Social Europe: A long march?

9 April 2009
Read in:
A number of recent cases have highlighted how "social dumping" can occur if companies register abroad so as to be exempt from collective labour agreements applicable in the countries in which they are operating. The European Court of Justice, called on to decide between the freedom to establish a business and the employment laws of individual member states, has been hesitant to rule in favour of employees. Disappointment at judgments may derive from the fact that the vocabulary of social protection often cannot be translated at a European level, writes Marc Clément: fundamental concepts are understood in different ways by Europeans, depending on their national culture. Before we ask the question of a social Europe, a legal solution to the co-existence of social Europes (in the plural) must be attempted.

When you start to list the building blocks that are missing in the construction of Europe, the lack of a common social approach is certainly one that springs most readily to mind — at least, in France. The recent publication by Jean-Claude Barbier of La Longue marche vers l’Europe sociale [The long march towards a social Europe]1 does have the merit of clarifying and amplifying the actual terms that I used in my first sentence. What do we actually mean when we speak of a “social Europe” and why is it that, when we speak of its place in the construction of Europe, we have to use different terminology for each of the EU countries?

It is not my purpose here to provide a complete survey of Barbier’s invaluable study; my intention is rather to show how the line of inquiry that he pursues is essential for anyone who is not content with mere slogans and who wishes to find ways for the European Union to extend its social dimension. It seems to me that we should make a distinction between two levels within what may be called “social Europe”. There is a weak version and a strong version and each depends on the degree of interdependence and social cohesion that one might imagine to exist between European countries.

Social Europes

The weak version of social Europe is a consequence of economic Europe. It is perhaps worth reminding ourselves at this point that the internal market, the basis of the European Economic Community that in 1993 became the European Community had, since 1957, been based on four founding principles. These were the free movement of persons, goods, services and capital. It soon became clear that opening frontiers implied that systems for social protection and national legislation would have to be adapted. For example, free movement of persons means that rights to social protection cannot be the sole preserve of a country’s nationals.2 The dynamics of social Europe derive from the process of identifying specific problems resulting from the implementation of the single market; they can be seen in moves to coordinate the policies of member states and, should this coordination be insufficient, to reduce obstacles by means of minimum standards. The point is not to develop a European social model at the level of the EU but rather to act in such a way that differing social models can coexist within the framework of an open economy. Take health insurance for example: coordinating services enables health cover to be extended to individuals who are members of another country’s scheme through a system of compensation operating between the two countries’ social security funds.

This is a Europe that leads to seeing economic principles as being in opposition to principles relating to social rights, when the latter are given the status of fundamental rights. It is also the kind of Europe that takes open approaches to coordination, with intergovernmental processes that aim at convergence in the social policies of the member states. Barbier demonstrates both the indisputable advantages of such approaches, but also their limitations: the more they become just exercises in style and communication, the less they constitute useful tools producing practical outcomes.

Within this weak interaction between the social protection systems of member states, it is certainly the case that a central role is played by decisions of the European Court of Justice. Indeed, problems arise precisely because Europe is also a community of laws. The open approach to coordination does not give rise to lively national debates; it is distant and remains the province of a small group of specialists. However, when it is the Court of Justice that pronounces a ruling, then both citizens and those active in the social field understand that the matter is a serious one and that there will be practical consequences that will be felt on a day-to-day basis. Barbier clearly identifies the crucial role of the European Court of Justice (ECJ) in this area but, reading the pages concerning the ECJ (and the European Central Bank), one still tends to feel that one is dealing with an institution that hands down decisions that are, if not actually irrational, at least unpredictable. Hence, it is worthwhile explaining the way that the judges create community law.

Once precedence of European law over national laws was affirmed in 1964, European judges found themselves faced with fundamental rights and, in particular, with social rights. It must be stressed that it is not only in the area of social law that the problem of taking account of fundamental rights arises. For example, as early as 1974, the ECJ had to rule on the conflict between European law and property law.3 Since Maastricht, article 6 of the Treaty of the European Union (TEU) treaty includes in paragraph 2 fundamental rights under the EU’s judicial system.4 Such being the case, it would be difficult to argue that the ECJ should treat social rights as a special case on the grounds that they do not form part of fundamental rights! Consequently, the court is required to identify, though not necessarily to rule on, conflicts between these rights.

