Southern and Eastern Europe in the Eurozone: convergence or divergence?

ABSTRACT A number of studies have concluded that, contrary to expectations, European monetary union has not reduced income disparities among the 12 Western European member countries. In fact, incomes per capita between Southern and Northern Eurozone members have diverged since the Eurozone was created, in contrast to earlier trends which had seen rapid convergence in living standards across Western Europe. The paper revisits this issue and investigates whether something similar occurred in the five Eastern European countries (Estonia, Latvia, Lithuania, Slovakia and Slovenia) which joined the Eurozone between 2007 and 2015. The conclusion is that, despite similarities with Southern Europe in some areas, overall developments in Eastern Europe were different and led to convergence rather than divergence. Important reasons for this were Eastern Europe's relatively high levels of institutional quality, a politically motivated determination to anchor these countries to the West and, possibly, the legacy of pre-war history.


Introduction
After a promising start, Europe's monetary union entered a fraught period during and after the so-called 'Great Recession' of 2008-09. Not only was that recession the deepest since World War II, but the area, unlike other developed countries, suffered in addition from a second, if shallower, recession in 2012-13. These were years during which it seemed that the whole project might unravel under the pressures wielded by financial markets. That this did not happen owes a lot to the July 2012 statement by Mario Draghi, president of the European Central Bank (ECB), that the bank 'is ready to do whatever it takes to preserve the euro' [emphasis added]. 1 Financial markets, in particular, have since then been much less volatile, but the tensions and the cross-country divergent trends that were brought to light at the time have not fully gone away.
The reasons for the divergent trends that have characterized the Eurozone, are of course, numerous. One possible source for them could have been asymmetric shocks, much feared by the literature on common currency areas and on Europe's monetary union itself (e.g. Bayoumi & Eichengreen, 1993). In the event, however, it does not appear that such shocks have been particularly prominent. By far the biggest shock to hit Europe was a symmetric one, the Great Recession itself, which led to the largest falls in output since World War II in each of the 12 founding member countries 2 (arguably, in Finland it also coincided with an asymmetric shock). 3 As for the minor cyclical fluctuations which preceded 2008-09, these seem to have been symmetrical as well. 4 Alternatively, processes of cumulative causation could have led human and physical capital to concentrate in more prosperous areas , increasing income per capita gaps between rich (i.e. Northern) and poor (i.e. Southern) countries. Yet, this is something that does not really seem to have occurred in Europe, be it in the long period of integration prior to the creation of monetary union, or in the shorter period that followed it. Foreign direct investment (FDI) inflows into the Mediterranean countries in per cent of GDP, for instance, were marginally higher, relative to inflows into the Northern countries in the years of monetary union than in the decade that preceded it. It is true that labour did move from South to North, particularly after the Great Recession, but the size of these movements has, so far, been very small (Arpaia et al., 2014).
A somewhat more plausible reason for the divergences in performance experienced inside the area could be linked, instead, to what has been called 'asymmetric behaviour' (Boltho & Carlin, 2013). In several so-called 'Club Med' countries public and private sector economic behaviour, already in the early years of monetary union, turned out not to be in line with what membership of such a union required. Thus, wage growth was way above the very sluggish developments in productivitythis weakened external competitiveness. In addition private debt accumulation was particularly rapid (as was the rise in house prices). When the recession(s) came, the house price bubble burst while debt pressures forced the private sector into sharp deleveraging and the public sector into severe austerity. In addition, institutional reforms, much needed in Southern Europe, were slow or non-existent. While such trends were much less in evidence in the Northern countries, the de facto 'beggar-thy-neighbour' behaviour which some of them followed (behaviour also incompatible with monetary union) led to massive current account surpluses which made external adjustment in the South more difficult. Partly as a consequence, the output of several Southern Eurozone countries was, in early 2019, still below, or just barely above, its pre-recession 2008 Q1 peak in contrast to the much more favourable developments seen in the Northern member economies. While since the 1950s Western Europe had seen rapid convergence in living standards between poorer and richer countries, this process was reversed for the Eurozone soon after monetary union was created (Christodoulakis, 2009;Boltho & Carlin, 2013;Wagner, 2014;ECB, 2015).
