At the Brussels European Council of 14 December 2007, the EU Member States decided to convoke a ‘reflection group’ on the future of the EU. Accordingly, the group will have ‘to take into account likely developments within and outside Europe’. The group, which will be chaired by the former Spanish Prime Minister Felipe González, will have to deliver its report in June 2010. The aim of this paper is to contribute to the debate on the future of Europe to be generated by this European Council initiative.

In particular, this working paper aims to establish a hypothesis about where the EU will be in 2017. A number of possible scenarios for the EU’s future can be easily envisaged. In ascending order from the least to the most integrationist, the following scenarios are suggested: (1) termination; (2) variable geometries; (3) status quo; (4) incremental integration with variable geometries; (5) incremental integration without variable geometries; and (6) political union.

Devising possible scenarios does not, however, tell us anything about the likelihood of their actually happening. In order to devise hypotheses on probable rather than possible EU scenarios in the next 10 years, we need to take into consideration the most important determinants of the EU’s process of integration. In my opinion, these determinants are the following: (1) enlargement to new Member States; (2) the degree of diversity such enlargement might bring about; (3) the increase in the heterogeneity of interests that enlargement might produce; and (4) the delegation of core sovereignty competences to the EU. All these factors can either hinder or foster EU integration, depending on how they work in the future and how they interact with each other.

This working paper shows that it is not unrealistic to see the EU being enlarged in the next 10 years, at least to Croatia, Macedonia, Albania and Montenegro and at the most to all these countries plus Bosnia Herzegovina and Serbia. If this is the case, then the EU’s legal, political and economic diversity would increase, because all these countries are already very different from the rest of the Union. The presence of nationalist political parties in most of them, coupled with ethno-social cleavages, would probably increase the degree of heterogeneity of interests within the EU. However, as regards competences, it is not foreseeable for the Member States to cede new core competences (in particular, in the field of defence and foreign policy) to the EU within the next 10 years.

The combination of all these factors, for the reasons explained in this paper, will probably cause the EU’s integration process to evolve to a point which would be placed between the current status quo and the variable geometries scenario. In other words, the main hypothesis put forward is that the EU would move in the next 10 years towards a situation in which it would progressively dilute into a free market zone, coupled with a number of policies that will be managed through variable geometry mechanisms.

The de-integration movement predicted in this paper is not, however, inexorable. First, because, as EU history shows us, it could be the precursor of a more intense integration after 2017. And secondly, and more importantly, because political leadership, if strong and sufficiently well oriented, could counterbalance such a trend.


What will the EU look like, from a political standpoint, in 10 years time? Will it be a ‘federation’, ie, a political union much like the ones in some of our respective nation-states? Will it just be an integrated space from an economic point of view with very little or no political integration? Or will it have disappeared? These are the sort of questions this paper attempts to deal with. Asking questions about the future shape of the EU is a legitimate enterprise. One could argue that it has always been. Exercises of the kind are now familiar to the EU. However, this effort is maybe more relevant today due to the collapse of the so-called Constitutional Treaty and to the uncertainties regarding the Lisbon Treaty fate after the Irish ‘no’ in June. Both episodes have made the alarm bells ring for Europe’s future in many national capitals. Rightly or wrongly, there is the perception that everything is open now for Europe, that anything could happen in the years to come.

To deal with the issue of the EU’s future, this paper is divided into two parts. In the first part I discuss the possible EU scenarios for 2017. In the second part I discuss not the possible, but the probable scenarios. The first analysis implies conceiving all possible scenarios one might think of, from the least integrationist to the most integrationist, irrespective of the likelihood of them actually happening. The second exercise mainly implies commenting on the variables that affect the likelihood of the scenarios happening. Therefore, I shall set forth a predictive model relative to the degree of integration that the EU can achieve in around 10 years time, and I shall discuss it in theoretical terms. This will help me put forward a hypothetical scenario for 2017.

Six Possible Scenarios for the EU in 2017

A scenario is an ‘illustration of the future which can help to understand change and the conditions of change’ (Langer, 2005, p. 3). Departing from this definition, I devise six possible scenarios in which the EU can find itself in 10 years. All of them belong to a single dimension, which I shall call ‘degree of integration’. In this dimension, the different scenarios are ranked from the ‘least integrationist’ to the ‘most integrationist’. The scenarios are as follows:

  • Scenario 1: Termination
  • Scenario 2: Variable geometries
  • Scenario 3: Status Quo
  • Scenario 4: Incremental integration with variable geometries
  • Scenario 5: Incremental integration without variable geometries
  • Scenario 6: Political union

Figure 1. The six scenarios placed along the axis ‘degree of integration’

image001 59 1

Scenario 1: Termination

According to this scenario, the UE would terminate. To use a less technical expression, it would cease to exist, it would disappear. It is interesting to note that it is very rare to find references to this scenario in the enormous amount of (scientific and other) literature addressing the issue of the EU’s future. For example, Langer (2005) does not ‘consider the scenario that the European integration will be terminated and the Member States return to the mere intergovernmental mode of relations. Although there are presently indications from opinion polls, referenda, media reports, expert opinions etc which could be interpreted in such a direction, I do not believe that the end of the EU in one or two generations would be a realistic scenario’. In turn, the European Commission Forward Studies Unit, in its ‘Scenarios Europe 2010’ (European Commission, Forward Studies Unit, 1999), one of the most exhaustive analyses on this issue, does not even contemplate the scenario we are now speaking about. And we could go on.

The Commission’s reluctance to speak of this kind of scenario is understandable. It would be more than controversial for the first Community institution to speak of the demise of the hand that feeds it. Less understandable, however, is the scientists’ reluctance to do so. If, as Langer proclaims, there is no shortage of indications pointing in the direction that the end of the EU could occur one day, why not contemplate it?

EU termination raises a number of practical issues. Though they are basically of a legal nature, it is nevertheless interesting to at least say a word about them. The EC Treaty does not foresee its own termination and neither does the EU Treaty. In the florid world of international law, some treaties do contemplate their own termination, whereas others do not. Therefore, it is not uncommon to find treaties that do not include provisions on termination. This does not mean that these treaties are immortal. As a matter of fact, parties to these treaties sometimes terminate them. Because this happens to be the case more times than one can think of, the Vienna Convention on the Law of the Treaties of 1969[1] has wisely established provisions on Treaty terminations for such cases. In my view, these provisions would be of application to the EU context.

In principle, the Vienna Convention establishes that there is no need for a specific reason to terminate a Treaty. Therefore, the respective consent of all parties to the EC Treaty and the EU Treaty, that is, the Member States (which are, to use the EU jargon, the ‘masters of the Treaty’) would suffice to put an end to the EU[2]. In other words, if all the Member States agreed that the EU should terminate, it would cease to exist.

However, this coincidence of wills may not take place. It is easy to imagine a situation in which most of the Member States, but not all, want to put an end to the EU. Suppose, for instance, that a majority of the Member States think that the very existence of the EU is an obstacle to the re-unification of an extended Europe that would comprise Russia as well (the old dream of having a Greater Europe from ‘Lisbon to Vladivostok’, as it were). The rest of the Member States would oppose the EU’s termination. Therefore, there would be no common agreement on the issue. The solution that the Vienna Convention offers for cases of non-agreement about termination is that those parties to the Treaty that want it terminated can adduce a ‘fundamental change of circumstances’ that would have occurred with regard to those existing at the time of the Treaty’s signing.[3] This rebus sic stantibus clause could work in my example as follows: the EC Treaty was signed in 1957, well before the collapse of the Berlin Wall in 1987. After the collapse of the Berlin Wall, the historical circumstances that gave rise to the creation of the EC have completely changed. Therefore, under the new circumstances (basically, the fall of the Soviet Union and the return to democracy of central and eastern Europe) the Treaty would have to terminate, since it was forged in a completely different historical context.

Of course, Member States belonging to the minority would oppose such an argument. For these cases, the Vienna Convention foresees a number of procedures for an independent third party to find a solution.[4]

The EU’s termination would of course have consequences. The Vienna Convention says in this regard that the Treaty’s extinction ‘does not affect any right, obligation or legal situation of the parties created through the execution of the Treaty prior to its termination’.[5] Therefore, the EU would have to fulfil all the obligations that it had entered into before termination of the Treaty. If it did otherwise, the EU’s legal responsibility would be of course engaged. Further, beyond the legal consequences, there would be other kind of consequences: to cite but the most obvious ones, Member States would suffer economic costs derived from the extinction of the single market. This aspect of the consequences (legal and other) of termination is not a minor one: it moves many authors to argue that termination would be so costly that Member States would never even consider the option.[6] This argument is relevant; however, the point that I want to make here is that whatever the costs of termination, they would not rule out by themselves the possibility of the EU’s termination. A different thing is how likely termination would be but, as pointed out in the introduction to this paper, the issue of probabilities shall be dealt with later on.

Scenario 2: Variable Geometries

Variable geometries would be the second least integrationist scenario. To start with, it is important to correctly depict what is to be understood by this scenario. Note, first of all, that I use the expression in the plural. With this I aim to contain a number of different situations that could be codified under the general label ‘variable geometries’. For example, Shaw (2003) and Ehlermann (1995) speak at the very least of four different cases of variable geometries, or as they call it, ‘flexibility’. First, we find cases of ‘multi-speed Europe’, the transitional periods that are often accorded to particular Member States in Community legislation being its primary example. Secondly, we find cases of reinforced co-operations. The EMU (European Monetary Union) would be the most notable example of this kind of situation. Third, we find cases of opting-out. The UK’s opting-out included in the Protocol and Agreement on Social Policy, annexed to the Maastricht Treaty, would be a case in point. And fourth, we find cases of international agreements among Member States adopted outside the EU’s institutional framework, like the Schengen Agreement of 14 June 1985, originally signed by Belgium, the Netherlands, Luxembourg, Germany and France. The last set of cases is the most radical form of variable geometry, since by definition it implies the very absence of the EU’s institutional structures.[7]

What unites all these different situations, and some others one could probably think of, is that not all Member States go together as regards the adoption of a particular measure or policy. Thus, in the first set of cases, the ‘time’ variable would make it possible to speak of the existence of a variable geometry situation: the MemberState or MemberStates that have been granted a transitional period would not go hand in hand with the rest of the States. In the second, the key would be that the MemberState in question would not have adopted a specific policy, whereas in the third one, it would have expressly declared its wish to drop out of a particular policy.

