Differentiated Integration and Flexibility in the EU under the Lisbon Treaty: challenges for the Trio Presidency (ARI)

Differentiated Integration and Flexibility in the EU under the Lisbon Treaty: challenges for the Trio Presidency (ARI)

Theme [1]: The entry into force of the Lisbon Treaty brings a number of institutional innovations. The Treaty includes a number of modalities of differentiated integration which did already exist and it adds some new forms.

Summary: The entry into force of the Lisbon Treaty brings a number of institutional innovations. Among these, the Treaty includes a number of modalities of differentiated integration which did already exist (such as enhanced cooperation or opt-outs) and it adds some new forms: ‘oriented’ enhanced cooperation; opt-ins and permanent structured cooperation on defence. Given the subsidiary character that these forms have vis-à-vis the standard Community methods (all states through the same legal instruments), the Trio cannot anticipate an agenda of application for any of them. Nevertheless, some issues may crop up in the agenda that require action by means of these instruments.

Analysis: The term differentiated integration refers to a variety of forms of cooperation and / or integration in which not all members of the EU take part, and which therefore do not have uniform legal and political effects for all of them. All of these modalities owe their existence to the answers they provide to this question: how can one move forward if there are countries which can and want to do so but there are others who do not want to or cannot? Flexibility is the operating principle that allows the EU to conceive of and put into practice methods of differentiated integration, as opposed to institutional, procedural and formal ‘rigidity’ that implies exclusive application of the EU system in all cases and for all countries.

Debate on differentiated integration and flexibility intensified with the failure of the EU Constitution and the troubles the bloc faced in getting the Lisbon Treaty ratified. This is not a new issue, however. In fact, proposals for and discussion of differentiated integration go back to the 1970s, to the time Britain joined the EU, and since then they have been a fixture, updated depending on the circumstances.

In the current context, there are essentially two circumstances that raise interest in this issue. On one hand, the size of the EU (27 members and rising) introduces a growing
 ‘heterogeneity’ (size, per capita income, geopolitical orientation, etc.) that causes European leaders and commentators to wonder if, at times, some of these member states have the will and / or ability to move forward in some policies and even with the process of integration in general (that is to say, if they really want to accept EU policies across the board and have the means to do so). Along with this traditional question, a non-inclusive idea is being hinted at more and more openly: that one group of states might want to provide themselves with a public good from which the rest is excluded.

Meanwhile, this growing heterogeneity among states coincides with a greater ‘internal diversity’ in political options regarding the EU. In some of the older member states, the existence of Eurosceptic parties, groups or tendencies has been constant; such is the case of Britain or France (where part of the left or right have at several times objected to the EU project). Joining them are some of the new states that became members in 2004, such as the Czech Republic (where Vaclav Klaus voices strong criticism of the EU), Poland (where conservative parties hold a reactionary vision of the EU) and even new anti-EU splinter groups in traditionally pro-European countries such as the Netherlands and even Italy. Indeed, there is a greater internal plurality that also means the absence of national consensuses on the EU and its integration project.

Theoretical conceptual schemes for differentiated integration

The theoretical schemes that have been presented over the years tend to be rather ambiguous approaches which often overlap and have little ability (beyond visual images) to get to the heart of the real problems involved in differentiated integration. As they are mixtures in this sort of discourse, a brief description of the concepts most often used is useful.

