supply per capita for the full sample, Asia-Pacific, America, and advanced economies was significantly and positively disrupted by the U.S. trade tensions (see Table 2). However, this positive disruption is accompanied by higher carbon emissions (see Table 3), indicating that impacted nations retreat from their renewable energy commitments in response to trade tensions to offset the effects. (ii) In the long run, the total energy supply per capita for the full sample, Asia-Pacific, America, and advanced economies is significantly and negatively impacted by the U.S. trade tensions (see Table 2), showing that sustained trade tensions adversely affect energy supply. This negative disruption is accompanied by lower carbon emissions (see Table 3), highlighting that impacted nations develop more energy-smart and efficient technologies over a prolonged exposure to trade tensions. As noted by Lau et al. (2023), such renewable energy infrastructures lead to a substitution effect between fossil and green energy, where countries reduce their per capita energy consumption or ecological footprint through green and energy-efficient technologies. However, renewable energy infrastructures also require significant initial capital investments, a highly skilled workforce, and novel technology (referred to as the technology base effort), which may disrupt long-run energy supply and security. (iii) Interestingly, compared to the U.S.-China trade tensions (see Tables provided in the appendix), the U.S. trade tensions (see Tables 2–3) have a more pronounced and significant impact on all economies and blocs, underscoring the notion that the U.S. trade tensions target a wider range of economies and blocs and have a greater effect than the latter, particularly regarding energy security and sustainability.

Table 2.Panel ARDL model highlighting the effects of the U.S. trade tension on energy security (TES)

Full Sample Asia America Europe
mg pmg mg pmg mg pmg Mg pmg
Ec -1.0533*** -1.0455 *** -0.9103 *** -0.8994 *** -0.9978 *** -0.9832 *** -1.1279 *** -1.1236 ***
(0.0026) (0.0439) (0.0994) (0.0991) (0.0499) (0.0412) (0.0612) (0.0605)
Short Run (TES) 0.0108 *** 0.0107 *** 0.0074 ** 0.0088 *** 0.0179 *** 0.0154 *** 0.0109 *** 0.0104 ***
(0.0015) (0.0007) (0.0032) (0.0018) (0.0036) (0.0010) (0.0016) (0.0008)
Long Run (TES) -0.0116 *** -0.0099 *** -0.0063 * -0.0095 *** -0.0212 *** -0.0167 *** -0.0114 *** -0.0083 ***
(0.0026) (0.0017) (0.0036) (0.0035) (0.0063) (0.0039) (0.0041) (0.0024)
Constant 0.3687 0.3666 1.5229 *** 1.5354 *** 0.8542 * 0.8046 * -0.4073 -0.4111
(0.2836) (0.2731) (0.5805) (0.5852) (0.4624) (0.4243) (0.3613) (0.3367)
Hausman Test 0.6 45.59 *** 0.85 0.82
No. of Countries 36 8 6 19
Obs. 1080 240 180 570
Africa Advanced Emerging
mg pmg mg pmg mg pmg
ec -1.0722 *** -1.0654 *** -1.1922 *** -1.1906 *** -1.0197 *** -1.0109 ***
(0.2241) (0.2203) (0.0673 (0.0584) (0.0514) (0.0507)
Short Run (TES) 0.0044 0.0045 *** 0.0150 *** 0.0152 *** 0.0097 *** 0.0093 ***
(0.0064) (0.0009) (0.0011) (0.0013) (0.0017) (0.0007)
Long Run (TES) -0.0077 -0.0046 -0.0121 *** -0.0119 *** -0.0114 *** -0.0089 ***
(0.0084) (0.0060) (0.0019) (0.0029) (0.0033) (0.0021)
Constant 1.2339 1.2247 -0.8753 *** -0.8677 *** 0.6689 ** 0.6569 **
(1.0166) (0.2069) (0.2046) (0.3256) (0.3109)
Hausman Test 0.29 -0.01 0.99
No. of Countries 3 7 29
Obs. 90 210 870

Notes: Values in parentheses are the standard errors. All the variables are expressed in percentage change or returns. ***, ** and * indicates statistical significance at 1%, 5%, & 10% levels, respectively. Also, ec denotes the error correction term. The acronym mg refers to mean group while pmg refers to pooled mean group. The lack of significance of the chi-square values of the Hausman test indicates that the pmg model is preferred.