The aim of this brief paper is to see whether the five Eastern European countries (Slovenia, Slovakia, Estonia, Latvia and Lithuania) which joined the Eurozone between 2007 and 2015, followed processes similar to those of Southern Europe and experienced divergence from Northern European levels of GDP per capita or, alternatively, saw convergence. To simplify the presentation, GDP weighted totals will be shown for this group of five countries (often called 'East' for short in what follows) as well as for two original Eurozone areas: a Northern one made up of Germany, France, the three Benelux countries, Austria and Finland (often called 'North'), and a Southern one constituted by Italy, Spain, Greece and Portugal (often called 'South'). 5 For these two latter groupings, the time span will cover the years 1999-2018. For Eastern Europe most observations will start in 2009. It is readily admitted that in 2009 only Slovakia and Slovenia were members of the monetary union while the three Baltic countries had not yet adopted the euro (Estonia did so in 2011, Latvia and Lithuania only in 2014 and 2015 respectively), but all three had instituted either formal currency boards (Estonia and Lithuania) or informal ones (Latvia) in the early 1990s and fixed their exchange rates to the euro between 1999 and 2005 (Staehr, 2015). 6 They were, therefore, in some ways at least, de facto members.
Section 2 looks at macroeconomic trends (e.g. growth, debt accumulation, external competitiveness), while Section 3 examines the reform and governance record. A brief concluding section tries to bring the various strands together.

Macroeconomic trends
Monetary union, as originally envisaged in the so-called Delors Report (European Council, 1989), had as one of its main aims 'converging standards of living', thus further promoting a trend that had established itself from the start of European integration thanks, largely, to increasing intra-EU trade. Between the early 1960s and the late 1990s the gap in GDP per capita (measured in purchasing power parities) between the Southern and Northern parts of the Eurozone, as defined in the Introduction above, had shrunk from 37 to 14 per cent. A more formal investigation confirms this. Even the simplest specification of a standard catch-up model provides a satisfactory explanation of what happened and shows that income per capita convergence clearly took place between 1950 and 1999 in the group of 12 Western countries that founded the Eurozone in 1999 (Table 1). 7 This finding refers to so-called β-convergence; there is evidence that σ-convergence (or a decline in income per capita dispersion) also took place during the period (Kaitila, 2014).
Since the early 2000s, however, these developments have given way to divergence, with the Southern European economies clearly under-performing their Northern counterparts so far this century. Figure 1a provides a bird's eye view of the sharp break in trend that occurred between 2000 and 2005 (already before the Great Recession in other words) and Table 1 confirms this more formally. The same specification that showed clear β-  1950-1999 1999-2018 2009-2018 Original 12 Eurozone Const. 11.3** (7.1) convergence in living standards between 1950 and 1999, now shows divergence between 1999 and 2018, and this 'despite initial expectations that the single currency would act as a catalyst for faster real convergence' (ECB, 2015, p. 32). Nor has there been much evidence of a decline in income per capita dispersion (or σconvergence) since the euro was created (ibid.).
The East European member countries, however, appear to have had a different experience. Convergence with Northern Europe was present, whether one looks at the 1999-2018 period or at the shorter (and arguably more relevant) 2009-18 years (Table 1). Interestingly, the rate of convergence recorded, between 2½ and 3 per cent per annum in both periods, is higher than the 2 per cent figure that has been shown to apply to regions in a large number of countries (Barro & Sala-í-Martin, 2004). The 2009 recession was even more pronounced in Eastern Europe than it was in Northern or Southern Europe. But while Southern Europe's recovery was muted (GDP in 2018 was still below its pre-recession peak), both Northern and, especially, Eastern Europe saw rapid recoveries. It can, of course, be expected that poorer economies, as all the five new member countries are, would grow more rapidly than the mature economies of the West, but the contrast with the, also somewhat poorer, Southern states is striking, 8 as is the comparison with another group of East European EU member countries which did not embrace the euro (Bulgaria, Croatia, the Czech Republic, Hungary, Poland and Romania) (Figure 1b). Though slightly poorer in 1999 than their Eurozone Eastern neighbours, these economies also showed rapid, but not quite as rapid, catch-up.
