Above all, The Great Transformation tells of the conflict between the imperatives of a capitalist world economy and the pursuit of social welfare within nation-states. Polanyi's account of the 1920s and 1930s analyzes the incompatibility of international capitalist arrangements with both democracy and the social reforms that had been won by the European working classes.
—Fred Block and Margaret R. Somers, “Beyond the Economistic Fallacy”Footnote 1Economically speaking, the Union remains, with its dense web of directives, and often dubious prebends, far from a perfect Hayekian order. But in its political distance from the populations over which it presides, it approaches the ideal he projected. What he did not anticipate, though it would perhaps not have surprised, and certainly not disconcerted him, is the disaffection that the regime he envisaged has aroused in the masses subject to its decisions.
—Perry Anderson, The New Old World Footnote 2Although regional integration can be seen as governments' attempt to recover some of the capacity that has been lost at the national level to intervene in markets, it is often exceedingly difficult to move forward with transnational social policy or, more generally, with redistribution, because national welfare states differ not only in their generosity but also in the way they organize redistribution. As a result, governments have frequently been unwilling to transfer to the transnational level the capacity to tax and redistribute. Such forms of political integration become easy prey to governments' desire to defend historically grown and sometimes hard-won institutions. However, some observers have recently argued that international courts might step in where there is political stalemate. Because they are beyond political haggling, judges might be the actors who infuse social content into regional integration, but there are limits to what courts can achieve. In particular, while court rulings can strengthen individual rights, they cannot create the norms of reciprocity and solidarity that underpin redistribution. In fact, legally enforced access to national social policy programs might in the end lower the level of social protection at the national level.
With regard to the European Union (EU), Caporaso and Tarrow have recently argued that decisions of the European Court of Justice (ECJ) have led to a re-embedding of markets.Footnote 3 Drawing on Polanyi's The Great Transformation, they argue that, after a period of economic liberalism and market building, the pendulum is now swinging back toward social policy and correcting markets. Just like Polanyi envisioned, they argue, free markets are so disruptive a force that eventually the demand for more regulation emerges. The term “double movement” famously captures this dynamic. Caporaso and Tarrow focus neither on social movements that rally against free markets nor on elected governments that are pushed by voter demand for social protection. Instead, they identify the ECJ as the agent restricting the reach of the Common Market and contend that the ECJ has gradually built up an array of social rights that apply to EU citizens. From their perspective, the Luxembourg judges have taken on the task of re-embedding the market.
In this article, we take a different position. Although we find a Polanyian perspective useful and support the claim that any empirically sound assessment of European integration has to recognize the rulings of the ECJ, our conclusions differ starkly from Caporaso and Tarrow. Instead of re-embedding markets, the EU is beginning to resemble Hayek's blueprint of “interstate federalism,” where individual (economic and social) rights are located at the central level while the capacity for taxation and interpersonal redistribution remains entirely decentralized. What appears to be the nucleus of supranational social policy might turn out to be a recipe for less social protection and redistribution at the national level.
There are two reasons for this. First, by granting non-nationals access to social transfers while being unable to oblige them to contribute financially puts pressure on the generosity given to all entitled persons. As economic liberals have aptly observed, divorcing rights from obligations limits the capacity for redistribution. Second, easing the free movement of labor is only one subdimension of European social and economic integration. Admittedly, the ECJ has granted social rights not only to workers who cross borders but also to their dependents and, thus, infused social content into the principle of free movement. Yet, this is only part of the story. To get the full picture, we need to look at three dimensions of European integration: market-shaping integration, market-enhancing integration, and the creation of a European area of nondiscrimination. Taking these three dimensions into account shows that political initiatives to re-embed markets have become extremely difficult as EU members have grown ever more economically diverse. At the same time, integration through law (as opposed to political integration) continues apace and limits national governments' ability to correct markets.
Despite its focus on the EU, this article addresses the more general question of what type of federation regional integration produces. While it is generally accepted in the comparative political economy literature that federalism reduces the size of government,Footnote 4 recent research has further refined this general insight in showing that the diversity of the constituent units shapes decisions to centralize social spending. In particular Beramendi has shown that regional differences in wealth and inequality render any centralization of spending capacities unlikely.Footnote 5 Hence, while an international union of relatively homogenous countries—like the initial six members of today's EU—could in principle centralize redistribution, a group of more heterogeneous countries will find it very difficult to agree on such a scheme. If regional integration takes place among economically diverse countries—and this may actually be the constellation in which it promises the largest welfare gains—it is bound to resemble a Hayekian type of federalism.
Hayek maintained that interstate federalism will necessarily limit the political capacity to intervene in markets as long as the power to tax and to redistribute remains the prerogative of the constituent units while supranational agents guarantee the free movements of goods, capital, services, and labor.Footnote 6 Regional integration generally follows this pattern as national governments defend their right to tax and to spend. Under conditions of heterogeneity, regional cooperation creates an unusually decentralized type of federalism, in which taxation and social spending are the exclusive prerogative of the member states.Footnote 7 What makes the EU an interesting, although not unique, case is that the ECJ can push ahead with integration—removing obstacles to factor mobility—even in the absence of political agreements. Since a number of international courts replicate features of the ECJ design,Footnote 8 our findings are relevant beyond the seemingly special case of the EU. As other economically heterogeneous regions follow in the footsteps of the EU in creating interstate federalism, they might well witness the same pattern of political gridlock and judicial dynamic that facilitates economic liberalization rather than social protection and redistribution.