The fourth one (international treaties) involves a more complex kind of cases. One could conceive at least two different situations here. The first situation would be one in which some Member States make an international agreement outside the Community’s institutional framework, since inside that framework it is impossible to reach an agreement. And the second situation would be one in which all Member States make an international agreement outside the Community framework, for some reason (for example because the Community decision-making rule does not fit them: imagine that decision rule is majority voting and that they want to adopt the measure by unanimous agreement).

As the list of examples suggests, there are already cases of ‘variable geometries’ in the EU framework. In fact, one could even say that variable geometries are sprawling in the Community context.[8] They are developing into something more frequent than some would like them to (Toth, 1998). However, it is still possible today to say that the cases of variable geometries are the exception, the rule being the cases in which the so-called ‘Community method’ is in action.[9]

Therefore, a correct way to depict this scenario would be one in which the current exception became the rule, and the rule the exception. In other words, the second scenario would reflect a situation in which there would be more instances of ‘variable geometries’ than instances in which the Community method is at play. To use a metaphor, the EU would look like a ‘gruyere cheese’, with many holes of variable geometries within a never-ending decreasing surface of Community orthodox integration.

To be sure, we could find different degrees of integration within the ‘variable geometries’ scenario itself. It is obvious that it would not be the same, as far as our measure ‘degree of integration’ is concerned, that the cases of ‘multi-speed’ Europe were the most frequent cases, and those of ‘enhanced cooperation’, the least, and vice-versa. Accordingly, this scenario could adopt more and less radical forms. In this sense, the most radical expression of this scenario would be one in which only one policy was managed through the ‘Community method’ and the rest of the policies through reinforced co-operation and International Treaties. The policy I speak about is the single market policy. The reason for this choice is not arbitrary: I do not see how this policy could be managed through variable geometries. It is not by chance that the EU and the EC Treaties have specifically excluded the possibility of having a reinforced cooperation in this field.[10] Beyond that, every other policy could in fact be dealt with by variable geometries. This leads us to the following conclusion: in its most radical expression, this scenario would be one in which the EU would have diluted into a free market zone, with no other supranational features, plus a number of policies which the Member States would adhere to as a function of their respective needs and interests.

From a different angle, it is important to underline that the ‘variable geometries’ scenario gives rise to a certain paradox. On the one hand, it is true, as I have argued before, that the multiplication of the instances of variable geometry means, in principle, a lesser degree of integration. However, on the other hand, it is also true that inside a specific variable geometry setting, one may find instances in which Community deepening is enhanced.[11] And, as we know, deepening is one of the forms integration can adopt. Suppose, for example, that Member States cannot agree on a measure related to energy policy establishing a Community independent agency, with both regulatory and control powers. Then a majority of them agree to create a reinforced cooperation between them with exactly the same content as the original Community measure. Couldn’t we say that integration has been strengthened in this way? In more general terms: what is the justification for placing variable geometries as the second least integrationist scenario?

The answer to this question is of course contingent upon the definition we adopt of integration. By this word, or phenomenon, I understand the situation in which all Member States, with no exception, adopt measures using the Community institutional setting. This definition contains two different aspects. In the first place, there is a merely numerical question. In this sense, only the presence of all Member States would indicate integration. In the second place, there is an institutional aspect. It is not only that all Member States have to be present; it is also that all of them have to have used the ordinary decision-making procedures and institutions established in the Treaties. This is what is usually referred to, in the Community jargon, as the ‘Community method’. In the absence of any of those elements, Member States would be placing themselves outside the Community method, and therefore there would be a lesser degree of integration.

From this perspective, it is justified to place ‘variable geometries’ as the second least integrationist scenario and ‘status quo’, for example, as a more integrationist one. As I have said before, the current Community situation is still characterised by the fact that the Community method is the rule and variable geometries the exception. This is, again, independent of the concrete degree of deepening that specific cases of variable geometries may adopt. For example, in the current scenario, one major instance of variable geometry is the EMU. Only 13 Member States belong to the Euro. Imagine that in the second scenario I have depicted here, the ‘variable geometries’ scenario, all Member States belong to the Euro, except the UK. We could still argue that the current scenario would be a more integrationist one, since, independently of the degree of deepening of each of the policies one considers, the Community method is more frequent than variable geometries on the whole.

In sum, the definition of ‘integration’ that I am addressing here is contingent. Its main restriction is that it does not capture the vertical dimension of the phenomenon of integration, but only the horizontal one. Despite this constraint, it is still useful, due to its greater simplicity. Identifying integration also with deepening would involve us in a very difficult discussion about the ways of measuring the different degrees of integration. It is therefore necessary to acknowledge that there are other possible (but always more complex) definitions of what integration is, that could give rise to a different location of our scenarios on the ‘degree of integration’ scale. However, from the perspective of the specific definition of integration that I am using here, it is clear that ‘variable geometries’ would be the second least integrationist scenario we can conceive.

Scenario 3: Status Quo

The expression ‘status quo’ denotes the freezing of the EU’s current degree of integration. In the first place, the EU, at present, can be depicted as a very complex mix of supranational and intergovernmental features, coupled with a certain dose of variable geometries, as said in the previous section. Legally speaking, the supranational features are mainly composed of the EC Treaty, which contains, as is known, all the provisions relative to the European Community. They are also composed of a legal order which has the characteristic of being ‘supreme’ to the national legal orders, constitutions included. From an institutional perspective, the supranational features are formed by the existence of three main institutions, the European Commission, the European Parliament, the European Court of Justice and an important number of administrative agencies. Further, from the perspective of the decision-making rules, many Community legal acts are adopted through majority voting. This is a key supranational feature, which has been the battleground of profound and sometimes tragic political and also academic disputes.[12] Finally, as far as policies are concerned, market policies constitute the core of the policies managed through supranational means. Of course the EU has developed, over the years, a wide range of policies which were originally ‘flanking’ market policies, but that now have a certain autonomy from their original economic orientation. I refer, for example, to the EU environmental policy, its social policy and its consumer policy. Important as they are, this does not rule out my previous assertion: market policies continue to be the backbone of Community supranational integration.

As regards the EU intergovernmental features (which are basically established in the EU Treaty), they are mainly composed of the existence of the two so-called ‘intergovernmental pillars’, the FCSP, the Foreign and Common Security Policy (second pillar) and the PCCJM, Police Cooperation in Criminal and Justice Matters, (third pillar). Basically, all decisions are unanimously taken inside both pillars; the role of the Community supranational institutions and agencies is downplayed, the main actors being the Member State’s governments; and the decisions that emanate from these pillars belong more to the category of classical international law than of Community law: for example, the European Court of Justice has its powers seriously curtailed in both pillars.

The third aspect which characterises the EU’s current state of affairs is an exceptional but ever growing use of variable geometries. This point has already been developed in the previous section, and it is not necessary to add more here to what has already been said.

In the second place, when I characterise this scenario as a freezing of the current EU situation what I am essentially saying is that the EU is reformed in the next 10 years. There would be neither more Treaty reforms nor other ‘informal’ arrangements containing de facto Treaty amendments in this period. From a substantive perspective, the bottom line that would characterise this scenario would be that the current equilibrium between supranational and intergovernmental features plus variable geometries would be kept from here to 2017.

This said, it must be clarified that I do not include in this picture a total and absolute EU paralysis, as if the EU had become quadriplegic. Of course, the EU’s ordinary affairs would continue to be managed. We could have more policies managed through variable geometries, or even new policies developed through the ordinary Community method. But the basic equilibrium between supranational and intergovernmental features (coupled with a certain degree of variable geometries) that is characteristic of today’s Europe would not be altered.

It is important to mention at this point the current process of Treaty reform that the EU is now undertaking. The so-called ‘Reform Treaty’ or ‘Lisbon Treaty’ has been signed by the Member States in December 2007. Two considerations can be made regarding this new reform. First, it is not clear whether the Treaty will be ratified, especially after the Irish ‘no’ last June. And secondly, even if the Reform Treaty goes ahead, it is not very risky to forecast that the ratification process will take a long time. Due to these uncertainties, it is better to conceive the Reform Treaty as an example of the scenario that I shall discuss in the next section, ‘incremental integration with variable geometries’.

Scenario 4: Incremental Integration with Variable Geometries

Contrary to what would happen in the ‘status quo’ scenario, in this scenario the EU would be reformed. The reform or reforms of the EU Treaties would have two main traits. First, they would enlarge the number of issues that would be managed through the so-called ‘community method’. This would happen through two ways: either because parts of the intergovernmental pillars would be communitarised, or because new competences would be attributed to the EU. A mix of both would also be conceivable under this scenario.

The enlargement of the supranational features of the EU would take place in a very incremental way. The evolution of this axis would thus follow a path-dependant (Margolis & Liebowitz, 1998) kind of logic. Therefore, under this scenario, there would be some evolution but no big changes in the degree of EU integration. Each reform of the Treaties would contain a marginal but nonetheless visible increment of the supranational features of the EU.