Table 3.Panel ARDL model highlighting the effects of the U.S. trade tension on emissions (CEM)
Full Sample Asia America Europe
mg pmg mg pmg mg pmg mg pmg
ec -1.0480*** -1.0386 *** -0.8848 *** -0.0119 *** -1.0164 *** -0.9948 *** -1.1143 *** -1.1103 ***
(0.0397) (0.0386) (0.1180) (0.0045) (0.0596) (0.0498) (0.0422) (0.0403)
Short Run (CEM) 0.0152 *** 0.0153 *** 0.0081 ** 0.0097 *** 0.0235 *** 0.0209 *** 0.01698 ** 0.0171 ***
(0.0021) (0.0009) (0.0040) (0.0021) (0.0041) (0.0013) (0.0028) (0.0012)
Long Run (CEM) -0.0162 *** -0.0157 *** -0.0083 * -0.0119 *** -0.0298 *** -0.0252 *** -0.0162 *** -0.0155 ***
(0.0027) (0.0021) (0.0045) (0.0045) (0.0076) (0.0051) (0.0036) (0.0029)
Constant 0.6519 0.6463 * 2.2024 *** 2.2397 *** 1.7979 *** 1.7256 *** -0.7354 -0.7144 *
(0.4014) (0.3906) (0.7991) (0.8042) (0.5129) (0.4710) (0.4559) (0.4309)
Hausman test 0.07 1098.86 *** 0.66 0.12
No. of Countries 36 08 06 19
Obs. 1080 240 180 570
Africa Advanced Emerging
mg pmg mg pmg mg pmg
ec -1.1267 *** -1.1192 *** -1.1478 *** -1.1442 *** -1.0239 *** -1.0148 ***
(0.1649) (0.1609) (0.0330) (0.0319) (0.0477) (0.0463)
Short Run (CEM) 0.0068 0.0072 *** 0.0218 *** 0.0217 *** 0.0136 *** 0.0128 ***
(0.0089) (0.0019) (0.0017) (0.0005) (0.0025) (0.0011)
Long Run (CEM) -0.0099 -0.0082 -0.0209 *** -0.0207 *** -0.0151 *** -0.0131 ***
(0.0097) (0.0063) (0.0019) (0.0038) (0.0034) (0.0025)
Constant 3.0126 *** 3.0115 *** -0.6083 * -0.6190 * 0.9562 ** 0.9348 **
(1.1485) (1.1639) (0.3215) (0.3412) (0.4772) (0.4618)
Hausman test 0.06 0.00 0.71
No. of Countries 03 07 29
Obs. 90 210 870

Notes: Values in parentheses are the standard errors. All the variables are expressed in percentage change or returns. ***, ** and * indicates statistical significance at 1%, 5%, & 10% levels, respectively. Also, ec denotes the error correction term. The acronym mg refers to mean group while pmg refers to pooled mean group. The lack of significance of the chi-square values of the Hausman test indicates that the pmg model is preferred.

IV. Conclusion

This study contributes to the literature on trade tension, energy security, and renewable energy transition by analyzing the effects of the U.S. trade tensions and the U.S.-China trade tensions on energy security and sustainable transition. On a global scale, the results emphasize that these trade tensions are accompanied by higher short-run energy security and increased carbon emissions, indicating that in response to trade tensions, countries tend to retreat from their renewable energy commitments to offset the shocks. Conversely, in the long run, the opposite holds true, highlighting the broader trade-offs between short-run energy security and long-run sustainability. The study draws policymakers’ attention to the importance of energy technology, diversification, energy supply chain decoupling strategies, and the Carbon Border Adjustment Mechanism (CBAM), among other measures, in building resilience to trade tensions and ensuring long-term benefits. The highlighted limitations (refer to section II) may serve as motivation for future studies.