We discuss Polanyi's core concepts of embeddedness and double movement to see how these can contribute to understanding regional integration today. We then turn to the example of European integration, first documenting the enormous heterogeneity among EU member states in terms of social models and varieties of capitalism. These differences have reached unprecedented levels in the wake of Eastern European enlargement and not only render common political projects such as social policy harmonization unlikely, but also enlarge supranational actors' opportunities to push “integration through law” because interest diversity reduces the probability of a legislative override of ECJ decisions. Subsequently, we outline the asymmetry between three dimensions of European integration: market-shaping integration, market-enhancing integration, and the creation of a European area of nondiscrimination. We show that the main effect of European integration remains the freeing of individuals from collectively imposed obligations. It therefore does not—in a Polanyian sense—enable societies to socially embed the market. In addition, we argue that the forces that enlarge the opportunity structure for granting individual rights vis-à-vis member states while at the same time constraining the likelihood for ambitious social integration are not specific to the EU, but rather reflect a general logic of regional integration under conditions of political-economic heterogeneity. Finally, we link our findings to the rise of EU-skeptical sentiments among those most vulnerable to the markets and to the risk of “negative politicization.” We contend that the failure to transnationally re-embed markets leads to political consequences that threaten progress toward political integration.
Embeddedness and the Political Nature of the Double Movement
Polanyi wrote The Great Transformation during World War II. In this book he tries to understand why democracy collapsed during the 1920s and 1930s and gave rise to fascism and “wars of an unprecedented type.”Footnote 9 Polanyi's answer is that “the origins of the cataclysm lay in the utopian endeavor of economic liberalism to set up a self-regulating market system.”Footnote 10 Economic liberalism, in Polanyi's core argument, entails the “stark utopia” that everything should be tradable—even labor, land, and money—although for him these “are obviously not commodities” and to “include them in the market mechanism means to subordinate the substance of society itself to the laws of the market.” He goes on to call this the “commodity fiction,” which is vital for a market society dominated by the principle that “no arrangement or behavior should be allowed to exist that might prevent the actual functioning of the market mechanism.”Footnote 11 In a market society, markets have to be shielded against political interventionism and against any tampering with the price mechanism, which is best safeguarded by separating politics and the economy.
According to Polanyi, embeddedness means the degree to which social relations are protected from and are sustainable independently of markets.Footnote 12 In contrast, disembedding the economy means commodifying the “very substance” of society, that is, land, money, and human beings. Their prices are determined through market exchange rather than through norms of appropriateness or political decisions: “Nothing must be allowed to inhibit the formation of markets, nor must incomes be permitted to be formed otherwise than through sales…. Neither price, nor supply, nor demand must be fixed or regulated; only such policies and measures are in order which help to ensure the self-regulation of the market by creating conditions which make the market the only organizing power in the economic sphere.”Footnote 13 In a market society, solidaristic forms of engagement such as reciprocity and redistribution, which used to coexist with markets, are subjugated to market exchange. Hence, disembedding markets means “no less than the running of society as an adjunct to the market. Instead of economy being embedded in social relations, social relations are embedded in the economic system.”Footnote 14
In the end, however, disembedded markets are unsustainable because they threaten to destroy society. Left to its own, the “satanic mill” of the market leads to a degree of social dislocation that no society can bear for long. Eventually, a countermovement against liberalization will seek to decommodify fictitious commodities. Polanyi uses the term “double movement” to capture the dynamic quest for free markets and the demand for protection.Footnote 15 Insofar as the countermovement succeeds, however, it undermines the efficiency of the market and might therefore exacerbate the economic situation of those who looked for social protection in the first place.Footnote 16 Protective measures are “the only means to save society from destruction through the blind action of the market mechanism” but these measures are “directly responsible for the aggravation of slumps and the restriction of trade.”Footnote 17 The double movement does not simply balance efficiency and equity but refers to contradictory impulses. In fact, during the interwar period, these antagonistic forces proved fatal for democracy.
While governments in the 1920s defended the market system—with the gold standard, free trade, and competitive labor markets at its core—the demand for social protection grew at the same time. Governments' commitment to fixed exchange rates even at the price of falling wages, rising unemployment, and human misery radicalized the left in Europe. Due to universal suffrage introduced in many countries after World War I, the labor movement was in a position to push for more far-reaching interventions, which threatened to undermine the market. In this situation, Polanyi says, the tension between capitalism and democracy became insuperable: “Labor entrenched itself in parliament where its numbers gave it weight, capitalists built industry into a fortress from which to lord the land…. The captains of industry were subverting the population from allegiance to their own freely elected rulers, while democratic bodies carried on warfare against the industrial system on which everybody's livelihood depended.”Footnote 18 Given this irreconcilable conflict, the time was ripe for the “fascist solution,” which saved the market but sacrificed democracy.Footnote 19
For Polanyi, the faith in self-regulating markets did not survive the cataclysm of the 1930s and 1940s. At the very end of his book, in a passage that many neo-Polanyians take as their point of departure, he outlined the contours of a new economic system. He maintained that in the future, markets would not be self-regulating: prices for land, labor, and money would no longer be exclusively fixed on the market but instead reflect political choices. Abandoning the gold standard would free governments from subjugating domestic policies to international commitments and, in turn, facilitate international cooperation. Polanyi summarized this vision as “economic collaboration of governments and the liberty to organize national life at will.”Footnote 20 These insights foreshadowed the Bretton Woods system of “embedded liberalism,” which reconciled international trade with the domestic capacity to intervene in markets and to correct market outcomes.Footnote 21
