ARENA Working Papers
WP 99/12



European Integration and the Changing Paradigm of Energy Policy

The case of natural gas liberalisation

Professor Svein S. Andersen
ARENA, University of Oslo

Energy policy and EU integration [1]

Despite its weak institutional core (Kohler-Koch 1996, Kohler and Eising 1999) the EU has a substantial ability to initiate and carry out reforms and transformations. The focus for this article is the development towards a common EU energy policy, with an emphasis on the transformation of EU natural gas policy (and markets) over the last decade. The objective is to develop an neo-institutional sociological perspective to better understand such transformations. The article emphasises the relationship between key theoretical ideas from neo-sociological institutional theory and their implications for transformations of policies and institutions. It is not a policy analysis in a conventional sense, but rather an attempt to link Europeanisation of policy to European integration. While such issues are often discussed separately, this linkage has often been a concern for students of EU energy policy (George 1985, Matlary 1997, Usherwood 1998). The EU energy policy is in many ways a problematic case, but also characteristic of the general EU development, with its periods of stagnation and conflict – and the eventual break through for quite radical transformations during the 1990s.

Up to the 1980s EU energy policy was regarded as a spectacular failure (George 1985:100). It has been characterised by strong conflicts between a common policy, on the one hand, and divergent national policies, on the other. Despite its inclusion in the very first treaties on the European Community energy was until very recently more or less unaffected by common market legislation, and few energy policy decision were taken at the central EU level (Schmitter 1996). A number of EU policies and regulation affects energy production, transportation, sales and consumption in many ways (general rules as well as more targeted ones, like taxes on mineral oil). Still, the EU is far from a common energy policy. However, since the late 1980s a number of important EU initiatives have been taken to strengthen the supernational influence over particular energy sectors (Matlary 1991, Lyons 1992, 1994, Padgett 1992). Results have been mixed, but in some areas EU level policies and regulation are emerging. Such changes are in line with, but unique versions of, global trends towards liberalisation. In the EU they take place within a special institutional context which at the same time reinforce, elaborate and modify such external influences.

Europeanisation as institutionalisation of a policy paradigm

The empirical focus is the emergence of a common EU energy policy during the 1990s, but not as energy policy in a traditional sense. Energy issues have been redefined and increasingly linked to three general perspectives. First, the EU's internal market programme where competition policy plays a major role. Second, EU's attempts to establish a common environmental policy with fiscal measures as key instrument. Third, the European Energy Charter and the Charter Treaty which were attempts by EU to create international market regimes that could support reform in the former East Bloc and thereby secure EU's energy supplies (Andersen 1995, Dore and DeBauw 1995). The most important changes in the last few years are related to the internal market directives on the electricity and gas markets (Hancher 1998).

The theoretical perspective is that of neo-institutional sociology. Europeanisation of policy is discussed as the institutionalisation of new policy paradigms at the EU level. Europeanisation of policy is not only a question of moving decision-making authority across levels, but also tied to the re-institutionalisation of policy-paradigms on both the EU and national level (Andersen 1995). Such paradigms represent the institutionalised principles of policy action. They structure the way in which policy-makers see the world and their role within it. Policy paradigms consist of both practises, in the form of policies, and means-ends designations, in the form of tenets of governmental action. As cultural frames they create and sustain certain patterns of action, and they symbolise those patterns as efficient (Hall 1992, in Dobbin 1994:19, Andersen 1993). The concept of policy paradigm challenges a neo-realist perspective, linking the general perspective of neo-sociological institutional theory (Meyer 1994, Meyer et al 1987, 1997, Scott 1994) to the study of policy sectors. This relationship will be further developed below.

In the area of energy successes reflect the `pull' of increased EU integration despite diverse and conflicting interests, to be exploited by political entrepreneurs, primarily the Commission (Nylander 2000). These changes involve three stages that are closely linked. First, the introduction and legitimisation of an alternative EU-level model or frame, where arguments about appropriate organisational principles and form play a key role. Second, the gradual specification of the content of this frame, towards to the formulation of an EU level policy paradigm where implications of alternative organisational forms are debated and negotiated. Third, the national implementation and modification which further give substance to the new EU level paradigm. The relationship between such processes may very; in terms of being sequential or parallel, and with respect to the type of interaction that takes place between them. A key point here is that this sequence differs from what would be expected within a neo-intergovernmentalist model.

