Is and was the EMU an Optimum
Currency Area?
Minke van Bergen 1764330
1Supervised by prof. dr. J.M. Berk
2November 2012
Abstract
An Optimum Currency Area (OCA) is a geographical area in which it is optimal to have a single, shared, currency. This paper analyzes whether the European Monetary Union (EMU)
can be classified as an OCA. It is therefore tested whether the EMU complies with the nine criteria as set by the Optimum Currency Area theory, both at its origination in 1999 and today
(2012). The nine criteria are tested separately, by using the concept of Principle Component Analysis among others, and the results show that not all criteria are met.
Key words: Optimum Currency Area, European Monetary Union, Economic and Monetary
Integration
JEL classification: E42, F15, F33 and F41.
1
Student at the University of Groningen, Faculty of Economics and Business, the Netherlands. Correspondence to minke.van.bergen@gmail.com.
2
2
1. Introduction
Already twenty years ago, in 1992, the Maastricht Treaty decided on the creation of a European Economic and Monetary Union (EMU), which would lead to the creation of a common currency among several European nations. In 1995 the euro was given its official name and in 1999 the EMU officially came in place, with the euro becoming an accounting currency first. On 1 January 2002 the euro became the only legal means of payment in Austria, Belgium, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal and Spain, thereby completely abolishing the individual national currencies. The economic benefits enjoyed by these countries attracted other EU members and the EMU today also consists of: Slovenia (since 2007), Cyprus (since 2008), Malta (since 2008), Slovak Republic (since 2009) and Estonia (since 2011).
1.1 The Maastricht Treaty
The Maastricht Treaty was signed in 1992 and came into force in 1993. The Treaty created the European Union (EU) and led to the creation of the euro. In the Maastricht Treaty, five main convergence criteria were laid down to which a country should comply before they were allowed into the third stage of the EMU and could adopt the euro (De Grauwe, 2009):
1. Price stability: a country’s inflation rate may not be more than 1.5% above the average of the three lowest inflation rates of the EU member-countries.
2. Sound public finances: a country’s government budget deficit may not be higher than 3% of its GDP.
3. Sustainable public finances: a country’s government debt may not be higher than 60% of its GDP.
4. Durability of convergence: a country’s long term interest rate may not exceed the average rate observed in the three member-countries with the lowest inflation rates of the EU (see the price stability criteria) by more than 2%.
3
These five criteria were inspired by the idea that potential monetary union members should have some degree of homogeneity before initiating the union, as otherwise it will be terribly difficult to make a single currency work for the entire union, and meant to ensure (macro)economic convergence of the member countries before forming a monetary union (European Commission, 2011).
When countries form a Monetary Union, they introduce a single currency and monetary policy for the whole union. Member-countries give up their individual monetary policy instrument and can no longer control the interest- and exchange rates of their currency. As member-countries do not have the possibility to adjust to shocks through monetary policy, problems could arise when an individual country faces an economic downturn while the rest of the union does not. When such an asymmetric shock hits, the country hit by the shock can no longer decrease the interest rate to boost the economy as other member-countries (not experiencing the shock) may have other interests (stable money growth, low inflation). Therefore, the country hit by the shock has to use other instruments to boost the economy. These potential problems associated with a monetary union, became particularly relevant in the financial crisis that started in 2007 and the subsequent European sovereign debt crisis.
1.2 The financial crisis
After the U.S. housing market bubble burst at the end of 2006, a subprime mortgage crisis emerged. The values of securities tied to U.S. real estate pricing plummeted, damaging financial institutions globally. Uncertainties about the solvency of banks, declines in the availability of credit and the damaged trust of investors had a great impingement on the global stock markets: securities made large losses in the period 2008-2009 and economies slowed down worldwide. Because of all this, international trade declined and credit tightened (IMF, 2009).
4
they could not depreciate their currencies to soften the shock (question remains whether that would have been enough).
When in 2009 the European economy slowly recovered, Greece initiated a new crisis. In January 2010 the many doubts regarding the Greek government finances led to a heavy downturn of the Greek public loan rates and the riskiness of government loans became the center of attention of financial markets. Many (southern) European countries were experiencing large budget deficits and even greater government debts. The Greek situation escaladed rapidly; Greek public loan rates kept declining rapidly, furthermore Greece’s credit rating was downgraded to ‘junk’ by Standard and Poor’s, which had never happened to a euro member-country before (Bloomberg, 2010). The euro weakened and stock markets throughout the euro-area plunged (as banks were now very reluctant to take on any risks, they got rid of European public loans).
After various bailout packages, IMF supports and write-downs, we are still in the middle of a debt-crisis. Not only Greece is in trouble, also Spain, Ireland and Portugal. In January 2012 Standard and Poor’s downgraded the credit ratings of Austria, France, Malta, Slovenia, Slovakia, Cyprus, Italy, Portugal and Spain (Standard and Poor’s, 2012). Moreover, this year even the economically larger countries, like the Netherlands will not be able to meet with the strict EMU standards as they were laid down in the Maastricht Treaty. This raises the question whether the European Monetary Union was built on the right foundation as the euro now seems to ‘pass on the disease’. Maybe the Maastricht criteria are not sufficient to guarantee an optimal working of the EMU; maybe extra criteria are needed, or maybe the criteria used to evaluate potential entrants should in fact be revised.
5
ensures more efficient monetary policies executed by the European Central Bank (ECB), as it ensures that the (monetary) interests of the individual countries are more aligned.
This thesis evaluates whether the EMU complies with the criteria defined by the OCA theory. The objective of this master thesis is to investigate whether the European Monetary Union (EMU) is an optimal currency area at the moment (2012) and if it was an optimal currency area at the origination in 1999.
The remainder of this paper is organized as follows. In section 2, the optimum currency theory is studied as well as the research of other scientists in this field and their relevance for this paper is considered. Section 3 derives the model(s) that will be used, while the data are discussed in section 4. Furthermore, in section 5 the results of this paper are presented and, last but not least, section 6 gives a conclusion.
2. Literature Review
A monetary union brings both benefits and costs to the member countries. The benefits include less exchange rate uncertainty, reduced transaction costs, price transparency, increased efficiency, exchange rate stability and increased policy credibility (De Grauwe, 2009). The costs include not being able to change the currency’s price (revaluate or devaluate), to change the interest rates, or to control the national money supply (De Grauwe, 2009). The theory of optimum currency areas tries to provide conditions member countries need to comply with before the formation of a monetary union in order to decrease these costs associated with a monetary union.