For authors like Caporaso and Tarrow, the postwar embedded liberalism compromise is presently being recreated at the European level, mainly through ECJ rulings. We believe, in contrast, that the EU's economic regime is closer to that of the interwar period. Albeit imperfectly, it resembles the “Conditions of Interstate Federalism” that Hayek outlined in 1939. In this far-sighted text, Hayek maintained that the main purpose of interstate federation is to secure peace. As a welcome side effect, however, such a federation would limit governments' ability to tax and spend if taxation and spending were to remain the prerogative of the constituent states.Footnote 22 Factor mobility in an interstate federation would prevent undue government interference in the economy and take away the power to affect prices. For the benefits to materialize, it is necessary that interstate grants and transfers are strictly limited and that the central state cannot impose taxes. Hayek summarizes the beneficial consequences of fiscally decentralized federation as follows:
There seems to be little possible doubt that the scope for the regulation of economic life will be much narrower for the central government of a federation than for national states. And since … the power of the states which comprise the federation will be yet more limited, much of the interference with economic life to which we have become accustomed will be altogether impracticable under a federal organization.Footnote 23
In some important respects, the EU comes closer to the model of competitive federalism than most federations.Footnote 24 While supranational actors have the means to protect the market against government intervention and to ensure price stability, the EU cannot raise taxes and transfers among the member states are very limited, while factor mobility is rising. In comparison to most national federations, the EU is atypical since the constituent states' expenditures are not funded by revenue-sharing schemes or tax transfers from the central authority.Footnote 25 Also, the EU lacks the capacity to impose uniform employees' codetermination procedures upon firms or to create a common collective bargaining system at the European level. In fact, it comes closer to what Weingast calls “market-preserving federalism” than most other federations do.Footnote 26
Echoing Hayek, economic liberals welcome the separation of market-making and market-shaping capacities in the EU as a protective device for individual freedom:
The Europe-wide economy has been substantially integrated, with historically unprecedented liberties of resource flows and trade across traditional national boundaries. Reform requires the establishment of a strong but limited central authority, empowered to enforce the openness of the economy, along with the other minimal state functions. In this way, and only in this way, can the vulnerability of the individual European to exploitation by national political units be reduced.Footnote 27
In other words, the specific type of economic federalism that exists in the EU is an effective way to prevent the re-embedding of markets. As the quote shown above indicates, economically liberal Nobel laureate Buchanan clearly thinks the setting is Hayekian. Some political consequences of the “unprecedented liberties” can already be detected: since the re-embedding of the market is not forthcoming, people most vulnerable to competition start to look for more radical political answers. Those who stand to lose from free markets have grown Euro-skeptic at best and nationalistic at worst—a finding that resonates well with Polanyi's interpretation of political developments during the interwar period.
Increased Heterogeneity of EU Member States
The disembedding impact of European integration is rooted in the varying dynamics of different integration forms. What perpetuates these diverse dynamics, however, is the economic, social, and institutional heterogeneity of European varieties of capitalism.
As Table 1 illustrates, vast differences in the levels of prosperity exist among member states. The table lists the purchasing power standard (PPS) adjusted per capita gross domestic product as a measure for economic welfare.Footnote 28 The average of today's EU with twenty-seven members (EU-27) was set at the value of 100. Differences in wealth reflect levels of productivity and are evident in the varying wage levels. The welfare level of five countries—Luxembourg, Ireland, the Netherlands, Austria, and Sweden—is more than 20 percent above the average. These very rich countries compare to ten countries whose level of prosperity lies below 80 percent of the average: Portugal, Malta, and eight of the acceded East European transformation countries. Poland, with its population of 39 million, generates a value added per capita that equals 58 percent of the EU-27's average.
TABLE 1. The heterogeneity of European varieties of capitalism
Notes: GDP per capita in PPS is the gross domestic product (GDP) per capita in 2008 according to the purchasing power standard (PPS) and as a percentage of the EU-27 average (Eurostat 2010a, 97). Total tax burden (including SSC) as percentage of GDP is the sum of each state's revenue from taxes and social security contributions (SSC) as a percentage of GDP in 2008 (Eurostat 2010b, 290). Social protection expenditures as percentage of GDP is the sum of all social protection expenditures as a percentage of the GDP for 2006 (Eurostat 2010a, 336). Collective bargaining coverage is the percentage of employees covered by wage agreements in 2006 (ETUC 2010, 5). Degree of trade union organization is the percentage of union-organized labor in the entire labor force (excluding retirees) in 2006 (ETUC 2010, 4). Degree of employer organization is the percentage of the labor force whose employers are members of an employers' association in 2006 (Visser Reference Visser2009).
Less than 30 percent of the annual gross domestic product (GDP) is raised as taxes and social security contributions in Ireland, Latvia, Romania, and Slovakia, but the figure is between 45 and 50 percent in Denmark and Sweden. This corresponds to the heterogeneous setups and sizes of European welfare states. The leanest welfare state (Latvia) and the most expansive one (France), measured by the sum of all social protection expenditures as a percentage of the GDP, differ by a factor of 2.6. Moreover, it is hard to imagine more diversity among the institutional premises of organized social partnership than those that exist now. With regard to the degree of union organization in the labor force (excluding retirees), national figures range from 8 percent (France) to 75 percent (Sweden). The data in the table show a similar situation for the degree of organization of employer associations, as well as for the percentage of employees covered by collective agreements above the firm level.
Member-state heterogeneity fuels political conflict in the EU. Hooghe and Marks show that an unprecedented degree of politicization over European integration has taken place in recent years.Footnote 29 Several highly politicized controversies over European issues can be traced back to conflicts involving opposing economic interests, and an increasing number of them arise from different levels of economic welfare. For example, the controversies over the Services DirectiveFootnote 30 and over the series of ECJ judgments starting with Viking Footnote 31 and Laval Footnote 32 deal with the question of how far companies from member states with low wages and low levels of labor protection are allowed to exploit the competitive advantage gained by lower wages and labor standards when posting their employees to the high-wage economies of the old member states.