What could explain the glaring difference between South and East? Asymmetric shocks in this instance too are unlikely to have been a cause. Since accession to the European Union in 2004 (and even before), cycles in the five Eastern countries here considered have been strongly correlated with those of the Eurozone (Fidrmuc & Korhonen, 2006;Stanisic, 2013). Amplitude may have differed (in 2009 GDP in the Baltic area fell by almost 15 per cent), but the direction of change was similar. Explanations linked to different behaviour are more likely to throw light on the issue. Many of the problems Southern Europe encountered in the recent past are linked to the very rapid accumulation of both private and public debt during its Eurozone membership years. Figure 2a and 2b show how these two variables grew in the three areas. It will be readily seen that there are similarities between the Southern and Eastern lines both in timing and in shape, but clearly not in size.
The 'easy finance' of the years that preceded the Great Recession saw a rapid accumulation of household and corporate debt in both the two poorer areas. And one consequence of this debt build-up was a pronounced rise in housing prices (Figure 3), which stood in sharp contrast to what happened in Northern Europe. 9 In Eastern Europe, some of this may have reflected the artificially low prices that characterized the early years of transition (Huynh-Olesen et al., 2013), but some was, no doubt, speculative, as was clearly the case in Southern Europe. The burst of the resulting bubbles was thus inevitable and strongly contributed to the double-dip recession of the South and to the deep 2009 recession in the East. In fact, the Eastern bubble was even greater than the Southern one and the subsequent slump was of roughly equal magnitude. It should be noted, however, that a good deal of this was due to what happened in Lithuania where house prices in nominal terms were multiplied by five between 2000 and 2007. 10 Excluding Lithuania a bubble is still evident, but it is no more pronounced than the one recorded in the Southern Eurozone countries. Interestingly, house prices are moving up again in the East, while they are still declining in some Southern countries. While South and East saw broadly similar private sector debt trends, the levels which debts reached were very different. Thus, when the 2009 recession forced adjustments throughout Europe, Eastern households and corporations, thanks to much lower indebtedness, found themselves under lesser pressure to retrench. Something similar happened in the public sector. Public debt/GDP ratios had been declining in both South and East prior to 2008, but then grew quite rapidly in both areas as a consequence of the recession. In many ways, this was, of course, a welcome development since the public sector sustained activity. In Southern Europe, however, debt ratios soared to 120 per cent of GDP, while in the Eastern group as a whole, even at its peak, the debt ratio never reached the 60 per cent Maastricht Treaty benchmark. Given such lower indebtedness, East European governments were never forced into the sharp austerity that befell Italy, let alone Greece, Spain and Portugal. The governments of these four Southern countries saw a swing of their structural balance equivalent to more than 6 percentage points of GDP between 2009 and 2015 (as against the less than 2 percentage points change in the Northern Eurozone) (IMF, 2019). There was austerity in the East as well in these years, but, thanks, in part no doubt, to lower debt levels, the government's structural balance improved by 3½ percentage points of potential GDP, a much lower figure than the one recorded in the South.
In summary, while debt trends in Eastern Europe were not that different from those of Southern Europe, debt levels were very different. Some of this may have reflected structural features. Emerging markets, as those of Eastern Europe have been since the 1989 fall of the Berlin Wall, have less developed financial institutions and systems. This, in turn, limits debt levels (even if Western banks tried aggressively to overcome such constraints). Prudence, perhaps, may also have contributed. Governments were keen to limit public debt because of impending entry into the Eurozone, while private agents may have been reluctant to incur excessive debts in a world still influenced by behaviour inherited from the communist past.
Be this as it may, Eastern Europe was able to avoid the pressures of financial markets which, in the West, threatened the Southern countries. While the spread between domestic and German interest rates on long-term public bonds reached sky-high levels in the South in the years 2011-13, 11 it was at significantly lower levels in the East and declined from some 4¼ percentage points in 2009 to less than ½ per cent in 2017-18, 12 as against the 2¼ figure still recorded in the South in these two years. Financial markets clearly felt that public budgets were in good hands and that defections from the euro were unthinkable, in contrast to the widespread fears of, for instance, a socalled 'Grexit' in the years 2011-12, or of a possible Italian defection in 2018. Not surprisingly, therefore, the unfavourable effects of debt accumulation on macroeconomic developments were much less pronounced.