  • Core (vanguard): this notion means that a group of countries take part in all possible schemes and modalities of integration and / or cooperation, in such a way that they automatically define the centre of gravity around which policy in Europe pivots. The idea of a ‘vanguard’ or ‘pioneering group’ is halfway between the ‘hard core’ (static) and the ‘multi-speed Europe:’ the vanguard seeks to open up a path that other countries will follow later.
  • Concentric circles: this concept was more in fashion in the 1980s and 90s, and has since fallen into disuse. Although it is applied both to the EU and its relationship with its surroundings, it actually serves more to relate the various international European organisations with the idea of integration. Starting off from a central core, cooperation and integration among states would be organised through different bodies, the greater or lesser intensity of which would generate circles around this central point. The circles would essentially define areas of cooperation and / or integration and their legal-political intensity (EU / European Economic area / Council of Europe / Organisation for Security and Cooperation in Europe).
  • Directory: more than an academic approach, this is an idea that is occasionally raised in political circles and involves (unlike the more neutral concept of a central core) a conscious will to direct, guide or influence policy in Europe and the EU. Obviously, it is not a ‘formalised’ concept stemming from an organisational or institutional model.
  • Variable geometry: this is a descriptive model of the system in which, within a common integration scheme, different groups of states participate in different policies.
  • Multi-speed Europe: unlike the previous concept, all states would take part in all EU policies, also within a common integration scheme, but would embrace these policies at different times, depending on their ability to implement them.
  • Europe à la carte: the areas of specific policies and ways of cooperation / integration are presented as if on a restaurant menu, from which countries can choose at will, depending on their interests.  
  • In relation with the previous concept, the idea of selective or differential membership is being increasingly used in order to describe the United Kingdom’s position within the EU (because of its selective repeal of parts of the EU acquis), but also as a possible model for incorporating Turkey, for instance. Here, it should be mentioned that Norway has adopted much of the EU acquis without being a member.

Existing modalities

Of course, the existing models of differentiated integration have not followed these theoretical models, but rather have responded to the need to provide ad hoc solutions to the specific demands of certain states in given circumstances. Formulas for differentiated integration were incorporated for the first time into the Single European act and have been retained, in their different variations, up to and including the Lisbon Treaty.

Voluntary exclusion (opt-out)
The opt-out is characterised by the following features. In the first place, it is defined under primary law (normally, through protocols) which, furthermore, spells out specifically which states accept this voluntary exclusion (in other words, it is not a generic option that is open to each and ever y state at all times). Also, the opt-out is applied to issues and areas that are clearly defined (it is not something that is available for any EU policy). Strictly speaking, the first case was Britain’s opt-out from the European social charter adopted as a protocol to the Single European act. The best known case, however, was the British and Danish opt-out from the third phase of Economic and Monetary Union (Sweden, which in a referendum rejected joining EMU, has no de jure opt out, although it does have a de facto one). Curiously, article 139 of the current Lisbon Treaty refers to states which have not fulfiled the criteria for advancing to the third phase of EMU as “member states with derogation”. Opt-outs have gone from the monetary realm to the Schengen acquis (once it was incorporated into the Treaty), from which the United Kingdom, Ireland and Denmark are excluded.

Finally, the last case of an opt-out came with the Charter of Fundamental Rights, which was incorporated into the Lisbon Treaty with a protocol. The United Kingdom and Poland have excluded themselves partially (the Czech President Klaus has demanded similar treatment for his own country).

Opt-outs have traditionally been viewed with wariness, the idea being that they undermine the EU integration model. But the cases mentioned earlier show the positive side: opt-outs allow integration to be applied to a sector without depending on the will of those who exclude themselves. Furthermore, such countries are the ones who assume the cost of their own exclusion (although Britain’s opt-out from the social charter did involve a lot a free-riding on the higher social standards that the charter introduced for other member states. Nevertheless, the UK did finally renounce its opt-out).

Enhanced cooperation
Unlike opt-outs, enhanced cooperation is a mechanism for affirming a will to advance and by definition it refers only to participating states. Primary law defines a framework but does not identify a priori the areas for specific application (although those in which it can be applied are identified: they are non-exclusive areas of jurisdiction which do not affect the internal market or economic, social and territorial cohesion) nor the specific countries which participate and which are excluded a priori. On the other hand, the pre-requisites for activating this policy are spelled out: there have to be at least nine member states interested (it was eight under the Nice Treaty), it must be open to including new participants, although “conditions of participation” can be established (articles 328.1; 333.1 of the Lisbon Treaty).