Further conflicts of interest result from the institutional heterogeneity of European welfare states and varieties of capitalism that would exist even if the levels of welfare among EU member states were more or less equal.Footnote 33 The EU not only includes several “worlds of welfare capitalism,” but also different forms of corporate governance and of industrial relations.Footnote 34 Differences between these were found, for example, at the heart of disputes over both the European Takeover DirectiveFootnote 35 and the “golden shares” rulings, in which the ECJ reinforced the free movement of capital.Footnote 36 Generally, within the EU-27, opposing interests that go deeper than those of the pre-existing European communities of the six, twelve, or fifteen members have to be balanced out.
What is decisive for the purpose of this study is not just that member states' worlds of welfare and varieties of capitalism differ so much that talk about the European social model is misleading,Footnote 37 but, more importantly, that this heterogeneity shapes different integration dynamics differently. In short, proponents of redistributive policies have to direct their demands to the national level, while proponents of liberalization asymmetrically profit from the supremacy of European law. As a result, the EU's radical version of fiscal decentralization will presumably persist.
Three Dimensions of Economic and Social Integration
To illustrate how member state heterogeneity affects various forms of integration differently, we distinguish among three dimensions that all affect the extent to which markets are socially embedded: (1) market-shaping integration, (2) market-enforcing integration, and (3) the creation of a European nondiscrimination area. While the first depends primarily on political decisions, the latter two will proceed even if the political ability or will to act is absent. All three dimensions advance integration, but their scopes and speeds vary.
Market-Shaping Integration
In the EU, three pillars of market-shaping integration—by which we mean integration aiming at the social embedding of markets—exist. First, in some areas of social policy the EU has the competence to legislate. Usually, legislation takes the form of binding directives or regulations that are subject to judicial review. A limited number of competencies to correct markets was already included in the 1957 Treaty of Rome, which laid the foundations for the EU. These focused on gender equalityFootnote 38 and on portability of social security entitlements to facilitate the free movement of labor.Footnote 39 At the time, however, the national capacity to develop social policy was not in doubt and governments had little incentive to transfer additional competencies to the European level. All major welfare programs—such as pensions, unemployment insurance, or health care—remained firmly under national control. With later treaty amendments, the community gained more competencies, mainly in labor law. In particular, working conditions, informing and consulting workers, and equal treatment of women and men became a subject of European regulation.Footnote 40 In those areas where the community had gained competencies and the Council of Ministers decided with qualified majority, a fair degree of national social policy has been harmonized through the EU. Although limited in scope, European law has repeatedly made changes necessary, even in mature welfare states.Footnote 41
Attempts to expand European competencies beyond (some aspects of) labor law and gender equality have largely been unsuccessful. As a consequence, and this is the second pillar of market-shaping integration, governments began to resort to types of “soft” coordination in the late 1990s. In 1997, the Amsterdam Treaty introduced the Open Method of Coordination (OMC), which was spelled out in detail later the same year at the Luxembourg summit.Footnote 42 Core elements of the OMC consist of periodically defined common European policy goals to be implemented by each of the member states using measures suitable to each country's situation. Mutual exchange between member states is supposed to provide insight into the “externalities” of domestic decisions, trigger common perceptions, and facilitate reforms. In order to encourage learning, the commission reports on the progress made, and “peer reviews” take place regularly.Footnote 43 However, aside from this exercise in “naming and shaming,” no sanctions are available if a member state fails to comply. Both the ECJ and the European Parliament are excluded from the OMC.
In 2000, the OMC became closely linked to the “Lisbon Strategy,” which announced that the EU would turn into the most competitive economic region of the world within ten years. While this clearly has not materialized, there is a good deal of disagreement in the literature about what the OMC has achieved.Footnote 44 Certainly, the OMC has allowed member states to “agree to disagree” while nonetheless cooperating. In this way, new areas of social policy have become subject to an exchange of opinions and the desire for mutual learning, while soft coordination has widened the EU's social policy agenda. However, there are limits to what the OMC can achieve. First, it cannot enforce policy change and depends for its success on the voluntary compliance of governments. Yet national actors do not seem ready to alter their own economic strategy to meet the concerns of competitors, either with regard to wages or taxes. Second, the gist of the Lisbon Strategy is to promote employability and to make people fit for the market. Activation, lifelong learning, and “flexicurity”—whatever else their merits are—do not seek to protect labor from markets but to include in it as many people as possible. If the OMC proves successful, it will further commodify the “fictitious commodity” of labor.
