EU Aggregate Demand as a Way Out of Crisis? Engaging the Post‐Keynesian Critique

Post&#8208;Keynesians have delivered an important advance in providing explanations of the Eurozone crisis, not the least in demonstrating how the formation of the European integration project lacked the means to manage effectively the macroeconomic imbalances between core and peripheral spaces across the region. Through a critical engagement with such descriptions, this article argues that to account more adequately for the formation of the asymmetrical and crisis&#8208;ridden forms of development across the Eurozone it is necessary to focus on the uneven and combined development of Europe's peripheral spaces and their integration into an expanded free trade regime since the 1980s. It is through a focus on the structuring condition of uneven and combined development shaped by capitalist social relations of production and attendant class struggles that we can locate the origins of the present crisis.

There is general agreement that the Eurozone crisis since 2010 has had a devastating impact on countries in the European periphery, such as Greece and Portugal, expressed in record levels of unemployment, social inequality and widespread social deprivation. There is, however, no agreement on the causes of the crisis or on how the crisis can be resolved. One set of lay explanations for the Eurozone crisis may focus on cultural arguments. Apparently, 'lazy Greek workers' have been put forward in some circles as the cause of the crisis, even though Greeks are amongst those who work the most hours in Europe (McDonald, 2012). 1 Another, more scholarly, set of explanations emphasizes the way different national institutional set-ups and growth models have been brought together in the Economic and Monetary Union (EMU). The crisis 'originated in the structural strains generated when different types of political economies were joined in a currency union' (Hall, 2012, p. 357). Export-led growth models of Germany and other northern European states, built on high levels of wage co-ordination, benefited from monetary union, while countries with demand-led growth strategies and no wageco-ordination ran into trouble (Hall, 2014(Hall, , p. 1228). 2 In short, the EMU's institutional design is regarded as causative in the incompatibility of different national political economies. Yet another set of explanations argues that unsustainable capital flowsin particular moments of capital flight 'from a country's banks and sovereign debt markets' (Jones, 2015, p. 822)are the major cause of the Eurozone crisis. Greater controls of capital flows are then the logical solution to the problem (Schelkle, 2017, pp. 314-315). While the cause of the crisis is here identified in relation to capital flows rather than institutional incompatibilities, the main focus remains on analysing different national political economies in an additive and externally related manner rather than as part of a larger whole.
A major, yet less often discussed, contribution to the question of the causes of the Eurozone crisis has been made by a range of post-Keynesian scholars together with some critical historical materialists such as Costas Lapavitsas. Unlike the above literature that tends to focus on diverse national institutional configurations, these scholars emphasize unevenness across the European political economy, as a whole, from the onset of EMU in 1999. National political economies are, thus, not regarded as distinctive, separate entities but as closely related parts within a larger whole. The purpose of this article is first to engage critically with this contribution to our understanding of the crisis. Thus, in the following section, we outline and then critique the post-Keynesian account. 3 Second, we present an alternative historical materialist explanation emphasizing how the roots of the crisis go much further back, immediately to the very moment when countries such as Greece and Portugal joined the EU during the 1980s and indeed before. Rather than reducing the causes of unevenness within European political economy to the institutional set-up of the EMU, our argument demonstrates that the uneven and combined development of capitalism has always constituted the geopolitical dynamic of the European integration project. This is because within historical and contemporary processes of capitalist expansion there can be no 'complete parallelism' of economic, social and political developmentor a situation of 'even' developmentbecause of the permanency of different structures and dynamics of state formation, labour movements, and the relationship of forces between classes (Mandel, 1967, p. 155). Most acutely, uneven development pertains to the very expansionary essence of capital and the existence of competition between 'many capitals' searching for surplus profit and increases in the rate of surplus value (Mandel, 1975, p. 27). Our argument, then, shares the view that uneven and combined development is a historically specific geographical expression of capitalism and a necessity within capital accumulation (see Smith, 1984Smith, , pp. 4, 202, 1986. But the contribution of what follows hones this insight to make the point that uneven development is very much the hallmark of geographical space resulting from the necessity of capital accumulation within the EU, constituting a regional contradiction of both expanded equalization and sharpened differentiation in conditions of production.