A widespread assumption in the EU literature is that the emergence of common policies reflect member states' preferences in the context of wider international pressures that undermine existing national models:

In a strict version of the neo-intergovernmentalist argument, states act on their preferences and as a result new normative models emerge and become codified in EU legislation. States initiate and control processes as well as outcomes. In another version states will develop and accept unplanned and unintended consequences of earlier commitments, even if it means changing earlier preferences, as long as domestic preference changes and EU level outcomes are controlled by member states (Moravcsik 1993, 1998). A third modified version of neo-intergovernmentalism is that states will accept initiatives from supernational institutions developing unforeseen consequences of established policies and legislation in the form of agenda setting and proposals by the Commission, the European Parliament or through court decisions (Matlary 1997, Peterson 1995, Tallberg 2000). In the latter version it is not clear that the states control the EU level process, or that changing state preferences are exogenous to the EU process. However, it is assumed that supernational initiatives can be accepted even in cases where sector effects are negative, as long as overall member state gains are higher than losses.

In all three types of argument, EU policy and legislation are regarded as projection of state interests, and their legitimacy rests on a cost-benefit analysis. All three versions of the neo-intergovernmentalist argument have clear empirical implication for sector studies. The first assumes that observed state interests in the sector explain processes and outcomes. The second version is more difficult to verify empirically since it presumably involves complex weighing of interests, transcending the sector, which may be difficult to observe and assess. The third version differ from the other two with respect to how Europeanisation is brought about by allowing a certain autonomy to EU level process, including state preferences being influenced by the EU level process, although outcomes are assumed to be restricted by state interests. However, it is difficult to verify complex cost-benefit analyses in cases where reluctant acceptance of sector changes are weighed against complex and long term interests in the EU project.

In the case of EU energy policy a number of studies suggest that the first and second version of the neo-intergovernmentalist models are not in accordance with data (Andersen 1995, Matlary 1997, Usherwood 1998, Eldevik 1999). The legitimisation of a new EU level frame takes place despite widespread and intense opposition from a majority of member states. Eventually the formulation of state (as well as industry) preferences reflected the new frame. This happened only when the new frame was established, and initially arguments about appropriate forms of organisation seemed to weigh as much as, or more than, calculations of costs and benefits. Only gradually state (but also industry and court) actions can be seen as shaping and elaborating the emerging policy paradigm.

The ambition here is not to disprove empirically a basic intergovernmental argument about why certain initiatives or outcomes prevail despite initial strong opposition from member states. Verification and falsification face the same problems of empirical measurement. Instead the discussion focuses on how transformation of policy takes place. The argument is that the intergovermentalist framework is weak with respect to three aspects of the transformation that has taken place in EU energy policy. This is particularly the case for two versions of intergovernmentalism which assume that state interestss are driving and controlling the process. The third version provides a certain role for supernational actors and processes, but even this argument has serious limitations when analysing EU transformations.

First, with its emphasis on local problem solving the intergovernmentalist perspectives little to say about the relationship between Europeanisation of policy and wider international trends. Second, it is not clear how diverse state interests, articulated within different policy paradigms, can be integrated within a common EU model at all. Third, when EU level policies emerge, what are implications for convergence and divergence of institutional arrangements in member states? These questions are central within a neo-sociological perspective, underlining the legitimisation of authoritative frames as a precondition for rational action as well as institutional transformation (Meyer 1994, Andersen 1993, Dobbin 1994).

The theoretical implications of an neo-institutional perspective will be discussed in relation to the EU liberalisation of the natural gas sector over the last decade. It is a case where the Europeanisation of policy has taken place despite initial intense opposition from a majority of countries:

The development of the gas directive can be summarised as follows: The first initiatives (1990) pushed by the Commission only had support from the UK. The other 11 members (at the time) of the Council of Ministers reacted strongly. It is one of the few examples where the Council has refused to handle a Commission proposal according to the (two-hearing, qualified majority) procedure. In 1994 the initial directive draft failed. Despite this initial reaction the issue was kept alive and over time a new directive was developed and eventually passed in 1998. Gradually the strategic reorientations of states and companies as well as the impact of court decisions came to play a key role in the further elaboration of the new frame.

As a new EU-level policy paradigm emerged, the traditional national paradigms were radically transformed in the sense that traditional statist models of energy policy were replaced by a model emphasising market functionality. And, as in most cases of EU energy policy, the initial distribution of (national and industry) interests were such that this transformation cannot be explained without reference to the broader context of European integration. It reflects the weight of the general internal market (where energy initially was not included), but also global liberalisation trends affecting the energy sector. The new directive has already led to important changes and made it impossible to preserve other parts of (traditional) national energy markets in the long run. Such changes have increasingly been reflected in new strategies both from (often reluctant) member states and parts of the industry.

This process and its outcome raise several questions of general interests:

  1. What is the relationship between the external influence of `global' models on each country, and to what extent does the EU represent institutional form which reinforce, channel and modify such impulses through the construction of new policy paradigms?
  2. What are the mechanisms through which integration pressures of the EU context affect energy? How are new policy paradigms developed?
  3. What is the relationship between EU level and national level changes? To what extent, and why; do we find convergence and divergence; i.e. extent to which a general and idealised model is reflected at the central EU level or in national arrangements?

