2. Literature review

European Economic and Monetary Union (EMU) was expected to foster greater macroeconomic stability, prosperity, and convergence. The introduction of the single currency would stabilize exchange rates and lower interest rates across the union. The policymakers assumed that, by eliminating the exchange rate uncertainty and reducing the cross-border transaction costs, a common currency would increase capital mobility and intra-regional trade, thereby boosting growth and helping per capita income levels to converge between poorer and richer countries (Aglietta and Brand, Reference Aglietta and Brand2013).

The EMU succeeded in establishing a credible monetary policy framework and deepened the financial integration, but many of the national governments failed to exercise sufficient fiscal discipline as well as undertaking sufficient structural reforms. In the first years of the euro, countries benefited from a stability-oriented monetary policy framework with low interest rates, low expected inflation, and a stable, common exchange rate. However, national fiscal policies were pro-cyclical and the structural reform momentum faded, leading to macroeconomic imbalances, and widening competitiveness gaps. With weak financial supervision, rapidly increasing cross-border capital flows ultimately served as a destabilizing force (Franks et al. (Reference Franks, Barkbu, Blavy, Oman, Schoelermann and Pradhan2018).

The Maastricht treaty established convergence criteria for countries to join the euro, focusing on nominal and fiscal indicators of harmonization such as: inflation, long-term interest rates, exchange rate stability, fiscal deficit and the government debt-to-GDP ratio. The criteria were aimed at achieving price stability and lowering the dispersion of inflation rates, while reducing excessive deficits before locking the exchange rates (Bini-Smaghi et al., Reference Bini-Smaghi, Padoa-Schioppa and Papadia1993; Emerson et al., 1992). The expectation was that the price level of tradeable goods would be equalized across countries, and that inflation would generally converge, though with some scope for countries that were catching up in level of productivity and incomes, to having higher inflation as non-tradable prices adjusted more slowly. The fiscal criteria aimed to reconcile a common monetary policy with decentralized fiscal policies, preventing spillovers from national policies (Andrle et al., Reference Andrle, Bluedorn, Eyraud, Kinda, Brooks, Schwartz and Weber2015; European Commission, 2013).

This expectation is confirmed by a several of studies, such as Busetti et al. (Reference Busetti, Forni, Harvey and Venditti2007), Engel and Rogers (Reference Engel and Rogers2004), Rogers et al. (Reference Rogers, Hufbauer and Wada2001, 2007)), and Weber and Beck (Reference Weber and Beck2005), all of which agree that prices were less dispersed and inflation rates among Euro area countries have converged in the mid-1990s. Contrary, research investigating the post-1998 period, including ECB (2003), Engel and Rogers (Reference Engel and Rogers2004), Fritsche and Kuzin (Reference Fritsche and Kuzin2008), Honohan et al. (Reference Honohan, Lane, Ventura and Begg2003), Rogers (Reference Rogers2007), and Weber and Beck (Reference Weber and Beck2005) concludes that the advent of the Euro resulted in the weakening of inflation convergence among the Euro area countries and in an increase in their price dispersion. An exception is Honohan and Lane (Reference Honohan and Lane2004), who report sharp convergence in inflation rates, since 2002.

Pirovano and Van Poeck (Reference Pirovano and Van Poeck2011), test for the stability of inflation differentials. The authors found that for the period from 1999 to 2006, data do not reject the hypothesis of convergence for the majority of countries. On the other hand, by considering the whole sample period 1999–2010, there is evidence of persistent divergence in inflationary dynamics for the Euro-area countries. Although the authors conclude that the inflationary dynamics are more unstable in the Euro area in recent years, that is, during the years characterized by the Great Recession, it is worth stressing that it is difficult to draw sound conclusions about the evolution of inflation differentials, when such limited sample periods are considered.

Beck et al. (Reference Beck, Hubrich and Marcellino2009) have analyzed the dynamics of regional inflation in a group of Euro-area countries. They aim to identify the empirical relevance of the national factors, in explaining the inflation differentials over the period from 1996 to 2004. The main conclusion of the investigation is that local, structural factors, such as limited competition in labour and good markets, have played a dominant role in explaining the variability of inflation.

Altissimo et al. (Reference Altissimo, Ehrmann and Smets2006) have studied the underlying factors of inflation differentials in a currency area. On the empirical ground, they found two key results: (a) persistent inflation differentials affect the euro area; (b) a pre-eminent role, in explaining the evolution of the inflation differentials, is played by the different responses of Eurozone countries to common, Euro-area shocks.

A degree of inflation convergence is undoubtedly necessary for a well-functioning monetary policy. With a common monetary policy and decentralized fiscal policies that are bound by a common set of rules, it is difficult for policymakers to deal with large cyclical inflation differentials across countries. However, under a single currency, some inflation differentials may be needed to facilitate real exchange rate adjustment. Particularly, any accumulated competitiveness gaps will need to be unwound by inflation differentials, instead of exchange rate adjustments.

