PROGRESSIVE REGRESSIONーMetamorphoses of European Social Policy wolfgang streeck

 Has any polity in world history undergone such a rapid and far-reaching series of transmogrifications as the European Union?(1) Founded as an organization for joint economic planning among six adjacent countries, in the context of the state-managed capitalism of the post-war era,(2) it grew into a freetrade zone, increasingly devoted to neoliberal internationalism under the rubric of the ‘Internal Market’. As the number and heterogeneity of member states grew, ‘positive integration’ was replaced by ‘negative integration’, in effect market-building: the removal of national regulations impeding trade, in an ever-broader sense, within the union. After the fall of the Soviet bloc in 1989, the EU became in addition a geostrategic project, closely intertwined with American strategy in relation to Russia. From a handful of countries jointly administering a small number of key economic sectors, the EU developed into a neoliberal empire of 28 states, obliged under union treaties to allow for freedom of movement for goods, services, capital and labour, and to refrain from ‘anti-competitive’ intervention in their economies.

  Social policy has been a touchstone for these successive iterations of the ‘European project’. What follows will analyse the trajectory of European social policy over the longue durée, as it has mutated from a projected federal social-democratic welfare state to a programme for competitive adjustment to global markets(3).But first, the terms ‘European’ and ‘social policy’ both require a word of explanation. The  European system of government, as organized, or disorganized, in the European Union, is a strange animal. It consists, first, of the domestic politics of its member states, which have with time become deeply intertwined. Second, member states, still sovereign, pursue nationally defined interests through national foreign policies in intra-European international relations. Here, third, they have a choice between relying on a variety of supranational institutions, on intergovernmental agreements among selective coalitions of the willing, or on both at the same time. Fourth, since the start of European Monetary Union, involving only 19 of the EU’s 28 member states, another arena of European international relations has emerged,  consisting of mainly informal intergovernmental institutions such as the Eurogroup, the gathering of Eurozone finance ministers, looked upon with suspicion by the supranational EU. Fifth, all of this is embedded in a matrix of nationally varying geopolitical locations and geostrategic interests, related in particular to the United States, on the one hand, and to Russia, Eastern Europe, the Balkans, the Eastern
Mediterranean and the Middle East on the other. And sixth, at the centre of the European system there is an ongoing battle for hegemony between the two most important member states, France and Germany—a battle which both deny exists.

   All these factors make ‘European’ policy arenas and processes difficult to demarcate, and responsibility for outcomes hard to attribute. The upshot is a confusing assortment of issues and interests, motives and structures, levels and sectors of policy-making, with actors operating in different incorporations and capacities—which, in the absence of a European public sphere, is worked out mostly behind closed doors. Social policy is a case in point. There are at least three different ways in which it may be—or may be imagined to be—‘European’. The first is as a common denominator of national social policies, represented in the notion of a ‘European social model’ that somehow encompasses the essence of the variegated national ‘social models’ in Europe. A second meaning refers to the social policies of the European Union, over and above national policies. Here, European social policy may complement, supersede, regulate, coordinate, perhaps protect, perhaps restructure the social policies of EU member states, adding an additional, supranational layer to the social-policy regimes of member states. And third, European social policy may, under some sort of federalist social democracy, absorb and replace—in other words, ‘integrate’—national social policies, on the way to a unified European welfare state, with identical social policies for Europe as a whole. Reference to the so-called ‘Social Dimension’ of the European Single Market often fails to keep the three meanings apart, which allows proponents to switch between them as they see fit.
  Moreover, assessing the empirical validity of different representations of a ‘European social model’ requires expert knowledge in comparative politics and law, given the wide divergence in the institutions, the politics and the substance of social policy across EU member states—all the more so after the accession of Scandinavia, Greece and, in particular, the former Communist countries of Eastern Europe. In practice this means that in public debate, anybody can imagine ‘Europe’ to be whatever he or she likes. This holds also for the extent to which European social policy in the second sense, as a set of supranational directives, is actually implemented by the Union’s 28 member states, with their very
different economic, political and legal systems. As a result, idealizing ‘Europe’ and European policy is much easier than assessing its realities, which has become a domain for insider specialists, most of whom tend
to be ‘pro-European’.

  A further consequence of the deep differences between EU member states is that the definition of ‘social policy’ itself needs to be drawn with a very broad brush. In the following, ‘social policy’ therefore refers to the entirety of authoritative political interventions designed to limit, if not eliminate, the vulnerability of wage earners and their families, or generally of less well-off citizens, in relation to the vagaries of markets in a dynamic privatecapitalist economy. In addition, an important function of social policy is to secure legitimacy for the regime of wage labour and managerial rule over the labour process. This can happen in two ways: by partially exempting actual and potential wage earners from market pressures—here we may
speak of social protection by de-commodification—or by enabling them to obey market signals more profitably, through public support for private adaptation to changing market conditions. The two approaches, one setting limits to the market, the other supporting compliance with it, may sometimes be difficult to distinguish, and indeed all social-policy regimes more or less include both kinds of measures, commodifying and decommodifying, to different degrees and for different purposes. Bearing these points in mind, we will examine the mutations of European social policy through five distinct stages. As will be seen, its trajectory followed closely the general trend of capitalist political-economic development in this period, sometimes driven by it, sometimes reinforcing or modifying it, but never checking or reversing it. 

  The story of European social policy begins in 1957 with the European Economic Community—the ‘Europe of the Six’. The eec built upon the European Coal and Steel Community, a technocratic instrument designed by French civil servants on the model of planification, to manage the then ‘key industries’ of coalmining and steel-making. Its main purpose was to prevent Germany from once again using its Montanindustrie to rebuild its military power; another was to guarantee French access to German coal. In addition, the ECSC would help to coordinate industrial change in these core sectors, historically organized by powerful and strike-prone trade unions. To this, the EEC added the idea of a ‘common market’ for the
six countries, initially realized only with significant exceptions (it would take until 1992 for what came to be called ‘the completion of the Internal Market’). In this context the EEC also moved to abolish discrimination in working conditions and social-security benefits on grounds of nationality, as a way to aid the mobility of workers between the six member states—in effect, to open North European labour markets to, mostly, Southern Italian surplus labour.

