Exchange rate pass-through in the Caucasus and Central Asia

Abstract

This article estimates the extent and speed of exchange rate pass-through (ERPT) in seven Caucasus and Central Asia (CCA) countries using monthly data over the January 1995–May 2020 period. The estimations are performed using the local projections method. We find that the average pass-through in the CCA is about 10% on impact and about 25% after 12 months. There is no evidence of asymmetric ERPT with respect to the size and the sign of exchange rate changes. The pass-through is broadly unchanged in fixed versus floating exchange rate regimes. There has been a downward shift in the speed of ERPT in the aftermath of the global financial crisis as CCA countries have entered a relatively low inflation environment. The pass-through estimates could be used by the CCA monetary authorities for inflation projections. The absence of nonlinearities in the pass-through with respect to the exchange rate regime suggests that transition from fixed to floating exchange rate regimes in the region is not likely to impose additional inflationary costs.

Keywords:

• Exchange rate pass-through
• inflation
• exchange rate regime
• local projections

JEL Classification Numbers:

• E31
• F31

Acknowledgements

The author thanks Marwa Alnassa, Yevgeniya Korniyenko, Subir Lall, Nathan Porter, Sergejs Saksonovs, and Hamid Reza Tabarraei for helpful comments and suggestions. Liliya Nigmatullina provided excellent production assistance.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1 The sample does not include Turkmenistan due to lack of sufficiently long time series.

2 Goldberg and Knetter (1997) define ERPT as ‘the percentage change in local currency import prices resulting from a one percent change in the exchange rate between the exporting and importing countries’.

3 The results are not affected by the COVID shock, as restricting the sample to the pre-COVID period provides qualitatively similar ERPT estimates.

4 We chose the 15% threshold for the definition of large exchange rate changes since it corresponds to one standard deviation of the annual exchange rate growth in the total sample. We have also tried 10%, 20% and 30% thresholds and the results remain qualitatively unchanged.

5 It is important to note that the coefficient λ^h does not affect the ERPT. It measures the shift in the intercept for all CCA countries between the two exchange rate regimes (high and low depreciation), while the ERPT measures the response of inflation to a 1% depreciation in each of these regimes.

Additional information

Notes on contributors

Tigran Poghosyan

Tigran Poghosyan has been with the IMF from 2009. Prior to the IMF, he worked at the Central Bank of Armenia and was a visiting researcher at the Deutsche Bundesbank. His research has featured in IMF research and policy publications, books and peer-reviewed journals. He holds two PhDs in Economics; from CERGE-EI and the University of Groningen.