I. Emphasizing the Importance of Aggregate Demand: The Post-Keynesian Assessment
From the end of the economic crisis in the early 1990s and especially in the first half of the 2000s, GDP growth rates had been comparatively high in Greece, Ireland, Portugal and Spain and often higher than in Germany as well as the Eurozone and EU averages (Bieler and Morton, 2018, p. 223). As a result, when looking at the volume index of GDP per capita in purchasing power standards, as expressed in relation to the European Union (EU28) average set to equal 100, by 2006 Greece had almost caught up with the EU average at a level of 96, with Portugal at a level of 83 having done slightly less well (Eurostat, 2017). However, this growth proved to be unsustainable. When the global financial crisis struck, economic growth quickly vanished because it was driven largely by debt that could no longer be accessed or financed (Hart-Landsberg, 2016, p. 2). By 2016 purchasing power standards in Portugal had fallen back to 77 and in Greece even further to 68 compared with the EU average, set at 100 (Eurostat, 2017). Post-Keynesian analyses tend to identify the institutional set-up of EMU as the main cause of the crisis and thus uneven development across the European political economy. 'EMU came with an economic policy package that is rule bound and has proven exceptionally dysfunctional during the crisis' (Stockhammer, 2016, p. 366). The related undermining of national policy autonomy in responding to economic crises is singled out as especially problematic (Stockhammer and Köhler, 2015, p. 38). With exchange rates fixed between countries as a result of the common currency and national fiscal policy severely restricted within the Stability and Growth Pact, the only way to increase competitiveness has been downward pressure on wages and work-related conditions (Lapavitsas et al., 2012, p. 158). Thus, the EMU was by default designed from the beginning as a model that would struggle to generate enough demand and almost inevitably relied on financialization and the creation of national and personal debt for economic growth: This has resulted in two distinct growth models, which are both unstable: debt-driven growth and export-driven growth. Both allow for growth, but are intrinsically unstable, because they require increasing debt to income ratios. In the case of the debt-driven model it requires domestic debt; in the case of the export-driven model it requires foreign debt of the trade partners. It is these rising mountains of debt that erupted in the crisis (Stockhammer, 2016, p. 369).
Hence, the main problems of EMU are understood to be the result of insufficient demand and, in particular, the asymmetries in the formation of aggregate demand across the European political economy shaped by the institutions of EMU (Patomäki, 2012, p. 79).
The uneven dynamic between the two different models is reflected in trade statistics. Export-driven growth models such as Germany have experienced an export boom in recent years, with almost 60% of its exports going to other European countries (Trading Economics, 2013). However, such a growth strategy cannot be adopted by all states equally. Some countries also have to absorb these exports. This is what many of the peripheral countries now in troublesuch as Greece, Portugal, Spain and Irelandhave done (Laskos and Tsakalotos, 2013, p. 86). The large profits resulting from German export success, of course, needed new fixes of investment to generate more profits. State bonds of peripheral countries seemed to be the ideal investment opportunity with guaranteed profits, backed by sovereign states. Thus, 'Germany has been recycling its current account surpluses as FDI [foreign direct investment] and bank lending abroad' (Lapavitsas et al., 2012, p. 31; see also Stockhammer and Köhler, 2015, p. 42). In turn, these credits to the periphery were used to purchase more goods in the core or to invest in non-tradable sectors such as construction. 'Trade imbalances have fuelled credit growth and bubbles in the credit-led economies as the current account surpluses of the export-led economies were flowing to credit-led economies and providing further liquidity' (Stockhammer, 2011, p. 10). Ultimately, post-Keynesians contend that the creation of credit links back to increasing income inequality over past decades. Seeing their relative income decline, poorer households increasingly had to rely on credit to maintain living standards. In aggregate terms, 'given the rise in inequality the credit expansion in the personal sector were both necessary for supporting aggregate demand and employment and at the same time unsustainable' (Sturn and van Treeck, 2013, p. 126). The introduction of the Euro and the related low interest rates in peripheral countries facilitated this financialization of the European political economy, as it made cheap credit more available. Peripheral states, unable to compete with the productivity levels and strong export performance in Germany, ended up with large current account deficits. In the long run, such development strategies based on capital inflows were unsustainable.
When it comes to potential solutions to the crisis, wage growth is regarded as essential for the stimulation of aggregate demand (Lavoie and Stockhammer, 2013, pp. 4-5). 4 'We argue that an active co-ordination of wage policy has to be at the centre of a new policy mix. This requires institution building and a very different role for labour unions in economic policy making' (Stockhammer, 2011, p. 5). Considering unevenness within the regional political economy of Europe, emphasis is placed on wages growing in line with productivity but the increase, it is argued, needs to be above productivity growth plus inflation in the export-oriented countries in general (Stockhammer, 2011, p. 19)and particularly in Germany (Stockhammer, 2016, p. 375)in order to re-balance matters. A post-Keynesian proposal in this respect is the 'European wage standard', based on national collective bargaining that is co-ordinated at the European level and intended to cause a convergence at the regional level of national wage shares, and, thereby, halt falling wage shares. As argued by Emiliano Brancaccio, 'above the minimal level of growth, the standard would link the growth of nominal wages with respect to labour productivity to the balance of trade so as to foster a return to equilibrium between countries with trade surpluses and those with deficits' (Brancaccio, 2012, p. 56). This focus on demand management through wage policy is complemented by demands for restructuring and shrinking the financial sector, a wealth tax, mechanisms to redistribute wealth across the various regions of the EU as well as a fiscal policy with the goal of ensuring full employment (Stockhammer, 2016, pp. 375-376). The vision for the European future is clearly a federalist solution with a common economic and wage coordination policy. Finally, it is argued from a post-Keynesian perspective that one solution to the Eurozone crisis lies in the embedding of a democratic global Keynesianism. As Heikki Patomäki states, 'global Keynesianism aims to regulate global interdependencies in such a way as to produce stable and high levels of growth, employment and welfare for everyone and everywhere, simultaneously' (Patomäki, 2012, p. 175). A wage-led recovery should be based on a 'Global Keynesian New Deal', including a re-regulation of the financial sector, the reorientation of macroeconomic policies in export-led growth countries towards a stabilization of domestic demand and, third, 'the reconstruction of international macroeconomic policy co-ordination and a new world financial order' (Hein and Mundt, 2013, p. 175).