  Otherwise, social policy did not play much of a role. (4) Trade unions remained firmly entrenched in the core industries, and national governments were mindful of the need to stay on good terms with them. Generally this was a period of rapid economic growth, rising wages, and near full employment, while markets were still far from fully integrated. Social policy was considered a national affair, to be taken care of by a growing economy. German ordoliberals, who had lost out at home against Social Catholicism and trade-union ‘cartelization’ of the labour market, resumed their battle at the European level, with a project to found the Common Market on a legally based competition regime that would ltimately render government intervention in most sectors illegal (in the long run, over half a century, they would triumph). In the meantime, member states could pursue their social policies independently without suffering competitive disadvantage, since their economies were still fundamentally national.

  Not entirely fitting this pattern were two clauses in the Treaty of Rome that had been inserted on the insistence of France. One noted that women and men should by law be paid equal wages for equal work in all member countries; the other, that the Community should work towards the ‘harmonization’ of its members’ social-security systems. The two provisions reflected French concerns that its equal pay and social-security regimes, a legacy of the 1930s front populaire, would impose a disadvantageous cost burden on the French economy. That worry turned out to be unjustified in the conditions of a fast-growing mixed economy and rapid catch-up on the part of other European countries. Moreover, while equal pay for women was unenforceable (or at least, unenforced) in France itself, ‘harmonizing’ national social-security systems turned out to be so fraught with difficulties, both technical and political, that, even though it
remained in the Treaty, it was never attempted.

  For the first two decades of the postwar era, social policy was thus embedded in the state-administered national capitalisms of ‘Europe’s’ six founding members.(5) Stage Two began in the early 1970s when European 5 See Andrew Shonfield, Modern Capitalism: The Changing Balance of Public and Private Power, London 1965; Andrew Shonfield and Zuzanna Shonfield, In Defence of the Mixed Economy, Oxford 1984. See also Wendy Brown, Undoing the Demos: Neoliberalism’s Stealth Revolution. Cambridge, ma 2015. social policy emerged as a political arena in its own right, largely in response to the widespread upsurge in labour militancy. Faced with waves of unofficial strikes, following the student and anti-war protests of 1968, European leaders moved to revamp national social policies and collectivebargaining
regimes to restore industrial and political peace. The first wave of European social policy—the 1972 Paris Summit, the 1974 Social Action Programme—was part of this response.(6) The effort was led by centrist and centre-left governments: Wilson and Heath in Britain, Pompidou in France, the left wing of the Italian Democristiana, and the German socialliberal coalition under Willy Brandt. Since Germany appeared to have contained the industrial disorder of the late 1960s better than the other countries, it was widely believed among policy-makers that Europeans could and should learn from Bonn—learn, that is, how to share power in order to keep it, and how to restore profitability through concessions to workers and unions. The European Community, just then being enlarged to include the UK, Ireland and Denmark, was seen—apparently as a matter of course—as the right instrument for a supranational modernization of national European social-policy regimes.

  European social policy in this phase may be explained as a socialdemocratic project to save the postwar settlement between capital and labour by updating its institutions. A major aim was to encourage centralized collective bargaining between strong unions and employer associations, at national but also, in a longer perspective, European level. For the latter, one needed politically unified European trade unions and an equivalent organization of employers. The European Trade Union Confederation (ETUC) was duly set up by the big national trade-union federations in 1973, and given official status in the treaties. The European Commission did its best to help European trade unions overcome the long-standing division between communist and social-democratic confederations, and was to some extent successful. But otherwise, collective-bargaining reform efforts were futile. National structures and traditions were too firmly established. Strike rates remained high in the 1970s, and inflation differed dramatically between countries, in particular between Germany on the one hand and Italy, France and the UK on the other. Industrial-relations reform on the German model lost its urgency after 1979, when the us Federal Reserve crushed inflation worldwide by raising interest rates to a level that was bound to cause high nemployment, in the us and abroad. Shortly thereafter, Britain under Thatcher followed suit. With a non-accommodating monetary policy, unions lost the capacity to extract concessions from governments which, after the Thatcher and Reagan experience, were no longer worried about unemployment standing in the way of their re-election.  Collectivebargaining reform ceased to be an issue for European social policy; when it returned twenty years later, under monetary union, it took the opposite direction—towards the  decentralization and individualization of wage-setting.
  Another leg of the social-democratic project was worker participation, both on the boards of large firms and on the shop floor. Here too, the idea was to learn from Germany. For many years, under the presidencies of Roy Jenkins and Gaston Thorn, the European Commission tried to pass legislation obliging member states to introduce parity ‘co-determination’ on the boards of publicly held companies. The Commission’s proposals met with robust resistance from European employers. But they were also opposed by some of the more powerful trade unions (outside Germany), which protested against being forced into collusion with the class enemy or being made to compromise their long-established rights to autonomous collective bargaining. Throughout the 1970s, member states debated various company-level workforce-articipation regimes. Ultimately, some—like Germany and Sweden—managed to pass legislation while others, notably Italy and the uk, failed. Meanwhile at European level, legislative proposals became progressively less ambitious. Parallel attempts to institutionalize worker voice on the shop floor, at the point of production, suffered a similar fate. In the end, only a few countries passed national legislation. At European level, the Commission’s draft ‘Vredeling Directive’ was defeated by concerted pressure from business, represented for the first time by American law firms, which used the opportunity to get a foothold in Brussels, where they now constitute a formidable force.(7)