A second set of problems which had bedevilled the Southern countries of the Eurozone has been a loss of external competitiveness. Labour productivity in Southern Europe has hardly grown since 1999, in contrast to what was happening in the Northern countries of the area (Figure 4a). Conversely, wage growth was relatively rapid (Figure 4b) reflecting, in the early years at least, the effect of large capital inflows boosting non-tradeable activity (residential construction in particular) and non-tradeable prices (Gabrisch & Staehr, 2015). Partly as a consequence, EU Commission estimates of intra-Eurozone real exchange rates using unit labour costs as a deflator show a massive appreciation between 1999 and 2009-11 ( Figure 5). Since then, the area's very subdued growth has imposed a measure of 'internal' devaluation, but the gap in the real exchange rate between the North and the South of the Euro area in 2018 was still of the order of 9 percentage points compared to 1999. OECD data, calculated on a similar basis, show an even larger gap of nearly 15 percentage points. 13 Unit labour costs are an imperfect measure of competitiveness since non-price factors are an increasingly important determinant of international trade flows (Storm & Naastepad, 2015). But Southern non-price competitiveness was almost certainly also eroded by a concomitant shift in employment and value added towards lowtech and medium-low tech sectors in manufacturing (ibid.), in contrast to what had happened in the ten years that preceded monetary union. Table 2 shows how in the 1990s it was the North which showed very little structural change towards R&D-intensive activities while the South did better. 14 It is plausible to think that the uncompetitive real exchange rates that the Northern countries faced in the 1990s and the Southern countries more recently, squeezed profits and, therefore, slowed down investment and the upgrading of production. Overall, therefore, the South may have lost competitiveness on both the cost/price and non-price front, as indirectly indicated by OECD estimates of gains and losses in market shares for exports of goods and services. These show the four Southern countries gaining such shares in the 1990s, but losing them massively in the next decade (Table 3).
Eastern Europe's Eurozone seems to have broadly avoided both these problems. It is true that wages have grown since 2009 at a rate well above that of both the Northern and Southern Eurozone areas. Productivity, however, has also grown rapidly. Hence, the EU Commission's real exchange rate estimates do not suggest that competitiveness was greatly impaired until 2015-16. In fact, there was significant depreciation between 2009 and 2012. And structural change seems also to have been quite rapid, thus further helping external competitiveness ( Table 2). The share of R&D-intensive activities in total value added rose sharply both in the early years of the century and even more so after 2009. A favourable real exchange rate from 2009 may have played some role in these developments. And the OECD estimates of changes in export performance mentioned above show large gains in export market shares since 2009. A summary judgment would suggest that, compared with what happened in Southern Europe, Eastern European firms were able to keep their unit labour costs broadly in check at least initially. Recent, or impending, membership of the Monetary Union appears to have represented a credible threat (or constraint) on company and trade union behaviour.
This suggests a more general conclusion. Planning a difficult entry into the euro area for the Southern countries that did not meet all the conditions stipulated by the Maastricht  Treaty, concentrated the minds of politicians and government officials in the late 1990s. Efforts were thus made to curb public expenditure and debt and to maintain wage moderation. Something similar may have been at work in Eastern Europe, particularly in the area of fiscal policy (Bökemeier, 2017). Slovakia and Slovenia succeeded in meeting all the Maastricht criteria before entry, while the Baltic countries followed an even more austere set of policies while shadowing the Eurozone from the late 2000s. Though output fell by nearly 15 per cent in 2009 in all three of them, against declines of less than 5 per cent in the Northern and Southern Eurozones, none of them resorted to devaluation, despite encouragements from some well-known Western economists (e.g. Krugman, 2008). The employment costs were huge, but the credibility gained on financial markets must have helped the subsequent adjustment.
In the Southern countries once entry had been achieved, some of the earlier efforts gave way to a more relaxed attitude. While before entry and in the early years of monetary unification IMF data suggest that the Southern Eurozone's real effective exchange rate declined by some 12 percentage points, subdued wage developments gave way to faster pay inflation thereafter, while productivity virtually stopped growing. Similarly, public debt rose again in some countries and private sector discipline was impaired.
So far at least (though, admittedly, the period is much shorter), there seems to have been less backsliding (or post-euro reform fatigue) in Eastern Europe than there was in Southern Europe. Public and private debt levels, in particular, are contained and mostly declining. External competitiveness is, however, more of an issue, at least in the three Baltic countries. EU data show that since 2012 the area's real exchange rate has risen by nearly 10 percentage points, bringing it well above its 2009 starting point. OECD calculations of real exchange rates on the same (unit labour costs) basis show a slightly faster appreciation since 2012, but a level in 2018 that is unchanged from 2009. Both sets of calculations suggest that Slovakia and Slovenia have broadly maintained reasonable levels of competitiveness, but this is much less the case in Estonia, Latvia and Lithuania where wage growth has been particularly rapid. If Southern experience is any guide (and absent reform of the Eurozone's present rules), this may eventually force the authorities in these countries to engineer some form of 'internal' devaluation, which could, in turn, slow down what has so far been a successful convergence process.