Nevertheless, there are at least two key flaws in this post-Keynesian assessment. First, post-Keynesians assume that capitalism can overcome crises successfully, if the right state policies are adopted. This overlooks the persistence and necessity of uneven development in capital accumulation and, specifically, the regional political economy and geographical space of Europe. As Nicos Poulantzas (1978, pp. 95-97) reminds us, the social formation is the nodal point of the expanded reproduction of capitalism marked by specific historical conditions within regional conditions of uneven development, which means that modern capitalist space acquires a different meaning than it had in the past, albeit one still crystallized by the state. Organized around wage labour and the private ownership of the means of production, capitalist social relations of production are enormously dynamic, as both labour and capital have to reproduce themselves through the market. While workers compete with each other to sell their labour 'freely', capitalists are in constant competition with each other over profitability and market share. 'Under free competition, the immanent laws of capitalist production confront the individual capitalist as a coercive force external to him' (Marx, 1867, p. 381). While capitalism is very dynamic, however, it is also crisis prone. The more goods that are produced, the more potential surplus-value is generated that needs to be realized, making it more difficult to bring together excess labour and excess capital in a fruitful way. Such capitalist disequilibrium, linked to the declining rate of profit and how that depresses productive investment and leads to both a crisis of over-production and a fall in the rate of capital accumulation, creates crisis conditions marked by the devalorization of capital (Marx, 1884, p. 545). In response, capital is in constant search for new markets in a drive towards outward expansion along uneven and combined developmental lines (Bieler and Morton, 2018, pp. 38-41). Therefore, whilst post-Keynesians correctly identify how the development trajectories of national political economies are further drawn together in uneven relations through the novel institutional architecture of the single currency, they overlook the fact that this has simply reinforced and exacerbated forms of unevenness that have characterized European political economy for much longer. Free trade policies, as initially embedded within the EU Customs Union since 1968 and then especially the Internal Market from the mid-1980s onwards, when free trade was extended from trade in goods to trade in services and finance, have generally tended to reproduce longstanding developmental asymmetries and polarizing conditions between countries. As demonstrated below in respective sections on Greece and Portugal, less developed countries remain locked into low value-added industrial sectors or even experience deindustrialization as industries in countries with lower productivity levels cannot compete with the higher productivity levels in partner countries, and which benefit disproportionately from trade. 'Unevenness is not … a result of market imperfections, but is in fact a product of the way competitive markets work in the real world' (Kiely, 2007, p. 18). Rather than simply focusing on the institutional set-up of the EMU as generative of unevenness across the European political economy, a return to the role of capitalist social relations of production and, therein, so-called free trade, in causing the structuring conditions of uneven and combined development is required (Bieler and Morton, 2014, p. 41).
The focus on the 'right' state policies indicates the second key problem of post-Keynesian analysts. They are in no doubt that the main cause of a decline in the share of wages in national income is the decline in the bargaining power of trade unions (Stockhammer, 2013). Therefore, to ensure wage growth, trade unions need to be empowered and play a clear role, together with strong employers' associations, plus government involvement if needed, in national wage bargaining (Hein and Mundt, 2013, p. 178). In other words, post-Keynesians invoke national institutions linked to the post-World War II class compromise that had seemed to guarantee steady wage increases and strong domestic demand levels in industrial countries such as Germany and Sweden during the 1950s, 1960s and 1970s. Such a technocratic vision, however, completely misses the dynamics underlying working-class gains in the post-1945 European political economy. Asbjørn Wahl (2011) makes clear that strong bargaining institutions were not the result of benevolence by employers or state policies but rather a result of class struggle. Against the background of the cold war and based on strong labour organizations at the national level, forged in industrial conflicts at the beginning of the twentieth-century, labour was able to balance the social power of capital in society. The result was an unstable equilibrium of compromises, in which capital retained the right to own the means of production in exchange for rising wages and an expansive welfare state. The Keynesian economic consensus at the time also underpinned European integration with its objective of obtaining social cohesion across the EU. Keynesians, too, understood that the market in itself will not lead to even development across Europe and, therefore, included the so-called Structural Funds as well as the Cohesion Fund as additional measures to support less developed regions. Nevertheless, against the background of global restructuring and the increasing transnationalization of production and finance, especially from the early 1980s onwards, the Keynesian consensus was replaced by variegated neoliberalism (Macartney, 2010). Capital, now organizing across transnational networks of production, renounced national class compromises based on its superior structural power vis-à-vis trade unions. At the European level, the emergence of a new compromise of 'embedded neoliberalism' occurred. This predominantly focused on neoliberal deregulation and liberalization as embodied in the Internal Market and EMU, while social policies were reduced to flanking measures (van Apeldoorn, 2002, pp. 158-189). When the EU was enlarged to include Central and Eastern Europe (CEE) in 2004, key redistributive policies were not fully extended to the new members (Bohle, 2006, pp. 69-74;Ó'Beacháin et al., 2012, p. 227).