  By the 1980s, the Commission’s efforts to create a European system of labour relations to accommodate strong unions was beginning to look like a holdover from a social-democratic era that had come to an end at national level. Employers chafed at a ‘Europe’ they identified with an outdated social-partnership approach to the problems of low profitability and slow growth. With Keynesianism on the way out and monetarism on the horizon, with Thatcher firmly in the saddle in the uk and Mitterrand turning to ‘supply-side’ economics and neoliberal institutional reform, they no longer saw the need to make concessions to a weakening and increasingly disorganized working class. Moreover, as ‘globalization’ became a realistic prospect, older ‘Fortress Europe’ ideas lost out inside unice, the European peak association of industry.(8) Even French capital began to look to the free-market world outside Europe, where it was no longer necessary to make concessions to domestic workforces which could be disciplined just as well by market pressures as by collective representation—and at lower cost.
  It was at this point that the European Commission began to seek new fields for a social policy that would be less contested than capital–labour relations and not yet occupied by national policy. It found two. The first was workplace health and safety, an area in which national business associations and trade unions shared an interest in eliminating low-quality competition by enacting mandatory standards for Europe as a whole; moreover, the machine industry in particular lobbied for harmonization at the highest level, as this would make parts of the installed fixed capital in European industry obsolete. The Commission’s second theme was equal opportunity for women in the labour market. The rapid rise in female labour-force participation since the 1970s had not yet been reflected in national social policies, which resulted in a gap in regulation through which the Commission could hope to make inroads in national systems. Here, European legislators were able to draw on the historic equal-pay provision in the Treaty of Rome. They also enjoyed the support of employers seeking an expanded and more ‘flexible’ labour supply—and happy to have government pressure applied on employment ‘insiders’ as represented by male-dominated trade unions, resistant to labour-market liberalization. 

  A series of directives on both health and safety and equal opportunity were passed in the 1980s, in a flurry of legislative activism. However, this petered out as gaps in regulation were closed, shared interests of employers and unions were exhausted, and national social-policy regimes caught up with the times. Otherwise this period saw the federal state-building goals of the early 1970s gradually pushed aside by the more firmly rooted project of liberal market-building, which was in fact a better fit with member states’ national sovereignty and the integrity of their domestic political institutions. Common economic policies expanded from selective sectoral planning to the institutionalization of free trade across national and
sectoral borders; modes of political regulation moved away from welfarestate federalism, evolving from positive to negative integration.

  By the mid-80s, Brussels had fallen into disrepute among Europe’s business classes, reflected in Thatcher’s untiring denunciation of the eu as bureaucratic, anti-competitive and even socialist. This began to change when Mitterrand and his Finance Minister, Jacques Delors, took up ‘Europe’ as an international lever for national neoliberal reform.(9) For this it was necessary to revive the integration process, in a direction that both required and contributed to renewed business confidence in it. One motive was certainly to preserve European integration as a tool for France to bind Germany into a (French-dominated) continental order. More immediately, however, the intention was to use a reconfigured European Community as an external constraint on domestic economies, not least in France, where the trade unions and the Communist Party were increasingly considered an impediment to economic progress.
  To bring capital back in, Delors was moved to the Presidency of the European Commission, where he focused his first term (1985–90) on what came to be called ‘the completion of the Internal Market’, scheduled by him for 1992. This was a project very much to the taste of the Thatcher government as it involved ‘negative’ integration, achieved by EU-wide economic deregulation.(10) Already on the horizon was the next liberalization project, European Monetary Union, which would be the capstone of the completed Internal Market.(11) To reassure the unions, the Commission held out the prospect of a ‘social dimension’ to be attached to the Internal Market, once it was fully in place. Arguing that one cannot ‘fall in love with a market’, Delors (he had not met Jeff Bezos!) claimed that the stability of the Internal Market would ultimately depend on it being embedded in a proper social-welfare state, which even business would at some point have to realize.
  For progress on these fronts, Delors first had to clear away the defunct European welfare state-building project—above all, ending the conflict over workers’ rights to participation and representation, without antagonizing either business or labour. In the end, prefiguring a general trend in European social policy, this was achieved by limiting EU legislation to multinational firms that chose to be incorporated in European, as opposed to national, company law, and by allowing them to choose from a wide range of models of workforce representation.(12) The Works Councils Directive of 1994 required large firms with bases in more than one European country to negotiate with a multinational delegation of their workforce over the creation of a supranational representative body, a ‘European Works Council’ (ewc), with legal rights to information and consultation—but not co-decisionmaking.(13)
  Delors’ promised ‘social dimension’ was to be vested in the Social Chapter attached to the 1992 Maastricht Treaty.(14) The Chapter, from which the uk temporarily opted out,(15) came with a Social Protocol that provided a privileged role for the ‘social partners’, the peak confederations of business and labour. It built on the so-called Social Dialogue, an informal institution involving the Commission, UNICE and the ETUC that had been set up by Delors early in his presidency and UNICE was hailed by himself, ETUC and the social-democratic Left as the beginningunice of an era of true pan-European tripartism. Under the Maastricht Treaty, if business and labour at European level agreed on a social-policy proposal, the Commission had to adopt it as its own and submit it to the European legislative process. If ‘the two sides of industry’ failed to reach agreement, the Commission had the option of proposing legislation independently—which however it was under no obligation to do.
  As a political arena, the tripartite procedure took the place of the EC’s Economic and Social Committee (ECONSOC), which in principle decidedeconsoc by majority vote. Since ECONSOC included representatives of national and municipal governments in addition to the Commission, business and labour each held only one quarter of the vote. This resulted in business sometimes being outvoted, which contributed to its discontent with ‘Europe’. By comparison, the new procedure gave business a de facto veto on European social policy, as long as it could successfully lobbyeconsoc the Commission to abstain from introducing its own legislation in the event of disagreement between the social partners. By the mid-1990s, it was possible to claim that Delors had built what one commentator Europhorically called a ‘corporatist policy community’(16) around the European Commission—a supranational layer of ‘social partnership’ that brought together business and labour, while generating centralized policies and institutions for a multi-level social-policy regime.
  EU officials may have hoped to use this ‘corporatist policy community’ for more than just codifying the commonalities of national frameworks and rendering them mutually compatible. What they overlooked, however— or perhaps repressed, for the sake of Euro-optimism—were older insights about tripartite corporatism: that it works only when employers are under a legal or political obligation to negotiate in good faith; when trade unions have both the right and the capacity to strike; or when a government credibly threatens to legislate if corporatist bargaining gets deadlocked. These insights were not, however, overlooked by Delors, who was very careful not to upset the eu’s new relationship with capital by sanctioning employers for their ‘strategic inactivity’. Considering their de facto lock on new social-policy initiatives, what some celebrated as European-level corporatism turned out to be such in appearance only, not in substance.(17)