Institutional quality
A further startling difference between the Northern and the Southern Eurozone can be seen in the broad area of institutions and governance standards. At the institutional level, the 'Varieties of Capitalism' literature (Hall & Soskice, 2001) which stresses the importance of institutional complementarities in strengthening countries' economic performance, had already suggested that Central-Northern Europe was better placed than Southern Europe in facing the problems posed by globalization. Looking at Eastern Europe in this context, a recent investigation found that the five Eastern Eurozone members, as well as several other Central and Eastern European economies, had since the mid-1990s been moving towards institutional set-ups closer to those of the socalled 'Liberal Market' (or Anglo-American) and 'Coordinated Market' (or Continental European) economies, while eschewing the less efficient so-called 'Mediterranean' model (Ahlborn et al., 2016). This may well have facilitated integration into the single currency bloc.
Turning to governance, earlier research had shown that for a variety of indicators preexisting gaps between a relatively transparent and efficient North and a less open and transparent South had actually increased rather than decreased since the inception of monetary union (Boltho & Carlin, 2013). And, similarly, progress in reforming and liberalizing the Southern economies, progress that had been rapid in the 1980s and 1990s, had stalled since the Eurozone was formed. Thus, the gap between best practices in the North and South (proxied by the so-called 'Freedom Index' published by the Canadian Fraser Institute), 15 rather than continue shrinking, as one might have expected now that the Southern economies were even more closely tied to their Northern partners, seems, in fact, to have widened ( Figure 6). A more detailed analysis had shown that in the early 2000s there had been some structural reform in the product markets of the Southern countries, but none in their labour markets (Alesina et al., 2008), just when the abandonment of the exchange rate instrument would have called for greater employment flexibility. While it is rather unlikely that these unfavourable developments were actually induced by Eurozone membership, they may well have contributed to the poor performance of the Southern countries over the last two decades. In this broad area too, developments in Eastern Europe appear to have been very different from those in Southern Europe. At the beginning of the millennium, the Eastern European economies were, not surprisingly, more regulated and distorted than their Southern counterparts. Undoing communism is, after all, a slow and time consuming process. Only a decade later, however, the five Eastern countries could boast of levels of the 'Freedom Index' that were higher than those of Southern Europe. And in more recent years the indicator has risen further in the East to levels that are now close to those of the North (and marginally higher than those reached by the other six Eastern EU members). In the South, by contrast, the gap with Northern better practices has hardly narrowed.
Turning to the six broader indicators of governance calculated by the World Bank 16 yields even more startling differences. At the outset of monetary union, Southern Europe lagged Northern Europe in each one of them. Nearly two decades later, these gaps, rather than shrinking, have increased further in five out of the six dimensions. 17 Particularly striking are the trends in the presence (or absence) of the Rule of Law and Control of Corruption indicators. These have been shown to be particularly important when looking at past growth performance in rich and poor countries respectively (Aixalá & Fabro, 2008). While Northern Europe saw stability, or only slight regression in these two indicators between 1998-2000 and 2017-18, Southern Europe witnessed pronounced declines from the early 2000s onwards. Here too, the Eastern European experience is strikingly different from the Southern one. The East lagged behind the North in 2009-10 in five of the six indicators, in some cases by large margins. By 2017-18, five out of these six lags had declined. 18 And in five out of the six dimensions the East had levels that were superior to those of the Southern countries, even though at the beginning of the century, it stood significantly below the South in all of them.
Worthy of note in this context is the experience of the already mentioned six other Eastern European countries that are members of the EU, but not of the Monetary Union (Bulgaria, Croatia, the Czech Republic, Hungary, Poland and Romania). In all the six dimensions these countries have levels of governance that are similar to those of the four Southern Eurozone members and significantly lower than those of the five Eastern Eurozone members. Interestingly, progress was particularly rapid in these latter countries in the early years of the century, but there were no relapses once Eurozone membership had been achieved, unlike the Southern experience (Figure 7). Elsewhere in Eastern Europe improvements alternated with setbacks, particularly so in the recent past in Hungary and, to a lesser extent, in Poland.