2.1 The OCA theory
6
area to another, this creates problems in both regions: the first region faces declining demand and unemployment, while the other region faces excess demand and inflationary pressures. With factor mobility, labor could move from the unemployment region to the excess demand region, thereby stabilizing the economy. In a later article, McKinnon (1963) expands Mundell’s theory by introducing the concepts of size and openness. McKinnon argues that Mundell overlooked the issue of factor immobility among industries (in light of the previous example of an asymmetric demand shift, the first region could have decided to start producing the high demand products of the second region thereby solving the asymmetric demand problem, but when this is not possible factor mobility is the solution), which is considered to be especially relevant in case of ‘small areas trying to develop industries in which economies of scale or indivisibilities are very great instead of efficiently moving factors elsewhere’ (1963, p. 724). In addition, McKinnon points to the fact that open economies tend to prefer fixed exchange rates, as this reduces the costs associated with frequent exchange rate movements. Changes in the exchange rate in open economies do not significantly affect real competitiveness and frequent exchange rate adjustments make the overall price index vary more in open economies than in relatively closed economies. Moreover, the smaller the economy, the more open it is likely to be (as for small economies it is probably not efficient to produce everything they need, instead it is more beneficial for them to engage in foreign trade) and thus the size of an economy also matters. Kenen (1969) also provides an expansion to the optimum currency area theory by raising the importance of the degree of commodity diversification. According to Kenen, economies with high product diversification are better candidates for currency unions than are those with less diversification, since the diversification provides protection against shocks and frequent changes in the exchange rate are therefore unnecessary. To illustrate this point, imagine a country participating in a currency union, which is not diversified and produces only one commodity that it also exports. When the demand for this commodity decreases (negative demand shock), the exports of the country will decline. As the exchange rate cannot depreciate, the adjustment must be done through a reduction in wages, an increase in prices or an increase in unemployment. While when a country is diversified, a shock hitting one sector can be accommodated by the other sectors.
7
extensions. Mongelli (2002) defines four phases in the development of the OCA theory: the pioneering phase, the costs-benefits phase, the reassessment phase and the empirical phase.
According to Mongelli, the pioneering phase started in late 1950s-early 1960s and included, among others, the articles by Mundell (1961), McKinnon (1963) and Kenen (1969) mentioned in the previous subsection. The costs-benefits phase of the early 1970s consisted of a reconciling of the OCA properties and the creation of a framework analyzing the costs and benefits of introducing a shared single currency. The works of Corden (1972), Ishiyama (1975) and Tower and Willet (1976) are prime examples of this phase.
The reassessment phase of OCA took place in the 1980s and early 1990s. The OCA theory became particularly relevant in this period, as certain European countries started to integrate their economic and monetary policies. However, when this process took place the OCA theory of Mundell (1961), McKinnon (1963) and Kenen (1969) was outdated and a reassessment of the theory was made. During this reassessment phase there was a reassessment of the costs and benefits of introducing a shared single currency and of some of the OCA properties, which led to the emergence of a ‘new’ theory of optimum currency areas as proposed by Tavlas (1993). According to Tavlas (1993, p. 669): ‘the theory of optimum currency areas has been modified, (…), in line with developments in expectations formation, the time inconsistency and credibility problems, labour mobility under conditions of uncertainty, and exchange rate determination’. In the ‘new’ theory there are fewer costs to monetary integration, as the loss in the autonomy of domestic policies is smaller, and some more benefits of which a more credible inflation policy is an example. The famous ‘One Market, One Money’ report of the European Union (EU), also originated in this phase
(European Commission, 1990). This report found that the EMU would reduce both inflation and output variability (which is desirable as it reduces uncertainty and stabilizes the economy); asymmetric exchange rate shocks would be eliminated and the wage and pricing behavior in the union would be subject to greater discipline, thereby offsetting any effects the suboptimal monetary policy could have on the individual member countries.
8
survey, the nine factors that are tested for when investigating the optimality of the EMU as currency area are:
1. Price and wage flexibility
With nominal prices and wages that are flexible within and among the countries of a currency union, the transition path after a shock will less likely be associated with unemployment and/or inflation and therefore exchange rate changes are less needed (Friedman, 1953).
2. Labor market integration
Labor market integration facilitates labor mobility, which in turn is needed to adjust to asymmetric shocks. Labor mobility can reduce the need to alter real wages, and subsequently the nominal exchange rate, between union members (Mundell, 1961).
3. Factor market integration
Just as labor market integration facilitates labor mobility, factor market integration facilitates factor mobility. Factor mobility can reduce the need to alter the real prices of the factors of production, and subsequently the nominal exchange rate, between union members after an asymmetric shock hits (Mundell, 1961). Since the two main factors of production are capital and labor, this criterion finds some overlap in both the previous (labor market integration) and the upcoming (financial market integration) criterion.
4. Financial market integration
The financial market can be subdivided in the capital market (also included in the factor market) and the money market. The difference between these two markets lies in the maturities of the securities and claims provided: the capital market provides long term finance, while the money market provides short term finance. Financial market integration facilitates the mobility of financial resources, which in turn can ease the financing of external imbalances, and reduce the need for changes in the exchange rate, in the aftermath of a shock (Ingram, 1962).
5. The degree of economic openness
9
that changes in the international prices of tradables are transmitted faster to the domestic cost of living (McKinnon, 1963).
6. The diversification in production and consumption
The importance of this criteria is already mentioned before, when discussing the article by Kenen (1969). Mongelli (2002, p. 3) states: ‘More diversified partner countries are more likely to endure small costs from forsaking nominal exchange rate changes amongst them, and to find a single currency more beneficial’.
7. Similarity of inflation rates
When inflation rates of certain countries are similar over time, the terms of trade (value of a country’s exports divided by the value of its imports) between these countries will remain reasonably stable, which in turn will bring balance to the current account transactions and trade, thereby reducing the need for changes in the exchange rate (Fleming, 1971).
8. Fiscal integration
With fiscal integration, countries can more easily transfer funds to member countries that are affected by an asymmetric shock. This way, it is easier to adjust to the shock and there will be less need for exchange rate changes (Kenen, 1969).
9. Political integration
Political integration makes it easier to achieve the eight criteria mentioned above. Moreover, it brings the member countries’ preferences in line with each other, through which a currency area becomes simpler (and cheaper) to maintain.
2.2 Two Schools
10
economic integration, the development of a country will be boosted in such a way that it is more likely to satisfy the OCA criteria than before the formation of a monetary union. As this second school line of thought might be relevant for the interpretation of the empirics of this thesis, it will be elaborated a little further.