Accepted: October 20, 2025 AEST

References

Baker, S. R., Bloom, N., & Davis, S. J. (2016). Measuring economic policy uncertainty. Quarterly Journal of Economics, 131(4), 1593–1636. https:/​/​doi.org/​10.1093/​qje/​qjw024
Google Scholar
Dunford, M., & Han, M. (2025). Energy Dilemmas: Climate Change, Creative Destruction and Inclusive Carbon-Neutral Modernization Path Transitions. International Critical Thought, 15(1), 76–100. https:/​/​doi.org/​10.1080/​21598282.2025
Google Scholar
Freund, C. (2023). Advances in technology affect trade and vice versa. Finance and Development, 60(2), 15–17.
Google Scholar
Gao, Y., Hafeez, M., Kouki, F., Sher, F., & Akbar, M. W. (2024). Unraveling the impact of financial stress and trade policy uncertainty on advancing renewable energy transition in the USA. Energy and Environment. https:/​/​doi.org/​10.1177/​0958305X241262504
Google Scholar
Kim, J., Jaumotte, F., Panton, A. J., & Schwerhoff, G. (2025). Energy security and the green transition. Energy Policy, 198(2), 114409. https:/​/​doi.org/​10.1016/​j.enpol.2024.114409
Google Scholar
Lau, C. K., Gozgor, G., Mahalik, M. K., Patel, G., & Li, J. (2023). Introducing a new measure of energy transition: Green quality of energy mix and its impact on CO2 emissions. Energy Economics, 122, 106702. https:/​/​doi.org/​10.1016/​J.ENECO.2023.106702
Google Scholar
Pakrooh, P., Nammouri, H., & Ben Jabeur, S. (2025). Energy transition and CO₂ emissions: Fresh insight from low, middle and high-income countries. Research in International Business and Finance, 76, 102880. https:/​/​doi.org/​10.1016/​J.RIBAF.2025.102880
Google Scholar
Pesaran, M. H., Shin, Y., & Smith, R. P. (1999). Pooled Mean Group Estimation of Dynamic Heterogeneous Panels. Journal of the American Statistical Association, 94(446), 621. https:/​/​doi.org/​10.2307/​2670182
Google Scholar
Pesaran, M. H., & Smith, R. (1995). Estimating long-run relationships from dynamic heterogeneous panels. Journal of Econometrics, 68(1), 79–113. https:/​/​doi.org/​10.1016/​0304-4076(94)01644-F
Google Scholar
Rogers, J. H., Sun, B., & Sun, T. (2024). U.S.-China Tension. SSRN Electronic Journal. https:/​/​doi.org/​10.2139/​SSRN.4815838
Google Scholar
Romer, P. M. (1990). Endogenous technological change. Journal of Political Economy, 98(5), S71–S102. https:/​/​doi.org/​10.3386/​w3210
Google Scholar
Salisu, A. A., & Ndako, U. B. (2018). Modelling stock price–exchange rate nexus in OECD countries: A new perspective. Economic Modelling, 74, 105–123. https:/​/​doi.org/​10.1016/​J.ECONMOD.2018.05.010
Google Scholar
Stepanov, I., Teschner, N., Zemah-Shamir, S., & Parag, Y. (2024). How sustainable is the energy transition? Implications of trade on emissions and energy security. Renewable and Sustainable Energy Reviews, 206, 114844. https:/​/​doi.org/​10.1016/​J.RSER.2024.114844
Google Scholar
Zhao, X., Zhai, G., Nasim, A., & Ma, X. (2025). Geopolitical risks, trade openness, and energy security: Empirical evidence from 41 countries. Energy Economics, 149, 108777. https:/​/​doi.org/​10.1016/​J.ENECO.2025.108777
Google Scholar
Zuo, Q., & Majeed, M. T. (2024). Does trade policy uncertainty hurt renewable energy-related sustainable development goals in China? Heliyon, 10(15), e35215. https:/​/​doi.org/​10.1016/​J.HELIYON.2024.E35215/​ASSET/​F41E5E42-F5F6-4427-A943-D2194E412BBE/​MAIN
Google Scholar

Appendix

Table A.Panel ARDL model highlighting the effects of the U.S.-China trade tension on energy security (TES)
Full Sample Asia America Europe
mg pmg mg pmg mg pmg mg pmg
ec -1.0394*** -1.0384 *** -0.8852 *** -0.8877 *** -0.9742 *** -0.9704 *** -1.1199 *** -1.1161 ***
(0.0447) (0.0438) (0.0986) (0.0939) (0.0597) (0.0665) (0.0598) (0.0598)
Short Run (CEM) 0.0099 0.0136 *** 0.0573 * 0.0467 ** 0.0251 ** 0.0316 *** -0.0085 -0.0058
(0.0110) (0.0052) (0.0317) (0.0189) (0.0118) (0.0076) (0.0132) (0.0056)
Long Run (CEM) -0.0309 -0.0288 ** -0.1304 ** -0.0722 *** -0.0521 *** -0.0674 ** 0.0093 -0.0017
(0.0200) (0.0124) (0.0573) (0.0272) (0.0118) (0.0285) (0.0231) (0.0175)
Constant 0.3234 0.3563 1.7492 *** 1.7254 *** 0.7640 ** 0.7953 ** -0.5208 -0.5154
(0.2833) (0.2681) (0.6415) (0.5883) (0.3713) (0.3989) (0.3175) (0.3255)
Hausman test 0.02 1.34 -0.40 0.53
No. of Countries 36 08 06 19
Obs. 1080 240 180 570
Africa Advanced Emerging
mg pmg mg pmg pmg mg
ec -1.0717 *** -1.0659 *** -1.1726 *** -1.0717 *** -1.0659 *** -1.1726 ***
(0.2249) (0.2297) (0.0412) (0.2249) (0.2297) (0.0412)
Short Run (CEM) -0.0304 -0.0319 *** 0.0272 ** -0.0304 -0.0319 *** 0.0272 **
(0.0197) (0.0010) (0.0125) (0.0197) (0.0010) (0.0125)
Long Run (CEM) 0.0232 0.0356 -0.0472 *** 0.0232 0.0356 -0.0472 ***
(0.0411) (0.0429) (0.0113) (0.0411) (0.0429) (0.0113)
Constant 0.9871 0.9677 -0.7634 *** 0.9871 0.9677 -0.7634 ***
(1.0916) (0.9753) (0.2339) (1.0916) (0.9753) (0.2339)
Hausman test 7.34 *** -0.10 0.33 7.34 ***
No. of Countries 03 07 29 03
Obs. 90 210 870 90

