Optimum currency area theory: evidence from post-Soviet countries and implications for Eurasian Economic Union

ABSTRACT

In this article the theory of optimum currency area is applied to post-Soviet and other selected countries. The study finds smaller exchange rate variability when the economies are closely linked by bilateral trade, are subject to similar shocks both on aggregate and at the industry level, have similar inflation rates, are open and smaller in economic size, and have higher labour migration as proxied by remittance flows. The estimation results also substantiate that the US dollar plays a dominant role as an anchor currency. Next, the study shows that economic fundamentals suggest limited prospects of a common currency for post-Soviet countries, particularly for the Eurasian Economic Union (EAEU). It is also found that Moldova, among the post-Soviet countries, better approximates an optimum currency area with Russia. Further, when the government debt-to-GDP ratio is taken into account, only Kazakhstan from the EAEU member countries stands out as having positive prospects for forming a common currency area with Russia.

KEYWORDS:

• Optimum currency area
• exchange rate variability
• exchange market pressure
• optimum currency area index

Acknowledgements

We thank the anonymous referee for valuable comments. We acknowledge financial support from the American University of Armenia (AUA) through faculty research grant 2016–2017. Gayane Barseghyan also acknowledges financial support from AUA through professional development grant 2018. The article was presented at the Second World Congress of Comparative Economics in St Petersburg; at the Armenian Economic Association 2016 annual meeting in Yerevan; at the seminars at the University of Mons and at the AUA. It reflects the views of the authors and not those of the institutions with which the authors are affiliated.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. We control for labour migration by proxying it with the average remittance flows between the countries.

2. For a more detailed literature review an interested reader is invited to refer, for example, to Mongelli (2002) and Tavlas (1993).

3. The lower exchange rate variability might reflect the fact that countries, to some extent, fulfil criteria of an OCA.

4. The Eurasian Economic Community here refers to Russia, Kazakhstan, Kyrgyzstan, Tajikistan, Belarus and Ukraine. For further integration initiatives in the Commonwealth of Independent States (CIS) during the 2000s, see Kan et al. (2011, Table 1).

5. Equity share is determined as the average contribution of GDP and population values to the total GDP and population in the common monetary area. The distribution methods are based on a variation of the exact method to obtain equity share of participating states.

6. This is widely used, although it is not an ideal proxy for intervention. However, Suardi and Chang (2012) find that the ‘co-movement between monthly reserves changes and intervention is governed by intervention amount’, and they conclude that ‘the conditional correlation is stronger when intervention frequency and amount increase’, which is likely to be quite relevant for our sample of countries.

7. More details on the variables calculations and data sources can be found in Appendix A.

8. The product groups are based on two-digit HS codes.

9. Likewise, two different currencies which are less variable against each other.

10. For the intervention index and pressure, the number of observations is 105, since no data were available for Iran, Turkmenistan and Uzbekistan.

11. Unfortunately, data are available for only a 5-year period, 2010–2014. However, this still turns out to be of statistical importance, along with being economically important to omit.

12. We perform the Hausman test for endogeneity, both for trade and output, but reject the exogeneity hypothesis for trade.

13. With a p-value = 0.0574, we marginally reject the null hypothesis of the coefficient of the instrumented variable being zero.

14. Here, formally, we refer to the explained variation of the respective OLS model as measured by its R-squared (Table B2, Appendix B).

Data on the asymmetry of output disturbances; trade linkages and the dissimilarity of commodity composition of export.

OLS estimation results. Dependent variable: exchange rate variability.

15. Although data are available only for a 5-year period, 2010–2014, the coefficient on the average remittance flows between pairs of countries still has the expected negative sign and is statistically significant in all specifications.

16. In Appendix B, we report the exchange rate variability OLS estimations both with a full (153) and restricted (105) sample (Tables B2 and B3, respectively). The results prove to be quite robust with respect to the omission of these three countries.

OLS estimation results for smaller sample. Dependent variable: exchange rate variability.

17. It is worth highlighting that the levels of variability are very high in comparison with those obtained for the euro area as given in Horvath (2007). This applies to both actual and predicted variability.

18. The correlation coefficient is equal to 0.23.

19. Belarus had adopted a crawling peg vis-à-vis the Russian rouble at the beginning of 2001 and, in June 2009, pegged its rouble to a currency basket with equal weights assigned to the US dollar, the euro and the Russian rouble.

20. From 1999 to November 2014, the Bank of Russia had a managed floating exchange rate regime, smoothing excess volatility and using the US dollar and euro basket as the operational indicators. Since 10 November 2014, it has abandoned the permissible range of the dual-currency basket, although foreign exchange interventions can still be conducted to maintain financial stability. For more details, one can refer to the ‘History of the Bank of Russia FX Policy’ from the official website of the Bank of Russia.

Additional information

Funding

This work was supported by the American University of Armenia [AUA professional development grant]; American University of Armenia [Faculty Research Grant 2016–2017].