The second line of literature states that a country is more likely to satisfy entry-criteria ex post than ex ante. The leading article in this area is the paper by Frankel and Rose (1996). This paper investigates the ‘relationship between the degree of bilateral trade intensity and the cross-country bilateral correlation of business cycle activity’ (p. 21) and finds that countries with a high intensity of trade with other potential currency union members tend to have business cycles that are more tightly correlated with those of the other potential members. This result is also supported by work of Artis and Zhang (1997), who find that before the European Exchange Rate Mechanism (ERM) came into existence, the business cycles of most countries were linked to the US, whereas after the origination of the ERM the countries move into the German business cycle. In addition, Corsetti and Pesenti (2002) find that even if the monetary union does not promote economic integration and intra-industry trade, there can be endogeneity in the OCA theory. They show that when the private sector chooses optimal prices in a monetary union, these prices make a currency area the optimal monetary policy for national policy makers as well, thereby making national outputs more correlated. Moreover, Fidrmuc (2008) emphasizes the importance of intra-industry trade for the synchronization of business cycles. In his paper he shows empirically that intra-industry trade induces converging business cycles for OECD countries and also for the countries in the EMU.
In the ‘One Market, One Money’ report of the European Commission (1990), the Commission states that in the EMU most shocks in demand will be symmetrical. The reason for this is that trade between the EU countries is to a large extent intra-industry trade and is characterized by the existence of economies of scale, differentiated products and imperfect competition. This creates a trade structure in which the member countries buy and sell the same product groups from and to each other, and creates more synchronized business cycles between the EMU member states. The European Commission thus supports the view that closer integration leads to more synchronized business cycles and less frequent asymmetric shocks.
11
lower correlation between countries’ business cycles, a point supported by Eichengreen (1992). According to Krugman this makes regions more vulnerable to regional shocks, a conclusion which he bases on the experiences of the US. However, the empirics are not conclusive regarding this view as Warin, Wunnava and Janicki (2009) actually find empirical evidence of intra-industry linkages in the EU-15. Growing horizontal integration, ‘based predominantly on market access and consumer income’ (p. 82), allows for the equalization of shocks in demand and shocks in trade and deepens the market integration. According to Warin, Wunnava and Janicki: ‘Europe is indeed becoming an optimum currency area in terms of allocation of capital’ (p.82).
However, the endogeneity issue can be interpreted in a broader sense. De Grauwe and Mongelli (2005) investigate four types of endogeneity: the endogeneity of product and labor market flexibility, the endogeneity of economic integration, the endogeneity of financial integration and the endogeneity of symmetry of shocks. By looking at the empirical literature available, they try to identify the strength of these four endogeneities. They conclude that the different endogeneities are at work in real life, but their strength is debatable. Issing (2001), in his article, refers to an endogeneity of political integration, as the existence of the EMU itself will affect the political structure drawn up at the origination of the EMU. Moreover, Blanchard and Wolfers (2000) introduce the endogeneity of labor market institutions in their paper explaining the unemployment in Europe by the interaction of shocks and institutions.
There is much more literature to be found on this subject of the OCA theory in combination with the European Monetary Union, but it seems to be of little interest these last years. Exactly 10 years after the euro became publicly used, it is interesting to investigate whether the European Commission was right and the EMU has ‘grown’ into an optimum currency area. And if not, it may be interesting to find out why that is and how the EMU should proceed.
3. The Model
12
criterion and which criterion gives the most trouble. A problem that could arise when using this approach is that it can be difficult to attach weights to the different criteria. When for example one or two criteria are not met, while all the other criteria are met, it can become difficult to draw conclusions.
3.1 Price and wage flexibility
Price flexibility will be measured by the elasticity of prices with respect to demand, following Moore (1922):
(1)
with P representing nominal price and Q representing demand.
Whereas wage flexibility will be measured, following Kertesi and Köllô (2000) and Klau and Mittelstadt (1986), by using the elasticity of wages with respect to unemployment:
(2)
where W represents nominal wage and U represents unemployment.
The elasticities can be interpreted in the following manner (Parkin, Powell and Matthews, 2002):
Elasticity ( ) value: Interpretation:
Perfectly inelastic
Inelastic
Unit elastic
Elastic
Perfectly elastic
13
3.2 Labor market integration
Following Bayoumi and Eichengreen (1996, p. 24) labor market integration will be measured by using ‘the explanatory power of the first principle component’ to ‘analyze trends in cohesion’. This will be done, following Silvia (2004), by looking at the nominal hourly labor costs and unit labor costs.
First, several correlation matrices will be made reflecting the mutual or reciprocal relationships between nominal hourly labor costs in the different EMU countries over different time periods. Thereafter, a principle component analysis (PCA) will be executed to reveal whether there exists multilateral conversion between the nominal hourly labor costs in the different EMU countries. A principle component analysis is a multivariate data analysis that basically reduces the number of underlying variables by counting correlated variables as one, thereby identifying the most important common factors explaining the coherence in a correlation matrix. These common factors are thus the principle components and the most important component is named the first principle component. The explanatory power of this first principle component indicates the extent to which the underlying variables are integrated. For this criterion this means that the first principle component indicates the extent of labor market integration between the countries under investigation, and increasing explanatory power of the first principle component over time indicates increasing convergence in the nominal hourly labor costs of the different countries.
To check for robustness the same procedure will be followed with unit labor costs (ULC). Unit labor costs represent the average cost of labor per unit of output and present an excellent robustness check as any labor costs differentials attributed to different levels of labor productivity are filtered out in this measure.
3.3 Factor market integration
14
as foreign direct investment (FDI) flows (to measure capital mobility). This will be done in the same manner as labor market integration, thus by making a correlation matrix for the long-term interest rates and FDI values and then performing a principle component analysis.
3.4 Financial market integration
Following Abiad, Leigh and Mody (2009), this criterion will be measured by using equation (3) below, as measured by Lane and Milesi-Ferretti (2006).
(3)
Where IFI stands for international financial integration, FA represents the gross stock of foreign assets, FL is the gross stock of foreign liabilities, is country and is time in years. Similar to the labor market integration criterion, financial integration will be evaluated by making a correlation matrix for the IFI and then performing a principle component analysis.
As a robustness check, the Chinn-Ito index will be shown in a table and a graph, as measured by Chinn and Ito (2008). This index measures a country's degree of capital account openness by creating an index ‘aimed at measuring the extensity of capital controls based on the information from the IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER)’ (Chinn and Ito, 2008, abstract), where a higher index value indicates greater financial openness. According to Kawai and Lamberte (2010, p. 245) ‘Countries with a score of 2.6 have complete unquestioned capital account convertibility’. When the index-values for the 14 countries under investigation are equal, financial integration is verified.
3.5 The degree of economic openness
Following Reuveny and Li (2003), this criterion will be evaluated by using equation (4).
(4)
15
Openness value: Interpretation:
Perfectly open
(Relatively) Open (Relatively) Closed
Perfectly closed
So to satisfy this criterion it must be that: .
