In 1982, John Ruggie opened a debate on the relationship between international free trade and domestic social compensation, giving rise to the term “embedded liberalism.”Footnote 1 Part of the impact of Ruggie's article was due to the analogy he drew between Polanyi's The Great Transformation Footnote 2 and the post–World War II international trade regime, while another was its role in launching the constructivist turn in international relations theory. But perhaps the most important reason for the article's influence was that it offered an early benchmark for the challenges to embedded liberalism that would emerge from the transformations in both domestic and international political economies since Ruggie wrote.
One of these transformations lies within the political economies of capitalist countries. “On the domestic front,” Ruggie recently wrote, “the broadly Keynesian macroeconomic policy tools with which the [embedded liberalism] compromise was associated succumbed long ago to attacks from monetarism, rational expectationism, supply side economics, and other approaches more consistent with neoliberal and even laissez-faire views and preferences…” A second change lies on the international front. Embedded liberalism, Ruggie continues, “was predicated on the existence of separate and distinct national economies, engaged in external transactions, conducted at arm's length, which governments could buffer effectively at the border by point-of-entry measures like tariffs, non-tariff barriers and exchange rates.”Footnote 3 In the current age of globalization that simple duality no longer exists. A third transformation lies in the growing intervention of international institutions into areas once considered the province of domestic politics.
These trends are particularly striking in Europe, where relations among states, international institutions, and citizens have become multilevel;Footnote 4 where transnational governance networks have been formed;Footnote 5 where international legalization is far advanced;Footnote 6 and where the European Court of Justice (ECJ) interprets European Union (EU) treaties and regulations in ways that profoundly affect domestic economic, social, and cultural life.Footnote 7 In this article we will examine how the ECJ has intervened between the European free-market regime and domestic structures to begin to create what we regard as a structure of supranational embedded liberal compromises.
The application of the embedded liberalism framework to a court may raise eyebrows among some of our readers.Footnote 8 But based on Article 234 of the European Treaty (formerly 177), the ECJ has expanded its capacity to declare actions of both states and nonstate actors contrary to European treaties and regulations.Footnote 9 Litigation in the court, writes Cichowski, “enables individuals and groups, who are often disadvantaged in their own legal systems, to gain new rights at the national and EU level.” The Court's decisions can “expand the scope or alter the meaning of treaty provisions” and through access to it, “EU citizens who may be excluded from EU politics can gain new power and voice through the mobilization of transnational public interest groups.”Footnote 10 As such, the ECJ plays a key role in linking supranational legalization with transnational mobilization.
In this article, drawing, as Ruggie did, on the foundational work of Polanyi, we add to the embedded liberalism framework in three ways: first, rather than focus on the international trade regime and on the post–World War II period, we focus on the EU today; second, rather than concentrate on multilateral free trade, we turn to the free movement of labor in the EU and its implications for social regulation; and, third, rather than examine adjustments by national authorities to protect their citizens from the vagaries of open international trade, we study how the ECJ has been shaping a transnational regime for the protection of rights.
We first turn to the role of states and markets in Europe's first great transformation, drawing on Polanyi's concept of “embeddedness,” and specifying the term in three complementary ways to show how Ruggie applied it to the postwar international political economy. We then turn briefly to the interactions between domestic and European social policy. We will argue that, while the relationship between European integration and national social policy regimes is controversial, the ECJ has emerged as a regulatory arbiter of compromises between international openness and social concerns. The third part of this article focuses on the court's decisions regarding the free movement of labor. We will present evidence that the court has both worked to perfect markets and gone beyond market-making to embed the market in what it considers the legitimate social purpose of protecting the rights of workers and their families. We close with a reflection on the future of contention over social rights in the EU, which we see as analogous and contributory to the enduring struggle over embedded liberalism in the international political economy.
States, Markets, and Great TransformationsFootnote 11
For Polanyi, the growth of a market society in the early nineteenth century was driven by the ideology of liberalism that found coherent expression in a legislative and regulatory program based on the naturalization of the market and on market discourse as the “common sense” of the emerging capitalist system. This made it possible for the first industrializers to release their economies from their mercantilist and corporatist strictures and ignore the severe costs of the transformation for both traditional and new subordinate groups. In Polanyi's terms: “Economic society was subject to laws which were not human laws.”Footnote 12 The self-regulating market,” he famously wrote, “was now believed to follow from the inexorable laws of Nature, and the unshackling of the market to be an ineluctable necessity.”Footnote 13 This intellectual revolution was the essence of Polanyi's “movement”—not in the narrow sense of a “social movement,” as scholars of contentious politics define the term today,Footnote 14 but as a social metaphorFootnote 15 to guide the political economy. It created a “common sense” in many sectors of society that free markets are the natural way to organize an economy.
Polanyi never believed that the market could be disembedded from society; this is why he said that the “ineluctable necessity” of the “unshackling of the market” was “believed”—it was not a social reality. “When Polanyi wrote that ‘the idea of a self-adjusting market implied a stark utopia,’” writes Fred Block, “he meant that the project of disembedding the economy was an impossibility.”Footnote 16 Markets, concludes Block from his exhaustive analysis of Polanyi's work, are “always embedded.”Footnote 17 The problem, however, is that the precepts of the disembedded market were believed in by European policymakers and bankers, who insisted on their defense even as the international financial system was crumbling. National states retreated into a countermovement that took the form, at best, of protected markets, and, at worst, of the authoritarian involution that drove Polanyi from Europe. The believed-in myth of disembedded markets came into conflict with the realities of national protection, producing the collapse of the international economy in the 1930s.