Of course, neoliberal restructuring was variegated and differed from state to state within the EU. Nonetheless, in the end even Germany as the dominant economy was affected. The so-called Hartz IV reform of 2003 led to the largest cuts in, and restructuring of, the German welfare system in the post-war era (Bruff, 2010, pp. 414-416;Jessop, 2014, p. 252). Collective bargaining has become decentralized and ultimately eroded, and the labour market as such has been liberalized as a result of 'the manufacturing firms' efforts to cut costs and boost external competitiveness' (Baccaro and Benassi, 2017, p. 101). In fact, it had been the downward pressure on wages and working conditions of German workers that facilitated the German export boom (Lapavitsas et al., 2012, p. 30). Moreover, co-ordinated wage bargaining at the European level, as envisaged by the post-Keynesians, had already been attempted between 1999 and the mid-2000s but eventually failed, especially because it had been impossible to enforce the mechanism in Germany (Erne, 2008, pp. 86-90, 96-9). Some observers blame German trade unions for this, accusing them of prioritizing the interests of German workers at the expense of workers elsewhere. Indeed, in the so-called Pforzheim Agreement of 2004, German trade unions accepted the increasing use of opening clauses, allowing companies to deviate from collectively negotiated working conditions and wage increases on the basis of company level agreements between employer and trade union. Consequently, 'for the whole period from the adoption of the Euro in 2000 right up to 2012, the average negotiated salary increase in Germany was 5.5 per cent below productivity increase, and even more drastic the effective average salary increase was a further 9.3 per cent below the negotiated average salary' (Bieler and Erne, 2014, p. 162). Others, however, point out that this failure was to a significant extent because German trade unions were no longer able to match the power of capital. From the 1990s onwards, German production structures, too, had increasingly been transnationalized with Central and Eastern European states, especially, offering a skilled and yet comparatively cheap workforce. Moreover, the powerful metalworkers' union, IG Metall, which organizes workers in transnational manufacturing in Germany, lost a strike in 2003 over the equalization of conditions in the metal industry between East and West Germany, that is, precisely at the moment when the controversial Hartz IV reforms, mentioned above, were imposed on German workers. The increasing use of labour from temporary agencies by large companies during the 2000s further undermined the collective strength of German trade unions (Bieler and Erne, 2014, pp. 161-165). Unsurprisingly, 'Germany now has the largest low-wage sector in the Eurozone (22 per cent of all employees) and has experienced an above average rise in income inequality' (Lehndorff, 2015a, p. 160), underpinned in the current conjuncture by a resurgence of rules-based ordoliberal reasoning and domestic political pressures (Bulmer, 2014).
In short, the post-Keynesian proposal of reflating demand levels through wage increases underpinned by the strengthening of trade union rights goes completely against the dominant trend of neoliberal restructuring, based on changes in the balance of power in society in favour of capital over labour. As Stockhammer and Köhler themselves admit, the post-Keynesian 'suggestions have strong normative elements with little regard for the political preconditions of [their] implementation' (Stockhammer and Köhler, 2015, p. 45).

II. Uneven and Combined Development in the European Political Economy
Uneven and combined development between northern and southern European countries goes back much further than simply to the establishment of EMU in the early 2000s. For Poulantzas (1976, pp. 10-13), Portugal, Greece and Spain exhibited a specific historical form of state marked by dependent industrialization that was peculiar to these countries' emergence within the European arena of uneven development. From the very beginning of their EU membership -Ireland in 1973 and Greece, Portugal and Spain during the 1980sall four countries have been further integrated into asymmetrical relationships of uneven and combined development. Unevenness is ultimately based on different productivity levels and concentrations of capital (equalization) and while Germany is mainly involved in capital-intensive sectors of global value chains (GVCs), peripheral European countries have historically been strong in labour-intensive sectors (differentiation). For example, the Portuguese economy is characterized by lowtechnology, labour intensive production structures, often poorly organized and based on human resources with low levels of qualification (Rodrigues and Reis, 2012, p. 198). These features persist despite a sustained effort in the post-dictatorship period to converge on capital-intensive forms of industrial accumulation through strategies of 'catch-up'.