3.6 The diversification in production and consumption
Following Jovanovic and Gilbert (1993) and Chandra, Boccardo and Osorio (2007), the Herfindahl Index (HI) will be used as the measure of production and consumption diversification. The HI lies between 1/N and 1, with 1/N indicating full diversification (with N being the total number of export/import categories, 1/N indicates that each industry has the same export/import value)and 1 indicating full specialization (as all exports/imports are from the same industry):
(5)
where is the share of total exports (for measuring production diversification) or imports (when measuring consumption diversification) attributed to the industry.
3.7 Similarity of inflation rates
This criterion will be evaluated by comparing the inflation rates of tradables of the different EMU members over the years by the same means as labor market integration and factor market integration: a correlation matrix and principle component analysis.
3.8 Fiscal integration
16
(6)
(7)
where GB indicates the factual government budget and PGB indicates primary government budget.
Next to the factual government budget, the primary government budget is used as a robustness check. As the primary government budget measures the factual government budget not including the interest payments over the government debt, the use of the primary government budget allows an investigation of the fiscal behavior of countries that is not influenced by the size of the government debt.
A fiscally integrated area consists of countries having similar values of MS. To test whether this is the case, a similar approach will be used as in section 3.7.
3.9 Political integration
This criterion will not be tested in a formal manner. To evaluate whether the EMU complies with this criterion, the concept of political union will be introduced and the original and current situations in the EMU will be highlighted.
4. Data
17
number of years, in this paper 5 years, and ‘rolling’ it over time. That is; the first data point will be determined by the average over the years 1990-1994, the second by the average over the years 1991-1995 etc. and the last data point will be determined by the average over the years 2007-2011. A cumulative moving average on the other hand, does not use a fixed time frame. The cumulative moving average is calculated by expanding the time frame with every year. That is; the first data point will be determined by the average over the years 1990-1994, the second by the average over the years 1990-1995, etc. and the last data point will be determined by the average over the whole dataset (1990-2011). This time-varying manner is chosen as it highlights the longer term trends in the data. The reason for using both averaging concepts side by side, is the matter of robustness. The year 1990 is chosen as the starting year as this was the year in which the first draft of the ECB Statute was prepared. Moreover, in this year the ‘One Market, One Money’ report was completed, the first stage of EMU started and an intergovernmental conference was launched to prepare for stages two and three of the EMU. Summarizing, this year was the official start of the European monetary and economic integration.
The fact that the investigation will be done using data over different countries over different years indicates a panel data set. The introduction of panel data in this research allows the use of both cross-sectional and time series information to test the workings of the EMU. In particular, it leads to a large number of observations, which increases the degrees of freedom and reduces the collinearity among the explanatory variables. The introduction of panel data thus improves the efficiency of the tests that will be done.
In table I a summary is presented of the different datasets used and the sources of these datasets.
4.1 Price and wage flexibility
To measure price flexibility, both the percentage change in real demand and the percentage change in nominal price level are needed. To measure the demand level, the following formula is used:
18
where Y stands for total demand, C stands for total consumer spending, I stands for total capital investment, G stands for total government spending, X stands for exports, M stands for imports and i and t represent country and year respectively. Total consumer and government spending is measured by ‘Final consumption expenditure (constant LCU)’, total capital investment is measured by gross capital formation (in constant local currency units), exports and imports are measured by ‘Exports of goods and services (constant LCU)’and ‘Imports of goods and services (constant LCU)’. All datasets are from the World Bank (2012) and are available for all 14 countries under investigation and cover the years 1960-2011.
The percentage change in the demand level of a certain country is calculated as:
(9)
The percentage change in the nominal price level is measured by ‘Inflation, GDP deflator (annual %)’ from the World Bank (2012). The World Bank defines this dataset as ‘Inflation as measured by the annual growth rate of the GDP implicit deflator shows the rate of price change in the economy as a whole. The GDP implicit deflator is the ratio of GDP in current local currency to GDP in constant local currency’ (World Bank, 2012). This dataset is available for all 14 countries and covers the years 1961-2011.
To measure wage flexibility, both the percentage change in wages and the percentage change in the unemployment are necessary. For the wages, the ‘Gross wages and salaries (current euros)’ dataset is acquired from Eurostat (European Commission, 2012). According to Eurostat, ‘Gross wages and salaries, just as compensation of employees, include both remuneration in cash and in kind, but unlike the latter, they do not include employers´ social contributions’. This dataset is available for all 14 countries, covers the years 1990-2011 and is measured in millions of Euros from 1-1-1999/millions of European Currency Unit (ECU) up to 31-12-1998.
19
The percentage change in the wage of a certain country is calculated as:
(10)
And the percentage change in the unemployment level of a certain country is calculated as:
(11)
4.2 Labor market integration
For this criterion, data on nominal hourly labor costs and unit labor costs are needed.
The OECD provides an annual dataset of the level of labor compensation per hour, which will be used as a proxy for the nominal hourly labor costs (OECD, 2012). According to the OECD it is an excellent proxy for the nominal hourly labor costs, with the difference that ‘the costs of recruitment, employee training, and plant facilities and services, such as cafeterias, medical clinics and welfare services, are not included’ (OECD, 2012a). This dataset is available for 10 countries, covers the years 1990-2011 and is measured in Euros from 1999 onward. Data prior to 1999 is measured by converting the national currencies into Euros using the fixed exchange rate between the national currency and the Euro at its adoption. The fact that this dataset is not available for all 14 countries might be a problem; the countries that are excluded from this dataset are Belgium, Luxembourg, Portugal and the United Kingdom. However, data for the largest part of the countries is available and moreover the unit labor costs dataset, which is more reliable (for an explanation of this statement see section 3.2), is available for all 14 countries. This dataset is also provided by the OECD (OECD, 2012) and provides complete annual data for the years 1990-2011. This data is also measured in Euros from 1999 onwards and data prior to 1999 is converted by using the fixed exchange rate between the national currency and the Euro at its adoption.
4.3 Factor market integration
1990-20
2011 and are measured in Euros from 1999 onwards, whereas data prior to 1999 is converted by using the fixed exchange rate between the national currency and the Euro at its adoption.
4.4 Financial market integration
To measure this criterion, data on the gross stock of foreign assets and liabilities of each country are needed together with the GDP levels and the Chinn-Ito index of each country. The gross stocks of foreign assets and liabilities are derived from the ‘International investment position - annual data’ dataset from Eurostat (European Commission, 2012). As the international investment position (IIP) is calculated by subtracting the gross stock of foreign liabilities from the gross stock of foreign assets, this dataset provides data on the gross stock of foreign assets and liabilities of each country. For the annual GDP levels, Eurostat provides a dataset named ‘Gross domestic product at market prices’ (European Commission, 2012).Both datasets are available for 13 countries (with the exception of France), cover the years 1990-2011(with some exceptions) and are in millions of Euros from 1-1-1999/millions of ECU up to 31-12-1998.