  Although the Social Protocol had been a major bone of contention in the debates on the Maastricht Treaty, in the end it produced little. For the most part, it either spelt out minimum standards which the majority of member states already met, or built bridges over and above national systems through European-level institutions, which left member-state institutions untouched, like the compromises over worker representation in multinational firms. Moreover the national implementation of EU legislation was further stymied in 1995 by the accession of Sweden, with its tradition of industrial-relations autonomy from political and legal intervention; to protect that tradition, Swedish unions insisted they be allowed to implement European legislation by collective agreement, instead of national legislation. Meanwhile the tripartite channel of  social policy-making fell into disuse, as the Commission left it to the social partners to come up with joint initiatives and employers dragged their heels. To keep the Social Protocol alive, the ETUC signed several agreements with UNICE that were substantively irrelevant, in that hardly any member country needed to change its laws upon their enactment;  this contributed to shifting European social policy to symbolic politics.(18) Much was made by the conservative British media of the Directive on Working Time (2003), which perhaps for that reason was sometimes held up by British trade unions and centrist Labour leaders as an example of helpful and beneficial EU legislation, although the average 48-hour working week it proposed was subject to so many exceptions and opportunities for opting out that it hardly mattered.(19) Ultimately, Delors’ stateless and therefore toothless pseudo-corporatism was no more than an intermediate stage in European social policy’s transformation into a free-market regime, in which concessions to labour were left to the good will of— mostly multinational—employers.
  In fact, probably the most substantial piece of European social-policy legislation in this period was passed outside the Social Protocol. The 1996 Posting of Workers Directive involved an issue, migration, central not just to unions and employers, but also, and in particular, to member states. ‘Posted workers’ are sent by their employer to work in a country other than that of their employment. If wages and conditions in their home country are inferior to the host’s, this may give firms in the former a competitive advantage within the Internal Market, with a chance of undercutting the receiving country’s labour-market regime. It would also result in different employment conditions co-existing in one country (‘legal pluralism’), eroding the sovereign power of national states to determine the law of their land. This was first recognized in the early 90s by France, which reacted swiftly with national legislation obliging foreign European firms providing services in France to abide by French labour standards. The EU, not to be left out, followed suit by passing a directive allowing member countries to make their labour regime largely binding on all firms operating on their territory, including non-national ones.(20) In subsequent decades, as market liberalization proceeded, the Posting of Workers Directive remained a live issue, undermined by European Court of Justice decisions like Laval (see below) and tightened up to avoid industrial strife in receiving countries and protect the sovereignty of member states over their national labour constitutions.

  By the mid-90s, with the effective completion of the Internal Market, the supply-side turn of the second Delors Commission and the countdown to monetary union, the desire of national governments for ‘Europe’ to free them from domestic institutional ‘rigidities’ got the better of the social dimension.(21) Gone were the days when ‘Europe’ was to become a supranational welfare state integrating, and improving on, member countries’ existing welfare regimes. Less ambitious efforts at regulating national policies from above, in order to harmonize them ‘upwards’, also ran out of steam, as did more or less successful attempts—in company law, workplace representation—to complement national institutions with supranational ones.(22) As European social policy became increasingly symbolic—apart from where it was used in defence of  national sovereignty against undesirable side-effects of market integration, as in the Posting of Workers Directive—the stage was set for eu competition law to be utilized by national governments for a neoliberal restructuring of their political economies.
  The late 1990s and early 2000s were the era of the ‘competition state’.(23) With globalization on the advance, Europe’s mainly centre-left governments—Blair, Jospin, Schroeder—defined their central task as making their national economies more ‘competitive’ internationally by making them more competitive internally. For their ideologues, unions were guilty of representing only ‘insiders’, at the expense of ‘outsiders’, who allegedly had to be protected by abolishing ‘rigidities’ shielding the core workforce. This did not preclude governments and, occasionally, employers from seeking alliances with trade unions, where the latter were still strong enough to be either obstructive or helpful. But in contrast to postwar democratic corporatism—when tripartism aimed to establish and secure a bargained compromise between the conflicting interests of capital and labour—the aim now was to forge a common strategy for economic prosperity in competition with other countries, by making domestic political economies more ‘flexible’. This involved not so much direct attacks on the unions, now a much weaker force than they had been in the 1970s, but rather a re-orientation of what had formerly been social policy toward objectives such as ‘flexicurity’ and ‘employability’. The promise was that, while companies could shed workers more easily, job loss would not mean prolonged unemployment but would issue in prompt re-employment. The task of public policy was to provide a smooth transition from one job to the next, by helping workers build the flexible ‘human capital’ necessary for the labour markets of the neoliberal era. Deregulation at the lower end would open up a plethora of low-wage jobs, which workers who had been drawing unemployment benefit would be compelled to take. Competition-state labour-market policies revolved around notions like ‘activation’ and ‘social investment’, suggesting that the requisite skills would guarantee upward mobility once in employment.(24) Blairite social theorists dubbed this the Third Way, positioned between the ‘rigidities’ of postwar state-administered capitalism and the ‘flexibility’ of a free-market economy of the Reagan–Thatcher kind.
  Third Way reforms took place mainly at national level. The role of ‘Europe’ was reduced to assisting the slow—or not so slow—transformation of national social policy in a post-Keynesian, neoliberal direction—for example, by spreading ideas like ‘flexicurity’ and urging member states to voluntarily adopt ‘best national practices’.(25) Supranational policy tools included the comparative ‘benchmarking’ of national performance; the promulgation of minimum standards, low enough not to be a burden on national governments; and information on presumably effective policies as used in other European member states. European social policy in this phase relied mostly on soft law, in a neo-voluntarist framework of ‘governance’ typified by the so-called Open Method of Coordination, Euro-phemistically celebrated as a process of mutual learning and joint policy experimentation.(26)
  With the accession of most of Eastern Europe in 2004, raising the number of member states to twenty-five, this could hardly have been otherwise. By the early 2000s, the EU had simply become too heterogeneous for an integrated European social policy. Liberalization, more or less politically cushioned, was now the one and only formula on which all member countries could agree, given the incessant market pressures from the EU’s Internal Market as well as from outside ‘Europe’. Positive integration on a common social model had become impossible, as no such model could have found a majority in the Council. At the same time, any roll-back of liberalization was also ruled out by the EU’s implicit constitution, with its high majority thresholds and the de facto constitution-writing powers of the Court of Justice of the European Union (CJEU).(27)