In the first place, the Treaty defines the ‘situation’ in which enhanced cooperation can be activated (as a last resort, once it has been confirmed that agreeing on ordinary EU rules is impossible) and also the ‘procedure’ for doing so: the initiative affects the relevant states, and the proposal is to come from the European Commission or the High Representative if it is a question of foreign and security policy. The final decision is adopted by the Council by a qualified majority (or unanimity in the case of foreign and security policy) and the European Parliament must grant its consent (in the case of foreign and security policy, the Parliament will only be informed).

Enhanced cooperation has drawn attention because it brings together the appeal of ‘cartesian’ models favoured by academics with its concrete legal existence. However, its greatest defect is the scant operational effectiveness it has shown since being created in the Amsterdam Treaty (1996). So far only one case of ‘enhanced cooperation’ has been implemented: the European norm that allows couples from different member states to choose which national jurisdiction will apply in divorce cases (the participating states are Spain, France, Italy, Romania, Luxembourg, Greece, Austria, Hungary and Slovenia). Aside from whether it is actually used or not, some commentators say that the value of enhanced cooperation lies more in its value as a threat to get a particular measure approved through regular EU channels.

“Oriented” enhanced cooperation…
This is a variation of the previous concept, applicable to certain realms defined a priori within the area of freedom, security and justice: under the Lisbon Treaty, mutual recognition of sentences, judicial and cross-border police cooperation (82.3), the creation of a European Public Prosecutor’s office (86.1), and judicial cooperation (87.3). The most notable innovation is that its activation is almost automatic if the ordinary legislative procedure fails; only a unanimous ‘no’ vote from the European Council can block its implementation (the so-called ‘accelerator’). This automatic mechanism is the biggest advantage of a system designed with the goal of overcoming the prospect of a British veto.

…and its corollary, discretionary participation (opt-in)
The aforementioned modality is pre-configured to anticipate a British (and possibly Irish) opt-out. It carries with it the possibility of Britain deciding in a discretionary fashion if it wants to take part in those areas in which it has previously stated it wants an opt-out (sic) (Protocol 22 of the Lisbon Treaty). In the past, member states of the EU adopted two Regulations (establishing the Frontex agency and regulating the anthropomorphic data to be included in databases on electronic passports) and the United Kingdom appealed both before the European Court of Justice on grounds that it had been illegally excluded! The Court wisely rejected both lawsuits (sentences C-77/05 and C-137/05).

Permanent structured cooperation
Applying enhanced cooperation to the area of defence (permanent structured cooperation) involves some differences compared to enhanced cooperation in foreign and security policy. Fundamentally, decisions can be adopted by a qualified majority, and participation requires the reaching of more binding commitments to carry out more demanding missions and the fulfilment of higher-level military capability criteria. The French government has already presented a proposal to begin articulating permanent structured cooperation.

Besides this, in the area of defence there are modalities which, without being forms of structured cooperation in the strict sense, nevertheless apply the principle of flexibility. Thus, the Lisbon Treaty allows for member states to organise cooperation and coordination among their administrations responsible for national security (art. 73); for the European Defence agency (art. 45.2), in which any member state which wishes can participate (in actual fact, all do except for Denmark); and to organise missions outside the EU to preserve peace, prevent conflicts and boost international security – this can be assigned to groups of countries (art. 42). However, it should not be forgotten that there is flexibility within the structure of security policy. Thus, a joint action within the framework of security and defence in which only some EU countries take part is what gave rise to the EU mission in waters off Somalia to prevent piracy (EU NaVFOR Somalia).

