Interest rates, inflation, and exchange rates in fragile EMEs: A fresh look at the long-run interrelationships

ABSTRACT

This study attempts to establish the possible existence of the long-run interrelationship between interest rates, inflation, and exchange rates in five EMEs (Brazil, India, Indonesia, South Africa, and Turkey), what is so-called by Morgan Stanley ‘Fragile Five’. To do so, we utilize Li and Lee's [2010. “ADL Tests for Threshold Cointegration.” Journal of Time Series Analysis 31 (4): 241–254.] Autoregressive Distributed Lag test for threshold cointegration and apply it to the sample country's time-series data from 2013:m1 to 2018:m12. Overall, our results are threefold: First, there seems to be a long-run positive relationship between actual rates of inflation and nominal interest rates supporting the validity of the ex-post Fisher hypothesis for all the sample countries. Second, the results support the presence of a cointegrating relationship between interest rates and exchange rates for Brazil, India, and Turkey but not for Indonesia and South Africa. Lastly, without exception, exchange rates and actual rates of inflation in all the sample countries tend to co-move in the long-run, implying that the depreciation of their currencies creates an inflationary effect on domestic prices through raising the prices of imported goods. The results above are widely consistent with both theoretical expectations and the relevant empirical literature.

KEYWORDS:

• Macroeconomic policy
• emerging market economies
• fragile five
• autoregressive distributed lag
• threshold cointegration

JEL CLASSIFICATIONS:

• E31
• E43
• E52
• E58
• E60

ORCID

Hüseyin Şen http://orcid.org/0000-0002-9833-824X

Ayşe Kaya http://orcid.org/0000-0002-7025-1775

Savaş Kaptan http://orcid.org/0000-0003-1401-6620

Metehan Cömert http://orcid.org/0000-0003-3906-7272

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 The figure is based on the World Bank’s national accounts data, and OECD’s national accounts data files in 2017.

2 The year of adopting inflation targeting is 1999 for Brazil, 2016 for India, 2005 for Indonesia, 2000 for South Africa, and 2006 for Turkey, respectively.

3 These countries are: Belgium, Canada, Denmark, France, Germany, Greece, Ireland, Japan, the Netherlands, the UK, and the US.

4 In this context, it is worth noting that the literature contains some other studies that look at the topic from the different vantage points: Developed vs. emerging market and developing, inflationist vs. non-inflationist, short- vs. long run, and the like. However, it is noteworthy that such studies yielded inconclusive results. This may be attributed to a several factors: Model and sample specification, expectations, monetary regime shifts, the country-specific elements, and so forth. In short, the Granger-causal relationship between interest rates and inflation still remains a puzzle, especially for empirical researchers. So, it is obvious that further studies are needed to clarify the issue.

5 For a comprehensive survey on the Fisher hypothesis, see, e.g., Cooray (2002).

6 Indonesia, Korea, Malaysia, the Philippines, and Thailand.

7 Indonesia, South Korea, the Philippines, Thailand, Mexico, and Turkey.

8 Of course, there are also, some, but relatively fewer, counter-arguments claiming that inflation causes movements in exchange rates. In this regard, see, e.g., Clarida and Waldman (2008).

9 Working with monthly time series data, for instance, rather than annual data, can be justified on the ground that using annual data may lead to the aggregation-biased problem as argued by Rossana and Seater (1995). One plausible way to overcome this problem is to study with monthly data as suggested by Berument and Jelassi (2002). Following the suggestions in Berument and Jelassi (2002), in this study, we work with the available highest frequency data, monthly data, instead of lower frequency data, such as quarterly and annual data.

10 The reason for doing so is to remove base-year effect on prices.

11 NEER index helps us to understand the pass-through effect of an import-dependent country. The NEER index is also useful in analyzing the relationship between interest rates and exchange rates. It determines not only the trade competitiveness of a country, but also influences interest rates through monetary policy.