A group in the literature focuses on the combination of descriptive and regression analysis and has its main starting point in Honohan et al. (Reference Honohan, Lane, Ventura and Begg2003), whose approach we largely follow in our analysis. The authors tried to analyze inflation differentials in the Euro area over the period 1999–2001 through a multivariate panel data model estimated by Ordinary Least Squares (OLS) and Generalized Method of Moments (GMM). They proposed a fairly general specification for inflation differentials, including measures of cyclical position (output gap), fiscal position (primary surplus), price level convergence (lagged price level) and external position (lagged differenced nominal effective exchange rate). Their evidence points to a highly significant price level convergence effect, external position and output gap across almost all specifications. On the other hand, the evidence on fiscal position is inconclusive, being insignificant and significant, but contrary in sign to prior expectations, in the OLS and GMM regressions, respectively. In the wake of these results, the output gap is thought to capture most of the effect, that can come through the fiscal channel.

De Grauwe (Reference De Grauwe1996) has tried to investigate the essential subject of convergence of inflation rates prior to the acceptance of a country into the monetary union. He found that, in the year of 1991, the degree of inflation convergence among the countries participating in the European Monetary System achieved the highest level and he concluded that further narrowing of the differences among inflation rates is unrealistic. He also stated, that the inflation convergence criterion is too tight to fulfill. Few years later, in a continuation of his previous research, he found that a further drop in inflation differentials occurred after 1991. The author cautioned against the tight nominal convergence criteria, and he argued that the transition to a successful monetary union should put less emphasis on such convergence requirements and more on strengthening the future monetary institutions of the union.

Hofmann and Remsperger (Reference Hofmann and Remsperger2005) found that the real exchange rate together with the short-term real interest rate have a crucial effect on aggregate demand. That means, that in the long run, the inflation differentials subside because the impact of inflation differentials on the real exchange rate is accumulated. They concluded that persistence of inflation differentials among Euro-countries, who already experience a low and stable inflation rate, is small. Additionally, they believe that the credibility of ECB will aid this problem among the countries that currently have a high and unstable level of inflation rate. Girardi and Paruolo (Reference Girardi and Paruolo2013) examined the pre and post-EMU Phillips curve of EMU counties namely Germany, France, Italy and Spain. Aside from these member states of EMU, they included UK as the control factor in their test. They find that, during EMU period unemployment is proven to be influenced by the ECB, as was anticipated, where the Phillips curves of 4 EMU countries are similarly common, while the slope of UK’s Phillips curve is found to be positive. Nevertheless, EMU does not have control over the ongoing high level of the wage—and price systems.

During 1997–2001, there were four observations that emerged, according to Fase (Reference Fase2002). Firstly, inflation differentials were wide and systematic. Secondly, convergence started to grow due to the common monetary policy. Thirdly, Germany, France and Italy had continuously showed lower inflation rates than the rest of the Euro-countries. And lastly, the inflation differentials have escalated after the launch of EMU. The increased convergence that is mentioned above implies, that the differences in growth, productivity and income, decline over time. Amidst a real convergence process, the inflation differentials may increase significantly, and this may persist until real convergence is achieved. Meriküll and Rõõm (Reference Meriküll and Rõõm2014) agree with this, since they find that the Euro-changeover contributed to higher price expectations in Estonia. Fase (Reference Fase2002) argues that monetary phenomenon is not the primary source of this differential and, hence, cannot be assisted by monetary policy alone. However, differentials are rather developed as a consequence of a particular country’s sensitivity to external shocks, as well as an adjustment process of new economic structure and market mechanism. Finally, according to Neaime et al. (Reference Neaime, Gaysset and Badra2018), Greek’s economy is an example of a country that is sensitive to external shocks, since they are the only country that experiences deep recession among the rest of EU countries.

In addition, Lopez and Papell (Reference Lopez and Papell2012) have studied the convergence of the inflation rates of countries constituting the initial members of the euro area and they found robust evidence of convergence, after the introduction of the Maastricht Treaty and, especially, after the adoption of the common currency in 1999. They also found that, after the onset of the global financial crisis, the inflation rates of the euro area members mostly do not deviate from the path dictated by ECB monetary policy. Arestis et al. (Reference Arestis, Chortareas, Magkonis and Moschos2014) investigate whether the inflation rates of the countries that pursue inflation targeting, have converged in contradistinction to the experience of the OECD non-inflation targeters. Their results indicated that the inflation rates converge regardless of the monetary policy framework.

