Three waves of convergence. Can Eurozone countries start growing together again?

Marco Buti, Alessandro Turrini 17 April 2015



There is a widespread view that the monetary union is now acting as a straitjacket on countries in the periphery, that it’s no longer delivering convergence in living standards. In this vein, the Euro Summit of 24 October 2014 called for work to continue to "develop concrete mechanisms for stronger economic policy coordination, convergence, and solidarity", and invited the President of the European Commission, in close cooperation with the President of the Euro Summit, the President of the Eurogroup, and the President of the European central Bank, "to prepare next steps on better economic governance in the euro area".1

This column argues that convergence in the Eurozone has never stopped – what changed over the main phases of the monetary union was the type of convergence.

The type of convergence currently at work is neither nominal convergence (i.e., convergence in nominal variables like inflation and interest rates) nor real convergence as commonly understood (i.e., convergence in per capita incomes) but a convergence that concerns the structure of the economy.

Structural convergence is posing the basis for renewed real convergence. However, for this to happen, the right institutions and policies need to be in place at the  EU/EZ level and the national level.

The three faces of economic convergence

Different meanings have been given to the term ‘economic convergence’. In growth literature, convergence refers to the expected tendency for countries to grow faster the lower their GDP per capita level. Real convergence (i.e., narrowing differences in terms of per-capita GDP, relative endowments of productive factors and relative factor prices) is what neo-classical growth theory predicts.

The debate at the start of the Economic and Monetary Union (EMU) was also centred around convergence. In particular, the Maastricht criteria emphasised nominal convergence (defined in terms of nominal variables such as interest rates, inflation, exchange rates, government deficits and debt): the objective was not only that of creating a single currency, but also a stable currency (e.g., Buti and Sapir, 1998). It was expected that the adoption of a single currency would give strong incentives to carry out structural reforms to compensate for the loss of monetary policy as a stabilisation tool.

The academic debate put the emphasis instead on Optimal Currency Area (OCA) theory and the extent to which real convergence was sufficiently advanced to make the economies of the countries participating in EMU synchronised, so as to reduce the risk of unaddressed asymmetric shocks in a monetary union. The debate on the possible tension between nominal and real convergence surrounded also EU enlargement, with emphasis on the possible difficulties encountered by accession countries in respecting Maastricht criteria on inflation, interest and exchange rates in light of the operation of Balassa-Samuelson effects (e.g., Grauwe and Schnabl 2005).

Real convergence was implicitly assumed to work in the direction of making the economies participating in the monetary union more similar in terms of economic structures, thereby approaching EMU to the OCA requirements and easing the respect of the Maastricht nominal criteria.

However, the facts proved this expectation wrong. The first decade of EMU showed that structural convergence is not necessary a by-product of nominal and real convergence.

The three waves of convergence in EMU

The run up to EMU coincided with nominal convergence: differences in inflation rates among the countries participating in the monetary union narrowed. The average level of nominal interest rates and their dispersion across ‘EMU-ins’ dropped dramatically, in light of converging inflation differentials, reduced exchange rate risk, and reduced government deficits (see Figure 1).

Figure 1. Interest rates: Mean and variance across EA-12 countries

Source: elaborations on AMECO data

The years preceding the completion of EMU were also characterised by a strong record of real convergence, with higher growth rates registered in countries with lower income per capita.

After EMU completion, nominal convergence was virtually achieved, with the notable exception of government debts. The fall in interest rates was accompanied by a major process of integration of financial markets, notably bond markets. The statistics were also showing that real convergence was still working as expected (see Figure 2). The belief that the OCA criteria were "endogenous" became popular, namely that these criteria, although not satisfied ex-ante, would have been satisfied ex-post, thanks to the fact that EMU participation would have implied increased trade integration and a stronger degree of co-movement of economic cycles (Frenkel and Rose 1998).

Real convergence during the first ten years of EMU was made even more visible by the massive capital flows running from the ‘center’ to the ’periphery’ of the euro area, in light of vanishing exchange rate risk and reduced default risk perceptions. This process was paralleled by widening current account deficits in the periphery and growing surpluses in the core (see Figure 3). Capital inflows coincided with accelerated growth rates. As shown in Figure 2, countries in the periphery, while receiving capital inflows and recording current account deficits, were also registering sustained domestic demand and reducing unemployment at a fast rate (see Figure 4).

Figure 2. Scatterplot across EZ countries of per capita GDP growth (1999-2007), GDP per capita level in PPP (1999)

Source: elaborations on AMECO data

The sustainability of the additional growth accruing to the periphery was however dependent on the quality of the investments made possible by cheaper capital. In this respect, it did not help that buoyant capital inflows were matched by persisting inflation differentials and competitiveness losses in the periphery. Investment in the periphery became increasingly profitable in non-tradable activities. Large-scale bubbles in the housing sector were fuelled in countries like Spain and Ireland. Symmetrically, countries in the centre started growing increasingly out of exports, also thanks to improvements in price competitiveness, while relatively high real interest rates prevented the overheating of asset markets, notably housing. Moreover, the satisfactory growth performance registered in the EZ periphery led to a situation of ‘reform anaesthesia’ – the important reforms to facilitate an effective adjustment in the monetary union were delayed because they were not seen as urgent (European Commission 2008).

Figure 3. Current account and unemployment developments, average for core versus periphery EZ (weighted average)2

Source: elaborations on AMECO data.

