A new test for optimum currency area with an application to the Central and Eastern European countries

ABSTRACT

This research introduces a new test for optimum currency area that is based on synchronization of monetary policy recommendations. The main advantage over the more traditional synchronization of business cycles is that it takes into account two known determinants of monetary policy: inflation and the output gap. As an application, the test is applied to the EU economies with a particular focus on the Central and Eastern European Countries. Some of these countries have recently joined the euro area, some are the members of ERM II, and the rest are debating whether to fulfill the convergence criteria of adopting the euro. The main findings are that within the last few years the current non-euro area members (with the exception of Bulgaria and Romania) fit the euro area as well as the core euro area countries.

KEYWORDS:

• Monetary policy
• taylor rule
• optimum currency area
• eurozone

JEL CLASSIFICATION:

• E52
• E58
• F45

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

This study relies on monthly data in the computation of the desired interest rates from 1999 to 2017. Data on inflation rates is the annual percentage change in the CPI index taken from the IMF – International Financial Statistics (IFS) from https://data.imf.org/. Data on the real effective exchange rate (used as a robustness check) and industrial production is also taken from the IFS. Data on seasonally adjusted unemployment and the 3 month interbank interest rate (the Euribor) is taken from Eurostat from https://ec.europa.eu/eurostat/. The monthly output gap is computed from the deviation of industrial production from trend. However, industrial production is only a small portion of GDP (20 to 30%)¹² and varies very little within and between European countries. Perhaps a better measure of the output gap is the inverse of the unemployment gap.¹³ This measure varies much more within and between countries, and there is statistical evidence that it is half as variable and inversely related to the output gap (see Okun 1963 1963). Further, it provides a better fit in the estimation than the industrial production gap. The unemployment gap is computed as the cyclical component of an H-P filter¹⁴ from the seasonally adjusted monthly unemployment rates.

Notes

¹ Notice, if policy recommendations are aligned, this does not mean that the realized common policy is proper or right, but simply there is a policy that is suitable for all members. This is in the spirit of OCA tests which mention nothing about whether actual policies are suitable for certain members.

² The Taylor rule is an interest rate reaction function based on economic conditions that are known to influence monetary policy. In particular it takes into account both inflation and the output gap. The rule has a track record of mimicking actual monetary policies of major central banks (see Clarida, Gali, and Gertler 1998).

³ Although as EU members they are obliged to adopt the euro, fulfiling the Maastricht criteria (in particular joining ERM II) is optional.

⁴ See the Reuters articles: Krasimirov (2018), Ilic (2017), Marinas (Marinas 2018, March 10), and Koranyi (2020). As of 2020 only Bulgaria and Croatia are members of ERM II.

⁵ Asso et al. (2010) describes the Taylor rule’s role in practice.

⁶ The authors first estimate a relationship between exchange rate variability and its potential economic determinants. They then use predicted (imputed) values for each country’s exchange rate variability (relative to Germany) as a measure of an OCA index. If the imputed exchange rate variability is low then the country is said to be a more suitable member of a currency union with Germany as it does not need to adjust (or vary) its exchange rate by much. Perhaps the logic of not using actual exchange rate variability as an OCA index is that some countries’ exchange rate regimes with Germany might not correspond to their actual needs (or desires) that are based on economic conditions. Many for example had a fixed exchange rate with the Deutschmark.

⁷ CIA World Factbook – https://www.cia.gov/library/publications/the-world-factbook/fields/2012.html

⁸ For its use see Longbrake (Longbrake 2008, 46), McCulloh (McCulloch 2007), and Ball and Tchaidze (2002).

⁹ The smoothing parameter used is 129,600 for monthly data as recommended by Ravn and Uhlig (2002).

¹⁰ Krasimirov (2018), Marinas (Marinas 2018, March 10), and Koranyi (2020).

¹¹ From Okun’s Law the unemployment gap is the inverse and half as variable as the GDP gap therefore its coefficient is −1 instead of .5.

¹² CIA World Factbook – https://www.cia.gov/library/publications/the-world-factbook/fields/2012.html

¹³ For its use see Longbrake (Longbrake 2008, 46), McCulloh (McCulloch 2007), and Ball and Tchaidze (2002).

¹⁴ The smoothing parameter used is 129,600 for monthly data as recommended by Ravn and Uhlig (2002).

Additional information

Funding

This work was supported by the Nazarbayev University [Social Policy Grant].