The Chinn-Ito index dataset is retrieved from Chinn and Ito (2008). This dataset is also available for 13 countries, as in this dataset the relatively small country Luxembourg is excluded, and covers the years 1970-2010.
4.5 The degree of economic openness
21
For export values, a dataset named ‘Exports of goods and services (current LCU)’ is used. The World Bank defines exports of goods and services as representing ‘the value of all goods and other market services provided to the rest of the world. They include the value of merchandise, freight, insurance, transport, travel, royalties, license fees, and other services, such as communication, construction, financial, information, business, personal, and government services. They exclude compensation of employees and investment income (formerly called factor services) and transfer payments’ (World Bank, 2012). This dataset is available for all 14 countries and covers the years 1960-2011.
The import values are measured by the dataset ‘Imports of goods and services (current LCU)’. Similar to the export dataset, this set is defined by the World Bank (2012) as: ‘the value of all goods and other market services received from the rest of the world. They include the value of merchandise, freight, insurance, transport, travel, royalties, license fees, and other services, such as communication, construction, financial, information, business, personal, and government services. They exclude compensation of employees and investment income (formerly called factor services) and transfer payments’. This dataset is available for all 14 countries and covers the years 1960-2011.
Lastly, the GDP (current LCU) dataset is measured as ‘the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. It is calculated without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources’. This dataset is available for all 14 countries and covers the years 1960-2011.
4.6 The diversification in production and consumption
For this criterion, data is needed concerning the annual import and export values of countries subdivided into industries.
22
products, n.e.s.; Other manufactured goods; Machinery and transport equipment; Commodities and transactions not classified elsewhere in the SITC). The dataset is available for all 14 countries, covers the years 1990-2011 and is in millions of Euros from 1-1-1999/millions of ECU up to 31-12-1998.
4.7 Similarity of inflation rates
The inflation rates used to evaluate this criterion are the same as the ones used to evaluate criterion 1: ‘Inflation, GDP deflator (annual %)’ from the World Bank (2012). The reason for using this deflator as inflation measure instead of using one based on consumer prices is that this deflator measures the inflation rates of tradables, whereas inflation measures based on consumer prices also measures services and other non-tradable products. The World Bank defines the GDP deflator dataset as ‘Inflation as measured by the annual growth rate of the GDP implicit deflator shows the rate of price change in the economy as a whole. The GDP implicit deflator is the ratio of GDP in current local currency to GDP in constant local currency’ (World Bank, 2012). This dataset is available for all 14 countries and covers the years 1961-2011.
4.8 Fiscal integration
To measure the degree of fiscal integration, datasets on the primary government budget, the factual government budget and on GDP levels are needed.
For the primary and factual government budgets, the ‘Government revenue, expenditure and main aggregates’ dataset is retrieved from Eurostat (European Commission, 2012). With this dataset, the primary government budget can be calculated as:
(12) whereas the factual government budget can be calculated as:
23
The dataset provides data for all 14 countries under investigation over the years 1990-2011 and is measured in millions of Euros from 1-1-1999/millions of ECU up to 31-12-1998. The GDP data is also retrieved from Eurostat (European Commission, 2012) as ‘Gross domestic product at current market prices’. This GDP dataset is chosen as it is, similar to the government revenue and expenditure dataset, measured at current market prices.
The percentage change in the primary government budget (PGB) of a certain country is calculated as:
(14)
And the percentage change in the GDP level of a certain country is calculated as:
(15)
5. Results
In this section each criterion will be carefully tested and the outcomes of these tests will be evaluated. Moreover, it will be indicated whether the European Monetary Union fulfills each of the nine criteria.
5.1 Price and wage flexibility
Price flexibility is measured by the elasticity of prices with respect to demand. In tables II and III the simple and cumulative moving averages of these elasticities can be seen respectively. As indicates flexibility, it becomes clear that this criterion is not satisfied today nor was it satisfied in 1999 as there are more elasticities below the value of 1 than there are above. Moreover, both Germany and Sweden have only simple moving average elasticities of below 1, while the cumulative moving average elasticities show series below the value of 1 for Belgium, Finland, France, Germany and Sweden.
24
than there are below, indicating wage flexibility. Moreover, in both tables no country can be found with an elasticity series of below 1. In addition, these tables show an improvement in wage flexibility over the years.
5.2 Labor market integration
The results of this criterion are shown by tables VI-IX and figures I-II. Tables VI-IX report the percentage of the overall variance of the hourly and unit labor costs that can be explained by the first principle component, whereas figures I-II show these values in a graph. Tables VI and VII show the simple and cumulative moving average results of the principle component analysis of the hourly labor costs respectively, whereas tables VIII and IX show the simple and cumulative moving average results of the PCA of the unit labor costs. It can be seen that the principle component analysis is executed not only for the whole group of countries under investigation, but also for the EMU ‘core’ (Austria, Belgium, Finland, France, Germany, Ireland, Italy, Luxembourg, the Netherlands, Portugal and Spain) and the control group (United Kingdom, Denmark and Sweden) separately, so as to identify any divergence between these two groups of countries.
The hourly labor costs tables (VI and VII) show lower values for the last period of the table than for the first period, indicating that hourly labor costs convergence has declined. However both tables show values of above 90 for almost all periods, which indicates that the hourly labor costs convergence has always been, and remained, pretty high and the decline in convergence can thus be considered negligibly small. When identifying the effect of the official start of the EMU (1999), it can be seen that including the year 1999 in the cumulative average results in an increasing trend in convergence for the EMU ‘core’ countries and a decreasing trend in convergence for the countries in the control group.
25
It can be concluded that this criterion is satisfied, as the high PCA outcomes indicate high convergence, both at the start of the EMU as today.
5.3 Factor market integration
The results of this criterion can be seen in tables X-XV and figures III-V. Tables X-XV show the percentage of the overall variance of the long-term interest rates, FDI inflows and FDI outflows that can be explained by the first principle component, whereas figures III-V visualize these results in several graphs. Tables X and XI show the simple and cumulative moving average results of the principle component analysis of the long-term interest rates respectively, whereas tables XII and XIII show the simple and cumulative moving average results of the PCA of the inward FDI flows and tables XIV and XV show the simple and cumulative moving average results of the analysis of the outward FDI flows. Similar to the previous criterion, the principle component analysis for this criterion is executed not only for the whole group of countries under investigation, but also for the EMU ‘core’ (Austria, Belgium, Finland, France, Germany, Ireland, Italy, Luxembourg, the Netherlands, Portugal and Spain) and the control group (United Kingdom, Denmark and Sweden) separately, so as to identify any divergence between these two groups of countries.