Ruggie, Keynesianism, and the Embedded Liberal Compromise
It was to prevent just such a disaster from reoccurring that the architects of the postwar Bretton Woods system built the embedded liberalism compromise that Ruggie discerned in his 1982 article. While they certainly did not launch an integrated countermovement against free markets, they accepted a tissue of exceptions, expansions, and special cases within the overall framework they constructed for multilateral free trade. The outcome was the longest period of economic expansion in history.
If Block is right in his interpretation of Polanyi—and we think he is—then all economies are embedded in political and legal arrangements, and much political conflict in modern states turns on which arrangements are made and for whose benefit. Since disembedded markets do not exist, there is no such thing as a market whose sole purpose is to maximize efficiency, and the particular balance of market-making and market modification in any society depends on the rights that employers and workers gain vis-à-vis one other and the state.
Neo-Polanyians such as McMichaelFootnote 18 see Polanyi's countermovement as a classical social movement, most recently in the form of the “global justice movement” against neoliberalism. But by “movements,” Polanyi meant more than classical social movements: he meant shifts in cultural and ideological paradigms such as the one that produced the move to liberal markets in early nineteenth century England. By a “countermovement,” he meant a broad cultural and ideological reaction against such paradigm shifts.Footnote 19 It is also worth underscoring that, for Polanyi, movement and countermovement both had statist, as well as nonstate facets. States embraced the teachings of Ricardo, Malthus, and (eventually) Hayek and Friedman, but they also had to respond to the instability created by liberalizing markets. States did so, as we have seen in current American and European elite responses to the economic crisis of 2008–2009, incrementally, around short-term and often contradictory goals, and with a political logic. While the state was a prime mover in the move to a market society, it was also an active agent of the countermovement, as Polanyi made clear in his long list of Britain's regulatory efforts to harness the market.Footnote 20 Market-making produced market-modification by embedding it in an underlying social logic.
In much the same way, but with a neo-Keynesian mentality, the elites who fashioned the Bretton Woods embedded liberalism compromise were responding to the dangers they saw in completely free international markets. No one would accuse them of being part of “a social movement” in the classical sense, even though they were far from ideologically neutral. Fundamentally believers of international liberal exchange, they were also cognizant of broader social purposes. To avoid the insoluble conflict between international openness and domestic protection that had destroyed the interwar system, they built safety nets inside the international trading and monetary system they constructed after the war.
We will argue that the ECJ has been occupying a similar role to that of the shapers of the great postwar compromise with respect to the market-making policies of the Single European Act and the Treaty on the European Union.Footnote 21 The EU, as we will show below, does not make social policy, but it is not an unalloyed agent of global neoliberalism. But it does attempt to shape market-making through regulations that aim to embed the market within its understanding of legitimate social purposes.
Three Meanings of Embeddedness
To lay the groundwork for this understanding, we need to begin with the concept of “embeddedness.” Polanyi employed the term in three ways.
• First, embeddedness can mean that markets are, from the start, constituted by politics and society, in the sense that markets could never exist in an apolitical or asocial space.
• Second, embeddedness can mean the social protections created to guard against the depredations that markets might cause, for example, unemployment, income loss, forced job changes, sickness, disability, inequality, and so on. This is the meaning that social-democratic-oriented scholarship has given the term.
• Third, embeddedness can also mean the positioning of markets within a broader set of social and political rules and cultural understandings that make them work not only more efficiently but also more equitably, with greater security for market and nonmarket participants and in tune with a variety of other social purposes, and so on.Footnote 22
Let us comment briefly on each of these meanings of embeddedness, which are sometimes conflated in the neo-Polanyian literature.
Embeddedness and the constitution of markets
We take this first sense of embeddedness for granted, like Polanyi and most political economists and economic sociologists working today. It is true that markets have achieved a high degree of structural differentiation in modern society. Advanced capitalist societies have banks, firms, factories, trading houses, stock market institutions, and physically specialized structures for production and exchange. The market is also motivationally distinct to some degree and is institutionally separated from the centers of government. Yet, as Polanyi argues, religion, culture, government, family, and social networks may be as important for the economy as monetary institutions. Economic sociologists have carried this view of markets further with the aim of integrating social structure with economic theory.Footnote 23
Embeddedness as compensatory social policy
This is how most welfare state scholars use the term. They examine programs such as social security and social insurance aid to working women and men for child support, education loans and subsidies for students, retraining for workers displaced by market competition, unemployment assistance, and so forth. These programs are usually deeply entrenched in politics, that is, they are institutionalized in government laws, administrative regulations, and court jurisprudence. Polanyi used this meaning of embeddedness in his enumeration of the compensatory social protections that followed the first wave of industrialization in England.
Embedding the market within political, social, and cultural understandings
Although we assume the first meaning and will refer to the second from time to time, with Ruggie, we are mainly interested in the third meaning of embeddedness: the positioning of markets within “legitimate social purposes.” As Ruggie argued about the Breton Woods arrangements, to say anything sensible about the content of international economic orders and about the regimes that serve them, it is necessary to look at how power and legitimate social purpose become fused to project political authority into the international system.Footnote 24
We can grasp the importance of this distinction when we think back to how “legitimate social purpose” was inserted into the construction of the Bretton Woods system. The goal of the architects of that system was to “minimize socially disruptive domestic adjustment costs as well as any national economic and political vulnerabilities that might accrue from international functional differentiation.”Footnote 25 We see the ECJ in much the same way. We will show that the court attempts to shape transnational understandings in a way that is analogous to the safeguards, exemptions, and restrictions that Ruggie saw built into the Bretton Woods system. The EU uses its regulatory tools not only with the aim of market-making, but to engage in both market-making and market modification in something resembling the compromise of embedded liberalism that Ruggie saw in the founding of the Bretton Woods system.