Social dialog represents the third pillar of market-shaping integration. The EU has long propagated corporatist policymaking as a principle at the EU level. Yet lip service to social dialogue had few real-world consequences prior to the 1990s.Footnote 45 With the Maastricht Treaty, this began to change. Social partner organizations at the EU level—trade unions, employers, and industrial associations—were given the opportunity to work out directives themselves that the council could subsequently turn into binding legislation.Footnote 46 During the 1990s, several so-called social partner directives came into being, and it seemed that a new instrument for embedding the market had been found. Since then, however, social dialogue has lost its momentum. During the past decade, the social partners—mainly employers—have no longer been committed to advancing binding agreements. Instead, they rely on the same type of “soft coordination” as governments do—substituting “declarations of intent” for legislative proposals. The reason is that trade unions and employer federations are—just like governments in the council—confronted with growing interest heterogeneity which they find increasingly difficult to overcome.Footnote 47
In sum, the EU holds a number of market-shaping instruments for promoting social policy harmonization. Yet, if these rely on hard law, their scope is limited mainly to gender equality and labor law. Soft law, in contrast, can be used in highly diverse policy fields because it does not necessitate change. Not only the Council of Ministers but also the social partners readily rely on it given the diversity of their members' interests. As a consequence, and just as Hayek had predicted in his vision of interstate federalism, the potential for “positive integration” and market-shaping interventions is still limited and will most likely stay so as the EU grows more diverse.Footnote 48
Market-Enforcing Integration
The dynamics of market-shaping and market-enforcing integration differ significantly, as Scharpf has shown in several articles.Footnote 49 The reasons are that market-enforcing integration profits from expansionist judicial lawmaking and that the heterogeneity of European varieties of capitalism does not impede, but even promotes, this so-called “integration through law.”Footnote 50 Already in the 1960s, the ECJ ruled that the economic freedoms provided market participants with individual rights vis-à-vis their member states (the direct effect), and that European law generally overrides national law (the supremacy).Footnote 51 Armed with direct effect and supremacy, European competition law became a live weapon in the hands of supranational actors, because, from that point on, it could potentially be used to liberalize economic sectors for which it had initially not been made, namely third-sector areas providing public goods and services, such as telecommunications, energy, and transportation.Footnote 52 It took, however, more than twenty years until the commission began to set up treaty violation proceedings in order to break up state monopolies in third-sector areas. Until the mid-1980s, more and more areas were being discovered in which this body of law could be applied. For example, current legal disputes involve public banking, public broadcasting, and hospitals, and in all of these cases the semi-public organization of the respective sectors has, beyond other aims, a social purpose. Although it is important to keep this form of judicially enforced liberalization in mind, we will focus on another line of ECJ decisions: the expansionist interpretation of the reach and scope of the so-called fundamental freedoms.
Initially, the “four freedoms”—the free movement of goods, capital, services, and persons—aimed at guaranteeing discrimination-free transnational access to the member states' markets. Since its pathbreaking Dassonville and Cassis de Dijon decisions, however, the ECJ has replaced the principle of nondiscrimination by the principle of nonrestriction.Footnote 53 According to the nonrestriction principle, every national regulation that restricts the transnational exercise of one of the four freedoms is in potential violation of European law, even if the regulation is discrimination-free, that is, imposed equally on nationals and nonnationals alike. This reinterpretation has far-reaching implications for the scope of market-enforcing integration. Interpreted as individual rights for restriction-free market action, the four freedoms can now be used not merely to eliminate disguised protectionism on the part of the member states, but rather to target a wide variety of member states' political regulations as obstacles to European law.Footnote 54 Furthermore, this line of ECJ rulings not only targets political regulation but also—due to the legal doctrine on the horizontal effect of European law—the actions of private bodies such as firms or trade unions.
Three examples from the ECJ case law of the past decade relating to corporate law, labor dispute law, and tax regulation illustrate that market-making integration systematically tackles national regulations with social content. The practical impact of market-creating “integration through law” on social matters goes far beyond the impact that both soft coordination and the European social dialogue have had in the past and are likely to have in the foreseeable future.
Our first example for the social impact of market-enforcing integration is the liberalization of corporate law initiated by the ECJ. Until the end of the 1990s, European law was not considered an obstacle to applying the so-called “company seat theory” or “real seat doctrine.” This doctrine stated that the legal status of a company was not based on the place it was established, but on the place where the actual company headquarters was located. In other words, if the seat of a company was in Germany, its internal matters were governed by German law. Assuming that—as Dammann puts it—headquarter relocation costs usually outweigh the advantages of a more attractive corporate law, firms usually had no choice but to accept the respective body of regulation.Footnote 55
The ECJ overturned the application of the company seat doctrine in its rulings on Centros, Überseering, and Inspire Art.Footnote 56 In the view of the court, this theory's application violated the European freedom of establishment, and the judges saw no overriding reasons of general public interest to justify such a violation. In particular, the court ruled that the establishment of foreign letterbox firms (in which the company seat has no practical meaning for the economic activities of the business) is protected by European law. In practice, this implies that entrepreneurs now have the freedom to choose whichever legal form among the entire EU-27 they deem appropriate when founding a company.Footnote 57
The freedom to circumvent national corporate law has consequences for core elements of member states' production models. For example, when a company's seat is in Germany but it does not choose the German legal form, management board codetermination does not apply if the company has grown beyond the size of 500 or 2,000 employees.Footnote 58 This is not just a theoretical possibility, as recent experiences demonstrate. The court's corporate law decisions have led to a boom of firms with foreign legal forms in Germany. In most of the cases, the respective firms do not exceed the number of 500, let alone 2,000 employees. However, codetermination is affected in several cases, and their number is increasing. Pütz and Sick find that from December 2006 to December 2010, the number of cases relevant to codetermination (that is, firms of more than 500 employees) increased from seventeen to forty-three.Footnote 59 In effect, the ECJ has transformed German supervisory board codetermination, generally perceived as a key element of Germany's Soziale Marktwirtschaft, from an obligatory to a voluntary institution.