The second aspect would be that these reforms would facilitate the conditions for making variable geometries. This evolution would also be incremental. Little by little, step by step, the possibilities of making variable geometries would be eased. Accordingly, we would expect an increment, not exponential, but also very visible, of variable geometries situations. However, under this scenario, the policies managed by the Community method would always outnumber the policies managed though variable geometries.

As commented above, a perfect illustration of this kind of scenario is the so-called Reform Treaty. The first thing to be stressed in this regard is that the Reform Treaty eliminates every single vestige of the Constitutional Treaty. It eliminates the word constitution; it even says that the Union ‘will not have a constitutional character’; it eliminates the expression ‘Union Minister of Foreign Affairs’ that the Constitution had included, and substitutes it for the more intergovernmental one of ‘High Representative of the Union for Foreign Affairs and Security Polity’; it eliminates the denomination ‘laws’, and keeps the old ones (directive, decision, regulation, etc); and it gets rid of the so-called ‘Union symbols’ (flag, anthem and motto).[13] Maybe more important from an institutional perspective: the Reform Treaty erases the Constitution provision on the primacy of EU law over the national legal orders,[14] a hard-won achievement of the ICG 2004.

Even though the Reform Treaty wipes out all traces of the Constitutional Treaty, it does in fact enlarge the cases that will be managed through the Community method. The main example of this is in the third pillar. As is known, the third pillar was partially communitarised through the Nice Treaty. The Reform Treaty envisages its full communitarisation thus creating an Area of Freedom, Security and Justice.[15] This notwithstanding, it should be pointed out that the Reform Treaty opens the door to the activation of reinforced co-operation in this area.[16] For its part, intergovernmentalism will still have an important presence with the Reform Treaty. In particular, the Common Foreign and Security Policy remains an area of intergovernmental cooperation.[17]

Incremental changes occur as well regarding the general regulation of reinforced co-operations. In particular, the Reform Treaty establishes that enhanced co-operation will require the favourable vote of at least nine Member States. This is, again, a perfect example of the sort of incrementalism that is so pervasive today in the EU context and that would characterise this scenario. Thus in the Constitutional Treaty, it was foreseen that reinforced co-operation would need at least the support of one third of the Member States. By definition, this was a more conservative rule, since as the EU enlarges, the number required for creating reinforced co-operation would proportionally be augmented as well. The Reform Treaty, instead, freezes this number at nine Member States. Therefore, the threshold would be the same irrespective of whether the EU keeps enlarging or not. In this way, reinforced co-operations are further facilitated.[18]

It remains to be seen what the fate of the Reform Treaty will be. It is, however, clear that if the Reform Treaty is finally ratified, it will perfectly match the scenario just discussed in this section.

Scenario 5: Incremental Integration Without Variable Geometries

This scenario is both similar and different to the scenario previously discussed. It is similar in the sense that in this scenario, as well as in the previous one, the enlargement of the supranational Community features would be enhanced in an incremental way. It is different in the sense that in this scenario variable geometries would be reduced.

The progressive reduction of the variable geometries situations would also be incremental. Therefore, variable geometries would not disappear from the Community landscape overnight. The reduction would be both institutional and material. Institutionally speaking, variable geometries would be made more difficult in each Treaty reform. And materially speaking, they would be used less and less. In the most radical expression of this scenario, variable geometries would disappear with time.

As hinted before, variable geometries, in the different forms they have adopted over the years, have been present in the Community context since its inception. What is new is the existence of a specific flexibility mechanism. As is known, this mechanism was incorporated in the Amsterdam reform of the Treaties, both at a general level and for concrete cases. Though the general flexibility mechanism has never been activated (Thym, 2006), some of the specific reinforced-cooperation mechanisms are active, the EMU being the most clear example. Is it possible then to conceive a back-up as regards variable geometries?

As shall be discussed later, it is not very likely that the EU will return to its ‘original position’, that is, to a time when everything was nice and simple and there were very few cases of differentiated integration. In my opinion we are entering an era in which we will have to learn to coexist with more complexity and therefore with more variable geometries. However, one should not rule it out as a possible scenario.

Scenario 6: Political Union

The establishment of a political union would be the most integrationist scenario. However, the term ‘political union’ is open to discussion, at least when transposed to the Community context. Do we want to say with this term that the EU would be transformed into a state,[19] or even in an ‘empire’, as part of the literature on European integration argues today?[20] Or are we thinking of a hitherto unknown and therefore new form of political union?[21] These would be the two extremes of the current debate over the would-be European political union.

To start with, it is important to de-dramatise this debate, and to treat it in a more scientific way. In this sense, it is important to recall that all democratic political unions distribute power along two main axes: a horizontal one and a vertical one. The horizontal axis is formed by three powers that constitute Montesquieu’s classical division: the executive, the legislative and the judiciary. In turn, the vertical axis is about the share of power between the centre and the periphery.

That democracies distribute power along the first –horizontal– axis is uncontested. That they also do it along the second –vertical– one is perhaps more provocative. But I do not know a single modern and democratic State that does not delegate at least a minimum degree of power to its lower units of government. France would be the classical case in point (Hazareesingh, 2002).

Furthermore, power may be more or less concentrated along both the horizontal and the vertical axes (see Figure 2). From this perspective, we can have Presidentialist, semi-presidentialist and parliamentarian systems as far as the horizontal axis is concerned. And as regards the vertical one, we encounter centralised, decentralised and federal or con-federal systems. As I said before, even the most centralised systems know at least a minimum degree of decentralisation.

Figure 2. Concentration of power along the horizontal and vertical axes

Horizontal axisPresidentialist
Vertical axisCentralised
Federal or con-federal

It is clear that if the EU became a political union, it would have to resemble, at least partially, one of the models hinted at in Figure 2, or a combination of these. The literature that argues that the EU would be an ‘unidentified political object’, to use Delors’s famous words, is pure rhetoric. A political European union would have to distribute its power along the two axes I refer to. Further, its concentration of power would be, within both axes, lesser or greater, depending on the concrete institutional arrangements that the Member States agreed to.

It is also clear, however, that the ‘state’ element would perhaps be more present than in most of the political unions that we know. This would perhaps differentiate the political European Union from most of the others. But arguing that because of the ‘state’ element the EU, if it became a political entity, would be something completely new is not a very convincing argument. Just think, for example, of con-federal political unions like Switzerland (Church & Dardanelli, 2005).

Taking the previous elements into consideration, what would this scenario look like? Of all the different combinations that Figure 2 could give rise to, I would tend to depict this scenario as a parliamentarian or as much as a semi-presidentialist regime with an intense degree of vertical decentralisation (therefore following the model of con-federal states). This European Political Union would have a parliament that would represent the European people (and not the Member States’ peoples, as happens now). It would not have a Commission but a government which would be chosen by the parliament. On the horizontal axis, the ‘state’ element I referred to above would be present in a number of different ways. For example, even though the European government would have a specific political colour, it would also be composed of members of different nationalities, in such a way that no MemberState could have more than one national in it. Finally, the EU judiciary would be above the Member States’ judiciaries. It would have the final say in EU affairs.

Regarding the vertical axis, there would be a clear-cut definition of the competences of MemberStates and the EU. Division of power would follow the American model rather than the German one.[22] That is to say, competences would not be chopped into pieces. When a competence was delegated to the EU, it would have the entire competence, and not part of that competence. Conversely, Member States would retain the whole of the competences they would keep.

The ‘state’ element would emerge again as far as the vertical axis is concerned. For example, Member States’ governments would be active players in the Community game. Therefore the Council of Ministers would be kept as it is today. Only decision-making procedures would change: unanimity would disappear and all decisions would have to be taken together with the European parliament. Depending on the subject, one or the other would have the final say.

The European Political Union would have, according to this scenario, a Constitution (and not a ‘Constitutional Treaty’, which tries to combine the best of the state and international organisation worlds with the result of creating a never-ending misunderstanding about the true nature of the EU). The EU would therefore have a constitutional status. In the international arena, it would act as if it were a state like any other. The sort of constitution that the EU would have would resemble more the US model than the Constitutional Treaty. Therefore, the basic rules of the European game would be drafted in a clear, elegant and sober manner. All the details would be developed by laws.

In short, under this scenario, the EU would resemble many of the political unions that we know today; it would, however, have its own specificities, as is common in all political unions we know today.

Probable Scenarios for Europe in 2017

Once the possible scenarios have been reviewed, I shall now move on to analyse the probable scenarios in which Europe might find itself in 10 years time. To do this I shall proceed in the following way. The first step will be to set up a model that helps us to predict which of the possible scenarios will be more likely to happen. In the second place, I shall discuss in depth the explanatory variables of which the model is composed. Thirdly, I shall analyse theoretically how these variables actually work in order to forecast what the EU’s future will be in the next 10 years. In view of the previous discussion, I shall establish in section four of this paper a hypothesis about the most probable scenario in which the EU might find itself in 2017.

It is important to note that the following discussion is of a theoretical kind. However, it is clear that the methodology I am using here allows taking this analysis a step further and subjecting it to further testing, and in particular, to statistical testing. Yet in order not to overload this analysis, the second step will not be taken here.

A Predictive Model of Europe’s Degree of Integration

The predictive model of the probable EU scenarios in 10 years time takes the following shape:

Integr = f (Enlarge, Divers, Heterints, Compts)

‘Integr’ is the dependent variable. It measures the EU’s degree of integration. The independent variables would be ‘enlarge’, ‘diver’, ‘heterints’ and ‘compts’. ‘Enlarge’ measures Europe’s widening to new Member States. ‘Diver’ measures the legal, political and economic diversity at the interior of the EU. ‘Heterints’ reflects the diversity of preferences of Member States. Finally, ‘compts’ describes the intensity of delegation of core sovereignty competences to the EU.