To be sure, since the passage of the Single European Act, the EU's central policy direction has indeed been the creation of a single market.Footnote 26 But its simultaneous efforts to regulate supercomputers, genetically modified seeds, the Internet, software competition, gender equality, and food quality are far more than bureaucratic icing on the cake of neoliberalism. We see another component of European regulation that counters—or at least modifies—the dominant thrust of institutionalized neoliberalism. We will use detailed empirical evidence from the case law of the ECJ regarding the free movement of labor to show that the court has used its growing discretion within the institutions of the EU to modify what some have seen as the disembedding of European markets with a re-embedding of social regulation at the supranational level.
We choose the regulation of the labor market as our policy focus because workers are considered no less a commodity than land or capital, but at the same time are human beings, with families, local commitments, and rights that the ECJ, in its decisions on the free movement of labor, has increasingly recognized through a sequence of judgments. Many of the conflicts and contradictions in the European integration project can be understood as the result of this Polanyian duality. Much like the continued conflicts in the Bretton Woods system and its successors at the international level, it was bound to result in continued struggle between the principles of unfettered liberalism and European social regulation, as we will argue in our conclusions.
In the section below, we argue that “regulation for efficiency” by itself implies a great deal of social content, but also that the social content of the market is likely to exceed what can be explained on efficiency grounds alone. By “regulation for efficiency” we simply mean that regulations are crafted so as to improve prospects for economic exchanges—in our case by transnational labor market exchanges. International regulations might do this by outright removal of barriers to exchange (for example, defining work in other EU countries as illegal), coordinating different regulatory environments (for example, social security), or harmonizing incompatible national regulations (for example, credentials regarding work certification). Under the guise of adapting existing economic practices to market-making, regulation can also produce an embedding of markets in social practices.
European Regulation and Domestic Social Policy
Before presenting our data drawn from the case law of the ECJ, we should take note of three schools of thought that are relevant to the nature and the social costs of European integration: “the EU as regulatory state” school, the “EU as agent of neoliberalism” school, and scholars who have pointed to the growing legalization of international politics.
The EU as Regulatory State
Best represented in the work of Majone, the regulatory school holds that while the EU is constitutionally limited to spending no more than 1.3 percent of the gross domestic product of the EU members, it is a strong international governance structure. How can both of these assertions be true? Majone's answer: the EU specializes in the making and implementation of regulations; in short, it is a regulatory rather than a Westphalian state—an international and increasingly supranational state that specializes in the management and control of international externalities in order to produce greater efficiencies. The making of rules and their broad oversight take place in Brussels and Luxembourg (where the ECJ sits) while the fiscal implications of such regulations are passed on to the member states.Footnote 27
Our response to Majone's work is that it is not so much wrong as incomplete, for even in the service of efficiency, regulation may promote health and safety values, environmental values, and values related to the management of risk—what we call “social efficiency.” Social efficiency requires that regulations be designed so as to narrow the gap between private costs and benefits and social costs and benefits. A firm that is required to pay for the costs of cleaning up after itself (internalization of the negative external cost) provides a good example of how this works. EU responses to globalization and market-making also promote gender equality, regional equality, environmental protection, and laws that take into account the solidarity of the family when the mobility of labor is in question, thus going beyond regulation for efficiency.
The EU as Agent of Neoliberalism
As Majone was writing about regulation, a number of authors argued over whether the EU can be classified as an agent of neoliberalism. For example, Van Apeldoorn employed the neologism “embedded neoliberalism” to describe a class compromise in which transnational industrial and financial capital are hegemonic, but where the core of neoliberalism has to be blended with elements of opposed projects, such as neomercantilism and social democracy.Footnote 28 Other scholars wrote about the neoliberal shift in an international political economy framework, attempting to make sense out of the shifts, initially domestic, in the direction of deregulation and privatization. As Cerny argued, this shift involved “pressures for trade liberalization, the shift of the international monetary system from fixed to floating currencies, the explosion of international capital mobility and integration of global financial markets, the expansion of multinational corporations, the growth of transnational production chains and network forms of business organization…”Footnote 29
In Europe, a huge literature, covering a multitude of topics, developed on social policy,Footnote 30 including works councils, laws related to corporate governance, equal pay for men and women, maternity rights, rights of migrants regarding work benefits, access to unemployment insurance, disability benefits, and coordination of social security benefits. Important theoretical contributions have been made to this literature by Leibfried and Pierson,Footnote 31 Scharpf,Footnote 32 Rhodes,Footnote 33 Ferrera,Footnote 34 Hemmerijck and Rhodes,Footnote 35 and FaistFootnote 36 Rather than try to cover this vast literature, we summarize it briefly below.
The issue of social policy remains controversial among EU scholars.Footnote 37 For example, while Scharpf argued that the combination of national heterogeneity and institutional rules at the EU level, particularly the veto in the Council of Ministers, has frustrated attempts to build a coherent social policy, Leibfried and Pierson have written about ways in which market compatibility requirements have introduced social policy through the judicial door. While Streeck and SchmitterFootnote 38 provided a pessimistic account of the effects of market expansion on organized interests,Footnote 39 their broader argument about the relationship between regional integration and deregulation “becoming one and the same”Footnote 40 has not held up well and many scholars have shown that with “freer markets, more rules” have followed.Footnote 41 In part responding to the claims of Streeck, Schmitter, and others, MajoneFootnote 42 distinguished between social policy and social regulation, a distinction we find useful. While the former has to do with policies that are generally redistributive and are made at the national level, the latter relates to rulemaking and implementation at the European level.