Tables X and XI show PCA values exceeding 90 percent for almost all periods, with no significant pattern indicating the effect of the EMU introduction, indicating a high convergence in long-term interest rates. When comparing the control group with the EMU ‘core’, it can be seen that the control group has much higher long-term interest rate convergence than the EMU ‘core’ in most of the periods.
26
react this strongly to the starting of the EMU, as these values reach a (much lower) peak in the periods following this year, thereafter also trending downwards.
The FDI outflow tables (XIV and XV) show roughly the same patterns as the inflow tables described above, with low overall convergence levels and higher convergence for the control group than for the EMU ‘core’. However, the year 1999 does not bring a peak for any of the country groups. If anything, it brings a dip. After the year 1999, the PCA values of all country groups increase for one period, thereafter trending downwards.
It can be concluded that the long-term interest rates are integrated, both at the initiation of the EMU in 1999 as today. The inward FDI flows increased to values around 70 percent when 1999 was included in the moving averages, indicating integration at the initiation of the EMU. However, when looking at the PCA values of the last period, it can be seen that today only the countries in the control group are integrated with values well above 50 percent, whereas the EMU ‘core’ is not integrated. The outward FDI flows tell a slightly different story, with values rapidly decreasing when the year 1999 is included in the moving averages, and low convergence for the EMU ‘core’ combined with high convergence for the control group today. Overall, it can thus be concluded that the long-term interest rates are integrated at both times (1999 and today), whereas the inward FDI flows were integrated in 1999 but are not at the moment, and the outward FDI flows have never been integrated. Therefore, this criterion is not satisfied.
5.4 Financial market integration
27
and Spain) and the control group (United Kingdom, Denmark and Sweden) separately, so as to identify any divergence between these two groups of countries.
The IFI tables and figure (tables XVI and XVII and figure VI) show PCA values exceeding 90 percent for almost all periods, with the highest values surrounding the year 1999. However, it can also be seen that today’s values are much lower for the EMU ‘core’, although upward trending. As these values are still above 60 percent and upward trending, the IFI variable indicates financial market integration both at the origination of the EMU as today.
As a robustness check, the Chinn-Ito index is shown in tables XVIII-XIX and figure VII. These show that all countries except Belgium were financially integrated from 1999 onwards, while Belgium was moving upward in 1999 and fully integrated in 2003. This variable thus also shows that the EMU was integrated both in 1999 and today.
Overall, it can thus be said that this criterion is fulfilled. The European Monetary Union was financially integrated in 1999 and is even more integrated today.
5.5 The degree of economic openness
28
5.6 The diversification in production and consumption
As mentioned in section 3.6, an HI of 1/N indicates full diversification (with N being the total number of export/import categories, 1/N indicates that each industry has the same export/import value) and 1 indicates full specialization (as all exports/imports are from the same industry). As the dataset used in this thesis consists of seven industries, 1/7=0.1429 indicates full diversification here. In tables XXII-XXV the time-varying results of formula (5) are shown, with tables XXII and XXIII showing the results for product diversification and tables XXIV and XXV showing the consumption diversification results. All four tables show virtually the same results, with most values between 0.2 and 0.35, indicating that there is no full specialization nor is there full diversification. However, since the values are much closer to full diversification (0.1429) than to full specialization (1), it can be concluded that the EMU countries have a high degree of production and consumption diversification. The same can be concluded for the control group Denmark, Sweden and United Kingdom. Moreover, this conclusion holds for both the EMU at its origination and today.
5.7 Similarity of inflation rates
The results of this criterion are shown by tables XXVII and figure VIII. Tables XXVI-XXVII report the percentage of the overall variance of the GDP deflator that can be explained by the first principle component, whereas figure VI shows these values in a graph. Table XXVI shows the simple moving average results of the principle component analysis of the GDP deflator, whereas table XXVII shows the cumulative moving average PCA results. The principle component analysis is executed not only for the whole group of countries under investigation, but also for the EMU ‘core’ (Austria, Belgium, Finland, France, Germany, Ireland, Italy, Luxembourg, the Netherlands, Portugal and Spain) and the control group (United Kingdom, Denmark and Sweden) separately, so as to identify any divergence between these two groups of countries.
29
for the EMU ‘core’, indicating more convergence in the control group. After the year 1999 is included in the cumulative moving averages, all values decrease, indicating less convergence. For the simple moving average the story is slightly different, as no real trend can be distinguished here. Overall, it can be seen that this criterion is not satisfied in 1999 (the simple averages show that inclusion of the year 1999 leads to a decrease in the PCA value for the EMU ‘core’, while the period before the PCA values were already pretty low) and it is satisfied today (as the simple moving average show that over the period 2007-2011 the PCA values were just below 60 percent and even higher in previous periods). However, the downward trend faced with today and the fact that the PCA values are not really high, indicates that there still is much work to do for this criteria.
5.8 Fiscal integration
The results of this criterion are shown by tables XXVIII-XXXI and figures IX-X. The tables report the percentage of the overall variance of the factual and primary budget sensitivities, explained by the first principle component. Tables XXVIII and XXIX show the simple and cumulative moving average results of equation (6) respectively, whereas tables XXX and XXXI show the simple and cumulative moving average results of equation (7). It can be seen that the PCA procedure is executed not only for the whole group of countries under investigation, but also for the EMU ‘core’ (Austria, Belgium, Finland, France, Germany, Ireland, Italy, Luxembourg, the Netherlands, Portugal and Spain) and the control group (United Kingdom, Denmark and Sweden) separately.
The simple moving average tables (XXVIII and XXX) show similar patterns: values that are fluctuating over time, with both the whole group of countries and the EMU ‘core’ having higher values today (2007-2011) than they had in the begin period (1990-1994) and the core group having higher values at the begin period than today. The cumulative moving average tables (XXIX and XXXI) show for all three groups lower values today than in the begin period.
30
increased convergence between the core EMU countries and decreased convergence between the control group countries. However, the cumulative moving average series show a different pattern, as both the EMU core group and the control group have overall decreasing series.
Remarkable is that all series decrease when the year 2002 is included in the average. However no other effect of the introduction of the euro can be detected. The official start of the EMU (1999) does not seem to have any real impact, as no striking trends can be discovered surrounding this year.