Having historically been leading sectors in Portugal's economy, an analysis of the Textile and Clothing (TC) and footwear industries, as well as motor vehicles, illustrates the difficulties faced in producing a shift in dominant accumulation strategies that could overcome historical forms of uneven and combined development across the geographical space of European political economy. Despite the TC and footwear sectors expanding during the 1980s and the early 1990s, this had little to do with a shift towards the sort of capital-intensive production found in Europe's core. After gaining unimpeded access to a larger European market through Portugal's accession to the then European Economic Community (EEC), the Small and Medium Enterprises (SMEs) that dominated this sector were able to exploit already very low labour costs alongside growing demand across consumer markets by functioning as subcontractors to larger, international TNCs, focusing on labour-intensive assembly of prefabricated parts. Far from a shift into more capitalintensive forms of textile, clothing and footwear manufacturing, these sectors instead became subordinated to the requirements of transnational capital that captively governs the relevant value chains of accumulation (Gibbon et al., 2008, p. 323). As for the motor vehicle industry, in 1995 an assembly plant was opened in Palmela near Lisbon as a joint venture between Ford and Volkswagen. This was used as irrefutable evidence of Portugal's convergence on more capital-intensive forms of accumulation. However, whilst 50 per cent of the components that were part of the finished vehicles were produced nationally, the remaining 50 per cent, mainly capital-intensive production components, were imported from other countries such as Germany, France and the UK. Further, many other highly profitable functions, such as product strategy, sales and marketing, and investment were controlled and carried out by divisional or head offices in the U.S. and Germany, respectively (Corkill, 1999, p. 163). Essentially, then, despite facilitating the type of large foreign direct investmentthe largest in Portugal's history to that pointthat was meant to provide the leverage to rapidly overcome historical forms of uneven development, the same configuration of hierarchy developed in the establishment of a new export-oriented production structure. As Marx (1894, pp. 364-365) forewarned, 'if capital is sent abroad, this is not because it absolutely could not be employed at home. It is rather because it can be employed abroad at a higher rate of profit'. In sum, pointing to ultra-modern advances in technology is never enough. This is because within the structuring conditions of uneven and combined development, a 'second wind' to accumulation on a world-scale can be derived from simultaneously reproducing new forms of both equalization and differentiation in production conditions. Despite Greece's policy programme of catch-up during the 1980s and especially in the 1990s, its economy experienced deindustrialization from the moment it joined the EU in 1981, unable to cope with the higher levels of competition within the EU free trade regime. This increased competition witnessed many large manufacturing firmsconstituting the core of Greece's industrial production structurego bankrupt (Giannitsis and Kastelli, 2014). Now, Greece's manufacturing sector is dominated by very small firms when compared with businesses in most other European countries (Linardos-Rulmond and Robolis, 2016, p. 10; Economakis et al., 2015, p. 432). An overview of Greek exports since the 1990s indicates the predominantly labour-intensive activities. In 1990, the top exporting industries in Greece were the TC and footwear sector (25.5 per cent of total exports), food products (17.9 per cent), chemicals and minerals (13.1 per cent), and metal products (7.5 per cent). By 2006, a significant shift had occurred, evidenced by the chemicals and minerals sector contributing to total exports with an increase to 16.2 per cent, followed by food processing (11 per cent), construction materials and equipment (11 per cent), and then the TC and footwear share of exports standing at only 8.7 per cent (Jordan, 2017, p. 136). This trajectory is only compounded by the fact that there has been no equivalent prospect of shifting to more capital-intensive production as had initially been hoped with the Palmela investment in Portugal due to a lack of significant foreign investment. In short, the dominant accumulation regimes that cohered in Greece's and Portugal's political economies have predominantly been based on labour-intensive forms of production.
This vulnerable position within Europe's political economy has become further precarious for Greece and Portugal as the deepening integration of emerging economies, especially China, during the 1980s and 1990s into the global economy, increased direct competitive pressures across key industries. In particular, the TC and footwear industries have lost market share to cheaper labour locations in Asia. Thus, the market share of Chinese goods in this sector in the EU15 increased sharply at the expense of several southern European countries, including Greece, though particularly Italy and Portugal (Leao and Palacio-Vera, 2011, pp. 11-12). For instance, Greece's share of the TC and footwear sectors in total exports fell from 25.5 per cent in 1990, to 8.7 per cent in 2006, without, as we have seen above, any significant growth in other sectors that would compensate for this shift. In the case of Portugal, heightened global competition precipitated a similar process of de-industrialization around 2000 as Greece had witnessed during the 1980s. Between 2000 and 2007 Portugal lost jobs in manufacturing at an average annual rate of 2 per cent. Approximately 60 per cent of the jobs lost were in the dominant export-oriented industrial sectors of Portugal's economy (Lagoa et al., 2014, p. 13). For example, the level of employment in the TC sector declined from 11.6 per cent of total employees in 1982 to 5.4 per cent in 2006. Demonstrating this trend further, 'textile importance in industrial exports declined from 19 per cent in 1982 to 8 per cent in 2006, while clothing exports, after increasing from 11 per cent in 1982 to 13 per cent in 1992, were to decline to 4 per cent in 2006' (St. Aubyn, 2014, p. 71) This was one of the fastest rates of deindustrialization in the EU (Lagoa et al., 2014, p. 13).