A recent example of this kind of conflict arose with the judgment in the Viking case.5 A Finnish shipping company wished to register its ship in Estonia in order, for reasons to do with pay rates, to use Estonian crew. Because seamen have a trade union organisation at European level, they are in a position to bring pressure to bear on the shipping company. In his judgment, the EU judge ruled that the freedom of establishment guaranteed by the TEU was threatened by the right to strike. It is important to understand that, in this judgment, the ECJ was identifying an obstacle to the right of establishment but did not make a ruling. It would be the national jurisdiction that would have to decide whether the right to collective action, recognized as a fundamental right, was properly exercised. The view taken by John Monks, the General Secretary of the European Trade Union Confederation (ETUC) was:

We would have welcomed a more clear and unambiguous recognition of the rights of unions to maintain and defend workers’ rights and equal treatment and to cooperate cross border, to counterbalance the power of organized business that is increasingly going global.6

This certainly is the problem raised by this judgment. Inside Europe, trade union coordination is essential as a means of standing up to transnational companies. If this kind of coordination is not possible, then there is a degree of asymmetry between the accepted right of the company to freely establish itself in the country of its choice and the ban on Europe-wide industrial action. It should be noted that the answer to this question will be given by the English High Court which is to resolve the dispute, since the seamen’s union, the International Transport Workers Federation, has its base in London. The problem really is the absence of a European decision: can we refer such a matter for resolution to a national jurisdiction when the question has a Europe-wide dimension? It is probably the case that the 27 jurisdictions of the 27 member states would today all come up with differing solutions. All that trade unions would then need to do would be to relocate their proceedings to whichever member state was most favourable, to whichever was the “paradise for trade unions”. The alternative would consist in drawing up European rules that would make such Europe-wide action permissible. That would amount to establishing the right to collective action on a European scale. It is certainly not a simple approach to take.

Whereas, with the Viking judgment, the uncertainty is created by the absence of any European legislation, it is quite a different matter when it comes to the Laval and Rüffert judgments.7 In these rulings, the ECJ was not simply dealing with two opposing principles or the opposition between the TEU and social rights. It had a directive at its disposal, namely Directive EC/96/71 concerning the posting of employees. In other words, European legislation had put in place a framework that ensured that the social law that applies to a posted worker is indeed that of the member state in which the worker is operating, but also that this social law must be explicit, that is, that its rules cannot be applied unless such rules form part of national law. Under the Swedish system collective agreements are private contracts; signing up to such a collective agreement is therefore not compulsory under Swedish national law. The ECJ has upheld the view that an Estonian agreement is equally binding as a Swedish agreement. This undermines the Swedish model, which permits unions to take very severe action against companies that have failed to adhere to Swedish collective agreements.

The numerous commentaries to which these rulings have given rise seem to me to be based on rather inaccurate analyses of the Court’s legal reasoning processes;8 such accounts result from a desire to prove a point about a liberal Europe based on treaties, rather than from any concern to carry out a rigorous scrutiny of the arguments on which a judge may have based his ruling. Contrary to what is sometimes argued, the Court had only two alternatives: either to base its findings on a directive that clearly indicates that the working conditions of posted workers are governed by the laws and collective agreements that have the force of law — the option that the Court actually chose — or to consider that this directive ought to give way to the right to collective action as defined at the national level. The latter option would, paradoxically, lead to the end of any social Europe since, at that point, no European regulation would be able to withstand the application of social rights at a national level. It also meant either recognising that social rights do not have the status of fundamental rights in the sense of Article 6 of the Treaty but instead amount to “local” fundamental rights, or else considering that European legislation violated the basic right to collective action, as defined in Sweden, by adopting such a directive.9 The two alternatives were impracticable, given that they meant that it was necessary to recognize that social rights had to be interpreted at a national level: what could you then have replied to any member state that was restricting the right to strike or was applying a limited version of the principle of equality of treatment between men and women?

What lessons are to be learnt from these judgements and what prospects might be opened up by incorporating the charter on fundamental rights in the Lisbon treaty? It is evident that this is a source of uncertainty for Barbier, who sees the charter as a means by which the ECJ’s case law might be developed. Nevertheless, for two reasons, I think that we should not pin our hopes on incorporating the charter into the body of European constitutional law.