Overall, it can be seen that the PCA values for the EMU ‘core’ were trending downwards in and surrounding the year 1999, both for the factual government budget sensitivities as for the primary government budget sensitivities. As these values were not high to begin with, this downward trending leads to values below the 50 percent, indicating that this criterion was not fulfilled in/surrounding the year 1999. Today, the PCA values of the EMU ‘core’ are above 60 percent and moving upward, therefore it can be concluded that this criterion is satisfied, however still work remains to be done.
5.9 Political integration
‘A monetary union can only function if there is a collective mechanism of mutual support and control. Such a collective mechanism exists in a political union.’ (De Grauwe, 2011, p. 19).
Perfect political integration takes place when countries form a political union. Simply stated, a political union is a group of nations or states that share one central government and that thus have given up national sovereignty. In practice, a political union can take many different gradations and dimensions.
31
2009). The European Central Bank has supranational monetary authority, while at the same time there is much predominantly national sovereignty in other policy areas. Member States have given up sovereignty over policy areas such as monetary, exchange rate, competition and trade. Moreover, when looking at areas such as agriculture and external trade policy, the EMU has also developed some degree of political integration.
In other areas, such as defense and foreign policies, it is obvious that the EMU has some way to go as they are mostly domestically regulated. And it has an even longer way to go in such areas as social security, wage and taxation policies, to name just a few, as they are completely domestically regulated. Moreover, the current European debt crisis is a good example of how the lack of political integration (in this example taxation policy) can have severe consequences for the workings of the EMU as a whole.
Overall, the EMU has taken many steps towards political integration, and is still taking steps. However the current European debt crisis is a good example of the insufficiency of these steps (as an example of the lack of taxation policy integration). The EMU is not yet politically integrated, nor was it politically integrated at its origination. More steps are needed to achieve a situation where national policies are replaced by supranational policies, thereby creating a stable and unified EMU that is equipped to handle all situations possible.
5.10 Overall
32
the criteria that were not met in 1999, were met in 2012. These results are in accordance with the school of thought that promoted the endogeneity of the OCA criteria and believed that countries automatically become more integrated when becoming part of a currency union. Table XXXII below gives an overview of the criteria that were tested and indicates whether the criteria are met by the EMU ‘core’ states at the establishment of the EMU (column labeled ‘1999’) and whether the criteria are met today (column labeled ‘2012’). ‘Yes’ indicates that the criteria is met in that particular year, whereas ‘No’ indicates that the criteria is not met in that year.
Table XXXII: Overview of the results*
*’No’ and ‘Yes’ indicate whether the criteria is met by the EMU core in that particular year
1999 2012
Price Flexibility No No
Wage Flexibility Yes Yes
Labor Market Integration Yes Yes
Factor Market Integration No No
Financial Market Integration Yes Yes
Economic Openness Yes Yes
Diversification of Production Yes Yes
Diversification of Consumption Yes Yes
Similarities of Inflation Rates No Yes
Fiscal Integration No Yes
Political Integration No No
6. Conclusion
33
The relevant literature was studied and it became clear that many papers were dedicated to the subject, but no paper did what this thesis aimed at doing. Many researchers devoted their time to the OCA theory, its implications and its empirical consequences. Also the relationship between the OCA theory and the EMU was highlighted by some researchers, however not all criteria were used, nor were they empirically tested.
The nine criteria that were tested in this paper were: price and wage flexibility, labor market integration, factor market integration, financial market integration, the degree of economic openness, the diversification in production and consumption, similarity of inflation rates, fiscal integration, and political integration. Five of these criteria (labor market integration, factor market integration, financial market integration, similarity of inflation rates and fiscal integration) were tested by applying the concept of Principle Component Analysis (PCA) to several defining variables, three criteria (price and wage flexibility, the degree of economic openness and the diversification in production and consumption) were tested by using universally accepted formulas, whereas the last criterion (political integration) was evaluated theoretically.
The results show that at the establishment of the European Monetary Union, the member states did not comply with all the criteria as they were brought forward by the Optimum Currency Area theory, nor do they comply with them today. On the other hand, much progress has been made in the area of integration between member states and several criteria that were not met in 1999 are met today. These results are in accordance with the school of thought that promoted the endogeneity of the OCA criteria and believed that countries automatically become more integrated when becoming part of a currency union, as explained in paragraph 2.2. It is not to say that the integration process is finished when the criteria are fulfilled. Further steps towards complete integration are needed for the future EMU to be better equipped when a new shock hits. Moreover, good monitoring of this integration process is necessary and more research should be devoted to this subject.
34
35
References
Abiad, A., D. Leigh and A. Mody (2009). Financial integration, capital mobility, and income
convergence. Economic Policy 24(58), pp. 241–305.
Artis, M.J. and W. Zhang (1997). International Business Cycles and the ERM: Is There a
European Business Cycle? International Journal of Finance and Economics 2(1), pp.
1-16.
Bayoumi, T. and B. Eichengreen (1996). Operationalizing the theory of optimum currency
areas. CEPR Discussion Paper 1484.
Blanchard, O. and J. Wolfers (2000). The role of shocks and institutions in the rise of
European unemployment: The aggregate evidence. The Economic Journal 110, pp.
1-33.
Bloomberg (2010). Greece's Debt Cut to Junk, First for Euro Member.
http://www.bloomberg.com/news/2010-04-27/greek-debt-cut-to-junk-at-s-p.html. Chandra, V., J. Boccardo and I. Osorio (2007). Export diversification and Competitiveness in
Developing Countries. World Bank, Washington.
Chinn, M. D. and H. Ito (2008). A New Measure of Financial Openness. Journal of Comparative Policy Analysis 10(3), pp. 309 – 322.
Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences. Lawrence Erlbaum Associates, New Jersey.
Corden, W.M. (1972). Monetary Integration, Essays in International Finance. International Finance Section 93, Princeton University, Princeton.
Corsetti, G. and P. Pesenti (2002). Self-Validating Optimum Currency Areas. NBER Working Paper 8783.
Csajbók, A. and A. Csermely (2002). Adopting the euro in Hungary: Expected costs, benefits
and timing. NBH Occasional Paper 24.
36
De Grauwe, P. and F.P. Mongelli (2005). Endogeneities of Optimum Currency Areas: What
Brings Countries Sharing a Single Currency Closer Together? ECB Working Paper
468.
Eichengreen, B. (1992). Should the Maastricht Treaty Be Saved? Princeton Studies in International Finance 74, pp. 1-74.
European Commission (1990). One market, one money. An evaluation of the potential benefits
and costs of forming an economic and monetary union. European Economy 44, pp.
1-347.
European Commission (2011). The Euro, Who can join and when?
http://ec.europa.eu/economy_finance/euro/adoption/who_can_join/index_en.htm. European Commission (2012). Eurostat.
http://epp.eurostat.ec.europa.eu/portal/page/portal/eurostat/home/.