  Today, a decade after the global financial crisis, European social policycjeu as a political project has disappeared from sight, even in the Third Way mode.(28) This does not mean there is no social policy anymore, nor that it has ceased to be contentious; on the contrary, battles over social protection within the European state system have intensified. At least two things have changed, however. The first is that conflicts over European social policy are now being fought out in confrontations between defenders of national welfare states and Europeanist ‘reformers’—the latter including such supranational institutions as the European Commission, the CJEU and the ECB, still set on the pre-2008 course of neoliberal marketbuilding. The second is that social-policy battles have been relocated to the fields of monetary, fiscal and immigration policy, which claim priority over all others. In the process, the status of social policy within Europe’s political economy has undergone a profound change. Under the postwar standard model of social democracy, social policy had been institutionalized as a semi-autonomous policy area, governed by a logic of social integration that challenged, and partly counterbalanced, the drive of economic policy for maximally efficient deployment of capital. Since then, social policy has lost its semi-autonomy. It has been absorbed into a more or less monistic neoliberal logic of rationalization-by-commodification,
in which it became instrumentalized for, and subsumed under, a societywide process of competitive restructuring. 

  Examples abound of the regressive subordination of social policy to neoliberal restructuring. A key mechanism is the claim to priority of European competition law over domestic social policies. Inventive policy entrepreneurs have repeatedly invoked treaty limitations on ‘state aid’ to end subsidies for non-profit service providers, so that private firms can compete with them on a ‘level playing field’.(29) Although the courts may not necessarily rule in their favour, the doors remain open for another try later. Member states that refuse to submit their public services to private markets are thus operating under a judicial sword of Damocles that may fall upon them any time. The CJEU’s novel interpretation of the ‘four freedoms’ of the Internal Market has played a crucial role here. Originally, member countries simply pledged that they would not discriminate against foreign workers, investors or suppliers of goods and services. By the early 2000s, however, the Court was being called upon to find in breach of the treaties any national practice that might restrict cross-border economic activity by making it ‘less attractive’ to international investors.
  Interpreted this way, the ‘four freedoms’ may override national parliamentary legislation on virtually any market-containing policies, requiring deep revisions in the institutional fabric of member states. For example, the Commission has made repeated attempts to get the CJEU to declare the minority public stake in Volkswagen to be in conflict with the ‘freedom’ of capital, on the grounds that it might deter foreign investors from buying Volkswagen shares. Flimsy as this argument may seem, it apparently speaks to the heart of market-minded CJEU judges. Two landmark cases concern the relation of national labour law to the European ‘freedoms’.(30) In its 2007 rulings on Laval and Viking, the CJEU found that national rights to collective bargaining and to strike had to be weighed against the right of EU-based business firms to operate in other eu countries, and might have to take second place. In Viking, a Finnish company operating ferries between Finland and Estonia chose to incorporate in Estonian law so as to escape Finnish collective wage agreements; in response Finnish trade unions took industrial action against it. In Laval, a Lithuanian construction firm tendered for work in Sweden, paying its (Lithuanian) workers Lithuanian wages, while refusing to sign the Swedish national wage agreement, as demanded by the Swedish construction workers’ union. On both occasions, the Court ruled that European internal-market freedoms took precedence over national trade-union rights.
  The Viking and Laval decisions have many facets. Lawyers close to the ETUC emphasize that the Court for the first time recognized collective labour rights as basic rights, whatever this may mean in practice. More importantly, however, the Court also ruled that these need not necessarily trump the ‘four freedoms’ and required national courts to apply a complicated test—proportionality, adequacy, justified by ‘compelling reasons of general interest’—in adjudicating the legality of national industrial action within the internal market. Although the two verdicts seem to have been moderated later in parts, they provide national courts with powerful legal instruments by which to restrain trade-union collective action in the name of market freedom.(31)
  As governed by ‘European’ or international law, immigration may also function in essence as social policy. The arrival of unskilled workers may undermine collective bargaining in low-wage sectors, to the extent that it still exists; it may also increase income inequality. In the process, it may furthermore weaken public perceptions of poverty and inequality as a problem—and, indeed, allow opponents of social protection to declare acceptance of domestic inequality a commandment of global solidarity with the ‘really poor’.(32) Immigration may also exert pressure on social-assistance budgets while weakening the willingness of citizens to be taxed for them, as a growing share of the expenditure may be going to newly arriving non-citizens. There is some evidence from Sweden that immigration can give rise to local educational segregation, as middle- and upper-class parents extract their offspring from schools that educate the children of immigrants and send them to more selective institutions.(33)