The ECJ decisions on Viking, Laval, and Rüffert—our second example—have recently received attention because they were interpreted as landmark decisions on the struggle between economic freedoms and social regulation in the European common market.Footnote 60 In the context of our discussion, two aspects are particularly important. The first is the reinterpretation of the Posted Workers Directive from 1996.Footnote 61 In Article 3, paragraph 1, this directive lists a number of mandatory rules for posted workers' minimum protection on matters such as pay, rest, and holidays, while Article 3, paragraph 7, explicitly states that this minimum protection in force in the host country “shall not prevent application of terms and conditions of employment which are more favourable to workers.” In Laval, however, the court referred to the list in Article 3, paragraph 1, as defining the ceiling on the maximum standards that member states are allowed to impose on posted employees from other EU member states.Footnote 62 With this judicial reinterpretation, the court effectively limited the host countries' room to maneuver in preventing races to the bottom in the field of labor standards, a problem that is set to become increasingly prevalent as heterogeneity among member states increases.
A second aspect of these judicial innovations is equally relevant for our discussion: the court further expanded the so-called horizontal or “third-party” effect of the European market freedoms, which implies that European law not only obliges member states, but also private bodies (such as industry associations and trade unions) to refrain from actions that might restrict market freedoms. In the decisions on the Viking and Laval cases, the court ruled that trade unions are obliged not to hinder or block transnational economic activity by collective action, such as strikes, unless the trade unions' demands were justified by overriding reasons of the public interest and passed the proportionality test.Footnote 63 Until Laval, few observers would have argued that restricting disputes among the social partners was among the aims of the EU. As a matter of fact, although Article 153, paragraph 5, of the Lisbon Treaty states that the treaty provisions do not apply to “the right to strike or the right to impose lock-outs,” the court has effectively subordinated the right to strike under the European fundamental freedoms.
Our third example for the social impact of market-creating ECJ case law concerns tax law, in particular the law on corporate income taxes. Common market integration opens up loopholes for tax avoidance. Firms transfer profits and losses across national borders to minimize the tax burden. In a series of decisions such as Cadbury Schweppes and Marks & Spencer, the ECJ ruled that the common market logic legitimized such tax-avoidance practices and that national efforts to restrict these were not justified by overriding reasons of the public interest.Footnote 64 By handing down these decisions, the ECJ has fueled inner-European tax competition, while diverging national interests have prevented the member states from being able to agree on a political solution to contain competition. The more heterogeneous the tax systems of the member states are, the more intense becomes the tax competition, and the more unlikely it is that political harmonization efforts succeed.Footnote 65
As Ganghof and Genschel have shown, the practical problem with the increase of inner-European tax competition is not its direct effect on company tax revenue.Footnote 66 So far, the broadening of corporate tax bases has prevented a dramatic fall of tax revenues. More important is the indirect effect of corporate tax competition on personal income taxes. Because firms can be used as tax shelters for personal income, the corporate tax has a protective function for the personal income tax (the so-called backstop function). As tax competition pushes nominal corporate tax rates down, the backstop function is undermined. In this situation, governments have two options: they can accept a widening tax rate gap between corporate tax rates and top personal tax rates, thereby opening up loopholes for top earners, or they can limit the personal income tax's progressivity. Corporate tax competition, in short, constrains the progressivity of the income tax. Therefore, it constrains member states' redistributive capacity. Rather than socially embedding the market, this line of ECJ case law undermines policies that aim at correcting market outcomes to achieve redistribution.
The three examples not only illustrate the expansionist dynamic of market creation by means of judicial lawmaking, but also show how market-enforcing integration affects social, Polanyian-style embeddedness at the national level. Authors such as Majone or Alesina and Wacziarag draw a clear line between market-enhancing policies, on the one hand, which by definition improve overall welfare and therefore “do not need a strong normative foundation” (implying that output legitimacy is sufficient to underpin such measures), and redistributive policies, on the other, which “can only be legitimated by majority decisions and hence place too heavy a burden on the fragile normative foundations of a transnational polity.”Footnote 67 This clear and elegant distinction, however, becomes decisively blurred as soon as one considers the indirect impact of extending the reach and scope of the four freedoms on social measures such as codetermination, collective labor law, and tax policy.
Unlike political integration that can easily fall prey to conflicting interests and political blockades, “integration through law” is not negatively affected by increasing heterogeneity. In fact, the opposite may hold true, for two different reasons. First, with a higher potential for conflicts and more cross-border economic transactions, the probability of legal disputes increases. Accordingly, the ECJ will be asked to adjudicate conflicting claims more often and, thus, will have further opportunities to tackle restrictions to the four freedoms.Footnote 68 Second, as heterogeneity among rule-takers rises, the probability of coordinated resistance to judicial interpretation of market freedoms should be in decline, and if it is true that the ECJ accounts for likely reactions to its jurisprudence, decreased likelihood of coordinated resistance should increase the court's zone of discretion.Footnote 69 We will come back to this point, but first turn to the third dimension of European social and economic integration.
Creation of a European Area of Nondiscrimination
Significant progress has also been achieved in a third dimension of integration that, like market-enforcing integration, proceeds by granting individual rights to market participants and, more generally, European citizens: the creation of a European area of nondiscrimination. We distinguish two subdimensions of this form of integration: protection against discrimination based on characteristics such as gender, age, and ethnicity, on the one hand; and discrimination-free transnational access to the social security systems of the member states, on the other.