Fidrmuc, J. (2008). The Endogeneity of the Optimum Currency Area Criteria, Intra-Industry
Trade, and EMU Enlargement. Contemporary Economic Policy 22(1), pp. 1-12.
Fleming, J.M. (1971). On Exchange Rate Unification. The Economic Journal 81(323), pp. 467-488.
Frankel, J.A. and A.K. Rose (1996). The Endogeneity of the Optimum Currency Area
Criteria. NBER Working Paper 5700.
Friedman, M. (1953). Essays in Positive Economics. University of Chicago Press, Chicago. IMF (2009). World Economic Outlook, Crisis and Recovery. World Economic and Financial
Surveys. International Monetary Fund, Washington, DC.
Ingram, J.C. (1962). Regional Payments Mechanisms: The Case of Puerto Rico. University of North Carolina Press, Chapel Hill.
37
Issing, O. (2001). Economic and Monetary Union in Europe: political priority versus
economic integration? Paper for the Conference 2001 of the European Society for the
History of Economic Thought.
Jovanovic, B. and R.J. Gilbert (1993). The Diversification of Production. Brookings Papers on Economic Activity, Microeconomics 1993(1), pp.197-247.
Kawai, M. and M.B. Lamberte (2010). Managing Capital Flows: The Search for a Framework. Edward Elgar Publishing Inc., Northampton.
Kenen, P. B. (1969). The Theory of Optimum Currency Areas: An Eclectic View. In R.A. Mundell and A.K. Swoboda (Eds.), Monetary Problems of the International Economy. University of Chicago Press, Chicago, pp. 41-60.
Kertesi, G. and J. Köllô (2000). Wage Inequality in East-Central Europe. Budapest Working Papers, no. 2000/7.
Klau, F. and A. Mittelstadt (1986). Labour market flexibility. OECD Economic Studies 6, pp. 8-45.
Krugman, P. (1993). Lessons of Massachusetts for EMU. In F.S. Torres and F. Giavazzi
(Eds.), Adjustment and Growth in the European Monetary Union. University of Cambridge Press, Cambridge, pp. 241-266.
Lane, P.R. and G.M. Milesi-Ferretti (2006). The External Wealth of Nations Mark II: Revised
and Extended Estimates of Foreign Assets and Liabilities, 1970–2004. IIIS Discussion
Paper 126.
McKinnon, R.I. (1963). Optimum currency areas. The American Economic Review 53(4), pp. 717-725.
Mongelli, F.P. (2002). “New” Views on the Optimum Currency Area Theory: What is EMU
Telling US? ECB Working Paper 138.
Moore, H.L. (1922). Elasticity of Demand and Flexibility of Prices. Journal of the American Statistical Association 18(137), pp.8-19.
38
OECD (2012). OECD.Stat Extracts. http://stats.oecd.org/index.aspx?. OECD (2012a). OECD Glossary of Statistical Terms.
http://stats.oecd.org/glossary/detail.asp?ID=4835.
Parkin, M., Powell, M. and K. Matthews (2002). Economics. Pearson Education Limited, Harlow.
Raines, P. (2000). The impact of European integration on the development of national labour
markets. ILO Employment Paper 2000/1.
Reuveny, R. and Q. Li (2003). Economic Openness, Democracy, and Income Inequality: An
Empirical Analysis. Comparative Political Studies 36(5), pp. 575-601.
Ross-Thomas, E. and A. Davis (2010). Greece's Debt Cut to Junk, First for Euro Member. http://www.bloomberg.com/news/2010-04-27/greek-debt-cut-to-junk-at-s-p.html. Silvia, S.J. (2004). Is the Euro Working? The Euro and European Labour Markets. Journal of
Public Policy 24(2), pp. 147-168.
Standard and Poor’s (2012). Credit FAQ: Factors Behind Our Rating Actions On Eurozone
Sovereign Governments.
http://www.standardandpoors.com/ratings/articles/en/us/?articleType=HTML&assetID =1245327305715.
Tavlas, G.S. (1993). The ‘New’ Theory of Optimum Currency Areas. The World Economy 16(6), pp. 663-685.
The Economist (2010). Sovereign risk and the banks: The safety-net frays. http://www.economist.com/node/15498231?story_id=15498231.
Tower, E. and T. Willet (1976). The Theory of Optimum Currency Areas and Exchange Rate
Flexibility. International Finance Section 11, Princeton University, Princeton.
Warin, T., P.V. Wunnava and H.P. Janicki (2009). Testing Mundell’s Intuition of Endogenous
OCA Theory. Review of International Economics 17(1), pp. 74–86.
Tables
Table I: Data Summary
Criterion Variable Measured by Source
Price flexibility ΔDemand Final consumption expenditure (constant LCU) World Bank (2012)
Gross capital formation (constant LCU) World Bank (2012)
Exports of goods and services (constant LCU) World Bank (2012)
Imports of goods and services (constant LCU) World Bank (2012)
ΔPrices Inflation, GDP deflator (annual %) World Bank (2012)
Wage flexibility ΔWages Gross wages and salaries (current euros) European Commission (2012)
ΔUnemployment Unemployment, total (% of total labor force) World Bank (2012)
Total labor force World Bank (2012)
Labor market integration Nominal hourly labor costs Labour compensation per hour OECD (2012)
Unit labor costs Unit labour costs OECD (2012)
Factor market integration Long-term interest rates Long-term interest rates OECD (2012)
FDI flow inward FDI flows by partner country OECD (2012)
FDI flow outward FDI flows by partner country OECD (2012)
Financial market integration IFI Gross stock of foreign assets European Commission (2012)
Gross stock of foreign liabilities European Commission (2012)
Gross domestic product at market prices European Commission (2012)
Chinn-Ito index Chinn-Ito index Chinn and Ito (2008)
Economic openness Exports Exports of goods and services (current LCU) World Bank (2012)
Imports Imports of goods and services (current LCU) World Bank (2012)
Income level GDP (current LCU) World Bank (2012)
Diversification in production Exports per industry International trade of EU, the euro area and the Member States by SITC product group European Commission (2012) Diversification in consumption Imports per industry International trade of EU, the euro area and the Member States by SITC product group European Commission (2012) Similarity of inflation rates Inflation rates Inflation, GDP deflator (annual %) World Bank (2012)
Fiscal integration (Primary) government budget Government revenue, expenditure and main aggregates European Commission (2012)
Table II: Price Flexibility*
*Simple moving averages
41 Table III: Price Flexibility*
*Cumulative moving averages
42 Table IV: Wage Flexibility*
*Simple moving averages
43 Table V: Wage Flexibility*
*Cumulative moving averages