Busetti et al. (Reference Busetti, Forni, Harvey and Venditti2007) have studied the convergence properties of in inflation rates among the countries of the European Monetary Union over the period 1980–2004. They detected two separate clusters, or convergence clubs: a lower inflation group that comprises Germany, France, Belgium, Austria, Finland and a higher in inflation one with Spain, Netherlands, Greece, Portugal and Ireland. Italy appears to form a cluster of its own, standing in between the other two. In a similar recent research Karanasos et al. (Reference Karanasos, Koutroumpis, Karavias, Kartsaklas and Arakelian2016) studied the convergence properties of inflation rates among the countries of the European Monetary Union, over the period 1980–2013. They detected three absolute convergence clubs in the pre-euro period, which comprise early accession countries namely Luxembourg, Austria and Belgium. A second sub-group included Germany and France and the third The Netherlands and Finland. They have also detected two separate clusters of early accession countries in the post-1997 period: the first with Germany, Austria, Belgium and Luxembourg, and the other one with France and Finland. For the rest of the countries/cases they found evidence of divergent behavior.

3. Causes of inflation differentials

In order to ascertain any of the potential policy implications of persistent inflation differentials, it is more than necessary to properly identify their underlying causes. However, this is not an easy task because, in a large monetary union such as the euro area several factors may contribute to inflation divergence. We will try to present a brief description of the main factors, which have been put forward to explain the existence of long-lasting inflation differentials in the euro area. In particular, a distinction is made between structural and cyclical factors related to the convergence process.

4. Methodology

With the purpose of establishing the relative contributions of some of the causes of driving inflation divergence in EMU, we ran multivariate regressions using annual data for the period from 1999 to 2018. A general specification for explaining inflation differentials can be written as follows:

(4.1) $$ {\pi}_{it}-{\pi}_t^{EMU}\hskip0.35em =\hskip0.35em {\alpha}_i+\beta \left({P}_{i,t-1}-{P}_{t-1}^{EMU}\right)+\gamma \left({x}_{i,t-}{x}_t^{EMU}\right)+{\varepsilon}_{i,t} $$

where

π_i,t is the inflation rate of country i for period t,

$ {\pi}_t^{EMU} $ is the average inflation rate for the euro area,

α_i is the unobserved country—specific effect,

$ {P}_{i,t-1}\hskip0.35em and\hskip0.35em {P}_{t-1}^{EMU} $ the national and euro area price levels respectively and

$ {x}_{i,t}\hskip0.35em and\hskip0.35em {x}_t^{EMU} $ are the vectors of national and eurozone variables, which play a key role in determining inflation differentials in the short run. In order to get a simpler version of the above equation, we can follow Honohan et al. (Reference Honohan, Lane, Ventura and Begg2003) and linearly combine into a time variable the euro area variables and obtain the following equation:

(4.2) $$ {\pi}_{it}\hskip0.35em =\hskip0.35em {\alpha}_i+\beta {P}_{i,t-1}+\gamma {x}_{i,t}+{\varphi}_t+{\varepsilon}_{i,t} $$

where

(4.3) $$ {\varphi}_t\hskip0.35em =\hskip0.35em {\pi}_t^{EMU}-\beta {P}_{t-1}^{EMU}-\gamma {x}_t^{EMU} $$

The explanatory variables that are included in the x_i,t vector, are a measure of Balassa–Samuelson effect, output gap, a measure of fiscal stance and the growth rate of the nominal effective exchange rate. At that point, we must note that the time dummies in the regression capture the euro—area common movements in inflation and in the regressors (Honohan et al., Reference Honohan, Lane, Ventura and Begg2003). This leaves the regression to explain inflation differentials coming from idiosyncratic national movements in the determinants. So, our empirical specification is as follows:

(4.4) $$ {\pi}_{it}\hskip0.35em =\hskip0.35em {\alpha}_i+\beta {P}_{i,t-1}+{\gamma}_1{BS}_{i,t}+{\gamma}_2{GAP}_{i,t}+{\gamma}_3{FISC}_{i,t}+{\gamma}_4{NEER}_{i,t}+{\varphi}_t+{\varepsilon}_{i,t} $$

where π_i,t is the quarterly inflation rate in country i, BS_i,t captures the Balassa–Samuelson effect through the difference between the CPI and PPI, GAP_i,t is the output gap, FISC_i,t is the fiscal stance, where we have included the deficit and the public debt as a measure for that, and NEER_i,t is the growth rate of the nominal effective exchange rate.

According to our model, equation (4.3) will be as follows:

(4.5) $$ {\pi}_{it}\hskip0.35em =\hskip0.35em {\alpha}_i+\beta PG+{\gamma}_1 PD+{\gamma}_2 BD+{\gamma}_3 OG+{\gamma}_4 ER+{\varphi}_t+{\varepsilon}_{i,t} $$

where PG productivity growth, PD is public debt, BD is budget deficit, OG is output gap and ER is exchange rate.