First, the charter is, unfortunately, associated with special opt-outs for Poland and the United Kingdom which, in my view, will result in limitations being placed on the ECJ’s ability to develop the rights inscribed in the charter.10 Secondly, fundamental rights are nothing new in terms of European law as we pointed out earlier: the charter itself is the result, point by point, of the Court’s case law. It therefore appears to me very likely that an ECJ judge will choose to follow the route that he has himself laid down by basing his conclusions on rights that are considered to be at one and the same time inscribed in the charter and the result of the constitutional traditions of member states, rather than allowing the charter itself to play any special role.

Does that mean that there is nothing more to be expected from this Europe of rights when it comes to better integration of the social dimension of Europe? Not necessarily, but it is clear that we should not expect the Court to deliver more than it can, because any response that there might be would have to be political and, in the European Union, any political response has to be expressed in terms of European standards. In that sense, the prominence given by the European Socialist Party in its manifesto for the European elections to the Viking and Laval judgements11 and their implications for collective bargaining is entirely justified. By choosing, at one and the same time to rule (in the Laval case) and not to rule (in the Viking case), Europe is showing how timid and hesitant it can be. In making its decisions, the ECJ has shown that strictly judicial logic imposed solutions that member states or those involved in social action did not find easy to accept. But you cannot ask a judge to play the role of a legislator. He may try to give some measure of judicial coherence within the limits of the EU’s laws and principles but he may not go beyond the framework set by the texts themselves. If they need to be revised, then it is a matter for the EU’s legislative bodies (the Parliament and the Council). This would amount to a complete revision of European social legislation.

A social Europe?

We have yet to find the magic formula that might produce the strong interaction needed to put in place a Europe-wide system of social protection. If we follow the line of reasoning put forward by Barbier, the key could only be multilingualism. If we accept that systems for social protection are the results of a political culture — and, on my reading of the book, in particular of the very painstaking analysis of the various national systems that Barbier uses as the basis of his argument, this seems very likely — then the task is enormous!

His argument is based on painstaking analysis of the vocabulary of social protection. The same terms do not have the same significance in the various languages and some of the terms are untranslatable. For example, the French concept of partenaires sociaux translates into English as “labour and management” or into Danish as arbejdsmarkedetsparter, a term that expresses competence solely in relation to employment rights.12 It is interesting to note that European terminology does evolve: nowadays “social partners” can be found in European legislation in English; for a British person this is a purely European concept and thus creates some measure of uncertainty. Does it represent some very small indication of a convergence in European social models or is it just a matter of contamination by Brussels jargon? Barbier’s sound arguments demonstrate the great difficulties facing any European approach to social protection when fundamental concepts are understood in different ways by Europeans, depending on their national culture. Such divergences between concepts are also a sign that social protection systems in Europe are rooted in political history and in a succession of choices that have shaped their specific features. This being the case, it is of the utmost importance that we should not artificially obscure such differences by using a technocratic “Brusselese”, which simply disguises lack of agreement on basic questions.

And so it seems to me that the bewilderment that was evident in the way that the Laval judgment was received may perhaps also result from a feeling of having been cheated by the effects of Directive EC/96/71, a feeling that could well derive from that basic inability to communicate concepts of social protection at a European level. The framework set by the directive is very close to the French frame of reference; at least this remains the case until such time as opt-outs might be allowed for individual companies. If this were to be permitted, then such opt-outs would have to be available to any foreign company that came to set up in France. French collective agreements acquire the force of law. However, such legislative validation of collective agreements is the opposite of the Swedish or Danish model. Thus, even the minimum level of harmonisation of the directive comes up against different political cultures: arbejdsmarkedetsparter do not expect anything from the state and, above all, do not want the state to become involved in their negotiations. Nor do they wish for any legislative confirmation of their efforts. The directive brings the state into the discussions — something which is perfectly normal for a French person but which, for a Scandinavian, offends against the very basis of the social model.

Language as a vector of political communication is thus at the heart of the process by which rules are drawn up for national solidarity and therefore also at the heart of social protection systems. There can be no redistribution without a feeling of having a common bond. This feeling of belonging is expressed in a political culture that generates institutions and political practices that are nowadays very varied amongst member states of the European Union.