In general terms, one would expect that the relation between the dependent and the independent variables would be of a negative kind. Therefore, as the value of the four independent variables increases, the value of our dependent variable should decrease. A graphical expression of this would be the following (see Figure 3):

Figure 3. Relation between dependent and independent variables (I)

image002 53 1

Further, suppose, for instance, that we gave values, from 1 to 10, to the different variables that we have in Figure 3. In this case, the more the mean value of the independent variables, the less the value of the dependent variable. Imagine, for example that the mean value of our four independent variables equalled 7. In this case, as the relation between de dependent and the independent variables is negative, the value of the degree of integration would be, say, equal to 3 (see Figure 4).

Figure 4. Relation between dependent and independent variables (II)

image003 37 1

A second general remark that is important to make is that I have excluded from my analysis any sort of consideration relating to the issue of voting. For example, Golub (2007) includes this variable in his analysis of the factors that explain EU decision-making speed. He argues that the more QMV, the greater the speed of the EU’s decision making process. However, in my analysis, QMV would belong to the dependent variable, and not to the independent one. The reason for this is that I am trying to explain the degree of EU integration, whereas Golub is trying to do the same regarding a more specific variable, decision making speed. In my case I assume that the extension of QMV would be an indication of more integration. Remember that I have defined ‘degree of integration’ as a strictly horizontal phenomenon. Viewed from the angle of this definition, it is clear that the greater use of QMV would reduce the chances of the Member States making use of variable geometries, which is the very negation of the variable ‘degree of integration’ as I have defined it. In any case, QMV would be a further expression of the so-called Community supranational features, which also indicate more (horizontal) integration.

Discussion of Variables

Degree of Integration. Much has been said up to now about our dependent variable, so I will just summarise some of the main points that have been raised so far. The variable ‘degree of integration’ has many faces: at the very least, a horizontal one and a vertical one. To simplify matters, I have used a very traditional definition of what integration is, that is, its horizontal definition. Therefore, there would be integration in those instances in which Member States have adopted a measure using the traditional community method. I have said before that this definition involved, above all, a quantitative question: the more the instances in which Member States have adopted together a measure, the greater the degree of integration. This would be the case also if Member States adopted measures using QMV and therefore outvoted some Member States. The important thing here would be that the decision has been actually taken, and therefore the emergence of an instance of variable geometry has been foreclosed. In other words, this definition of degree of integration would entail two steps: a first step, in which all Member States would participate in the decision of whether or not a certain measure has to be taken; and a second one in which the measure is adopted (albeit by overruling some Member States).

Enlargement. The ‘enlargement’ variable denotes the widening of the EU to new Member States. The greater the number of Member States that join the EU, the greater the enlargement and, as I have said before, the lower the degree of EU integration.

The argument according to which enlargement and degree of integration are related to each other in a negative way is very controversial. The scientific literature that addresses the effects of enlargement upon integration (or proxies of integration) is disputatious in this regard. For example, Golub (2007) argues that the influence of enlargement upon EU decision-making speed (which could be depicted as an indirect indication of integration) is of a positive kind. Although he does not include in his analysis the 2004 and 2007 enlargements, he concludes that enlargement eases the speed of EU decision making. This is in sharp contrast with, at the very least, conventional wisdom. The Reflexion Group for the IGC of 1997 is a very good example that illustrates where conventional wisdom is in this matter. The Reflexion Group said in 1995:

‘to ensure that the next enlargement does not weaken, change the nature or actually break up the Union, the reforms needed to cope with the challenges involved must first be made’.

Therefore, for the Reflexion Group, enlargement would hinder integration per se. Only the reform of the EU institutional landscape would help to cushion the blow that enlargement would be to the EU.

There are also middle-ground positions. For example, Schneider (2002) says the following: ‘the relationship between widening and deepening is basically curvilinear: moving from a status of exclusive sovereignty to some sort of formalised cooperation is easier for a large rather than a small number of states; yet in the event that a regional organisation gains momentum, adding members becomes less desirable because of the dramatically increasing transaction costs’.

Of all the three positions, Schneider’s appears to be the most convincing. It may well be that the addition of new members is indispensable for the setting up of an international organisation. However, after a number of years, and if the international organisation has spread its activities, new entries could augment the transaction costs in an unmanageable way. Transaction costs are the mechanism that explains why new increases in the number of Member States make integration stagnate. As new Member States join the EU, the costs associated with reaching agreements (in terms of time, gathering of information, bargaining, investment of resources, etc) also increase. The transaction costs increase delays in the adoption of agreements and this, in turn, hinders integration.

Schneider does not clarify in his work where the line is drawn between an efficient international organization, in terms of its number of Member States, and an inefficient one. However, the author wrote in 2002, that is, in the wake of the 2004 Central and Eastern Europe enlargement. His conclusion cannot be clearer: ‘with regard to the current negotiations (…) enlarging the EU will most likely increase the wish of some Member States to loosen the collaborative network’ (ibid., p. 197). This would provide an indirect indication that for this author, the line would probably have been crossed after the latest enlargements. We would be now in a time in which future widening would only reinforce the trend that Schneider forecasts.

Diversity. This variable indicates the degree of political, economic and legal differentiation between Member States. As we have already said, the assumption here is that the greater the diversity, the lower the degree of integration.

Diversity is not a variable that is usually contemplated in the analyses of the determinants of EU integration. It is usually assumed that the greater the enlargement, the greater the diversity. In other words: diversity is very often subsumed within the variable enlargement.

The connection between diversity and enlargement is not, however, self-evident. In fact, an enlarged EU does not necessarily have to be a more diverse one. It all depends on the direction that enlargement takes. It is clear that it is not the same for the EU to enlarge to, for example, include Turkey, or the Balkans, or that it enlarges to include countries like Switzerland or Norway. In the first case we would experience a further degree of political, legal and economic diversity than in the second case. As both variables do not necessarily mean the same, it is therefore essential to disentangle them.

It is also necessary to separate diversity from enlargement since the former acts as a control variable of the latter. If correctly disentangled, the effect of enlargement upon integration can be determined in a more precise way. We would be more sure, for example, that when we say that enlargement is playing a negative role upon integration, it is not the effect of other variables, and in particular, diversity, which is interacting there.

As hinted before, the relationship between diversity and integration is also of a negative kind. We would expect that as the value of diversity increases, the value of integration decreases. The mechanism that explains this relationship is that new Member States, if very diverse compared to the old ones (those States that have already been members for some time), will oppose the adoption of measures that are very different from their respective economic, political or legal backgrounds. Negotiations will therefore be protracted, the longer they take the more complex they will become (Fearon, 1998) and this will in turn hinder integration. Here the key element is the transition cost that new Member States will have to pay to adapt to very diverse political, legal or economic situations. The more this kind of cost rises, the more they will oppose new measures.

It is also important to mention that the passage of time does play a role as regards this variable. As time passes by, the Union gets less and less diverse, since, at least in theory, new Member States will little by little adapt their political, economic and legal backgrounds to the Community rule. In other words, the passage of time would make the negative impact of this variable upon the degree of integration decrease. If the Union paralysed enlargement, it would get to a point (which is impossible to specify in advance) in which the role of this variable upon integration would be much less significant.

Heterogeneity of Interests. This variable indicates the degree of heterogeneity of preferences between Member States. The assumption is that the more preference heterogeneity, the less integration.

Contrary to what happens with the variable diversity, heterogeneity of interests is present in the current discussion over the elements that influence the degree of EU integration. For example, Golub affirms that ‘what appears to bog down the [integration] process is the presence of a memberState government with extreme preferences’. In his analysis, Golub studies the influence that the Thatcher governments had upon the EU decision-making speed. He shows that while she was Prime Minister, the ‘hazard rate for adoptions was still 22% lower after 6 months of negotiations’. From this he derives the conclusion that what appears to be ‘a real threat to speedy EU decision-making is not more MemberStates but the election of extremists, whether in the old or in new Member States’.

The mechanism that explains the negative relation between this variable and the degree of integration is that the more distant the Member States’ respective points of departure in EU negotiations, the more costly it is to reach agreements. Transaction costs are once again the determinant. And, in this connection, Golub’s measure of extremist governments can be a useful indirect indication pointing at the presence of heterogeneity of interests. We shall see later on how this measure is specified in the context of this work.

Delegation of Core Competences. Delegation of core sovereignty competences to the EU is our last variable. I assume here that the greater the delegation of this kind of competences to the EU, the lower the degree of integration.

The issue of competences is also present in the literature that attempts to explain the degree of EU integration. This does not mean that the role that this variable plays regarding integration is uncontroversial. Golub, for example, argues that the expansion of the EU legislative agenda to new policy sectors does not slow down the EU’s decision-making speed. His analysis shows that this variable is not significant. Instead, authors like Ludlow argue in the opposite direction. For this author, one of the main determinants of the EU’s present complexity (which, as has been argued here, would be the very opposite of integration such as has been defined in this work) is precisely the never-ending growth of the EU legislative agenda (Ludlow, 2008).

A middle-ground position seems again adequate for understanding the actual impact that the delegation of competences has upon the degree of EU integration. As Golub argues, it is difficult to find good reasons to explain why the expansion of the EU legislative agenda, as such, may have an impact on integration. Quite the contrary, the expansion of EU competences, if agreed by all Member States, may act as an engine of integration, rather than a brake. In fact one of the most convincing arguments of the very existence of the EU is the one that says that, through the EU, Member States attempt to solve the problems of inertia existing at the national level. To put it in simpler terms, Member States would delegate to the EU those areas of competence that cannot be managed, for one reason or the other, at the national level. The expansion of EU competences is not detrimental to the EU’s integration, at least per se.