From the early 2000s onwards, even the prospect of shifting to more capital-intensive parts of production diminished. This is primarily a result of the end of the Cold War and the EU's eastward expansion to CEEa set of national political economies whose labour force was cheaper, more skilled, and, importantly, much less institutionally integrated into a system of collective bargaining than their Southern European counterparts. In fact, West European corporations have benefitted greatly from this unevenness between the old EU member states and the new members from CEE. Anticipating the enlargement of the EU in 2004, transnational capital reorganized value chains of accumulation to the new member states of CEE (Mamede, 2012, p. 38). By integrating suppliers from low wage CEE countries in their transnational production networks, corporations reduced their wage costs considerably through whipsawing tactics that put subcontractors and workers from different locations in competition against each other. German multinationals profited most from the availability of a cheap and well-trained industrial labour force in its Eastern hinterland. They established themselves predominately in the Czech Republic, Poland, Hungary, and Slovakia and used relocation threats to obtain concessions at home. 'The delocalization of manufacturing to emerging Europe has actually helped to create jobs in the home country by sustaining productivity in manufacturing, while contributing to the sharp fall in Germany's relative unit labour costs' (Simonazzi et al., 2013, p. 660). This increased their international competitiveness significantly, whilst also exacerbating unevenness of industrial competitiveness, and the inability to overcome it, between different regions.
This restructuring within European production networks and the eastward reorientation of German trade therefore came at a cost for Southern European countries, which were unable to find alternative export outlets and were increasingly unable to compete with China in their traditional production sectors. In response, a notable part of southern Europe's low-cost industry moved to the EU's South-East, where wage-levels are as low as in China (Stan and Erne, 2014). Between 2003 and 2008, the flow of Italian and Greek direct investments in Romania and Bulgaria (12.89bn US$), for example, exceeded German investments in these two countries (11.08bn US$) by far, especially if one bears in mind that the German GDP figures were about 40 per cent higher than the combined Italian and Greek numbers (United Nations Conference on Trade and Development, 2014). As a recent study confirmed, in order to avoid bankruptcy 'more than 12,000 firms (1.5 per cent of all Greek companies in 2012) have relocated from Greece to the Balkans and Cyprus during 2007-2012', and here especially Bulgaria (Kapitsinis, 2017, p. 703). As long as cheap credit within the Eurozone guaranteed the necessary aggregate demand despite ongoing wage moderation, the relatively high growth rates between the mid-1990s and 2007 in Europe's southern and eastern periphery seemed to indicate a "catching-up" process or equalization with core EU countries. The global financial crisis of 2008 rent asunder this illusion reinforcing the point of differentiation and that the 'determination of the state by the relations of production, or the primacy of struggles over the state, is inscribed in diverse temporalities and in historical forms marked by uneven development' (Poulantzas, 1978, p. 41).
In sum, despite post-Keynesian assessments, uneven development cannot simply be reduced to the institutional set-up of EMU and the introduction of the Euro in the early 2000s. Uneven development, rather, is the hallmark of geographical space resulting from the necessity of capital accumulation within the EU, constituting poles of both expanded equalization and sharpened differentiation in conditions of production, as witnessed in the Internal Market with the free movement of goods, services, capital and labour, locking states such as Greece and Portugal into a subordinated position within GVCs.