Two further indicators may also be considered important: the extent of corruption and the size of the underground economy. Both these are often thought to be very present in former communist states, particularly those, as the Baltic countries, still closely linked to Russia. And indeed, in 1999, perceptions of corruption, as measured by Transparency International, were much higher in the Eastern Eurozone members than they were in their Southern (let alone Northern) counterparts, while the extent of the shadow economy (at close to a quarter of GDP) was very similar (Medina & Schneider, 2018). By 2015 the size of Eastern Europe's underground economy was (quite significantly) below that of the South, as were, by 2018, Eastern perceptions of corruption. For both these indicators too, the five Eurozone countries have lower values than those recorded by the other six Eastern EU members.
A final indicator is a synthetic index of the overall 'Quality of Government', produced by the University of Gothenburg in collaboration with the EU Commission, and which uses many of the variables mentioned above (Dahlberg et al., 2018). This broadly confirms the findings that have been reported so far. Between 2009 and 2018, while the overall 'Quality of Government' remained broadly stable in both Northern and Southern Europe, it improved in the Eastern European Eurozone to a level now marginally superior to that of the South and clearly superior to that of the other Eastern EU members.
Did all this contribute to the successful convergence of the five Eastern countries? One possible channel of transmission could have come through inflows of FDI. Good governance is likely to attract footloose multinationals. And Eastern Europe has been highly successful in attracting FDI. UNCTAD data show that between 1999 and 2018, the five Eastern Eurozone countries (but also the six other Eastern countries that are members of the EU) saw FDI inflows equal to 3½ per cent of GDP, as against figures of not quite 3 per cent in the Northern Eurozone and only 1¾ per cent in the Southern Eurozone.
More generally, theory would suggest that better governance (in the form, for instance, of lesser corruption, presence of the rule of law, administrative efficiency, some liberalizing reforms, etc.) should contribute to higher levels of trust and a better overall economic performance. And, as illustrated above, there is a clear correlation between worsening standards in the South and improving standards in the East, with divergence and convergence respectively. Causation could, of course, go both ways. Was it planned Eurozone membership that spurred reforms and governance improvements, or was it the pre-existence of relatively high standards that convinced politicians that integration into the single currency was not only desirable, but also feasible? It may well be impossible to answer this question and there are probably arguments in favour of both hypotheses. Be this as it may, governance seems to have improved rapidly in the five Eastern countries here looked at, be this from the beginning of the century or from the time these countries were either joining, or planning to join, the Eurozone, in clear contrast to the trends witnessed in the South and also elsewhere in Eastern Europe.
Why should this have been? There is no obvious answer to the question. If anything, the legacy of over 40 years of communism and the plunge into an unfamiliar market economy, accompanied initially by huge recessions and hyperinflation, might well have led to the destruction of whatever trust and social capital existed in these societies, leading to a worsening in economic performance and in governance standards. Yet this need (and may) not have happened. Evidence for East Germany, for instance, shows that for a number of indicators often deemed to be proxies for the presence of social capital, East German behaviour in the 2000s is not inferior (and is, at times, superior) to that of West Germany, despite the 40 years of communism. The most plausible explanation for this finding relies on the country's pre-communist history (Boltho et al., 2018).
Could history also throw some light on the governance performance of the five Eastern economies here considered? Unfortunately, data on governance do not seem to be available for the pre-war years of these countries (two of which did not even exist in their present form before the early 1990s). Yet, indirectly, some indication of, for instance, the presence or absence of corruption, can be gleaned from data on educational achievements. Comparative evidence shows that mass education clearly reduces over time the perceived incidence of corruption (as measured by Transparency International) (Uslander & Rothstein, 2016). Comparable inter-war data on educational achievements are not easily available. Using a UNESCO study (UNESCO, 1957) on illiteracy (and making a number of adhoc adjustments and a few heroic assumptions), suggests that inter-war illiteracy (as a per cent of the over 10-or over 15-years old population) was, not surprisingly, very low in the countries that are today part of the Northern Eurozone and much higher in the Southern countries. What is perhaps less expected is that its level in Southern Europe may have been nearly 3 times higher than it was in the five Eastern countries here considered, 19 and this whether one uses 2016 GDP weights, 1930 population weights or unweighted averages. In this area too the six other Eastern countries, which are now EU members, lagged quite substantially behind the literacy levels achieved in what are now the Eurozone members.