Yet it all depends on the specific sector that Member States delegate to the EU. In particular, it may well be the case that the delegation of core competences to the national sovereignty does hinder integration. Clearly, it is not the same to delegate the management of market policies as to delegate foreign policy to the EU. In the former case, the opportunity cost of delegation is much less than in the latter. This is because delegating market policies to the EU is a key to the economic development of states, whereas the advantages of losing autonomy in the field of foreign policy are much less clear.

Opportunity costs are therefore the mechanism that explains why Member States are reluctant to cede core competences to the EU. Once they delegate them, it explains why they establish so many controls over their development. Precisely because they fear big opportunity costs if core competences are managed by the EU alone, they establish institutional mechanisms addressed to lose as little autonomy as they can. Such controls stifle integration. Therefore, because of them, the more core competences Member States delegate to the EU, the lower the degree of integration.

It is obvious that implicit to this argument is a discussion about what core competences are. Core competences are what Member States understand core competences to be. Thus the definition is necessarily contingent. Perhaps the most we can say about it is that core competences are those competences that define the basic traits of national sovereignty. Without those competences, a MemberState could not consider itself a state anymore, at least as we still currently conceive them.[23]

Independent Variables at Work


Will the EU enlarge in the next 10 years? If so, to what extent? In order to answer both questions it is important to start by distinguishing between two situations in which the new would-be Member States find themselves. A first situation is formed by those states that hold a formal status of ‘candidate-countries’. These are the states that are currently negotiating or will soon start negotiating with the EU the terms and conditions of their future accession to the EU. Turkey, Croatia and Macedonia belong to this category.

The second situation is formed by those states which hold a formal status of ‘potential candidate-countries’: Albania, Bosnia Herzegovina, Serbia and Montenegro. These states have already undertaken what is called in EU jargon ‘the Process of Stabilisation and Association’. In this context, all of them have adopted Stabilisation and Association Agreements with the EU. They have to fulfil a number of prerequisites before they can have the formal status of ‘candidate countries’.

Beyond this division, the specific situation of each of the previously mentioned states is the following. To start with Turkey, this is the state that has made the greatest headway in the process to accession. However, after the so-called ‘ports’ crisis, the European Council decided to suspend negotiations in December 2006 as regards eight specific chapters of the negotiation package.[24] Although the halt does not mean calling into question the whole accession process, it is clear that Turkey’s entry in the EU will be further delayed. In fact, the European Commission considers, in its recent communication on enlargement,[25] that the full normalisation of bilateral relations with the Republic

of Cyprus is a must if Turkey wants to resume negotiations.

From a different angle, it is important to note that Turkey’s accession is one of the most controversial issues the EU is currently facing (Ruiz Jiménez & Torreblanca, 2007). In fact, Member States’ leaders have made declarations against the accession of this country to the EU. This is the case of France: President Sarkozy stated in the presidential election campaign of April-May 2007 that he was against the possibility.[26] He even gave an alternative to Turkey’s accession, the so-called ‘Mediterranean Union’. According to Sarkozy, Turkey could belong to the Mediterranean Union, which in turn would have a special relationship with the EU. But this is the furthest that this state would integrate in the EU. The whole thing only gets more complicated if one takes into account that the current French constitution foresees a referendum on the accession of new Member States. Although this provision does not specifically mention Turkey, it is clear that it points above all to this state.[27]

Regarding Croatia, the Commission was pleased to announce in its Communication on enlargement of November 2007 that accession negotiations with this state were ‘advancing well and entering a decisive phase’. This notwithstanding, the Commission also mentions a number of areas in which this country still has to improve. In particular, the Commission mentions the following five: judicial reform, administrative reform, minority rights, refugee return and the restructuring of its steel and shipbuilding industries. In its specific progress report on Croatia of November 2007,[28] the Commission is particularly tough as regards the issue of corruption. According to the Commission, ‘corruption remains widespread’. This is by the way a general theme in the Commission’s evaluation of the Western Balkan countries’ merits for accession. Regarding the prospect of accession negotiations with this state, the Commission expects substantial progress, ‘provided that the country maintains the necessary reform momentum and meets benchmarks’. As with Turkey, no-one dares to give a date for its accession. However, most analysts agree that if it maintains the current pace of steady reforms, Croatia could be prepared for accession sooner rather than later (Rodin, 2006).

The Former Yugoslav Republic of Macedonia was granted the formal status of candidate country in December 2005. However, accession negotiations with the EU have not started yet. Therefore, Macedonia is still subject to the so-called ‘stabilisation and association process’. In this regard, the Commission says in its Communication on enlargement of November 2007[29] that it ‘has made progress but still needs to accelerate the pace of reforms in key areas’: police reform, judiciary reform, the consolidation of the rule of law, the protection of minority rights and unemployment. Again, the Commission repeats the formula that it uses with Croatia when it says that ‘corruption remains widespread’. The Commission does not indicate at all when the accession negotiations with this country will start, but the reading of the specific Commission report on this country only invites a moderate optimism.[30]

As regards the ‘potential candidate countries’, Albania signed a Stabilisation and Association Agreement with the EU in June 2006. The Commission says in its general communication on enlargement of November 2007 that Albania has made progress in a number of areas but still faces ‘major challenges’. In particular, the Commission says that Albania has to complete ‘long and overdue’ reforms in the electoral and judiciary areas. The Commission also points to the fact that the private sector is having problems in developing in this country, which in turn might call into question its economic development. According to the Commission, ‘corruption remains widespread’ in this country as well.[31]

Montenegro signed its Stabilisation and Association Agreement in October 2007. This country seems to be evolving along similar lines to those of Albania. In this connection, the Commission says, in its communication on enlargement of November 2007, that Montenegro is making progress though it still faces, as Albania does, ‘major challenges’. These challenges are: judicial and administrative reform, the conditions of refugees and displaced persons (in particular the Roma minority) and macroeconomic stability. The same formula the Commission employs for other Western Balkan countries is used as regards this country: for the Commission, corruption is widespread in Montenegro.

Bosnia Herzegovina recently signed a ‘Stabilisation and Association Agreement’ with the EU in June 2008. According to the Commission, progress in four areas (police reform, administrative reform, public broadcasting reform and achieving full cooperation with the International Criminal Court for the former Yugoslavia) has been decisive for the agreement. However, the Commission’s specific report on Bosnia Herzegovina of November 2007[32] says that its ‘complex institutional arrangements, frequent attacks on the Dayton-Paris peace agreement and nationalistic rhetoric’ reduce the chances of it entering the European club. With regard to its ability to align its policy and legislation with European standards, it has made only limited progress. The reading of the specific country report on Bosnia gives a sombre impression as regards the chances it has of solving the problems the Commission mentions in the short term.

The case of Serbia is more complex than the previous ones. Serbia was granted the status of potential candidate country in the Zagreb and Thessaloniki summits of November 2000 and June 2003, respectively. Further, the Serbian and EU authorities signed a Stabilisation and Association Agreement in April 2008.[33]

However, the main problem with Serbia has been and still is the issue of the status of Kosovo, above all after the latter declared its independence in February 2008. The External Relations Council of Ministers decided, in its session of 18 February 2008, to give a free hand to the Member States as regards Kosovo’s recognition. Since then many Member States have recognised the country, whereas others have not. Spain is for the moment in the ‘non-recognition’ camp.

After reviewing the situation of each of the specific countries that are either candidates or potential candidates to EU accession, what can be said about their prospects for joining in the next 10 years? As the Commission indicates, in its memo of 8 November 2006,[34] ‘it is too early to speculate about when the next accession will take place’. It adds that ‘at present, it appears unlikely that a large group of countries will in future accede simultaneously. The candidates and potential candidates vary considerably in terms of their political and economic development and administrative capacity. Further accessions are likely to occur in the medium to long term, given the present state of pre-accession preparations’.

It all then depends on what the EU authorities understand by ‘medium or long term’. However, 10 years appears a reasonable time-frame for these countries to enter the EU. For example, the 2004 enlargement (to Central and Eastern Europe plus Cyprus and Malta) took around seven years to materialise, if one takes as the departing point of the process the European Council of Luxemburg of 12 and 13 December 1997, in which the enlargement process began.[35] Of course, the current process of enlargement and the previous one are not fully comparable: it is a truism to say that every enlargement process has its peculiarities. But it is also true that in many senses both processes have points of contact. After all, excepting Turkey, the rest of the states that have applied for EU membership were on the other side of the Iron Curtain, as with most of the states that joined the EU in the 2004 and 2007 enlargements.

According to what has been said, Croatia is clearly a country that will join the EU within the next 10 years. Macedonia, Albania and Montenegro could follow Croatia’s lead and join a little later and also within the next 10 years. The prospects of Bosnia Herzegovina and above all Serbia are much more uncertain. They are the last to have signed Stabilisation and Association Agreements. The least that can be said is that the negotiation process to sign these agreements was arduous. In turn, Turkey is a very different case in which objective aspects such as the degree of achievement of the EU benchmarks for accession combine with emotional issues about the ‘essence and limits’ of Europe. Only if European public opinion were to change as regards the prospects of Turkey’s accession would we see it joining the EU. However, it is not very realistic to think that it will change in the next 10 years.

In summary, it is not unreasonable to forecast that the enlargement variable will be placed in a scenario in which at the minimum four and at the maximum six states will join the EU in the next 10 years.