Notes: Values in parentheses are the standard errors. All the variables are expressed in percentage change or returns. ***, ** and * indicates statistical significance at 1%, 5%, & 10% levels, respectively. Also, ec denotes the error correction term. The acronym mg refers to mean group while pmg refers to pooled mean group. The lack of significance of the chi-square values of the Hausman test indicates that the pmg model is preferred.

Table B.Panel ARDL model highlighting the effects of the U.S.-China trade tension on emissions (CEM)
Full Sample Asia America Europe
mg pmg mg pmg mg pmg mg pmg
ec -1.0279*** -1.0238 *** -0.8534 *** -0.8555 *** -0.9737 *** -0.9571 *** -1.1086 *** -1.1047 ***
(0.0415) (0.0402) (0.1119) (0.1088) (0.0671) (0.0778) (0.0459) (0.0432)
Short Run (CEM) 0.0323 ** 0.0287 *** 0.0710 * 0.0614 ** 0.0171 0.0315 *** 0.0323 * 0.0252 ***
(0.0143) (0.0073) (0.0410) (0.0262) (0.0208) (0.0113) (0.0186) (0.0068)
Long Run (CEM) -0.0726 *** -0.0486 *** -0.1295 * -0.0676 * -0.0421 -0.0777 ** -0.0720 * -0.0478 **
(0.0269) (0.0156) (0.0731) (0.0346) (0.0384) (0.0381) (0.0379) (0.0218)
Constant 0.6209 0.5888 2.3573 *** 2.3443 * 1.4257 *** 1.4816 *** -0.6655 -0.7349 *
(0.4037) (0.3819) (0.8679) (0.8043) (0.3293) (0.3777) (0.4875) (0.4228)
Hausman test 1.19 0.93 61.78 *** 0.61
No. of Countries 36 08 06 19
Obs. 1080 240 180 570
Africa Advanced Emerging
mg pmg mg pmg pmg Mg
ec -1.0806 *** -1.0695 *** -1.1300 *** -1.1316 *** -1.0022 *** -0.9997 ***
(0.1915) (0.2019) (0.0237) (0.0256) (0.0503) (0.0484)
Short Run (CEM) -0.0402 *** -0.0486 *** 0.0461 *** 0.0488 *** 0.0289 * 0.0212 **
(0.0149) (0.0144) (0.0153) (0.0129) (0.0174) (0.0084)
Long Run (CEM) 0.0145 0.0442 -0.0698 *** -0.0739 *** -0.0733 ** -0.0366 **
(0.0639) (0.0464) (0.0104) (0.0286) (0.0335) (0.0185)
Constant 2.5290 2.4279 ** -0.5582 0.9111 * 0.8477 *
(1.1719) (1.0697) (0.3490) (0.4804) (0.4522)
Hausman test 0.45 -0.02 1.74
No. of Countries 03 07 29
Obs. 90 210 870

Notes: Values in parentheses are the standard errors. All the variables are expressed in percentage change or returns. ***, ** and * indicates statistical significance at 1%, 5%, & 10% levels, respectively. Also, ec denotes the error correction term. The acronym mg refers to mean group while pmg refers to pooled mean group. The lack of significance of the chi-square values of the Hausman test indicates that the pmg model is preferred.


  1. It is assumed that all parameters (long-run and short-run) differ across countries or entities.

  2. It is assumed that long-run relationships are common across units, but the short-run dynamics and error-correction speeds differ. We find this assumption to be more fitting for the current inquiry, as long-term responses to trade tensions are expected to converge across countries. However, these assumptions are tested for each model using Hausman’s test to decide the best fit.