While the EU's modest budget ceiling makes redistribution difficult, its peculiar niche in the regional division of political functions lies in regulatory policies. If Majone is correct, EU social policy is a modest effort to control for market failures (and more directly for national regulatory failures) and is therefore not in competition with national welfare states.Footnote 43 Rhodes concludes that “core features of welfare state regimes [will] remain nationally specific, with supranational influence restricted to a limited number of areas deemed crucial for market integration.”Footnote 44 By the mid-1990s, with the publication of Leibfried's and Pierson's European Social Policy, there was broad agreement that the primary project of the EU was market-making, that the Mitterrand-Delors axis that backed a forceful European social policy had failed, and that the tangled history of attempts to legislate a social policy at the regional level had been mostly “a saga of high aspirations and modest results.”Footnote 45
Legalization and International Politics
Given the institutional obstacles preventing the Council of Ministers from making social policy, most EU social policy is a product of the ECJ and comes in the form of market-compatibility requirements. To some, this simply underscores the failure of the EU to legislate social policy, while to others it represents the leading edge of a regional policy, with social legislation endogenous to court-led jurisprudence. But European social policy is part of a much broader movement to legalization and the creation of rights, which has enhanced the role of the ECJ.Footnote 46 For example, Mabbett argues that EU rights extend beyond the market (from market citizen to citizen of Europe), using the EU's record combating discrimination and advancing disability rights as her case study.Footnote 47 Shapiro interprets the work of the ECJ as part of a dramatic increase in courts as well as judicial review on the continent of Europe.Footnote 48 He argues effectively that these institutions have spread for the most part because they are successful agents of dispute-resolution.
While some see the expansion of rights as all but inevitable, others are more cautious. De Burca,Footnote 49 while recognizing advances in the autonomy (from the economy) of EU law, is also quick to point out that the Court has been hesitant to tread where legislative institutions have yet to lay down markers. Others, for example, Palier and PochetFootnote 50 tie the expansion of rights to practices, such as the open method of coordination, that are not yet entrenched. Still, the fact remains that courts and judicial review are spreading as a means of third-party dispute resolution.Footnote 51
We conclude from this vast body of literature that while both the regulatory state approach and the “EU as agent of neoliberalism” arguments should be taken seriously, the debate about the EU's social implications requires more central attention to legalization and especially to the role of the ECJ. It is clear that social policy remains lodged at the level of the national state, but the international aspects of social regulation have increasingly migrated to the European level, where the court has assumed a key monitoring, legitimating, and adjudicating role.
But why has the ECJ assumed this role? Why not the EU's legislative institutions or some combination of interest groups in alliance with the Commission as a supranational bureaucracy? Part of the answer is negative and part is positive. Negatively, other EU institutions have major limitations. The Council of Ministers is paralyzed by the veto and thus captive to the preferences of the least cooperative member. Interest groups, numerous and fragmented, face insuperable collective action problems. The Commission, which has the right to initiate legislation, is tied to the Council of Ministers, which in turn is bound by the veto. Positively, the ECJ does not face these obstacles. It is politically insulated; it does not face a consensus requirement since it votes by majority; and it does not need a prior legislative base, though of course it does need some legal basis in the Treaty. Given high levels of economic interdependence among the peoples of Europe, the inevitable disputes that result from this interdependence, a growing body of law, and some delegated authority, the ECJ is the natural dispute resolution institution. As we will show below, this provides the Court with a unique strategic position with regard to elements of the opening of markets, such as the free movement of labor.
The Free Movement of Labor
Beginning with the Treaty of Rome, Article 48 [post-Amsterdam Article 39] established that workers in any member state have the right to move to another member state to work there, and to settle in another country with their families. The self-employed have the right of establishment under Article 52 and the right to provide services under Article 59. Furthermore, secondary legislation (Regulation 1408/71) lays down that persons residing in the territory of one member state where certain provisions apply to nationals are subject to the same obligations and enjoy the same benefits under its legislation as nationals of that state.Footnote 52 This principle, free movement of workers, is an economic right in that workers are entitled to move from country to country in search of work.
Nevertheless, the treaty itself provides only the barest outline of the social conditions surrounding worker movement, that is, the social embeddedness of workers. The general idea motivating this section is that ECJ judgments as well as EU legislation will not treat workers as abstract units of labor but—as our third definition of embedding implies—embed them within social networks. What this means operationally is that spouses, children, and family relations will be taken into account when making law about movement across national borders.
As we follow the developing law of the EU regarding free movement of labor, including case and statutory law, we will see three successive phases which illustrate the growth and thickening of labor market policy to include ever more social content.
• The first phase laid down the foundations for free movement, giving the relevant treaty provisions direct effect, and clearly establishing the meaning of worker.
• The second phase used the logic of market failure to expand the scope of free movement. This phase was more regulatory than deregulatory.
• In the third phase, the social standing of workers and people in general was taken into account. Here the jurisprudence of the ECJ and secondary legislation shaped labor markets beyond the common understanding of efficiency.