I explained before that diversity is a variable that is connected with the enlargement variable. Therefore it attempts to measure the degree of diversity that the EU will experience if enlargement to new states takes place. We have seen in the previous section that it is not unrealistic to forecast that the EU could enlarge to a minimum of four more countries and to a maximum of six more in the next 10 years. Thus, I assume these would be the enlargement scenarios upon which the diversity variable has to be projected.

The trouble with this variable is that it is difficult to forecast with any certainty what the level of diversity will be at the moment in which enlargement takes place. In this respect, the only thing we can do is to use the degree of diversity that these states now have as a benchmark of future levels of diversity. Of course this only gives a very rough approximation of how different these countries will be from the EU at the time of enlargement, principally because one of the main objectives of the Stabilisation and Association process is precisely to reduce diversity. This is, however, the only alternative we have.

I also said before that diversity was measured in terms of the political, legal and economic differences that the would-be new Member States have in relation to the current EU. Since the 2004 enlargement, the current EU is already very diverse. Therefore it is much more appropriate to use as a benchmark the EU-15, that is, the EU before the last enlargement. This is, incidentally, what the Commission does (although implicitly) when it assesses, in the progress reports it produces, the candidate or potential candidate countries.

The Commission’s progress reports are precisely the best source of information regarding the current degree of diversity that the candidates and potential candidates have. In this connection, progress reports give information about each of these countries’ political cultures when they comment on the state of their democracies, and more particularly, when they assess the functioning of specific institutions such as their constitutions, parliaments or governments. What we want to understand is, specifically, the extent to which these countries are developing a culture of free elections and competition between political parties.

Progress reports also give information about these countries’ legal cultures when they comment on the existence and development of the rule of law in these states, and more particularly, when they refer to the degree of judicial independence that is found in each of them. We want to know here if one of the key elements of the rule of law, its judicial system, is in place and if so the extent to which it is in place. Finally, progress reports also give information about their economic culture, particularly when they assess the extent to which these countries are embracing a market economy. We want to know here whether and to what extent these countries are adopting the main tenets of open capitalist economies.

As regards the four states that have more options to enter the EU in the next 10 years (Croatia, Macedonia, Albania and Montenegro) the findings are as follows. In terms of their political culture, all of them have adopted institutional frameworks that allow political pluralism and free elections. However, many efforts have still to be made in order for democracy to become more than mere form. Many countries have not yet developed a culture of political collaboration and agreement between the different political parties. The use of key democratic institutions –such as parliament– is in some cases epiphenomenal. And in others, the relations between the government and parliament are not sufficiently close: mistrust between both institutions remains the norm. In relation to legal culture, one finds one and the same comment all the time: the judicial system lacks independence in all these countries. Finally, as regards economic culture, and despite some problems, this is the area in which all these states seem to be having less problems in approximating to standard EU-15 culture. For example, macroeconomic stability is a key feature in all these countries. Sustained economic growth is even present in many of them. Market mechanisms are at work in all of them. Unemployment, the lack of a business culture and the strong presence of the old public sectors constitute, however, the other side of the coin.

Diversity is even more pronounced in the remaining two states, Bosnia Herzegovina and Serbia. These are the two states that have less chances of entering the EU in the next 10 years. In Bosnia Herzegovina, and as regards the first issue, political culture, the Dayton-Paris Agreement obligations are not being fulfilled. For example, the Commission denounces that as a result of the failure to reform the constitution, the 2006 election took place in violation of the Human Rights Convention (in particular as regards the protection of minorities). It is very difficult to speak of the existence of an institutional framework giving rise to political pluralism and free elections in Bosnia Herzegovina at present. Regarding the rule of law, judicial independence is also an issue in a country in which there is evidence of the lack of a well-developed judicial institutional infrastructure. Finally, as regards economic culture, this is the aspect in which Bosnia Herzegovina presents the best score, even though market resource allocation efficiency is still hindered.

In turn, Serbia has similar problems. In terms of political culture, and even though elections are considered by international observers to be in line with European standards,[36] the Commission considers that the current constitution contains provisions that do not fully accord with what a constitution of a member State would be expected to contain. In terms of legal culture, judicial independence is still an issue, with the appointment of judges and prosecutors still open to ‘political influence’. As regards economic culture, the country is doing well in macroeconomics, but market mechanisms are far from being in place due to distortions produced by ‘administered prices’.

In summary, all these countries are very diverse in their respective political, legal and (to a lesser extent) economic cultures compared to the EU-15. This is of course not a surprise: all these countries were, not many years ago, behind the Iron Curtain. Therefore it will take time and reform until the EU-15’s political, legal and economic cultures are embedded in their respective societies. The question is whether the EU will wait until diversity is severely reduced before it enlarges again or not. If one takes as a benchmark the 2004 and 2007 widening, the answer would of course be a negative one. If the EU had not waited, then the result would be a more diverse Union, even more diverse than the one we have now with 27 Member States.

Heterogeneity of Interests

Measuring the variable ‘heterogeneity of interests’ poses the same problem as in the previous section: it is very difficult to anticipate in advance the degree of heterogeneity of interests the Union will have in the future. As with the diversity variable, we also need to use heterogeneity of interests proxies and try to simplify our analysis as much as possible. In this connection, I shall circumscribe the scope of this variable to the new would-be Member States. That is, I shall analyse the degree of heterogeneity of interests the Union would have after enlargement. I have assumed before that the EU could enlarge in the next 10 years to a minimum of four new states and a maximum of six new states.

Regarding proxies, I take Golub’s measure of ‘extremist governments’ as a marker of the presence of a significant degree of preference heterogeneity within the EU. In particular, Golub uses in his analysis the Thatcher years as a measure of the effect that an extremist government might have on EU policy-making. Following Golub’s idea, I attempt to identify whether and to what extent we could witness the emergence of extremist governments in the new states once they have joined the EU. In order to do so I use two indicators: first, whether there are political parties that hold nationalistic platforms in these countries; and secondly, what the ethno-social composition of these countries is. Both features can hint whether there is a breeding ground for the emergence of extremist governments with radical preferences. In effect, I assume that if strong nationalistic parties exist in a given country, extremist ideas will have a greater chance of making headway. I also assume that the more complex the ethno-social composition of a country, the greater the internal tensions, and therefore the greater the chances for the emergence of extremist positions in that country. Of course, as with the diversity variable, the best we can do is to analyse what happens now in these countries in terms of both features and attempt to project this to the future.

It is also important to mention that time plays a role here as well. It is not simply optimism to assume that as time passes, nationalistic platforms have more chances to dilute, and ethno-social tensions to accommodate, especially if these countries gradually align their respective legal, political and economic cultures with European standards. The fact that there are strong nationalist parties and ethno-social divisions in these countries at present does not necessarily mean that they will have the same intensity in the future and that, once in the EU, both features will remain the same. However, it is also true that the fact that both features currently exist increases the likelihood that the extremisms with which they are associated will emerge even when these countries have joined the EU.

To start with Croatia, the main political parties are the Croatian Democratic Union (centre-right) and the Social-Democratic Party (heir to the old Communist party, now recycled as social-democracy). Officially none of these parties holds nationalistic platforms but their leaders’ rhetoric is punctuated with soft nationalistic ingredients. There is one party that holds clear and express nationalistic positions, the Croatian Party of Rights. In the last election it had seven seats in the Croatian parliament. However, in the latest election of November 2007, this has dropped to one. If we add this seat to the other three seats that another minor nationalist party (the HDSSB[37]) has gained, the nationalist parties only have four (out of 148) seats in the Sabor (the Croatian parliament).

Regarding the ethno-social dimension, Croatia seems to be a fairly stable country, since Croats dominate (they represent almost the 90% of the population). Serbs account for only 5% and the remaining 5% is made up of Bosniaks, Hungarians, Slovenes, Czechs and Romas.[38]

The situation in Macedonia is far more complex than in Croatia. First, the four main political parties are ethnic-based. There are two Macedonian parties (the VMRO-DPMNE[39], centre-right), and the SDSM[40] (centre-left) and two Albanian parties (the Democratic Union for Integration and the Democratic Party for Albanians). Therefore, irrespective of their ideological profile, all of them have obviously strong nationalist leanings. Secondly, from an ethno-social perspective, the Macedonian population is quite diverse: although Macedonians dominate (64%), the Albanian minority constitutes more than 25%. Macedonian society is therefore very much divided along political and social lines. Although the Macedonian constitution establishes a number of checks and balances to ensure that decisions cannot be taken disregarding the Albanian minority, the Commission’s reports indicate the existence of tensions at both levels.[41]

In Albania, the situation also has specific contours. There are two main political parties, the Democratic Party and the Socialist Party. The first governs at present in coalition with the very conservative Republican Party. In the 2001 election, the Republican Party was part of the Union for Victory, a coalition of parties that included the nationalist National Front. The National Front now has 18 seats in the Albanian parliament and is the third-largest political force in the country.

From an ethno-social perspective, Albania does not present major divisions. The population is 95% Albanian. Although estimates vary, it seems that the remaining 5% is made up of Greeks, Serbs, Romas and Macedonians.[42]

The situation in Montenegro is as follows: the ruling coalition is the Coalition for European Montenegro, made up of the Democratic Party of Socialists of Montenegro, the Social Democratic Party and the Croatian Civic Initiative. Except for the latter, none of these parties have an ethnic basis; however, they struggled for the independence of Montenegro from Serbia. The opposition is constituted by the so-called Serbian list. The Serbian list is composed of a number of political parties, two of which are ethnic based and can be considered nationalist (pro-Serb): the Serb Peoples Party and the Serbian Radical Party. Although the political situation is less polarised than in Macedonia (since in principle the two main parties of the Coalition for European Montenegro are not ethnic based) it is clear that the different coalitions represent at the very least very different views as to the way the relation with Serbia should be conducted.

