How do world commodity prices affect Asian commodity exporting economies? The role of financial frictions

ABSTRACT

This note examines how recent volatile fluctuations of world commodity prices due to the COVID-19 shock affect Asian commodity exporting economies. Our analysis shows that a drop in world commodity prices has a significant negative impact on output, consumption, and investment of sample countries. However, the impact on trade balance is positive in some countries and negative in others. The difference between these two groups is attributable to whether world commodity prices affect each country’s interest rate spreads.

KEYWORDS:

• Commodity prices
• Asian emerging economies
• financial frictions
• SVAR
• COVID-19 shock

JEL CLASSIFICATION:

• E32
• E37
• E44
• F32
• F36
• F41
• F44
• F62
• O16
• Q43

Acknowledgments

I am grateful to an anonymous referee for the encouraging comments. I am also grateful to Shugo Yamamoto, Toshiki Jinushi, Masahiko Shibamoto, and Kenya Takaku for their helpful comments.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

¹ For example, Shousha (2016) considers Argentina, Brazil, Chile, Colombia, Peru and South Africa as emerging economies and Australia, Canada, New Zealand and Norway as advanced economies. Drechsel and Tenreyro (2018) use data on Argentina. The sample counties included in Bastourre et al. (2012) are Argentina, Brazil, Bulgaria, Colombia, Ecuador, Malaysia, Mexico, Panama, Peru, Poland, Russia, Turkey, and Venezuela. Fernández, González, and Rodriguez (2018) uses data on Brazil, Chile, Colombia, and Peru for SVAR analysis.

² Specifically, we choose the five countries included in Table 3 on page 19 in ADB (2016).

³ The raw data are available at http://www.worldbank.org/commodities. Taking averages across the months in a given quarter, we convert monthly commodity indices into quarterly indices.

⁴ The source for the US CPI is the Federal Reserve Bank of St. Louis FRED Database: https://fred.stlouisfed.org.

⁵ Although, following World Bank (2016), we use the US CPI in the multi-country analysis due to data availability constraints, the use of import prices or any other cost-related prices might be interesting to consider wealth and substitution effects in a specific country analysis.

⁶ The WITS data base is available at https://wits.worldbank.org/.

⁷ See, for example, Sims, Stock, and Watson (1990), Hamilton (1994, 652), and Kilian and Lütkepohl (2017, 373–7).

⁸ As for GDP deflator, quarterly data is not available for Azerbaijan, Kazakhstan, and Mongolia. For these countries, we convert annual data to quarterly data.

⁹ The SVAR model contains two lags, and is chosen taking into account several lag length selection criteria. Specifically, we take AIC, SC, and HQ criteria into consideration. Since in the case of Azerbaijan the three criteria show that appropriate lag length is one, we checked how the impulse response with one lag differs from that with two lags, and confirmed that our main implications do not change.

¹⁰ The raw quarterly series are seasonally adjusted. As for the GDP deflator, quarterly data are not available for Azerbaijan, Kazakhstan, and Mongolia. For these countries, we convert annual data to quarterly data.

¹¹ The raw quarterly series are seasonally adjusted.

¹² Output, exports, and imports are obtained from IFS. The raw series are seasonally adjusted and transformed to real terms using the GDP deflator.

¹³ The raw annual series are obtained from IMF Global Debt Database. We convert the raw annual data to quarterly data.

¹⁴ Among many, for example, see Végh (2013) and Schmitt-Grohé and Uribe (2017).

¹⁵ ”Output” is omitted due to low performance.

¹⁶ The data source is IMF Global Debt Database. We convert the raw annual data to quarterly data.

Additional information

Funding

This work is supported by JSPS KAKENHI [16K03741, 20K01744].