  Finally, with the financial crisis of 2008, what remained of European social policy was subordinated to the supranational emergency legislation adopted in the effort to save the euro. Although monetary union had been introduced without political union, precisely to preserve the sovereignty of Europe’s countries, in the crisis it de facto turned into a social-policy union which imposed strict budget rules—‘austerity’—and deep ‘structural reforms’ on member states. Crisis-management tools, like the fiscal compact, the six-pack, the two-pack, and others, became in effect parts of the EU’s socio-economic constitution. The ‘rescue operations’ that took place after 2008 to secure the servicing of member states’ public debt were accompanied by ‘Swabian housewife’ lessons, to the effect that nobody can spend more than they have earned. The rise of what I’ve called the ‘consolidation state’ put pressure on countries’ social acquis, to the extent that fiscal discipline was deemed necessary to protect the commercial acquis of the creditor banks.(34) Not only were so-called ‘programme countries’ told to cut their social spending, for example on healthcare or pensions, so as to regain the confidence of financial markets. There were also detailed demands for institutional change, including decentralization of collective bargaining, aiming in effect at the de-unionization of national economies. Here, social policy came to be absorbed in fiscalstabilization policy, as dictated by ‘pro-European’ creditor governments.

       vi. future prospects
  The last time European social policy as such became a subject of general debate was in 2005, when the Treaty on a Constitution for the European Union was rejected in the French and Dutch referendums. One explanation— put forward by the French President, Jacques Chirac, among others—was that governments had not paid enough attention to social policy. Today, opposition to the EU-driven liberalization of national socialpolicy regimes comes from a Europe-wide movement of ‘populists’, often right-wing, and therefore vulnerable to moral condemnation by internationalist opinion, for which the democratic alternative to ‘xenophobia’ and ‘racism’ is the opening of national economies to external competition, regardless of the absence of a supranational European social policy to compensate the losers of liberalization. Concepts like ‘social Europe’ and the ‘European social model’ have almost completely disappeared.(35) ‘Europe’ and the ‘European project’ are today promoted as vehicles of universal peace, international human rights and civilized speech, rather than as an alternative to unbridled capitalism.

  Since the crisis, the subordination of European social policy to the imperatives of capital has become more visible. As a result, it is being drawn into an intensifying conflict over ‘Europe’ and ‘European integration’ as a whole. The first article of faith of integration—that European law must take precedence over national law and politics—is now being challenged by a broad ‘nationalist’ counter-movement.(36) Neoliberal attempts to turn social policy and the welfare state into instruments for the promotion of economic competitiveness are increasingly being answered by popular demands for national governments to be held accountable again for what happens to the lives, and the ways of life, of national citizenries. In some cases this has borne fruit. National governments are making concessions to voters and workers—for example, by postponing ‘reforms’, or running higher than permitted budget deficits, as in Italy, France and Spain—while the Commission looks the other way and the ECB lends tacit support. Also, now that the brief interlude of Germanstyle immigration-by-asylum has proved politically unsustainable in Germany itself, member states are reverting to regulating immigration on their own. Meanwhile, pressures for ‘structural reform’ have become politically deadlocked. Greece, its bank debts settled essentially without a precedent-creating haircut, is left to its own devices, while even the most determined ‘reformers’ know that Italy will never accept the sort of treatment to which the Syriza government was forced to submit.(37)
  Most recently, in the empire-building phase of European integration, the lines of conflict over social policy have turned international, from labour versus capital to poor countries in Europe’s South—and perhaps also in the East—versus rich countries in the North and West. Ironically, it was the common currency, imagined to be the final nail in the coffin of European nationalism, that re-animated international recriminations in Europe, in particular between the Mediterranean countries—sometimes including France—on the one hand and Germany and its Northern European allies on the other. Now the issue is no longer whether governments, employers or both should pay for family leave or social investment. Instead, it is how countries benefitting from the euro can be held responsible for paying compensation, or damages, to countries doomed to lose under a hard common-currency regime. Rather than the price of social peace and economic collaboration in a capitalist national economy, what is being negotiated here is the price of peaceful international relations between neighbouring countries, payable in ways that allow payees to save face and payers to hide the transaction from their public.
  Indications are that, in the absence of functional equivalents to the institutions of conflict management in social policy, bargaining tends to be hard and accompanied by strong collective emotions; inter-state social policy is likely to be more politically explosive than domestic social policy. In any case, relocating the compensation of market ‘losers’ to international relations—where it may deteriorate into bribes for local elites keeping their countries in imperial dependence—contributes to the reemergence, real or perceived, of the nation-state as a unitary, homogeneous actor with a solidified corporate interest, overwriting the pluralistic concept of national politics that had been characteristic of postwar ‘standard democracy’.

  European social policies have come and gone, changing with the dynamics of the capitalist economy, the power relations between capital and labour, the size and heterogeneity of ‘Europe’ as a polity. What set out as an international exercise in the technocratic public management of selected private industries turned, for a short time, into a socialdemocratic welfare state-building project. This challenged, and was challenged by, a neoliberal supranational market-building project which had by then begun to grow out of the mixed economy of postwar statemanaged capitalism in Europe. That conflict was resolved in the 1980s in favour of supranational market-building, with the advent of the ‘internal market’ in 1992 and the introduction of the stateless common currency at the end of the decade, constitutionalizing the Union’s ‘democratic deficit’ with respect to both monetary policy, at the supranational level, and fiscal, economic and social policy in member states. Added to this in the 2000s was an empire-building project, which used financial and fiscal support for peripheral countries to draw them into the orbit of the West European centre, helping ‘the West’ extend its periphery to the borders of Russia.

  European social policy today is no longer a (‘relatively’) autonomous policy field driven by interests, however weak, in conflict, however limited, with the imperatives of capital accumulation. Rather than holding up or modifying the course of capitalist development, European social policy in its many different versions was pulled into the general crisis of capitalist accumulation and of the postwar state system, and more recently into the battle over a post-neoliberal order and, indeed, the future of capitalism. What will happen to European social policy in its present imploded and submerged condition, now that postwar ‘European integration’ has stalled? It depends on what becomes of the West European state system with its overextension, its frustrated ambition for technocratic centralization, its widening disparities among regions and states, the growing inequality of its citizens, and the geopolitical ambitions of Paris, Berlin and, in a different way, Brussels. What seems clear, however, is that the project, reaching back to the 1970s, of a supranational European welfare state giving political definition to a ‘European social model’ has come to an end.