First Phase: The Foundation
The right to free movement of labor has been interpreted by the ECJ as a fundamental right with direct effect. Direct effect means that provisions of the Rome Treaty are directly effective (without national mediation) and that individuals can seek legal recourse if they think their personal rights have been abridged under the treaty. While this point may seem obvious, the legal standing of workers cannot be taken for granted once outside the context of the national state. National labor legislation generally applies to nationals, that is, to citizens of the state in question. If a national from state A moves to state B and assumes a position of work, it is not clear what, if any, labor legislation applies to this worker. The Treaty of Rome contains provisions relevant to the position of migrant workers but when the treaty was signed, it was not clear what practical effect this would have, since provisions of treaties do not usually directly create rights and obligations. Indeed, the treaty and subsequent treaties say many things about a variety of issues, only a small portion of which are given direct effect. It was not until the Van Gend en Loos caseFootnote 53 in 1963 that the principle of direct effect was promulgated for any part of the treaty and it was not until later that this basic judicial principle spread to other areas, such as gender equality policy.
The ECJ established direct effect for workers and construed efforts on the part of governments to establish conditions that, intentionally or not, had the effect of restricting the access of foreign workers as violations of the treaty. The court stated in RoyerFootnote 54 that Articles 48, 52, and 59 “which may be construed as prohibiting Member States from setting up restrictions or obstacles to the entry into and residence in their territory of nationals of other Member States, have the effect of conferring rights directly on all persons falling with [in] their ambit.”Footnote 55 The court also clarified the meaning of worker, refusing to allow the Dutch government to disallow the label “worker” to a person who worked only part-time in another country for a wage the Dutch government considered too low for subsistence. In doing so,Footnote 56 the court drew on earlier jurisprudence from a 1963 case in which it made clear that the definition of worker was a matter of EU law. This was not an arbitrary move on the part of the court but was instead seen as necessary for the implementation of the free movement provisions of the treaty.
If states could decide what it meant to be a worker, members could in effect escape the reach of EU law by defining workers in a narrow way. This foundational phase put in place the tools for the second and third phases examined below.
Second Phase: Market-Failure Jurisprudence
In a sense, the logic of market failure is the revealed master variable behind the jurisprudence of the free movement of workers. The very reason the treaties contain provisions for free movement of workers is the belief that the barriers to worker movement across borders are great, and that some of these barriers are legal and institutional, that is, not due to a “failure of preferences” or the simple absence of job opportunities. For a market failure to be categorized as such, preferences (to move) and opportunities (to find work) must be present, yet there is a failure of exchange, that is, of a labor contract between individuals in different countries.
Thus, when national practices exist, such as discriminatory treatment in favor of domestic workers, access to special benefits on the part of nationals, or failure to coordinate social security provisions across countries, Treaty provisions can be invoked by injured parties to claim redress. In this sense, the Treaty aims to correct failures of labor mobility when movement otherwise would have taken place. A fluid labor market requires nondiscriminatory treatment. It may also require laws to coordinate separate legal systems, whether they are discriminatory in intent or not. In this sense, laws or institutions are permissive rather than restrictive. They permit actions to take place that otherwise would not have.
Two striking examples of market-failure-driven jurisprudence concern the continuity of the working lives of workers who labor in different countries and the Court's case law regarding tying benefits to residence. Consider the continuity issue. What happens when a worker completes part of his or her life's work in one country and another part in a different country? It is quite possible that the minimum number of years required to collect benefits (or full benefits) will not be met in either country. A worker who works fourteen years in Italy and four years in Germany satisfies the minimum conditions neither for an Italian pension (fifteen to twenty years) nor a German pension (minimum of five years).Footnote 57 Here the market and nationally defined benefits are arranged in such a way as to prevent benefits from being collected for a transnational worker. Surely separate national treatment will result in labor market failures, and a less than optimal number of workers will cross national frontiers for this reason.
The ECJ has begun to tackle cases of this type. One case arose out of the denial by Dutch authorities of cash benefits to a Dutch woman, Ms. Klaus, who had worked successively in the Netherlands, then Spain, then the Netherlands, and then Spain once again.Footnote 58 She was denied benefits because of a provision of Dutch law stating that no cash benefits should go to a person who, at the moment of entry into the insurance scheme, was not capable of work. After being turned down by the Dutch social security institution, she appealed her case to a Dutch Tribunal who put questions to the ECJ under the Article 177 procedure. The court rendered a judgment that supported Ms. Klaus, saying that “the working life of the person concerned should be seen as a whole, and not just from the limited standpoint of a particular job in one country, at one period of time.”Footnote 59
What about the link between residence requirements and the distribution of pensions? Here the Council of Ministers and the ECJ have teamed up to provide an impressive legal structure facilitating worker rights in the face of residence requirements of entrenched welfare states. Some states require residence in the country of employment in order for pensions to be paid. This would mean that a Danish worker who had worked his or her entire life in Denmark could not choose to retire in Portugal, and collect benefits. Such territorially based provisions are obviously prejudicial to migrant workers.