From an ethno-social perspective, the situation in Montenegro is quite polarised: according to CIA estimates, 43% of the population is Montenegrin and 32% Serbian.

The two remaining countries, Serbia and Bosnia Herzegovina, have, according to my assumptions, less chances of entering the EU in the next 10 years. However, I have also said that in an optimistic scenario they could join the Union if they did well. From the perspective of the variable that we are now analysing, the following can be said of theses countries.

Starting with Bosnia-Herzegovina, this is by far the country with the most potential for the emergence of tensions deriving from its current political and social composition. It is not necessary to enter into the complex details that define the country’s political and social situation. Suffice to say that, politically speaking, the Dayton-Paris agreement establishes two entities (the Bosnia-Herzegovina Federation and the Republic of Srpska). In the first there is a majority of Bosniaks and Croats and in the second a Serbian majority. However, both entities have common institutions. For example, they have a rotating presidency of three members (one Bosniak, one Croatian and one Serbian); they have a bicameral parliamentary assembly, the House of Representatives and the House of Peoples. In the House of Representatives, 28 seats are allocated to the Federation of Bosnia-Herzegovina and 14 seats to the Republic of Sprska; and in the House of Peoples, five seats are allocated to the Bosniaks, five to the Croats and five to the Serbians. All these institutional arrangements speak for themselves about the complex political and social situation that is found in a country in which the Bosniaks represent 48% of the population, Serbians 37% and Croats 14%, according to CIA data.

Serbia is a different case. Putting aside Kosovo

, Serbia itself is quite stable both politically and socially. Socially speaking, and according to CIA data, 85% of the population is Serbian.[43] Politically, the three major parties in Serbia are the Democratic Party (of the current President, Tadic), the Serbian Radical Party and Kostunica’s Democratic Party of Serbia. In the latest legislative election, held in May 2008, Tadic’s coalition won 38% of the votes; the Serbian Radical Party 29% and Kostunica’s party 11%. In principle, the Serbian Radical Party, a nationalist right-wing party, has a more nationalist position than the other two; however, all three oppose Kosovo’s independence.

However, it is clear that to have a full social and political picture of this country we need to take into account Kosovo. Kosovo basically comprises Albanians.[44] Politically speaking it has its own institutions, even though they are supervised, as explained in a different section, by a UN protectorate, the UNMIK. The last parliamentary election, which took place in November 2007, was won by the Kosovo Democratic Party. On the eve of the election, the leader of the Kosovo Democratic Party, Hashim Thaci, declared that he would proclaim Kosovo’s independence ‘immediately after the 10th of December’.[45] It was proclaimed on 17 February 2008.

In the opinion of most analysts (Garrigues, 2007) the way out of this situation would be through an independent Kosovo that would eventually join the EU. However, if this is the case, it is not clear what Serbia’s stance will be. It is very likely that if Serbia wants to join the EU, it will have to accept both an independent and a European Kosovo. However, if both countries finally join the EU, it is not unrealistic to forecast that extremist positions will arise from both sides as regards the issues that concern them.

To sum up, the political situation in all of these countries is characterised by the presence of more or less strong nationalistic parties. The only exception could be Croatia but, as indicated, the major political parties pepper their political discourses with soft nationalist rhetoric. From a social perspective, the situation is relatively stable in only two states, Croatia and Albania. In the rest, ethno-social divisions are a feature, even though the degree of those divisions is greater or lesser depending on the country and on the way that current institutional arrangements are working. Given this situation, it is clear that in these countries there is a breeding ground, at least at present, for the emergence of extremist government positions, and therefore, for the increase of heterogeneity of interests within the EU. Therefore, if, as assumed in this work, Croatia, Macedonia, Albania and Montenegro join the Union in the next 10 years, this variable would be at least at an intermediate-to-high level; if Bosnia Herzegovina and Serbia are added to the list, the variable would definitely have to be placed at a high level.


The delegation of core competences to the EU is our last variable. As regards its functioning we can make at least three points.

The first is that core competences can be delegated to the EU in a number of ways. Of course they can be delegated through the reform of the Treaty; but a reform of the Treaty is not by any means the only way to do it. First, because the EU Treaties have provisions that allow for such delegations to be made.[46] And secondly, because such delegations can take place de facto and if supported by the European Court of Justice, they can be finally institutionalised.[47] Therefore, a first idea would be that new competences can be delegated to the EU without a reform of the EU Treaties.

The second point concerns the Lisbon Treaty. Basically, the Lisbon Treaty has delegated a whole new area of competence that belongs to what I have identified as ‘core sovereignty competences’: I refer to the so-called freedom and justice area. Thus, by communitarising the remaining third pillar (Police and Justice Cooperation in Criminal Matters) it has created a whole new area of community competence that will be called, once (and if) the Treaty is finally enacted, the area of Freedom, Security and Justice. If one takes a look at the cautious wording that the Treaty uses when it drafts this part,[48] one is better able to understand the point I wish to make: that Member States are reluctant to cede core sovereignty competences to the EU and that, once they do it, they establish myriad controls over the management of that policy that bogs down the process of integration.[49] If this assumption is true, one should expect that integration will be delayed in this area as a consequence of the changes introduced therein.

A third point regards the future. It is impossible to know what competences Member States will attempt to delegate or not delegate to the EU in the coming years. But taking into account the enormous delegation effort that has been undertaken in the last 20 years of Community existence, one has to acknowledge that there are not many more areas of competence left out of the influence of the EU. The only core sovereignty competences that remain to be ceded are defence and foreign policy. Although the Lisbon Treaty has introduced changes in both areas, both policies basically remain national policies. If the world’s current geopolitical circumstances do not greatly change in the next 10 years, it is not very risky to forecast that foreign and defence policies will be kept at the national level.

The conclusion as regards this variable is therefore the following. One the one hand, if the Lisbon Treaty is finally enacted, the EU will have at its disposal a full new area of Freedom and Justice but, taking into account the different checks and balances that Member States have introduced here, it will not develop rapidly. On the contrary, it is probable that this area will promote the opening up of new cases of variable geometries. On the other hand, it is not foreseeable that new core sovereignty competences are ceded to the EU in the future. This means that the ‘competences’ variable will probably operate at a low level.

Conclusions: The Main Hypothesis for the Next 10 Years of EU Integration

We can now put together our four variables in order to have a more complete picture of how they would work. The first and most important of our variables, enlargement, could be placed at an intermediate to high level. In principle I have assumed that in the next 10 years the EU would enlarge to four states in the most realist scenario and to six states in the most optimist one. This is not as big as the 2004 enlargement (that concerned 10 new Member States) but it is still not insignificant. In turn, our diversity variable would be expected to be at a high level. Taking into account that the political, legal and economic cultures of the candidate and potential candidate countries are still far from European standards, it is easy to imagine that, once in the EU, these countries would still be notably different from the rest of the Community countries, and above all, from the EU-15. Regarding the expected heterogeneity of interests, enlargement would probably also increase this aspect. As shown in this work, there is a breeding ground in many of the would-be new Member States for the emergence of governments with extremist positions. We would therefore place this variable at a high level also. Finally, regarding the delegation of core competences to the EU, the new area of Freedom, Security and Justice would probably be a further brake on horizontal integration. It is also sensible to think that no new competences will be delegated to the EU in the next 10 years. Therefore, this variable will be probably placed at a low level.

To sum up, the combination of our four variables would be probably placed at an intermediate to high level. If, as I have assumed, the relation between the independent variables and the dependent variable is of a negative kind, this means that the EU’s degree of integration should be placed at an intermediate to low level. If the intermediate level is constituted by the status-quo scenario, then the level of Community integration in the next 10 years would be placed at some point before the status quo scenario. Therefore, the main hypothesis this work puts forward is that in the next 10 years, and for the reasons previously mentioned, the EU’s level of integration will move between the status quo and the variable geometry scenarios (see Figure 5).

Figure 5. A hypothesis for EU integration in 2017

image004 40 1

If this hypothesis proves to be correct, there would be a number of consequences, the first and most important of which would be that the EU would slowly make its way towards more and more variable geometry situations. The second consequence would be that the Lisbon Treaty would be the last reform that the EU and its MemberStates would see in the next 10 years.

It is also important to take into account that there is nothing inexorable in the process of integration. Therefore, a process of de-integration may be the precursor of a boom in further integration. In fact, it is very likely that once the enlargement to new states has been digested, then the Union might experience new momentum. As I have indicated at many points in this paper, the passage of time could have beneficial effects for integration: for example, as regards the ‘diversity’ and ‘heterogeneity of interests’ variables.

There is only one thing that has not been integrated in this analysis and which could affect the hypothesis that I have put forward: political leadership. The de-integration tendency that is predicted here for the next 10 years of Community history could be counterbalanced if the new generation of political leaders that we have today in the main European capitals decided to hold a strong and common European vision. But that, of course, is a different story, a story that needs to be told somewhere else.

Antonio Estella
Jean Monnet Professor of European Union Law, Universidad Carlos III, Madrid 


Borzel & Risse (2000), ‘Who is Afraid of a European Federation? How to Constitutionalise a Multi-Level Governance System’, Harvard Jean Monnet Working Paper, nr 7/00.

Church & Dardanelli (2005), ‘The Dynamics of Confederalism and Federalism: Comparing Switzerland and the EU’, Regional and Federal Studies, vol. 15, nr 2, June, p. 163.

Ehkermann (1995), ‘Increased differentiation or stronger uniformity’ EUI Working Paper RSC, nº 95/21

Estella (2002), The EU Principle of Subsidiarity and its Critique, OxfordUniversity Press, Oxford.