1 This is a condensed version of a Max Planck Institute for the Study of Societies Discussion Paper, available online, based on a Keynote Address given at the fiftieth anniversary Conference of the Social Policy Association, Durham, 10–12 July 2017.
I am grateful to Ruth Dukes for bringing me up to date on the latest developments. All remaining errors are mine. return

2 The pattern here was remarkably close to what Polanyi, in an impressively prescient paper written in 1945, called ‘regional planning’—something he saw as a highly preferable alternative to a regional gold-standard regime: Karl Polanyi, ‘Universal
Capitalism or Regional Planning?’, London Quarterly of World Affairs, January 1945, pp. 1–6. 
3 For a similar reconstruction of the history of European social policy, see Antonios Roumpakis and Theo Papadopoulos, ‘From Social Regulation of Competition to Competition as Social Regulation: Transformations in the Socioeconomic Governance of the European Union’, in Dan Horsfall and John Hudson, eds, Social Policy in an Era of Competition: From Global to Local Perspectives, Bristol 2017, pp. 55–69. return

4 On what follows see Ruth Dukes, The Labour Constitution: The Enduring Idea of Labour Law, Oxford 2014, pp. 130 ff. return

5 See Andrew Shonfield, Modern Capitalism: The Changing Balance of Public and Private Power, London 1965; Andrew Shonfield and Zuzanna Shonfield, In Defence of the Mixed Economy, Oxford 1984. See also Wendy Brown, Undoing the Demos: Neoliberalism’s Stealth Revolution. Cambridge, ma 2015. return

6 Dukes, The Labour Constitution, pp. 137–45. return

7 In its initial version, the draft Vredeling Directive would have required all firms above a certain size, national and multinational, to set up arrangements for farreaching worker participation in management. It was finally withdrawn in 1986.

8 UNICE stood for Union des Industries de la Communauté européenne. In 2007 the organization was renamed  BusinessEurope. return

9 See the analysis of Bruno Amable, Structural Crisis and Institutional Change in Modern Capitalism: French Capitalism in Transition, Oxford 2017. return
10 Fritz Scharpf, ‘Negative and Positive Integration in the Political Economy of European Welfare States’, in Gary Marks et al., eds, Governance in the European Union, London 1996, pp. 15–39. return

11 Ashoka Mody, EuroTragedy: A Drama in Nine Acts, Oxford 2018. return

12 With everything removed that might have weakened the domestic position of employers, or subjected corporate governance to interference by workforces and unions, a European Company (Societas Europaea, SE) was established in 2004, as an option for multinational European firms seeking incorporation in European rather than national law. return
13 Details, procedural as well as substantive, are complex, not least because the Directive was written so as not to disturb or supersede existing national arrangements. Today, EWCS offer workforce delegates from different countries an opportunity to meet regularly, not just with management but also with each other, to exchange information and, perhaps, coordinate activities. How effective they are for workers is still being debated and seems to differ between countries, sectors and firms. Managements appear to use EWCS mostly as forums to build workforce identification with the firm as an international organization. return
14 For precise detail and the practical consequences of the legislation see Dukes, The Labour Constitution, pp. 125–30, 138–55. return
15 The heavily symbolic nature of European social policy is confirmed by the fact that the end of the UK opt-out—New Labour rejoined in 1997—did nothing to prevent the rise of inequality, the decay of collective bargaining and the deterioration of employment conditions in the UK over the years that followed. return

16 Gerda Falkner, ‘The European Union’s Social Dimension’, in Michelle Cini and Nieves Pérez-Solórzano Borragán, eds, European Union Politics, Oxford 2016, p. 277. return

17 On this see the early diagnosis by Wolfgang Streeck and Philippe Schmitter, ‘From National Corporatism to Transnational Pluralism: Organized Interests in the Single European Market’, Politics and Society, vol. 19, no. 2, 1991. See also an
extensive postmortem of EU ‘corporatism’ in the context of ‘strategic passivity’ and the rise of the European Roundtable of Industrialists: Armin Schäfer and Wolfgang Streeck, ‘Korporatismus in der Europäischen Union’, in Martin Höpner and Armin
Schäfer, eds, Die Politische Ökonomie der europäischen Integration, Frankfurt 2008.
18 Three directives were passed under the tripartite Protocol procedure—on parental leave (1995), part-time work (1997) and fixed-term work (1999)—followed by five bilateral agreements, implemented and monitored by the social partners directly, on telework (2002), work-related stress (2004), harassment and violence at work (2007), inclusive labour markets (2010), and a ‘framework of action’ on youth unemployment (2013). return
19 On the Working Time Directive, and the general tendency among some British trade unionists to overstate the benefits of eu labour law, see Mary Davis, ‘The Chimera of Labour Rights in the EU’, Full Brexit: Analysis, 15 July 2018. return

20 See Werner Eichhorst, Europäische Sozialpolitik zwischen nationaler Autonomie und Marktfreiheit: Die Entsendung von Arbeitnehmern in der EU, Frankfurt 2000. Given the diversity of national industrial-relations systems, implementing the new
directive required highly complex political and legal manoeuvres. Eichhorst maintains that most of the legal substance in the directive had long been international law under the Rome Convention.

21 The Delors Commission’s turn was laid out in the White Paper, ‘Growth, Competitiveness, Employment: The Challenges and Ways Forward into the 21st Century’, COM(93) 700, 5 December 1993. return

22 See also George Menz, ‘Whatever Happened to Social Europe? The Three-Pronged Attack on European Social Policy’, in Menz and Amandine Crespy, eds, Social Policy and the Euro Crisis: Quo Vadis Social Europe, London 2015, pp. 45–62. return
23 Bob Jessop, ‘From the Welfare State to the Competition State’, in Patricia Bauer, ed., Die Europäische Union—Marionette oder Regisseur? Festschrift für Ingeborg Tömmel, Wiesbaden 2004. return

24 See Bettina Leibetseder, ‘Social Investment Policies and the European Union: Swimming against the Neoliberal Tide?’, Comparative European Politics, vol. 16, no. 4, 2018, p. 597. return

25 See also the various charters of ‘social rights’ attached to the treaties, the most recent being the 2017 ‘European Pillar of Social Rights’, general enough in substance to be acceptable to countries as diverse as Sweden and Bulgaria. Little is
known about the real effects of European measures on the ground in member states, an issue rarely discussed in the research literature. An exception is the indepth comparison of the impact of EU social-policy directives in Gerda Falkner et
al., Complying with Europe: EU Minimum Harmonization and Soft Law in the Member States, Cambridge 2005. As  summarized by Falkner, the study found ‘major implementation failures’, as countries ‘privilege their domestic political concerns over the requirements of EU law’. Extending the analysis to newer member states from Central and Eastern Europe showed that EU standards ‘all too often remain a “dead letter”.’ Falkner, ‘The European Union’s Social Dimension’, p. 279.