5. Data

We use data for the 19-euro area countries over the EMU for the period 1999–2018. Our baseline proxy for the Balassa–Samuelson effects is constructed as the annual growth rate of labour productivity in the business sector. The output gap is obtained using the Hodrick–Prescott filter (with a smoothing parameter set to 30) on initial series of GDP at constant price. The indicator for the exchange rates variations is constructed as the growth rate of the nominal effective exchange rate weighted by the ratio of imports to GDP. All the data are taken from Eurostat and are in annual frequency.

7. Conclusions and policy implications

The main goal of this article was to identify the determinants of inflation differences in the euro area for the period 1999–2018. The stability of inflation differentials is an important condition for the smooth working of a currency area, such as the European Economic and Monetary Union. Unfortunately, the inflation rates have diverged more widely among the members of the Eurozone than expected, and the recent surge in inflation dispersion illustrates that there is no tendency for inflation differentials to decline or to disappear.

We estimated the factors of inflation differentials using the OLS and GMM method. Results based on a dynamic panel data model estimated using the OLS estimator and system-GMM estimator, defined in Blundell and Bond (Reference Blundell and Bond1998), suggest that the main sources of euro area inflation differentials are twofold. They partly stem from differences in the exposure to external shocks, that is, nominal effective exchange deviations, as well the Balassa–Samuelson effect (defined as productivity growth in the estimations). Inflation differentials seem to be the result of different position of the countries in the economic cycle (the output gap in the estimations). Our empirical estimates appear to suggest that a one-percentage-point increase in the positive output gap, typically leads to an increase of about 20 basis points in the inflation rate of EMU countries. However, and even surprisingly, fiscal stance does not seem to strongly affect the inflation differentials, only slightly.

Finally, the growing interest of the ECB, regarding national inflation differentials within the euro area, clearly reveals its concern about the potential implications of such lasting inflation disparities. Lasting inflation differentials may also complicate the task of the ECB for the definition of its single monetary policy: a “one size fits all” policy, and should thus induce a too restrictive policy for countries exhibiting the lowest inflation rates and simultaneously an excessively accommodating policy for the most inflationary countries. However, our analysis also reveals the key role played by productivity growth exchange rates in amplifying inflation differentials. Since the monetary policy of the ECB is geared at maintaining a low and stable inflation, productivity growth should be increased, and the real effective exchange rates should be decreased and become more homogenous among EMU countries, as suggested by Altissimo et al. (Reference Altissimo, Ehrmann and Smets2006). Inflation differentials may therefore become less persistent.

Undesirably low inflation rates have been very costly to the euro area economy. At the Zero Lower Bound (ZLB), the macroeconomic risks associated with undesirably low inflation can be particularly high, as shown by Arias et al. (Reference Arias, Erceg and Trabandt2016)). With the coexistence of low inflation and the ZLB, many of the other policy options may become difficult or detrimental for the economy. For example, as Eggertsson et al. (Reference Eggertsson, Ferrero and Raffo2014)) show, structural reforms that ordinarily could support growth, when they could be accompanied by interest rate reductions, may instead be unwise and prove contractionary at the ZLB when short-term rates cannot adjust.

The Recovery and Resilience Facility (RRF) is a new key instrument at the heart of NextGenerationEU to help the European Union emerge stronger and more resilient from the current and future crises. Also, if the inflation rates will diverge more widely among the members of the Eurozone than expected, then the RRF will help towards the normalization and the conversion of the inflation rates. The RRF confirms the EU’s aim to pursue a new growth strategy, based on the European Green Deal and on the concept of competitive sustainability. The four dimensions of competitive sustainability, namely the environmental sustainability, productivity, fairness and macroeconomic stability, identified in last year’s ASGS, should remain the guiding principles for the implementation of the Facility (European Commission, 2020). The proposed Recovery and Resilience Facility will offer large scale financial support for public investments and reforms, that make Member States economies more resilient and better prepared for the future. It will help Member States to address the economic and social challenges they are facing, in an even more critical manner in the aftermath of the crisis, in various areas such as social, employment, skills, education, research and innovation, health issues, but also, issues related to the business environment, including public administration and the financial sector.

Also, the RRF will help to have smart, sustainable and inclusive growth including economic cohesion, jobs, productivity, competitiveness, research, development and innovation, as well as a well-functioning internal market, with strong SMEs. The investments will aim to support companies that increase productivity, and also, the grants through the Facility will give us production externalities from public investment, and finally, we will have higher productivity of private inputs. The Member State economies will have projects that strengthen the resilience, increase productivity, leading to higher competitiveness of the Member States and reduce inequalities and divergences in the Union.

Crucially, it will also ensure that these investments and reforms focus on the challenges and investment needs related to the green and digital transitions, thereby ensuring a sustainable recovery. It will also help make the Union more resilient and less dependent by diversifying key supply chains (European Commission, 2020).