Currency and commodity return relationship under extreme geopolitical risks: evidence from the invasion of Ukraine

ABSTRACT

We examine the relationship between currency and commodity returns around the invasion of Ukraine in February 2022. We find that the expected positive contemporaneous relationship between currency and commodity returns reverses and becomes negative during this period of extreme geopolitical risks. In addition to commodity returns, currency returns around the invasion of Ukraine are significantly affected by geopolitical factors, particularly geographic distance to the war. Our results indicate that a war between two major commodity-exporting countries significantly affects global currency pricing.

KEYWORDS:

• Foreign exchange rates
• currency return
• commodity returns
• Ukraine–Russia war
• geographic distance

JEL CLASSIFICATION:

• F31
• F51
• G13
• G14

Acknowledgments

The authors would like to thank the Editor (David Peel) and an anonymous referee for their constructive comments on a previous version of the paper.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ We do not include currencies of Albania, Georgia, Moldova, Madagascar, Mozambique, Seychelles, Slovak Republic, Somalia, Uganda, Uruguay, and Zambia because there is no available data for control variables for these countries.

² On 20 January 2022, the Atlantic Council issued a report confirming that Russia had deployed critical combat capabilities to the border with Ukraine (https://www.atlanticcouncil.org/blogs/new-atlanticist/will-russia-make-a-military-move-against-ukraine-follow-these-clues/) and the US President reportedly predicted that Russia would move into Ukraine (“Biden Predicts Putin Will Order Ukraine Invasion, but ‘Will Regret Having Done It’”. The New York Times. 20 January 2022. Washington D.C. ISSN 0362–4331).

³ Data source: https://www.nato.int/cps/en/natohq/nato_countries.htm.

⁴ Data source: https://tass.com/politics/1418197.

⁵ Available from https://cran.r-project.org/web/packages/cepiigeodist/cepiigeodist.pdf.

⁶ https://data.imf.org/?sk=9d6028d4-f14a-464c-a2f2-59b2cd424b85&sId=1409151240976.

⁷ The News Sentiment Index of the Federal Reserve Bank of San Francisco is available at https://www.frbsf.org/economic-research/indicators-data/daily-news-sentiment-index/.

⁸ Overnight interbank interest rate is the most common interest rate with the shortest maturity across our sample countries. When overnight interbank interest rates are not available, we use the available interest rate with the shortest maturity.

⁹ Other potential determinants of exchange rates are changes in inflation rates, current account deficits, public debt, and terms of trade (Chen, Rogoff, and Rossi 2010). Given we are studying the short-term effect of the Russian invasion of Ukraine on the exchange rates and the above variables are available at a monthly frequency, at best, we do not include those controls. However, in the regression analysis, we control for time-invariant heterogeneity across currencies (by including currency fixed effects) and common exchange rate trends (by including month-year fixed effects), which could alleviate this issue.

¹⁰ Although Figure 1 shows a negative co-movement between commodity prices and exchange rates around the invasion, we cannot draw conclusions because co-movement between two nonstationary series, such as the S&P GSCI and the Currency indices, does not indicate causality.

¹¹ The online Appendix S1 shows that the main findings are not affected when we exclude Kazakhstan as a potential outlier (as shown in Figure 2), when we include the USD index returns as a control variable, when we control for the daily geopolitical risk index of Caldara and Iacoviello (2022) or control for local stock market return in US dollars.

¹² In an additional analysis, we distinguish between commodity currencies (Chen and Rogoff 2003; Norland 2020) and non-commodity currencies (discussed in Section S2 of the online Appendix). The main result is observed in both subsamples, but the effect is stronger for commodity currencies.

Additional information

Funding

This paper is based on work supported by the Spanish Ministry of Economy and Competitiveness [grant PID2019-105986-GB-C21]