It is therefore pointless, or indeed dangerous,13 to try to impose a European system for social protection until European citizens feel that they form part of the same community and are therefore ready to accept redistribution. The fact that generous systems of social protection are also based on strong national sentiment is often glossed over. But you cannot hope to understand the Danish system of social protection if you fail to take account of the importance for any Dane of their national community. Meeting your family at Copenhagen airport by waving a Danish flag or putting little Danish flags on birthday cakes expresses the fundamental role that the national community plays in the lives, including the private lives, of Danish people.

The crisis that Belgium is currently experiencing including, among the most extreme of Flemish demands, the idea that social protection in Belgium should be organized on a regional basis, likewise demonstrates — in a negative way — the extent to which the national community and social protection are closely linked. The Belgian situation shows that a national redistribution system will be undermined if the political system functions under a different territorial regime. If you take into account the electoral system in force in Belgium, the Belgian political parties are constituted on the basis of linguistic communities: PS et SPA for the Socialists, CDH and CD&V for the Christian-Democrats, MR and Open VLD for the Liberals, without any natural alliances making it possible to gain majorities on a national political programme. In other words, the electors belong to a single linguistic community. It is logical that, under these circumstances, solidarity across regional frontiers should face a challenge. In this respect, the Belgian crisis, as a negative example, must be very carefully scrutinized by the Union, since it tends to show that ever-increasing integration is by no means inevitable. That being so, we have to ask questions about the impact of greater and greater regionalisation in Europe when it comes to national social protection systems: the United Kingdom, Spain and Italy are all undergoing processes of extensive decentralisation and one might well wonder whether national social protection systems might themselves be weakened as a result.

In short, without a common political culture there can be no social protection, and without a European public space there can be no social Europe. Even if Barbier expresses reservations about the postnational approach proposed by Jürgen Habermas and has doubts about the effectiveness of constitutional patriotism, it seems to me that the point of departure is not so far away, particularly in his insistence on the role of language and of public space in the democratic process.14 One could equally well maintain that advocates of multilingualism aim to set up a European public space that would allow exchanges between European citizens so as to provide the right climate for the emergence of a Europe-wide form of social protection which, in the present state of a Europe that is constructed by law, can only take the form of standards.15 Barbier’s study comes to a stop on the threshold of this prospect. In that sense it does not contradict what Habermas or J.-M. Ferry16 suggest when they try to imagine what the future of the European Union might be if the conditions for the development of a European public space were combined, while at the same time seeking some alternative to the classic federalist project.

If we accept the argument in favour of multilingualism put forward by Barbier, there are many questions that need to be asked. What measure of understanding between the peoples of Europe will make it possible for us to develop social Europe? How will this social Europe link up with the different social Europes? Should we imagine that, in some far-off future, these social protection systems will fuse into one? Are there intermediate stages between the emergence of a social Europe and the situation that we have today?

What is the future for social Europe(s)?

The prospects that emerge from Barbier’s study, whether in the short or medium term, are no cause for rejoicing. Not, at least, if you have any hope of Europe-level social protection arriving any time soon. We do indeed have a long march ahead of us. Have we even started down that road? The answer would seem to be in the negative and, consequently, we have to ask ourselves whether there is any possibility of moving more quickly.

The first route has to be through strengthening exchanges. The Erasmus programme has had undeniable success. The steady increase in the number of students involved shows that spending time in another EU country is nowadays almost commonplace. But it is also undeniable that programmes that reach a larger public need to be developed. Putting in place a European public service would have the same kind of effect and would meet the need for strengthening multilingualism in Europe.

The second route involves developing the European political space. The French referendum at least had the merit of generating intensive discussion about Europe. The European question was at the centre of national politics. Are there any signs of a tendency, limited though it might be, towards “Europeanising” political debate in Europe? I would not rule that out if you consider the amount of space occupied in the media in France by European questions. The forthcoming elections to the European Parliament will make it possible to see what progress has been achieved. If the political parties appear to be concerned only with national communities and the direct interests of their electors, then there will be little chance that any mechanisms for social redistribution at European level will ever see the light of day.

