The establishment of the European Economic and Monetary Union (EMU) was admittedly a remarkable step in the direction of enhancing economic integration among European countries. The launch of the common currency was expected to lead to price stability, lower transaction costs, stronger intra-euro trade relationships and thus, to higher growth for country-members. This optimistic view is obviously related to McKinnon’s (1963) contribution to the theory of Optimum Currency Areas (OCA). However, a fundamental weakness of the EMU, such as the lack of homogeneity across member-countries, should not be ignored. A number of divergent factors, such as dissimilar national policies (apart from the monetary policy) and different national regulations on goods and labour markets, may increase the possibility of emergence of asymmetric shocks in the Eurozone. On the other hand, this view is related to Mundell’s (1961) original contribution to OCA theory.
Having in mind the aforementioned heterogeneity and the resulting asymmetries across countries, academics and policy makers focus on answering the question of whether the EMU achieved its goals. The main reservation in this analysis arises from the presence of asymmetries and the lack of autonomous monetary policy for member-countries. This is because an asymmetric shock could be managed by an exchange rate adjustment. However, in a monetary union, like the EMU, this is not the case. Thus, the main question is whether the common monetary policy (including the exchange rate policy) can achieve higher growth rates and higher economic and financial integration in the Eurozone.
De Grauwe (2009) argues that in the first decade of euro’s life and before the debt crisis arises, there is little evidence that the euro caused higher growth rates in the Eurozone. On the other hand, nobody can argue that the euro had negative impacts on growth. However, it is also true that the EMU suffers from significant design weaknesses, which became more evident and stronger during the sovereign debt crisis. What may be indicative of the progress of economic integration among EMU members is that real effective exchange rates deviate among them, thereby implying that their competitive positions have diverged (De Grauwe, 2009, 2010). Northern European countries such as Germany, Austria and the Netherlands, gain in terms of international competitiveness, while competiveness in international trade for Southern European countries, such as Greece, Italy and Spain, has deteriorated.
In this context, the present paper aims to find whether economic and financial integration has increased among countries after the creation of the EMU. To be precise, we investigate whether EMU countries as well as selected non-EMU countries are financially integrated with Germany, which is the leading country in the EMU as it has the highest influence on the common monetary policy. We initially expect that the euro has led to integrated commodity and capital markets in the Eurozone because of stronger trade linkages among its member-countries. On the other hand, given the high degree of heterogeneity across countries and the absence of (intra-euro) exchange rate fluctuations, it is doubtful that higher economic integration can be achieved among EMU countries (especially for those that are structurally different from Germany).
The existence of economic and financial integration between Germany and the rest of the Eurozone’s countries (and the non-EMU countries) is tested through two well-known international parity relationships, i.e. the Purchasing Power Parity (PPP) and the Uncovered Interest Parity (UIP). The empirical validity of the PPP hypothesis implies that goods markets are integrated, while the validity of the UIP condition implies the existence of capital market integration between countries.
A survey on the related empirical literature implies that the introduction of the euro may have failed to increase commodity and financial markets integration among EMU countries. Koedijk et al. (2004) find evidence in favour of the PPP hypothesis within the euro area only when common mean reversion among countries is assumed. Setting Germany as the benchmark country and assuming heterogeneous mean reversion coefficients, their evidence is strengthened only for France, Finland and Spain, while they found no evidence regarding the validity of the PPP between the EMU and major non-EMU countries. Furthermore, Christidou and Panagiotidis (2010) and Wu and Lin (2011) report that the adoption of the euro has weakened the evidence in favor of the PPP. Similarly, Huang and Yang (2015) find that after the launch of the euro, the evidence in favour of the PPP is stronger for non-EMU countries rather than EMU countries. When it comes to capital markets integration (i.e. the UIP hypothesis), Kim et al. (2006) find that the degree of integration among European bond and stock markets has declined after the introduction of the euro.
However, the above studies have tested the PPP and UIP hypotheses only as independent parity conditions. This implies that the possibility that deviations from the PPP equilibrium are utilized by investors when forming expectations has been overlooked. Motivated by the seminal papers of Johansen and Juselius (1992) and Juselius (1995), we expect that PPP deviations may interact with UIP deviations. In the present paper, we extend the empirical literature on economic and financial integration in the Eurozone by testing the PPP and the UIP jointly. To the best of our knowledge, we argue that the present paper is the first that tests jointly the PPP and UIP conditions between Germany (as the leading economy of the EMU) and the remaining EMU countries.
Another contribution of the paper is that, compared to the majority of the empirical studies in the literature, it uses more accurate price indices. Specifically, we utilise constructed Traded-goods Price Indices (TPI) instead of Consumer Price Indices (CPI) in order to avoid the presence of non-traded goods prices, which biases negatively the empirical validation of the PPP hypothesis. Moreover, we use state-of-the-art time series econometric techniques, which allow the presence of structural breaks in cointegration analysis. Admittedly, the launch of the euro in 1999 and the global financial crisis of 2007 have altered the behaviour of variables under consideration. Hence, these two facts have caused an equal number of structural breaks, which should not be ignored by our analysis. Finally, the use of Germany as a benchmark country allows us to shed more light on Germany’s leading role in the Eurozone. Does the degree of economic integration between Germany and the rest of the Eurozone’s countries allow the characterisation of Germany as the representative EMU country? Given Germany’s domination in the Eurozone, a number of policy-related issues arise for the future of the EMU.