Global Macroeconomic Repercussions of US Trade Restrictions: Evidence from a GVAR Model

ABSTRACT

We employ a global vector autoregression (GVAR) model to analyze international spillover effects of US trade restrictions, modeled as a reduction of US imports. Our sample consists of the US and 25 countries in the rest of the world, grouped into larger regions comprising European nations, non-European industrial countries and emerging economies. We find US trade restrictions to reduce trade volumes and income levels in the rest of the world as well as in the US. The trade balance deteriorates across all world regions except in the US, where it is unaffected by the trade restrictions. We also model the effects of a trade war in which the rest of the world responds in equal measure to the trade restrictions imposed by the US. We again find that export and import activity recedes both in the US and in the rest of the world, although the resulting effects are now strongest in the short run. The trade balance improves in the rest of the world but deteriorates in the US. In terms of the GDP response, the rest of the world is initially much harder hit by the imposition of the retaliatory trade measure than is the US.

KEYWORDS:

• Trade restrictions
• international spillover effects
• global vector autoregression
• trade war

JEL CLASSIFICATIONS:

• C32
• F14
• F42

Acknowledgments

The views and interpretations are those of the authors and do not represent those of the World Bank and its Executive Directors, nor the countries that they represent.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 There is a large and growing literature employing GVAR analysis to study international spillover effects. Chudik and Pesaran (2016) provide an overview of recent developments in GVAR modeling with respect to both the theoretical foundations and its numerous empirical applications.

2 Authors like Leybourne, Kim, and Newbold (2005) and Pantula, Gonzalez-Farias, and Fuller (1994) provide evidence of the superior performance of the WS test compared to the ADF test.

3 Stock and Watson (1996) provide a detailed description and analysis of the various structural stability tests.

4 To further check the robustness of the model around the financial crisis of 2007–2008, we conducted the same structural stability tests for a shorter sample of 1997Q2–2012Q4 (leaving out five years from both the beginning and the end of the sample). The results again indicate parameter stability at least at the 90% level, implying that the model is robust to the economic changes of that period. Details of the test results are available from the authors upon request.

5 Our results are robust to variations in the ordering of the trade volume and relative prices variables.

6 Depending on the specific set of trade policy measures implemented by the US government, the actual size of the contractionary impulse may be a multiple of the standardized shock analyzed in this paper, with the responses of all variables scaled up by the respective multiple. For example, Yalcin, Felbermayr, and Steininger (2017) simulate the expected impact of the protectionist measures contemplated by the Trump administration. Assuming the US to increase customs duties on imports vis-a-vis all other WTO member countries by 20% relative to the prevailing tariff levels results in a drop of US imports of between 20% and 30%. This figure rises to between 50% and 60% in a scenario in which the US is assumed to raise both tariffs and non-tariff barriers by 20% across the board.

Additional information

Notes on contributors

Bernd Kempa

Bernd Kempa: PhD, University of Toronto (1995), Post-Doc, University of Duisburg-Essen (1995–2006), Professor, Department of Economics, Viadrina University Frankfurt (2006–2008), Professor, Institute of International Economics, University of Muenster (since 2008).

Nazmus Sadat Khan

Nazmus Sadat Khan: PhD, University of Muenster (2016), Post-Doc, University of Muenster (2016–2018), Economist, World Bank Dhaka Office (since 2018).