In a nutshell, real convergence during the first EMU decade largely coincided with structural divergence.3 The structure of the economies at the centre and those at the periphery of EMU become increasingly different. As shown in Figure 4, the economies of the centre were relying more and more on exports and tradable activities, while those of the periphery were increasingly dominated by non-tradable sectors and construction.

In a sense, the well-known ’Walters' effect’ (i.e., the destabilising effect of real interest rate differences between a low-inflation centre and a high-inflation periphery; see Walters 1986) had implications that went beyond cyclical stabilisation. One could argue that a ’super Walter's effect’ operated during the first EMU decade, whereby not only cyclical positions, but also economic structures were driven by persistent real interest rate differences.

The financial crisis acted as a detonator of the imbalances accumulated during the first decade of EMU. Capitals started flowing in opposite direction, amid a general reappraisal of risk. The developments taking place since the run up to EMU were partly reversed – nominal divergence in interest rates linked to growing spreads between the centre and the periphery took place together with real divergence in growth and unemployment rates (see Figure 3). The correction of current account imbalances was accompanied not only by a compression of domestic demand in the periphery, but also by competitiveness gains and a re-composition of output towards tradable activities.

Can EZ countries start growing together again?

After years of protracted divergence, the most recent data point to an acceleration of growth and employment in most EZ countries in the periphery. The easing of bond market tensions, the reduced fragmentation of financial markets, and the reduced pace of the necessary fiscal consolidation in distressed countries are all contributing factors. Will this be enough? Will it last?

The current process of structural convergence is a necessary requisite for more real convergence going forward. As the economies in the periphery cannot rely again on growth patterns mainly driven by external borrowing, there is no alternative – the conditions for stronger non-tradable and export growth need to be in place. Though this process is a necessary condition for renewed convergence in living conditions across EMU, what has been achieved so far will by no means be sufficient.

The first reason is that much of the structural convergence observed is linked especially to the contraction in the non-tradable sector in periphery countries. A protracted expansion of the export sector requires ambitious reforms, supportive policy frameworks, and a sustained recovery in export markets. In this respect, the Commission has launched a three-pronged agenda to strengthen the recovery in Europe based on a renewed commitment to structural reforms, a commitment to fiscal responsibility that also takes into account growth objectives within the available fiscal space, and boosting investment via additional resources and the removal of bottlenecks.

Second, as discussed in a previous column (Buti 2014), the need for an export-driven growth strategy in the periphery may clash with the need to ensure sound public finances since price competitiveness gains in the periphery may imply debt deflation. A successful anti-deflationary stance by the monetary authorities will play a key role in easing such a trade-off.

Last but not least, sustainable convergence will require a credible institutional framework underpinning EMU and the preservation of a climate of trust. The financial crisis helped realign the market assessment of country risk with fundamentals. The market reaction proved, as so often, too strong and too late, but it did provide the basis for adjustment and structural convergence, including by means of much-needed reforms. Looking forward, the extent and sustainability of restored convergence will depend crucially on whether markets will maintain confidence in the economies of the EZ periphery. Recent EU level initiatives will help in this respect – strengthened financial market regulation and the establishment of a single supervisory mechanism for the banking sector will help ensure the compatibility between restored convergence and macro-financial stability. By taking a EZ wide view on supervision it will help lessen macro-financial risks. The establishment of a Capital Markets Union will contribute to the integration of equity markets across the EU, thus reducing the volatility of cross-border investment, enhancing risk sharing, and most importantly helping to reallocate excess savings within the EZ via equity rather than debt. Both developments should help tackle the super-Walters effect that hampered structural divergence in the pre-crisis period.

Figure 4. Cumulative growth rate of non-tradable/tradable value added, and construction / total value added (Weighted average for EZ core versus EZ periphery, averages for sub-periods: run up to EMU, first EMU decade, post-crisis)

Source: elaborations on AMECO data


Buti, M (2014) "A consistent trinity for the Eurozone",, 8 January.

Buti, M and A Turrini (2012) "Slow but steady? External adjustment within the Eurozone starts working",, 12 November.

Buti, M and A Sapir (1998) Economic Policy in EMU, Clarendon Press, Oxford.

De Grauwe, P and G Schnabl (2005) "Nominal vs. real convergence. EMU entry scenarios for the new member states", Kyklos 4, 537-555.

European Commission (2008), [email protected]: successes and challenges after 10 years of Economic and Monetary Union, European Economy

Frankel, J and A K  Rose (1998) "The endogeneity of the Optimum Currency Area criteria", Economic Journal 108, 1009-25.

Marelli, E and M Signorelli (2005) "Institutional, nominal, and real convergence in Europe", Banks and Bank Systems 5, 140-155.

Walters, A (1986), Britain’s Economic Renaissance, Oxford University Press, Oxford.


1 The issue of convergence in the Eurozone is discussed also in the Four Presidents' Analytical Note "Preparing for Next Steps on Better Economic Governance in the Euro Area", Informal European Council, 12 February 2015.

2 Centre: BE DE LU NL AT FI.  Periphery: EE IE EL ES FR IT CY LV LT MT PT SI SK. Centre and periphery EZ countries grouped according to their external position, as in Buti and Turrini (2013). With few exceptions, countries in the centre recorded a surplus over the 1999-2009 period, while periphery countries recorded a deficit.

3 Similar conclusions are reached in existing papers, e.g.,  Marelli and Signorelli (2005).



Topics:  Europe's nations and regions Macroeconomic policy

Tags:  Convergence; Eurozone; financial crisis; monetary union; EMU

Director General, DG Economic and Financial Affairs, European Commission

Head of Unit, DG Economic and Financial Affairs, European Commission