III. Class Struggle in the Eurozone: Going Beyond Technocratic Policy Solutions
When reflecting on how to overcome the Eurozone crisis, rather than simply pursuing a post-Keynesian focus on technocratic solutions, a focus on social class struggles within, against and beyond the crisis needs to be established. David Harvey (2005, p. 16) has identified neoliberalism as a strategy by (transnational) capital to re-enforce its control over the production process and restore class power. Ongoing austerity across the EU in the form of European integration, too, can be understood as a class project by capital to strengthen its position vis-à-vis labour. This does not imply that there have been no tensions or differences between various class fractions of capital. European export-oriented capital showed some concern about the dominant crisis response, considering that severe austerity imposed across the EU's periphery dampened international demand. There was not a questioning, however, of the overall need for austerity. For instance, in the Spanish case there is strong support for austerity amongst capital, directly linked to perceptions of an unsustainably large public sector and the perceived negative consequences for external competitiveness derived from trade union participation in wage bargaining (Albertos and Kuo, 2018, p. 37). In Germany and Sweden there is, furthermore, clear historical evidence amongst large export-oriented capital of a preference for, and implementation of, liberalizing 'co-ordinated' industrial relations institutions (Baccaro and Benassi, 2017, pp. 98-107;Kinderman, 2017). Some smaller firms in Europe's southern periphery objected to the decentralization of collective bargaining, but only because they believed that this enabled a more effective way of controlling the demands of labour whilst also dampening competitive market pressures (Bulfone and Afonso, 2017, p. 14). Larger capital did not share this reluctance (Bulfone and Afonso, 2017, p. 17). Therefore, ultimately the crisis has provided capital with the opportunity to justify cuts and roll back workers' rights (Lehndorff, 2015b) that they would not otherwise have been able to implement. In May 2010 and March 2012 Greece received financial assistance, Ireland was bailed out in November 2010, Portugal in May 2011. In March 2013, it was Cyprus's turn. Most recently, Greece was bailed out for the third time in summer 2015 under highly controversial circumstances surrounding a struggle between a Greek government led by the relatively new party Syriza and the Troika, consisting of the European Commission, European Central Bank, and International Monetary Fund, on behalf of the EU. Spain and Italy have also been heavily affected. Whilst not having to succumb to a formal bailout package, Italy and Spain had to present austerity programmes, developed nationally, before the extension of loans from European institutions were made to recapitalize their banks. For those under the 'guidance' of formal agreements, the bailout packages were made dependent on austerity policies, including (1) cuts in the funding of essential public services; (2) cuts in public sector employment; (3) pushes towards the privatization of state assets; and (4) the undermining of industrial relations and trade union rights through enforced cuts in minimum wages and a further liberalization of labour markets.
In more concrete terms, the EU's peripheral countries were initially obliged to cut back fiscal spending drastically. For example, 'the fiscal cuts imposed on Greece amount to 10.5 per cent of GDP for 2010 and 2011, and another 9.9 per cent until 2014. The consequence of this austerity has been a drop in GDP of 25 per cent since 2008, the bulk of which came after the memorandum agreement in 2010' (ELIAMEP, Crisis Observatory). But imposed austerity went also beyond direct cuts, with Greece being pressured to enact labour market deregulation and to make wage-setting 'more efficient'. As Lehndorff (2015a, p. 165) points out, new labour market legislation in Greece included 'the priority of company agreements over sector agreements, the suspension of the extension procedure of collective agreements, and a 22 per cent cut in the statutory minimum wage (31 per cent for under-25s)'. Similarly, as part of the bailout package for Portugal, the government agreed to stop extending collective agreements automatically to the whole industrial sector in 2011. Unsurprisingly collective bargaining coverage has fallen drastically, from a total of 116 industry-level agreements in 2010 to just 12 by 2012 (ETUI, 2016). Hence, while in 2010 1,309,300 employees were covered by collective industry-level agreements, in 2012 only 291,100 employees were. There have also been amendments to the Labour Code with the aim of creating greater flexibility for firms, thereby allegedly enhancing competitiveness and wage moderation. These changes have included reduced pay for overtime by 50%; the further flexibilization of fixed-term contracts by extending the probationary period from 6 months to a maximum of 3 years; and the relaxing of rules for redundancy and dismissal, especially for reasons revolving around economic circumstances (Clauwaert and Schomann, 2012, pp. 9, 11-2;Pine and Abreu, 2012, p. 25). In short, the crisis has been used by Troika officials to undermine the power of trade unions by cutting back their involvement in collective bargaining and industrial relations more generally whilst also creating more flexible labour markets, further enhancing the power of capital.
