Exchange rate pass-through in the Asian countries: does inflation volatility matter?

ABSTRACT

This paper presents a nonlinear relationship between exchange rate pass-through (ERPT) and inflation volatility. Through the lens of a threshold framework, we uncover a clear evidence of near to one ERPT to consumer prices once inflation volatility crosses a threshold level of 4.17. Clearly, there are significant differences in the degree of ERPT between the high and low inflation volatility in the inflation targeting (IT) and non-IT Asian countries.

KEYWORDS:

• Exchange rate pass-through
• inflation
• threshold
• inflation volatility

JEL CLASSIFICATION:

• C1
• E31
• F31

Acknowledgements

Financial support from Universiti Putra Malaysia [Grant number: GP-IPB/2014/9440900] is acknowledged.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Yanamandra (2015) find compelling evidence of complete ERPT of the Indian rupee both in the short- and long-run based on time series data. The author shows evidence of nonlinearity in term of whether the rupee is appreciating or depreciating (sign) as well as in terms of whether there are small or large (size) changes in the currency.

² Ben Cheikh and Rault (2016) consider economic activities as threshold variable in the exchange-rate–inflation nexus. They find a strong regime-dependence pass-through which is positively linked to economic activity. The ERPT is 0.15% when GDP is below a threshold value 3% while the ERPT elasticity increased threefold (around 0.47%) when the economy is above the estimated threshold value. Other threshold variables that have been used in the past include output gap, debts ratio, exchange rate change, exchange rate volatility.

³ Ben Cheikh and Louhichi (2016) choose inflation as threshold variable. They apply the Hansen’s (1999) grid search method to determine the optimum threshold values.

⁴ Basically, there are two popular nonlinear models usually applied in empirical analysis. The first is the Hansen TR where the transition across regimes occurred abruptly. The second is the smooth transition regression model where the transition between states is rather smooth (for application, Ben Cheikh and Rault 2016).

⁵ The likelihood ratio test of null hypothesis is based on the following F -statistics: F1 = (S₀ −S₁())/, where S₀ and S₁() are sum of squared errors under null and alternative hypotheses, respectively. The asymptotic distribution of F1 is nonstandard, and Hansen (1999) proposed to use a bootstrap procedure to compute the p-value for F1 under null.

⁶ Ben Cheikh and Louhichi (2016) define three inflation regimes: low inflation regime that includes counties with mean annual rates less that 2.76% (Japan), moderate inflation for those with inflation between 2.76% and 8.28% and high inflation regime, that is, mean inflation exceeding 8.28%.

Additional information

Funding

Financial support from Universiti Putra Malaysia [Grant number: GP-IPB/2014/9440900] is acknowledged.