In many member states, the improvement in protection against discrimination based on gender, age, sexual orientation, ethnic origin, or religious affiliation is attributable to the EU. This is particularly true for protection against on-the-job discrimination and against discrimination blocking access to labor markets. While such policies are not redistributive in a stronger sense, they have social content because they protect market participants who would be likely to perform worse if discrimination were legal. Both European politics and judicial case law are responsible for these successes. For example, the commission's negotiation skills are to credit for four antidiscrimination directives that were passed in the years 2000 to 2004.Footnote 70
Even if the relative importance of political agreements in creating European antidiscrimination policy is greater than in the measures we discussed in market-enforcing policy, we should not underestimate the major role case law has played in formulating European equal treatment policy. On the basis of comparatively few clauses in the European treaties—in particular, the principle of wage equality for the sexes and the free movement of workers—the ECJ has developed a far-reaching antidiscrimination jurisprudence that has had an impact in all member states (and that is among the reasons why some, such as Alesina and Wacziarag, argue that European integration has gone too far).Footnote 71 Examples of the most recent case law are the Mangold decision, in which the ECJ overruled an element of the German labor market reforms that aimed at making fixed-term contracts more flexible for older employees; the Maruko decision, which came down against the refusal to award survivor pensions to homosexual partners; and the Coleman decision, in which the ECJ expanded previous case law governing the workplace to also cover family members of employees.Footnote 72 Many other examples for the extensive equal treatment case law of the ECJ could be mentioned. We suggest treating this line of case law as being “social” in the wider sense, but not as an element of (or a substitute for) “Polanyian,” market-restricting policies at the European level: it aims to create free access to the market, unhampered by discrimination, in other words: free but fair markets.
We finally arrive at the dimension of European integration that inspired Caporaso and Tarrow to formulate their Polanyi in Brussels argument, namely judicially enforced, discrimination-free transnational access to the social security systems of member states, including their health care systems.Footnote 73 Caporaso and Tarrow convincingly distinguish three phases of this jurisprudence: first, a market-creating phase in which the member states were obliged to remove protectionist barriers against the free movement of workers; second, a phase in which the ECJ began to open up national social security systems for migrant workers; and, third, a phase in which the court opened up the respective social systems to family members as well, even if they did not live in the territory of the respective state.Footnote 74
We agree that this line of ECJ case law represents an independent and important (sub)dimension of European integration. Actually, we would go even further and emphasize that the ECJ has gradually liberated individual social rights in the EU from the long-standing and often criticized restriction that they refer exclusively to workers. The rights of common market citizens have increasingly been transformed into general rights of EU citizens. The ECJ has succeeded in making this transition by linking the freedom of movement—Article 21 of the Treaty on the Functioning of the European Union (TFEU)—to the ban on discrimination based on nationality (Article 18 of the TFEU). Any EU citizen who legally resides in another member country must not be treated any differently than citizens of that country. According to this legal doctrine, EU citizens have the right, in principle, to also claim social benefits in EU countries other than their original homeland, despite the fact that they may not have made financial contributions to the respective social security system. In the famous Grzelczyk decision, the ECJ emphasized that the requirement for members of one state to extend financial solidarity to those of other member states grows out of community law, as long as this requirement does not unduly burden the finances of the host country.Footnote 75 With this expansion of social rights, the court struck down the solidarity norm accepted until then by the member states, which linked the right to residency to a denial of any claims to social benefits during the duration of residency and to a requirement to have comprehensive health insurance coverage.
While increasing transnational access to national social security systems is a remarkable development, we doubt that this line of ECJ case law represents a European countermovement that re-embeds the market. From a migrant worker's point of view, this line of ECJ case law increases the availability of social benefits. In this sense, it undeniably has social content. However, it comes at a price. Social rights are only one side of the welfare state coin, the other, and equally important, side being social duties. As Caporaso and Tarrow also note, the court “weaken[s] the link between national payment and national consumption.”Footnote 76 Although Article 20 of the TFEU speaks of the rights and obligations that evolve from EU citizenship—there is little likelihood of imposing obligations on those who profit from newly attained social rights. Undermining the reciprocity between rights and duties puts both the effectiveness and the legitimacy of national social policy under pressure. As Menéndez puts it, European law imports a “nonsolidaristic logic” that may make Europe “more human, but less social.”Footnote 77
Governments may ignore the declining effectiveness and legitimacy of welfare state arrangements, or—as authors such as Bellamy, Menéndez, and Scharpf have argued—they may react by retrenching.Footnote 78 Such reaction may be unlikely as long as the transnational use of national social security and health care systems is the exception to the rule. It becomes more likely, however, with growing inner-European migration. Differences in wages and welfare state generosity provide incentives in particular for citizens of the new member states to move westward.Footnote 79 In sum, granting individual transnational access to social security and health care systems must not be confused with either the establishment of social policies at the European level or their protection at the national level. To us, the possibility that this line of ECJ case law will trigger welfare state retrenchment is at least as plausible as a perspective that interprets it as the nucleus of an emerging European welfare state.
Conclusion: Regional Integration Under Conditions of Heterogeneity
We have taken issue with the claim that the EU has entered a phase of reembedding markets. Neo-Polanyian analyses suggest that recent decisions from the ECJ have begun to shift the balance from liberalization toward social protection. In contrast to this interpretation, we have shown that the expansion of individual social rights is but one subdimension of integration through law and might eventually limit redistribution at the national level. Looking beyond this particular subdimension, we find that the ECJ drives forward market liberalization in areas such as labor law, codetermination, or taxation, while member state heterogeneity makes both political agreement on harmonization and political override of ECJ decisions unlikely. Under conditions of heterogeneity, political integration is prone to run aground, while integration through law continues apace.
Our argument resonates with recent work on fiscal federalism. In federations with a set of highly diverse constituent units—in terms of wealth, inequality, and economic specialization—the centralization of taxation and redistribution is unlikely.Footnote 80 The EU fits well with this model. Because the twenty-seven member countries are highly heterogeneous in terms of their welfare and production models, they have been unable to draw up common schemes of interpersonal redistribution and have instead relied on interregional transfers or soft coordination procedures. As a result, the EU comes closer than is usually warranted to Hayek's vision of “interstate federalism”: while the capacity to tax and redistribute remains the prerogative of the member states, nonelected agencies such as the ECJ and the European Commission prevent these states from placing barriers on the free flow of resources and goods across their borders.