Having taken advantage of the regionwide market, and having accumulated a pension, the worker must choose between the benefits and preferred place of living. Legislation passed by the Council of MinistersFootnote 60 has waived residence requirements and the ECJ has aggressively interpreted Council regulations so as to broaden the scope of their application. Again using the free movement provisions of the treaty as well as secondary legislation, the court has decided that a pension already acquired cannot be subject to a residence condition. Also one cannot be denied entitlement to a pension solely because of residence in another member state.Footnote 61
These are straightforward examples of case law that are intended to further the free movement of workers and peoples. But the logic of market failure can only be pushed so far. There is a “bridge case” that we think captures the limits of the market failure rationale. It applies to tourists rather than workers but the logic is the same. The case is Cowan v. Tresor Public.Footnote 62
In this case, Mr. Cowan, a British national, was mugged on a trip (holiday) to Paris. He applied for monetary compensation under a provision of the French criminal law. That provision allowed for compensation only for French nationals, or if there were a reciprocal agreement between France and the victim's country (which there was not). The French Administrative Tribunal referred the case to Luxembourg, where the French government argued that compensation was a right that is a manifestation of national solidarity. The ECJ rejected France's efforts to couple benefits with nationality since free movement of persons was involved. Instead, the court found in favor of Cowan, arguing that freedom of services implies the right to be protected from harm in the member states in question, and on the same basis as the nationals residing there.Footnote 63
In the economist's world, a fully developed vision of market failure would allow for the possibility that others would be discouraged by Cowan's experience (had Cowan been successfully denied by the French government) and would not take advantage of opportunities for tourism in light of the fear of being mugged without compensation. The ECJ apparently showed some concern that discriminatory treatment would create disincentives for tourists, thus raising the market failure flag, but this must surely be the most expansive interpretation ever of market failure on historical record. In any case, the court made no effort to explore the nature of these incentives. It was enough for the ECJ that Mr. Cowan was a Community citizen and that he was seeking access to services that were his right under Community law.Footnote 64 This suggests that Cowan's right to be compensated rested more on his citizenship in the EU, that is, on his membership in a political community, than on his economic status as tourist. Cowan, a British and EU citizen, is seen as politically embedded both with respect to French and EU citizenship rights. The Cowan case may represent the exhaustion of externality-driven jurisprudence in free movement.
The Third Phase: The Social Embedding of the Labor Market
Until recently, the ECJ's jurisprudence has been ancillary to the creation of a fluid transnational labor market. But the court has gone beyond this to take family and social considerations into account. One context in which family and other social considerations are in evidence relates to access to benefits for non-national workers with or without family.
Many countries have provisions for increased benefits for workers with dependents, including migrant workers with dependents, as long as they reside in the state in question. National laws relating to worker rights are potentially in conflict with the developing law of the EU. The court has begun to test these national legislative requirements and while it is still too early to know what the outcome will be, there are some indications. The free movement provisions of the Rome Treaty are potentially quite powerful and exchange of persons, particularly wage contracts across borders, may implicate a host of social phenomena not likely to be anticipated.
We examine a number of cases from the jurisprudence of the ECJ, selected so as to illustrate the social content of the free movement provisions. The first case is Commission v. Italy.Footnote 65 While this case does not directly involve family considerations relating to free movement, it does interpret free movement in a broadly social way, that is, in such a way as to take into account the social situation of workers who cross national boundaries. This case involved an Italian law that required that persons who rented or bought property that was itself renovated with public funds be Italian nationals, on the not unreasonable rule that consumption of benefits and payment for the goods should be linked. The ECJ rejected the position of the Italian government and made a quite broad defense of free movement, by arguing that the Rome Treaty's position on free movement “is concerned not solely with the specific rules on the pursuit of occupational activities but also with the rules relating to the various general facilities which are of assistance in the pursuit of those activities.”Footnote 66 In the same opinion of the Advocate General, it was noted that for free movement to be effective, access to broad benefits was necessary to foster integration of “self-employed workers and their families into the host country…”Footnote 67
In taking on this difficult issue, the ECJ weakened the link between national payment and national consumption. To be sure, this link was not complete, since migrant workers also might contribute via payroll taxes, sales taxes (value-added taxes), and property taxes. Nevertheless, one kind of solidarity (among nationals and their political institutions) was weakened and another was strengthened (between EU institutions and foreign workers).
From here, the ECJ's jurisprudence and legislation of the Council of Ministers tackled questions that more directly involved family considerations. The national legislation of almost all member states requires residence of family members on their territory in order to receive benefits. Following this rule would inhibit worker movement, since in many cases workers would be deprived of family benefits in both country of employment and country of residence (home country). As Cornelissen points out, “a frontier worker resident with his family in Belgium and working in Germany fulfils neither the conditions to entitlement required by Belgian legislation for Belgian family benefits (he is not insured in Belgium) nor those required by German legislation for German family benefits (his children are not resident in Germany).”Footnote 68 Anomalies such as this one are removed by legislation, specifically Regulation 1408/1971, which provides for removal of residence requirements for family members to receive benefits. This Regulation has been repeatedly tested and consistently decided in favor of workers' rights.Footnote 69
While these cases can be read as an extension of market-failure logic, in the sense that they made it easier for the Italian workers to work in Germany, a careful reading suggests that there is more than market-failure at work here. In the Bronzino case, there was little reason to think that the decision of the Italian workers to stay in Germany depended in any way on the payment of supplemental benefits for the children.Footnote 70 Indeed, the judgment of the court did not seem to rely on this line of reasoning.
An additional case illustrates the increasing detachment of worker rights from the actual movement of workers across borders and hints at the growth of an independent body of social rights only partially grounded in efficiency considerations. This case, Mary Carpenter v. Secretary of State,Footnote 71 has to do with rights of free movement by a person who never tried to exercise his or her right. This is a complex case which suggests the intricate relationship between law, markets, and social institutions. Ms. Carpenter was not a national of any EU state but rather a Philippine national. She visited the United Kingdom (UK) for six months and overstayed her permit and subsequently married Peter Carpenter, a UK national. Then Ms. Carpenter applied for permission to stay in the UK as the spouse of Mr. Carpenter. Her application was refused and the British secretary of state decided to deport her to the Philippines. Ms. Carpenter argued that her right to reside in the UK derived from Mr. Carpenter's freedom to provide services to other EU states (under Article 49EC). Her deportation would either require Mr. Carpenter to give up his business or separate his family. The ECJ decided that even if the derivative right of residence is not provided by secondary legislation, it can be imputed from the clause “protection of the family life of nationals of member states in order to eliminate obstacles to exercise of fundamental freedoms.” This is a good example of the embedded nature of economic relationships, even in a case where the citizen of the member state never tried to exercise his or her rights with regard to freedom of services.