These few points are meant to do no more than provide the beginnings of a list of political measures that might act as an itinerary taking towards a social Europe. They may seem to be simply stating the obvious. But it seems to me that we need to extend the diagnosis provided by Barbier by providing elements of a prognosis. Let us hope that creativity in this field may result in new kinds of cultural solidarity being devised. However, the slow maturation that will be required for a social Europe to emerge must not prevent us from paying attention to the social Europes and the ways in which they can coexist. For it is certain that we will have to live through a good many more years under a system in which the question of the link between the various social protection systems and the single market will still need to be asked. In that area too, creativity is what we will need.

  1. Jean-Claude Barbier, la Longue marche vers l'Europe sociale de Paris, PUF, coll. "Le lien social", 2008.
  2. Council Regulation EEC 1408/71 of 14 June 1971, on the application of social security schemes to employed persons, to self-employed persons and to members of their families moving within the EU.
  3. ECJ, 4/73, 14 May 1974, J. Nold, Kohlen und Baustoffgroßhandlung versus The European Commission.
  4. Articles 136 and 137 of the Treaty, concerning the jurisdiction of the Union in relation to social matters, indicate the areas in which the Union may legislate; excluded from that legislative area is the right to strike. This must not be confused with the fact that the right to undertake collective action is indeed a fundamental right and, as such, forms part of EU law -- but only as a fundamental right.
  5. ECJ C-438/05, 11 December 2007, International Transport Workers' Federation versus Viking ABP. For details of the Viking case, see the website of the European Confederation of Trade Union (ETUC): -- ed.
  6. ETUC press release, 11 December 2007.
  7. See Marc Clément, "Europe : comment lutter contre le dumping social ?" [Europe : how do we combat social dumping ?] Esprit, February 2008, and id., "L'Europe sociale est de retour" [Social Europe is back], For details of the Laval case, see: For the Rüffert case see: -- ed.
  8. See for example A. Supiot, "L'Europe gagnée par "l'économie communiste de marché'" [Europe taken over by "the communist market economy"], in Notre Europe (Débat Viking-Laval-Rüffert : libertés économiques versus droits sociaux fondamentaux où se situe l'équilibre ?) [Our Europe (The Viking-Laval-Rüffert Debate: economic freedoms versus fundamental social rights; where does the balance lie?)], who wrongly claims that the Court of Justice bans using the right to strike to force countries in a country A that are operating in a country B to abide by all the laws and collective agreements of country B. This is all the more incorrect inasmuch as the directive itself requires the service provider to abide by the laws and collective agreements that have the force of law in country B.
  9. We need to remember that the directive itself allows any kind of minimum standards to be imposed on European service providers if those standards have the force of law. To consider that the directive violates the right to collective action amounts to denying to European legislation any right to legislate on social matters.
  10. M. Clément, "Europe, peut-on choisir ses droits?", [Europe : Can we choose our rights ?]
  11. Manifesto of the European Socialist Party: "Les citoyens d'abord. Une nouvelle direction pour l'Europe" [People First. A New Direction for Europe], adopted by the Madrid Council, 1 and 2 December 2008 (proposal 18).
  12. J.-C. Barbier , op. cit., 207, these are the terms that translate the concept of partenaires sociaux in the first directive to use them, Council Directive EEC/76/207, of 9 February 1976, relating to the implementation of the principle of equality of treatment between men and women in connection with access to employment, training and professional promotion, and working conditions.
  13. See J.-C. Barbier, op. cit., 245 on his criticism of J.-M. Ferry's proposal for a universal European minimum income.
  14. J. Habermas, Droits et démocratie : entre faits et normes, Paris, Gallimard, 1996. [Between Facts and Norms. Contributions to a Discourse Theory of Law and Democracy. Trans. W. Rehg. Cambridge, MA: MIT Press.]
  15. M. Clément, "Pax europeana, la romanité de l'Europe par le droit", [The "Romanity" of Europe through law] Commentaire, Spring 2009.
  16. J.-M. Ferry, Europe, la voie kantienne. Essai sur l'identité postnationale, [Europe, the Kantian route. Essay on postnational identity] Paris, Cerf, coll. "Humanités", 2005.

Published 9 April 2009

Original in French
Translation by Mike Routledge
First published in Esprit 1/2009

Contributed by Esprit
© Marc Clément/Esprit Eurozine

recommended articles