There is also some evidence on the presence (or absence) of democracy between 1920 and 1938. This shows that the democratic performance of the Northern countries was significantly superior to that of both the Southern and Eastern countries, despite the turn to dictatorship in Germany and Austria in the 1930s (Figure 8). Equally, however, the score of the five Eastern countries is significantly above that of the four Southern countries (Center for Systemic Peace, 2018) (and also of the other Eastern countries that have joined the EU). This result uses inter-war Czechoslovak and Yugoslav data as proxies for Slovakia and Slovenia respectively. The democratic performance of Slovakia since the 'Velvet Revolution' is very similar to that of the Czech Republic, suggesting that the proxy may not be unreasonable; the performance of Slovenia since the break-up of Yugoslavia, on the other hand, has been much superior to those of Bosnia-Herzegovina, Croatia, Macedonia, Montenegro and Serbia. Had this also been the case in the inter-war years (and the data on illiteracy in Slovenia at the time would support such a conclusion), and had Slovenia been then independent, the gap between East and South in the 1920s and 1930s would have been even higher. 20 It could, of course, be argued that the democratic performance of the inter-war years has little bearing on contemporary standards of governance in countries that went through nearly half a century of communist dictatorship. Yet, the restoration of democracy in all the East European countries after the fall of the Berlin Wall was rapid and thoroughgoing. Using Freedom's House measure of democratic achievement shows a performance in recent years that is fully comparable to that of the Northern countries, with the Eastern Eurozone doing significantly better on this score than the South and the other six Eastern European countries not members of the euro, repeating, in other words, the pattern already seen for the inter-war period.
What these very tentative findings suggest, in other words, is that the Eastern Eurozone may have had governance standards already before the war that were superior (and, probably, quite significantly so) to those of the Southern Eurozone. This, in turn, could have contributed to its superior economic performance in recent years.
Alternatively (  wanted to avoid pressures from financial markets, let alone the bail-out loans and rescue packages by the European institutions and the IMF (and the attached draconian conditions). For Slovakia, in addition, successful Eurozone membership also ensured that its position as a major car-assembly workplace remained entrenched while for Slovenia it meant further detachment from the more chaotic situation prevailing in the rest of the ex-Yugoslavia. As for the Baltic States, any established institutional link with Western Europe was felt to give an additional degree of security to countries bordering the Russian bear.

Conclusions
A stark and simple conclusion emerges from the foregoing: the economic performance of the Eurozone's Eastern European member countries over the last decade was significantly better than that of its Southern member countries. Living standards in the latter diverged from those of the richer North, while living standards in the former continued to close their (admittedly large) gap with Northern Europe. External competitiveness weakened in the South while, at least initially, it strengthened in the East. And governance standards, very divergent from those of Western Europe after the fall of Communism, have rapidly converged to the higher standards of Northern Europe, in contrast to what happened in the South of the Eurozone. It is interesting to note in this context that the other Eastern EU members which have not so far adopted the euro have shown less progress in these various governance areas. Was it better governance that prompted membership by some countries but not others (a self-selection process, in other words), or was it membership that imposed improvements? Both forces were probably at work.
The rapid trade and financial integration with the West which all these countries pursued after the fall of Communism may well have been an important reason for both economic and institutional convergence. All shifted, at times drastically, their foreign trade away from the COMECON block and towards Western Europe. All wanted to attract FDI and on the whole succeeded in this (spectacularly so in the case of Estonia whose FDI inflows between 1999 and 2018 reached nearly 7 per cent of GDP).
One possible major reason for the differences in performance has been hinted at above. Countries that aim to qualify for monetary union will have to perform well on the Maastricht metrics to be allowed entry. Hence the three Baltic States in particular, which only joined between 2011 and 2015, had strong self-imposed limits on their public finance and inflation behaviour. Shadowing the Eurozone in order to achieve the conditions for entry is possibly more constraining than membership itself. Since entry, one might have expected some backtracking, as occurred in, for instance, Greece and Italy. And indeed traces of backtracking have appeared, at least in the three Baltic countries as far as wage developments are concerned, but, otherwise, euro-compatible behaviour has been broadly preserved. It is quite likely that the Eastern countries learned from the demonstration effects of what was happening in the South. Seeing Greece, Portugal and Spain (as well as Cyprus and Ireland) being rescued by outside intervention and Italy suffering from huge financial market pressures, may well have persuaded East European governments to pursue relatively orthodox policies and to continue with structural reforms.