In summary, during the past decades, EU law concerning free movement of workers and others has been expanded significantly by the ECJ and the economic aspects of the law have become increasingly infused with social content. The ECJ and national courts have interpreted the social objectives of the treaties and secondary legislation broadly and have used the jurisprudence of the court to fill in the gaps in the EU treaties. In all of this, it is not so much the case that social policy has been created de novo as that social policy has been progressively “read into” the rules of the market concerning free movement.
Discussion
In this article, we started from a premise that was argued most forcefully by Polanyi, namely that markets are always embedded within society. “Actually existing markets” are never the anonymous, arms-length, impersonal constructions of pure economic theoryFootnote 72 and few analysts would subscribe to the straw person of pure economic theory as an accurate description of the world. However, there are many who would also not subscribe to the strong version of embeddedness put forth here. In part, this may be due to the nonpublic, incremental, and piecemeal way in which the social content of the market has been inserted into market-making purposes in the cases we have studied. In part, it may be due to the fact that the ECJ's jurisprudence facilitates, rather than deters, commodification of labor. While the completion of the European Single Market and the implementation of the four freedoms came with much fanfare (“Europe 1992”), as Polanyi would have predicted, there was no equivalently dramatic and coherent countermovement to a European social policy, as he would also have predicted. Just as the UK in the nineteenth century countered the movement of liberal economics with incremental, scattered, and contradictory social protections, there have been few broad legislative initiatives by the EU equivalent to the grand pronouncements of the Single European Act. But the ECJ has interpreted existing treaty provisions and secondary legislation in an increasingly social way.
Polanyi was thinking of something deeper and less obvious than the simple claim that markets are embedded. Instead, he thought that market-making and market-embedding are ongoing processes in which market exchanges are “emancipated” from old structures at the same time as they are re-embedded in new ones. In the EU, the language used is often technocratic (regulation, deregulation, and reregulation) rather than sociological (embedded, disembedded, re-embedded) but the result is the same. Markets, politics, and social relations are part of the same comprehensive process. In part, this is because markets can't function without social and political content and in part, it is because people want things (security, protection from the disruption of markets) that spare versions of markets can't supply.
Our analysis may stir dissent from various quarters and if it does, to our minds, this would be a good thing.
A first disagreement is likely to come from European social democrats—both activists and scholars. Here it may be useful to return to a theme raised earlier in our article. Scholars such as Scharpf,Footnote 73 Streeck,Footnote 74 and Streeck and SchmitterFootnote 75 have argued that the quest to extend markets to the European level has weakened national social protections. Thus, the welfare states of Western Europe, which took more than a century to establish, are in danger of being torn down. Significantly, one of the conditions for the construction of European welfare states was selective closure of borders so as to prevent opt-in and opt-out behavior with respect to welfare shopping, which decreased the possibility of exit and increased the incentives for political voice, thus lessening competitive pressures on welfare states.Footnote 76
While we cannot hope to put this controversy to rest in this study, we think our article can stimulate refinements of arguments and more systematic empirical evidence on the relationship between national welfare states and regional integration. For one thing, the evidence to date is far from conclusively supportive of the idea that globalization undermines national welfare states.Footnote 77 A range of political options is possible, from economic protection, to openness with flexible labor markets, to the joining of openness and welfare compensation.Footnote 78 For another, for our argument to have force, national welfare states need not be recreated out of whole cloth at the European level. Our sense is only that a certain kind of coordination of existing welfare states is occurring—coordination that has its source in national policy externalities of the kind discussed in divergent social security programs. Whether national welfare states will be cut back, modified, strengthened (as compensation for global risks), or simply supplemented by social programs on a regional scale has yet to be decided.
A second line of criticism might be that the jurisprudence of the ECJ is too thin a gruel to sustain an international system of social protection. Some have pointed out that the case law is slight, has too many gaps, and has too little on which to draw from the Rome Treaty and subsequent treaties, apart from the four freedoms. Even the protections that do exist depend on market participation and do not extend to people as members of a political system. Further, a court-led social policy, cut off from popular politics, is vulnerable to permutations in top-down processes. Where are the social movements demanding that free movement of labor be embedded with social considerations, our fiercest critics may ask?
We respectfully demur from this line of criticism. Judges are indeed cut off from popular politics and—as some of our critics have argued—may have broadened the rights surrounding movement of labor as part of their professional careers. But first, as Polanyi showed about nineteenth-century Britain, not every part of a countermovement emerges from the base of society or has to be channeled through popular politics. It is worth recalling that one of the leaders in providing social protection to the working class was Bismarck—not a great friend of social movements!