26 Introduced by the 1997 Amsterdam Treaty, the OMC involves yearly, non-binding policy guidelines in areas such as employment strategy, pensions, immigration, education, culture and research; member states must report to the Commission
and to each other on how they have dealt with the guidelines. It has now been integrated in the European Semester, a complex arrangement for eu budgetary supervision and control. In ‘Neo-Voluntarism: A New European Social Policy 
Regime?’, European Law Journal, vol. 1, no. 1, 1995, I described neo-voluntarism as a (then just) possible ‘post-welfare state social-policy regime for Europe’, characterized by ‘cohesion by exemption,’ ‘unity by subsidiarity’, ‘governance by choice’
and ‘homogeneity by diffusion’. Its subsequent supersession by emu crisis policies under market pressure was not anticipated.
27 The CJEU has steadily expanded its jurisdiction since the late 1980s; the principle of the supremacy of eu law over national law was itself a creation by the Court, later incorporated in the 2009 Lisbon Treaty. In contrast to the us Supreme Court, CJEU judges are not subject to confirmation hearings by elected representatives. Instead, each member state nominates a judge, who is then ratified by all the other countries. return
28 To the same effect see Philip Whyman, Mark Baimbridge and Andrew Mullen, The Political Economy of the European Social Model, London 2012, p. 321: ‘A social Europe is an impossible dream.’ return

29 A recent case involved a private nursing home for the elderly suing the German Bundesland of Lower Saxony over its financial assistance to non-profit nursing homes. Interestingly, the issue is not whether such assistance is allowed under the
treaties—it no longer is—but only whether it had been granted sufficiently long ago that it would be protected under a grandfather clause.

30 Thomas Blanke, ‘Die Entscheidungen des EuGH in den Fällen Viking, Laval und Rueffert—Domestizierung des Streikrechts und europaweite Nivellierung der industriellen Beziehungen’, Oldenburger Studien zur Europäisierung und zur transnationalen Regulierung, no. 18, 2008; Christian Joerges and Florian Rödl, ‘Informal Politics, Formalized Law and the “Social Deficit” of European Integration: Reflections after the Judgments of the ECJ in Viking and Laval’, European Law
, vol. 15, no. 1, 2009; Simon Deakin, ‘The Lisbon Treaty, the Viking and Laval Judgments and the Financial Crisis: In Search of New Foundations for Europe’s “Social Market Economy”’, in Niklas Bruun, Klaus Lörcher and Isabelle Schömann,
eds, The Lisbon Treaty and Social Europe, Oxford 2012.

31 Subsequent Court decisions, among them AGET Iraclis (2016), have followed the Laval and Viking line. In addition, Stefano Giubboni argues that, through the back door, Laval has turned the minimum wage under the directive into a maximum
wage—part of a general trend in CJEU jurisdiction: ‘Freedom to Conduct a Business and eu Labour Law’, European Constitutional Law Review, vol. 14, no. 1, 2018.
32 In 2004, the Blair government allowed full freedom of movement into the uk for workers from the newly acceded countries—a reward for ‘New Europe’ governments that had supported the invasion of Iraq—although the treaties provided
for a multi-year grace period. Adding to the much higher immigration from Commonwealth countries, this certainly did nothing to halt the ongoing transformation of the UK into a low-wage economy with high wage inequality.
33 See Philip Mehrtens, Staatsschulden und Staatstätigkeit: Zur Transformation der politischen Ökonomie Schwedens, Frankfurt 2014; Lukas Haffert and Philip Mehrtens, ‘From Austerity to Expansion? Consolidation, Budget Surpluses, and the Decline of Fiscal Capacity’, Politics and Society, vol. 43, no. 1, 2015. return

34 Wolfgang Streeck, ‘The Rise of the European Consolidation State’, in Hideko Magara, ed., Policy Change under New Democratic Capitalism, London 2017. return
35 Bibliographic data show that books on ‘social Europe’ and the ‘European social model’ peaked in 2004 and 2005, at roughly 80 per year, after a steady increase from 1980 on, and then rapidly declined to about 20 in 2016. Journal articles rose from fewer than 10 in 2003 to more than 50 in 2009; in 2016 they were down to 25. return

36 Since the UK has remained outside monetary union, austerity in Britain is homemade rather than imposed by Brussels. Although London led the battle against EU welfare state-building, the English radical Right considers eu social policy as a prime example of European rule over Britain, regardless of the many exemptions its governments have managed to negotiate. Many on the British left, on the other hand, see what little EU social policy there is as a reason for remaining in the Union, fearing that a post-Brexit Britain would be even more firmly in the hands of free-market fanaticism than it is today. ‘Populist’ anti-eu sentiment in Britain, as  distinguished from elite anti-Europeanism, focuses primarily on immigration. To the extent that such sentiment spreads to the mainly pro-immigration Left, it views the EU as a device for the political classes of member states to evade domestic accountability. In this perspective, ‘taking back control’ means citizens ‘reclaiming their state’, to restore popular democracy and preserve a socialist option for the future. See William Mitchell and Thomas Fazi, Reclaiming the State: A Progressive Vision of Sovereignty for a Post-Neoliberal World, London 2017. return
37 Yanis Varoufakis, Adults in the Room: My Battle with Europe’s Deep Establishment, London 2017. return