A further and more political reason could also be tentatively advanced. All the former communist economies of Eastern Europe were, after the collapse of the Soviet Union, very keen to loosen their ties with Russia and strengthen their links with the West. European Union membership was obviously one step in this direction. Membership of the monetary union was seen as a further and very crucial step showing this commitment. Hence a greater willingness to adopt tough policies to secure entry, but also to maintain, once inside the club, the rules and behaviour that were expected. This would seem to be particularly true for the three Baltic countries (Staehr, 2015) which, for the sake of maintaining their pegs to the euro during the late 2000s, accepted huge recessions and massive increases in unemployment.
Finally, history may also have played a part. High standards of governance must surely be an important element in ensuring that countries hold onto economic commitments and pursue behaviour compatible with monetary union. In most areas these standards are, today, higher in the five Eastern countries than they are in the four Southern ones. A not implausible reason for this difference could be that already in pre-war times Slovakia and Slovenia (thanks possibly to their Austro-Hungarian heritage) and the Baltic states (thanks possibly to their cultural proximity to Scandinavia and Finland and/or, in the cases of Estonia and Latvia, to their Hanseatic past), were more open and transparent than were the four Southern countries (and possibly also than some of the six Eastern EU but non-euro countries). Data limitations prevent a full investigation of this issue, but the evidence on educational standards and on the presence or absence of democracy in the inter-war period, does point in this direction.
Notes 10. One possible reason for this may have been the very high (70 to 90 per cent) proportion of mortgage loans contracted at variable rates in the country [Rubio & Comunale, 2018]. The substantial fall in long-term interest rates, common to all the five Eastern Eurozone members between the early 2000s and 2008, may thus have had larger effects on mortgage borrowing in Lithuania. It should be noted, however, that more detailed research argues that house prices at their peaks were above the levels suggested by the fundamentals not only in Lithuania, but also in Estonia and Latvia [Huynh-Olesen et al., 2013]. 11. Greek 10-year bonds yielded 37 per cent in early 2012, giving a spread of more than 35 percentage points over equivalent German bonds; for Portugal, the maximum spread was 12 percentage points, while for Italy and Spain it was of the order of 5½ percentage points. 12. Excluding Estonia, for which data after 2009 could not be found. 13. The IMF's estimates of unit labour cost-based real exchange rates suggest that the gap between Northern and Southern Europe at the end of 2018, relative to 1999, was significantly larger (some 25 percentage points); data on this basis are, however, not available for the East European countries. 14. Column 3 in Table 2 stops in 2015 because almost no data are, as yet, available for more recent years. 15. An index that tries to measure the various ways in which governments intervene in an economy. 16. These are: 'Voice and Accountability; Political Stability and Absence of Violence; Government Effectiveness; Regulatory Quality; Rule of Law; Control of Corruption' [World Bank, 2018]. 17. The only exception is provided by the 'Political Stability and Absence of Violence' indicator which declined sharply from 2016 onwards in Northern Europe in the wake of several terrorist attacks. 18. The exception is the 'Regulatory Quality' indicator. 19. Slovakia and Slovenia did not exist at the time. Separate regional data were found for Slovenia (at the time a part of Yugoslavia); these show illiteracy rates significantly below those of the other regions of the country. Data for Czechoslovakia were used as a proxy for Slovak data, a procedure that is unlikely to bias the results unduly. 20. This broad conclusion is also supported by a further comparative and quantitative study on the presence of democracy over time (Vanhanen, 1984). While the study does not cover the Baltic States, it shows that the democratic performance of Czechoslovakia, but also of Yugoslavia, in the inter-war period was much superior to that of all the four Southern European countries here considered (or, for that matter, of the other Eastern countries members of the EU but not of the Eurozone).

Acknowledgement
The author is grateful for very valuable assistance and helpful comments to Pasquale Scaramozzino, as well as to Wendy Carlin and participants in a 2018 AIESSEC conference in Turin and a 2019 SSEM Conference in Milan. Extensive comments by two rather critical anonymous referees have also greatly improved the text.

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