European Commission, Forward Studies Unit (1999), ‘Scenarios Europe 2010. Five Possible Scenarios for Europe’, in Bertrand (Coord.), Michalski & Pench, Working Paper, July.

Fearon (1998), ‘Bargaining, Enforcement and International Cooperation’, International Organization, nr 52, p. 397.

Garrigues (2007), ‘Kosovo: la solución menos mala’, El País, 23/XI/2007.

Golub (2007), ‘Survival Analysis and European Union Decision-making’, European Union Politics, nr 8, p. 155.

Hazareesingh (2002), The Jacobin Legacy in Modern France. Essays in Honour of Vincent Wright, OxfordUniversity Press, Oxford.

Kortenberg (1998), ‘Closer cooperation in the Treaty of Amsterdam’, Common Market Law Review, vol. 35, 4.

Langer (2005), ‘Four Future scenarios for the European Union – Reflections from the Perspective of ‘Path Dependence’, Laboratoire europeen d’anticipation politique, 25/IX/2005,

Louis & Ronse (2005), L’ordre juridique de l’Union européenne, Bruylant, Brussels.

Ludlow (2008), ‘Value, Flexibility and Openness:The Treaty of Rome’s Success in Historical Perspective’, in Baquero & Closa (Eds.), European Integration from Rome to Lisbon: History, Law and Politics, Peter Lang, Brussels, forthcoming.

Mancini (1998), ‘Europe: The Case for Statehood’, European Law Journal, vol. 4, nr 1.

Margolis & Liebowitz (1998), ‘Path-dependence’, The New Palgrave Dictionary of Economics and the Law.

McCormick (2007), Weber, Habermas, and Transformations of the EuropeanState: Constitutional, Social and Supranational Democracy, CambridgeUniversity Press, Cambridge.

Rodin (2006), ‘Croatia’, Blockmans & Lazowski (Eds.), The European Union and Its Neighbours, Asser Instituut, The Hague.

Ruiz Jiménez & Torreblanca (2007), ‘European Public Opinion and Turkey’s Accession Making Sense of Arguments For and Against’, Real Instituto Elcano, Working Paper nº 20/2007.

Shaw (2003), ‘Flexibility in a Reorganized and Simplified Treaty’, Common Market Law Review, vol. 40, p. 279.

Schneider (2002), ‘A Never Ending Success Story? The Dynamics of Widening and Deepening European Integration’, in Steunenberg (Ed.), Widening the EU: The Politics of Institutional Change and Reform, chap. 10.

Toth (1998), ‘The Legal Effects of the Protocols Relating to the United Kingdom, Ireland and Denmark’, in Heukels, Blokker & Brus (Eds.), The European Union after Amsterdam, Kluwer, The Hague.

Thym (2006), ‘The Political Character of Supranational Differentiation’, European Law Review, nr 31 (6), p. 781-799.

Weiler (1991), ‘The Transformation of Europe’, Yale Law Journal, vol. 100, p. 2403.

Weiler (1998), ‘Europe: The Case Against the Case for Statehood’, European Law Journal, vol. 4, nr 1.

Zielonka (2006), Europe as Empire. The Nature of the Enlarged European Union, OxfordUniversity Press, Oxford.

(*) This paper has been funded by the BBVA’s Servicio de Estudios Económicos. I would like to thank Joaquin Vidal, David Mathieson, Miguel Jiménez and the participants of the seminar on ‘The Future of Europe’, that took place on 29 February 2008 in Madrid, for their support in preparing this paper and their comments at the seminar. The usual disclaimer applies.

[1] United Nations, Treaty Series, vol. 1155, p. 331.

[2] Article 54 b) of the Vienna Convention.

[3] Article 62 of the Vienna Convention.

[4] See article 66 of the Vienna Convention.

[5] Article 70 of the Vienna Convention.

[6] This argument is implicit in Weiler (1991).

[7] ‘Kortenberg’ (1998) (an alias of the author, a Community official wishing to remain anonymous) in fact seems to differentiate between variable geometries and international agreements adopted by Member States outside the Community institutional framework: ‘[If the solution of closer cooperation] had not been found, there was a serious risk of further instances of cooperation along the Schengen model, outside the framework of the Community treaties; and if these examples of cooperation became numerous, there was a risk that a schism would progressively emerge in the Community’

[8] However, as Thym remarks, the specific ‘reinforced cooperation’ mechanism introduced by the Treaty of Amsterdam has never been used so far. See Thym (2006). However, Spain and other MemberStates have recently proposed the adoption of a reinforced cooperation in the area of the law applicable to divorce. See El País, 26/VII/2008.

[9] I define later on what I understand by the expression ‘Community method’.

[10] See art. 43 e) of the European Union Treaty.

[11] In fact, article 43 a) of the European Union Treaty reads as follows: ‘Enhanced cooperation (…) is aimed (..) at reinforcing [the] process of integration’.

[12] Both disputes are reflected in Estella (2002).

[13] See point I.3 of the ICG 2007 Mandate, Council of the European Union, 26/VI/2007, 11218/07.

[14] See Declaration nr 17 of the Reform Treaty.

[15] See new Title IV of the Treaty on the Functioning of the European Union.

[16] See for instance article 69 B 3º of the Treaty on the Functioning of the European Union.

[17] See new Title V of the Treaty on European Union.

[18] At present, the EU Treaty establishes a threshold of eight Member States. However, if the EU enlarges to four or six new Member States, the threshold of nine Member States would in fact ease the possibility of making reinforced cooperations even when compared to what the EU Treaty currently establishes.

[19] Mancini (1998).

[20] The literature depicting the EU as an Empire is expanding. See in particular Zielonka (2006).

[21]Weiler (1998).

[22]Borzel and Risse (2000).

[23] See in this connection the insightful discussion about the concept of the modern state in McCormick (2007).

[24] According to the Additional protocol to the Ankara Agreement, Turkey had to extend the latter to the 10 Member States that entered the EU in May 2004, including Cyprus. More in particular, and according to the Council Decision of 23 January 2006, Turkey had to remove all restrictions to the use of its ports and airports by Cyprus vessels and airplanes. In November 2006, the European Commission confirmed that this had not been the case. This prompted the partial freezing of Turkey’s accession negotiations. See Commission Progress Report on Turkey of 6 November 2007 (SEC (2007) 1436).

[25] Communication from the Commission to the Council and the European Parliament ‘Enlargement Strategy and main challenges 2007-2008’ of 6 November 2007. COM (2007) 663.

[26] See Sarkozy’s proposal on Turkey at

[27] French Constitution, art. 88.5º: ‘Tout projet de loi autorisant la ratification d’un traité relatif à l’adhésion d’un État à l’Union européenne et aux Communautés européennes est soumis au référendum par le Président de la République’.

[28] Croatia 2007 Progress Report, SEC (2007) 1431, Brussels, 6/XI/2007.

[29] ‘Enlargement Strategy and Main Challenges 2007-2008’ Communication from the Commission to the European Parliament and the Council, COM (2007) 663 final, Brussels, 6/XI/2007.

[30] The FormerRepublic of Macedonia 2007 Progress Report, SEC (2007) 1432, Brussels, 6/XI/2007

[31] See Albania 2007 Progress Report, SEC (2007) 1429, Brussels, 6/XI/2007.

[32] Bosnia-Herzegovina 2007 Progress Report, SEC (2007) 1430, Brussels, 6/XI/2007.

[33] See Serbia 2007 Progress Report, SEC (2007) 1435, Brussels, 6/XI/2007. See also Kosovo 2007 Progress Report, SEC (2007) 1433, Brussels, 6/XI/2007.


[35] Enlargement to Bulgaria and Romania would take longer: these countries joined the EU on 1 January 2007.

[36] See Serbia 2007 Progress Report, p. 6.

[37] Croatian Democratic Assembly of Slavonia and Baranja.

[38] See CIA, ‘World Factbook: Croatia’ at

[39] Internal Macedonian Revolutionary Organization – Democratic Party for Macedonian National Unity.

[40] Social Democratic Union of Macedonia.

[41] According to the Commission: ‘(…) integration of ethnic minorities is quite limited. Most of them remain at a disadvantage with respect to education and employment. The increase in representation of the non-majority communities remains uneven among individual ministries. The number of non-majority communities’ members in the army remains low. Slow progress has been made within the police overall, and none in senior ranks, in particular within the criminal police and the department for security and counter-intelligence The strategy for equitable representation is not yet fully satisfactory, notably in terms of its targets and sanctions. Committees for interethnic relations have not been set up in all the municipalities concerned and are often not effective. Sustained commitment to implementing the regulatory framework for use of minority languages is required. There are not enough qualified interpreters for fully effective implementation of the law on use of minority languages in criminal proceedings and in local and central government’. Commission Progress Report on Macedonia, p. 15-16.

[42] See CIA ‘World Factbook’ at

[43] See ‘CIA: World Factbook’ at

[44] According to CIA estimates, they represent around 1.8% of Serbia’s population.

[45] See Agence France-Presse, 16/XI/2007, at The troika’s negotiation deadline was 10 December 2007.

[46] Art. 308 of the EC Treaty.

[47] The ECJ has done this on a number of occasions. See, for example, its ERTA saga case law. For extensive commentary on this see Louis and Ronse (2005).

[48] See for example new article 61 F of the Treaty on the Functioning of the EU: ‘It shall be open to Member States to organise between themselves and under their responsibility such forms of cooperation and coordination as they deem appropriate between the competent departments of their administrations responsible for safeguarding national security’.

[49] A good example of this is found in article 61 E of the Treaty on the Functioning of the EU: ‘This Title shall not affect the exercise of the responsibilities incumbent upon Member States with regard to the maintenance of law and order and the safeguarding of internal security’