Additional pressure was put on peripheral countries to privatize key national assets in order to improve the balance sheets. This pressure often bypassed democratic procedures. In August 2011 in relation to Italy, Jean-Claude Trichet, the then President of the European Central Bank, and Mario Draghi, who succeeded him in November 2011, urged 'the full liberalisation of local public services … through large scale privatisations', ignoring the fact that 95.5% of Italian voters had rejected the privatization of local water services in a valid national referendum fewer than 8 weeks earlier (Erne, 2012, p. 229; see also Bieler, 2015). The initial aim of privatizations in Portugal including various public utilities was to raise €5 billion. By April 2014 this target had been exceeded, reaching almost €9 billion, with more still scheduled. The state-owned enterprises that remain have been restructured through redundancies and pay cuts to bring down costs, and most have 'reached operational balance by the end of 2012' (European Commission, 2013, p. 28). Greece had initially agreed to a proposal for only €3 billion worth of privatization. This was eventually raised to €50 billion only months later. Against this background of sluggish privatization, the most recent bailout agreement was used to impose even more stringent monitoring processes. Greece had to commit itself to transfer €50 billion worth of Greek state assets to a new fund with little influence by Greek politicians (Wearden, 2015). 5 Arguably, the struggles over the most recent bailout agreement of Greece signify most clearly the dynamics of class struggle underpinning the Eurozone crisis. When the Greek parliament had decided to hold a referendum over the Troika bailout proposals, the European Central Bank withdrew the provision of additional emergency liquidity assistance to Greece, thereby indirectly forcing the Greek government to close the banks and its stock market as well as to limit ATM withdrawals and transfers of funds (Traynor and Hooper, 2015). 6 During the negotiations Troika officials had regularly dismissed any Greek counterproposals. While the Greek finance minister Yanis Varoufakis attempted, rationally, to explain along post-Keynesian lines why further austerity could not result in a revival of the Greek economy, the Troikarelying on capital's superior balance of powersimply imposed its conditions. This was aided by a curtailment of both temporal sovereignty, as governments no longer had time 'to deliberate about what their policies should be' (Flohr and Harrison, 2016, p. 315), and spatial sovereignty, as Troika officials dictated the terms of policy conditionality and the time frame of when it had to be fulfilled if the quarterly doses of financial assistance were to be maintained (Sotiris, 2017, pp. 177-178). In sum, inspired by Smith (1986, p. 100), the fluid geography of capitalist expansion and equalization marking the spatial political economy of Europe has elsewhere given way to a hardened geography of crisis dominated by differentiation.

Conclusion
European post-Keynesian analyses have made significant contributions to our understanding of the Eurozone crisis. They correctly highlight the fact that the EMU deepened unevenness across the European political economy and a lack of sufficient aggregate demand at the regional level has undermined the possibilities of overcoming the crisis. The post-Keynesian suggestion of empowering trade unions in order to secure higher demand levels in export-oriented countries such as Germany is in itself logical, but overlooks the conditions of class struggle underpinning the crisis. A focus on the structuring condition of uneven and combined development as well as the resultant agency of class struggles embedded within European integration can uncover the contradictory current dynamics underlying the Eurozone crisis. Uneven development has always been one of the key systematic expressions of the geographical expansion of capitalism and it is inherent in the very constitution and structure of the spatial political economy of Europe. The interdependent development of export-led economic growth models versus debt-driven economic growth models has been re-enforced by the institutions of EMU, but ultimately its origins go much further back, even beyond European integration and the expanded 'free' trade agenda reflected in the Internal Market project. In other words, before and since the beginning of EU membership, Greece, Portugal, Ireland but also Spain were drawn into relations of uneven and combined development, which inhibited their potential to catch-up beyond labour-intensive industries within the conditions of production shaping Europe. High economic growth rates in the early 1990s, further facilitated by easy credit through membership in the Eurozone were unsustainable in the end. The EU therefore provides robust evidence that 'the counteracting force of the uneven development of capital prevents the formation of an actual global community of interest for capital' (Mandel, 1975, pp. 332-333).
When the Eurozone crisis threatened to undermine capitalist accumulation in Europe, capital used the crisis for further neoliberal restructuring, increasing its power vis-à-vis labour. Post-Keynesians correctly point out that economically, austerity has not worked when considered in overall national terms. Nevertheless, as David Ruccio has demonstrated while analysing Latin America, the political economy policies of structural adjustment can be regarded as a success for capital, in strengthening the class dimensions of capitalism. 'What may be a failure from the standpoint of achieving full employment, price stability, and balance-of-payments equilibrium can be considered successful in terms of promoting the widening and deepening of capitalist class processes' (Ruccio, 2011, p. 204). Related to the Eurozone crisis and understanding austerity as a capitalist project, it does not matter that Greece, for example, has gone ever further into economic decline. In strengthening capital over labour, opening up Greek companies for privatization, austerity has facilitated new points of value-capture assisting capital in the search for further surplus accumulation. Thus, it is no surprise that magnified neoliberal restructuring has continued to be the main response to the Eurozone crisis in Europe.
Yet, this does not imply that there is no resistance or that alternatives are impossible. Counter-movements, however, cannot stem simply from theoretical reflections by post-Keynesian economists. They emerge out of concrete struggles, often in spaces where exploitation is at its severest. Be it Vio.me, a large metallurgy factory in the north of Greece occupied and run by its workers, be it the social health-care clinics across Greece (Daskalaki and Kokkinidis, 2017;Jordan, 2017, pp. 215-220), be it ongoing resistance against water privatization in Thessaloniki combined with attempts at transforming water into a commons (Bieler and Jordan, 2017), or be it the fight back for housing in Spain (Bailey et al., 2017, p. 210), it is through such counter-spaces of class struggle that, ultimately, the way towards a different future may emerge.