Parallel functional integration
The choice of the term ‘functional’ is not arbitrary. In the 1950s, the functionalist theoreticians of integration (Mitrany, etc.) embraced the principle that form (legal and political) would flow from the function that was sought and the top priority was in fact to define these functions and tasks. Thus, this modality involves creating schemes, which are parallel and concurrent with the goals of the EU, with participation by a number of members smaller than the total in the EU and which do not adopt its institutional model or its regulator y model. There are many examples, from the oldest ones (the Western European Union, finally absorbed by the EU itself ) to the most current ones, such as the Prüm Convention (2005) and the European Space agency (1975), the Schengen agreement (1985) or the Eurocorps (1992).

Pros and cons of differentiated integration

After this descriptive summary, it is a good idea to evaluate differentiated integration as a whole, albeit briefly. The disadvantages it entails can be summarised in the following:
Potential for reversing integration.

Most advocates of orthodox EU thought systematically conclude their analysis by saying that differentiated integration modes can undermine what has already been achieved and block further progress. But this argument is inaccurate because differentiated integration proposals have emerged rather as a reaction to the lack of drive toward integration on the part of the EU as a whole. The idea that there are integration-curbing mechanisms in these modalities is correct only if one contrasts integration with differentiated integration. However, the comparison is incorrect.

Rupture of legal unity within the EU
This argument is constructed in a parallel fashion to the previous one, and stresses the application of some EU norms only in some countries and to some EU citizens. At tactical level, there is no denying this. But if we are realistic, it should be viewed as a problem only in certain circumstances: if discrimination and / or legal insecurity arise. Without downplaying this possibility, it does seem clear that the existing modalities of differentiated integration discussed earlier have not caused major problems in this regard.

Encouraging free-riding
Perhaps the most solid objection refers to the possibility of free-riding: obtaining benefits from differentiated integration without assuming the costs of taking part in it. Some of the existing examples, namely EMU, show that these schemes actually externalise (transfer to countries not included) some of their costs (as seen in the lack of autonomous monetary policy of those which are excluded). Looking to the future, defence is often cited as an area in which free-riding will be inevitable. However, this argument can be countered in two ways. Firstly, some of the existing schemes (the European Defence agency and Eurocorps) have generated momentum in which participation is extended. Secondly, it is doubtful that a common, alternative scheme in which the possibility of free-riding disappears altogether can be attained.

The democratic nature of certain kinds of differentiated integration
Finally, many commentators are concerned by the inter-governmental tendency inherent in these modalities. Although it is hard to generalise, some of these escape the Commission’s capacity to take initiatives, and / or the control of the European Parliament, and / or the jurisdictional control of the Court of Justice, and / or could even skirt control by member states. This is a real problem, but the scope of it should be dealt with on a case-by-case basis, rather than prejudged as a possible generic situation.

As for the pro’s, they should be viewed with this in mind: differentiated integration exists and seems inevitable (in one form or another) in the future. Given this fact, the following advantages should be considered:

  • Differentiated integration does not curb progress, but rather just uniform progress by all the member states of the EU.
  • Differentiated integration serves to accommodate the diversity and plurality of the member states. What is more, if its creation, probably as part of primary law, is handled skilfully, the EU might be able to avoid downward negotiation by offering escape routes to minimalists (as happened with the United Kingdom with the Charter).
  • In general, costs are assumed by those which are not integrated and / or those which are excluded (in actual practice cases of free-riding have not arisen, although they cannot be ruled out in areas such as defence policy).
  • Finally, from the point of view of the Trio Presidency, it must be underlined that the 3 countries share a similar approach: Spain and Belgium participate in all existing modalities (with the exception of Belgium’s non-participation in the norm on applicable national jurisdiction for divorces) and Hungary has shown its willingness to take part as soon as possible in the most significant one (the euro). This means that they may easily share a similar approach towards these modalities and, eventually, they may steer any eventual prospective application.

Carlos Closa
Scientific Analyst, CSIC


[1] This ARI was originally written as a chapter of the book ‘Think Global – Act European II’ devoted to the Spanish, Belgian and Hungarian Trio Presidency and focused on the changing role of the rotating Presidency and the trio itself. It was published by Notre Europe, Egmont, GKI and the Elcano Royal Institute.