More generally, we suggest that the asymmetric potentials of disembedding markets at the national level and reembedding them supranationally are not idiosyncratic to European integration but represent a systemic effect that may also materialize in other regional integration processes, in particular in those that combine two features: first, political willingness to push integration beyond a customs union and, second, significant political-economic heterogeneity. In such a situation, economic integration is likely to face the risk of noncompliance, and political integration is likely to be stuck in what Scharpf has called the “joint decision trap.”Footnote 81 One solution to this problem may be to provide supranational agents with far-reaching powers, in particular in the field of judicial compliance enforcement. Such agencies, however, are likely to develop self-interest in extending the reach and scope of their influence, that is, they will encounter incentives to engage in expansionist interpretations of the treaties they protect.Footnote 82
The extent of the opportunity structure to engage in such expansionist treaty interpretations will be shaped by the presence and effectiveness of activation mechanisms and by how far the supranational court can grant individual rights. Courts need to be activated before expansionist jurisprudence can occur. In the context of the EU, the commission's readiness to take legal action against member states has turned out to be highly effective—in particular, the preliminary rulings procedure that provides the ECJ with exactly the cases it needs to extend the reach and scope of European law.Footnote 83 In addition, not every treaty-based international organization possesses the ability to grant individuals with rights vis-à-vis their member states. The World Trade Organization (WTO), for example, lacks such an instrument. However, European integration provides evidence that courts may construct such instruments on their own, as the ECJ's case law in the early 1960s has shown.
But supranational agencies do not act independently of the power-relational, institutional, and ideational environment in which the respective integration takes place. As the literature on judicial lawmaking has shown, international courts—like any other court—anticipate the likely reactions to their decisions and have an incentive to avoid enacting “court curbing mechanisms,” in particular legislative or constitutional override of their decisions.Footnote 84 Such ex post corrections to ECJ decisions, however, rely on the same intergovernmental logic that also shapes the potentials of ambitious political integration. In other words, it may turn out that the very same political-economic heterogeneity that made member states install strong independent agencies in the first place will also make their effective control unlikely, thereby enlarging their opportunity for performing “integration through law.” We therefore suggest that the particular asymmetry of political and judicial integration and its political-economic consequences are not limited to the EU, but may occur in other world regions as well. For example, if the WTO was able to grant individual rights to market participants—actually, some argue that it should be able to do soFootnote 85—a similar expansionist use of this instrument, combined with a low likelihood of member states' coordinated resistance, might occur.
As the EU example has shown, supranationally enforced individual rights are not bound to be purely economic in nature. Expansionist judicial lawmaking can also grant nondiscrimination rights, such as equal treatment rights with respect to gender or age, or social rights, such as the right to transnationally access the social security systems of the member states. However, there are limits to what can be achieved by judicial integration, since communities that are willing and institutionally able to engage in solidarism cannot be established by judges. Yet, to embed markets requires the reciprocity of social rights and obligations. Privileging rights over obligations might trigger retrenchment—and might therefore help to bring about Hayek's vision of interstate federalism rather than carrying Polanyi to Brussels.
We nonetheless think that a Polanyian perspective elucidates some recent developments in the EU. As we have argued, Polanyi held that economic liberalization brings about political conflict and countermovements that aim at re-embedding the market in order to prevent the subordination of society to the economy. Due to the lasting asymmetry between market-enforcing and market-restricting integration, “negative politicization” might be imminent, which could eventually hurt the creation of a European-wide political community. During the past two decades European integration has, in fact, become highly politicized while at the same time the gap between European elites and the public has widened.Footnote 86 The “permissive consensus” of the past has given way to a “constraining dissensus” and a widespread distrust of elite-driven integration.Footnote 87 In particular, those who have been negatively affected by stronger competition and integrated markets have adopted more skeptical views of integration.Footnote 88 “Those who have not [benefited] do not see their fate as shared with people from around Europe. Instead, they still view the nation and their own state as the appropriate unit to be defended against external forces, whether they are political enemies or forces of neoliberal globalization.”Footnote 89 When the countermovement—protection against market forces—is not forthcoming at the EU level and is losing ground domestically, losers grow Euro-skeptical, endorse nationalism, and sometimes support extreme right parties.Footnote 90
When it comes to implications for further research, evaluating the social impact of integration processes requires looking beyond social rights. Social progress is shaped not only by social integration, in a narrow sense, but also by the disruptive consequences of market-enforcing integration. And social rights, in order to be effective, have to come with social obligations. Strengthening social rights while at the same time weakening the ability to impose obligations may, in the end, trigger retrenchment. Second, as Caporaso and Tarrow have also argued, much of the integration research focuses too narrowly on political integration, while empirically, judicial integration may transform supranational law at least as much as political integration.Footnote 91 Third, we still lack knowledge on the individual and collective consequences of the European liberalization bias. For example, how strong is the causal link between integration and rising levels of income inequality, and are citizens who favor redistribution more likely to develop integration-skeptical attitudes? The answers to these questions might decide the future of public support for integration. Finally, and most importantly, European integration has to be understood as one case among many concerning regional integration processes. We have hypothesized that political-economic heterogeneity, the effectiveness of activation mechanisms, and the ability to grant individual rights are among the factors that shape the (im)balance between political and judicial integration. Empirical tests, however, would need to rely on qualitative comparisons with variance on both the dependent and the independent variable sides. The collection of such data represents a major challenge to recent and future integration research.Footnote 92