Secondly, while the ECJ is indeed cut off from popular politics, its jurisprudence is thickest in those areas in which social-economic demands are strongestFootnote 79 and where interest groups, nongovernmental organizations (NGOs), and social movements are most active. Conant's work is eloquent in making this point: it was only in the sectors in which strong domestic actors were pushing for clarification or extension of European treaties that the Court's decisions were ultimately effective.Footnote 80 Cichowski's book on gender rights makes a similar point: she shows how the institutionalization of social rights is occurring through the interaction between supranational litigation and transnational mobilization.Footnote 81
A third set of critics might argue that the ECJ's judgments on the free movement of labor is an “easy” case in the sense that if the expanding social content of the market can be demonstrated anywhere, it should show up in labor exchanges. Our response would be that we expect qualitatively different—but not absent—patterns of embeddedness to emerge in various sectors. For example, capital transactions may rest on a deep securitization of rules of exchangeFootnote 82 and on the development of personal networks of trust; the sale of sports stars may involve social understandings of community and identity and not merely efficiency; and in the area of gender equality, the court has taken a strong stand defending equality in labor markets, even when this is not economically efficient.
A fourth criticism can be raised as a question. If the ECJ is an agent of social policy, why is this the case? Unlike the U.S. Supreme Court, for which there is a voluminous literature on the attitudes and ideologies of the justices, we know little about the justices on the ECJ. The court decides by majority but votes are not published so “yeas and nays” cannot be assigned to individuals. Also, there are neither concurring nor dissenting opinions nor elaborate justifications for the judgments given, as in the U.S. Supreme Court. The result is that we cannot predict the direction of the ECJ's social jurisprudence from the attitudes of the justices. To impute those preferences from judicial outcomes is also an unproductive circularity.
We can only speculate as to why the ECJ has moved in the direction of social policy. First, as legal scholars tell us,Footnote 83 justices take the law and legal texts seriously, especially higher laws such as treaties. Consistency is highly valued and is behind the logic of precedent. The provisions of the Nice Treaty on labor mobilityFootnote 84 and on the free movement of labor are central. Second, while the court did not pioneer free movement of labor in the same sense that it led the way on gender equality, it quickly responded to the legislation of the Commission and Council of Ministers.Footnote 85 The Commission proved to be a valuable ally by supplying a concrete set of social regulations relevant to labor markets. This should not be surprising given that the Commission was an institutional magnet for social policy, even in failure.
A final critical reaction will certainly be that the EU is such a special regional institution, and the ECJ has such distinctive powers of adjudication over the actions of member-states that our case study has few implications for the international system as a whole. While we agree with authors such as KatzensteinFootnote 86 that the extent of legal and judicial integration in Europe is unique, and with Eberlein and Grande that “European integration has been distinct from other forms of regional integration, such as Asia-Pacific Economic Cooperation or the North Atlantic Free Trade Agreement,”Footnote 87 we hesitate to regard its mechanisms as unique. With Ruggie, we see a growing importance of state and nonstate actors in a complex international system.Footnote 88 With Grande and Pauly, we see increasing reliance on nonhierarchical and nonmajoritarian methods of political conflict resolution at all levels of the system—what they call “complex sovereignty” and what one of us has referred to as “complex internationalization.”Footnote 89 With Goldstein and her collaborators in a special issue of this journal, we see a growing process of international legalization in which the EU is an extreme, but not an isolated example.Footnote 90 We agree with the generalizing ambitions of the International Organization special issue that legalization is a general process that takes place in different forms and at different rates throughout the world.
Conclusion
In this article, we have argued that social policy is already “here” in the EU, that the lines between market and social policy are blurred, and that the logic of economic exchange cannot be kept separate from broad social considerations. We are not the first to have made this argument. But we think our particular contribution has been to show that—at least in the sector we have examined—the economy is always embedded, that it is part of an ongoing “instituted process,” and that, despite the oft-claimed uniqueness of the EU, the interplay between market forces and social norms is analogous to both the movement/countermovement dialectic that Polanyi perceived, and to the grand compromise between the open international economy and “legitimate social purpose” that Ruggie saw at the end of World War I.
While it is true that the Rome Treaty and its subsequent revisions have primarily attempted to institute a market, this process has involved far more than the creation of a system of private exchange. Polanyi would surely raise an eyebrow at the phrase “economic rights in the EU are limited to the market.” While rights are tied to economic activities such as work, trading, and production, these activities are not arms-length economic transactions. They are socially and politically embedded. The Court's jurisprudence has already gone beyond a strict economic definition of the market to take account of spouses, family, access to infrastructure (for example, public housing), and social considerations broadly conceived.
If we are correct that the “movement” to free European markets and the “countermovement” against it are both present in ECJ decision making, then we should not expect that the conflict between free markets and social protection will ever be resolved. Just as the postwar consensus that led to the Bretton Woods system has been followed by decades of conflict and uncertainty, continuing to today's failed Doha Round and to the muddled responses to the 2008–2009 financial crisis, the co-presence of neoliberalism and social protection in the European Union is certain to lead to continued struggle. Will neoliberalism “win,” in some fundamental sense? Will social protection “catch up” in an equally fundamental sense? We think not. Just as embedded liberalism has been under assault in the international financial system since its construction in the years following World War II, the conflict between neoliberalism and social regulation will continue as long as there is no authoritative, consensual, and determined understanding of the legitimate social purpose of the EU.
On the other hand, the general shift to legalization in international politics, the fact that the ECJ is the most legitimate supranational institution in the EU, and the growing socialization of national courts and lawyers into the acceptance of the integration of European law into domestic jurisprudenceFootnote 91 convinces us that the ECJ is not without resources in building a tissue of embedded liberalism. Like the countermovement that Polanyi pointed to, and its successor after World War II, the struggle over embedded liberalism will continue. It can already be seen in its latest incarnation in the European and American responses to the breakdown of the world financial system in 2008–2009. That, we suspect, will produce a new cycle of contention between the movement for free markets and the countermovement to embed those markets in a legitimate social purpose.