Before we enter the discussion of the natural gas directive and its implications, we will briefly discuss attempts to explain energy policy failures and successes and the relation to general EU integration.

Policy failures and problems of explanations

Various attempts have been made to explain Europeanisation of energy policy, or the lack of it, within general theories about EU development. Here we will crudely summarise a development divided into three stages: The first, up to the early 1990s, is characterised by attempt to explain by way of two competing theoretical perspectives, namely traditional intergovernmentalism and integration theory. The second stage, until the mid/ late 1990s, is characterised by attempts to combine elements from the neo-intergovernmentalist and neo-integrationist thinking. The third stage, which has developed over the last years, is an attempt to analyse the EU on the basis of institutionalist approaches, but so far little has been done along such lines in relation to energy policy. The purpose of this paper is to contribute within this third group of explanations; to apply neo-institutionalist sociological theory as an alternative framework for analysing Europeanisation of policy.

Historically, intergovernmentalism explains the development towards more EU integration (supernational authority) and common policy by pointing to the gradual emergence of common and overlapping interests among member states. This process is driven by rational actors building coalitions (Morgenthau 1973, Waltz 1979). Such a rational actor perspective would be consistent with the co-operation over coal (and steel) policy during the post World War II years. The Six (who had lost the war) saw co-operation on coal as a requisite for successful reconstruction, but also in the new spirit of a unified Europe. Within traditional integration theory the successful co-operation over coal was seen as a driving force of further EU-co-operation; first by spilling over to other energy sectors and later to other economic sectors (George 1985). The latter prediction was sadly wrong.

The lack of a common EU policy in the energy sector up to the late 1980s is consistent with a lack of overlapping interests between member countries (and national energy sectors). From the late 1950s to the early 1970s such co-operation did not seem necessary. During the 1970s national interests were too diverse. However, failures and successes of EU energy policy pose asymmetric problems of explanations. It is easier to explain why diverse and conflicting self-interests prevent common policies than to explain how a common denominator emerges (Scharpf 1988). This became increasingly clear as new initiatives emerged in EU during the 1980s.

New formulations of intergovernmentalism emphasise how EU institutions reduce transaction costs of formulating common policy, and that preferences may be altered through domestic processes that are exogenous to EU level processes (Moravcsik 1993, 1998). This has become the major perspective on EU integration. However, the focus is mainly on EU's treaty changes, and it can not account for the considerable variations observed across policy sectors. In the early and mid 1990s the question was often how and to what extent new energy sector initiatives could be accounted for by either neo-intergovernmentalism or neo-integration theory, respectively. As it turned out, the support for one or the other model was dependent on two factors; the time period studied and the nature of the case at hand.

The general finding was that EU energy policy was a failure, and that the major action was to be found at the national level (Schmitter 1996, Matlary 1996). How, then, to explain (some) successful EU energy policy initiatives since 1990? They have been pushed by the Commission which has exploited institutional rules on the EU level to take the initiative, and to redefine the energy sector in relation to the internal market, environmental policy and foreign policy. The areas to which energy is linked all represent areas where supernational institutions have been assigned tasks what the neo-functionalists call inherently expansive (Cameron 1991: 25). However, rather than Europeanisation of energy policy driving EU, it was the other way around. This is perhaps most obvious in the case of the internal marked and in environmental policy. The radical transformation that has taken place in the natural gas sector over the last decade is an example of an issue being redefined in terms of the internal market logic.

A key to understanding this development is how the EU – as an institutional framework and a set of central institutional actors – interact with underlying interests in the energy sector in the different stages of EU development. This interaction demonstrates the close link between EU decision-making, the problem of defining collective interests, on the one hand, and the dynamics of EU integration, on the other hand. From the mid 1990s this question was often posed more pragmatically; i.e. how specific cases or aspects of the EU development could be analysed as a combination of intergovernmentalist or integrationist mechanisms. Peterson (1995) and Peterson and Bomberg (1999) introduced a general multi-level model of EU policy-making to account for various types of decision-making, combing the two models. Intergovernmentalism, it was argued, is best suited to understand history-making decisions leading to treaty reforms. Such decisions frame EUs everyday policy-making where the central EU institutions as well as the expert level have a special significance, with a mix of intergovernmental and supranational influences.

Often decision-making within the EU is characterised by considerable political entrepreneurship exploiting the established EU institutional framework, leading to unexpected processes and outcomes. Such entrepreneurship is primarily exercised by the Commission, but also other actors may engage in this (Nylander 2000). Studies of energy policy in general, and the natural gas sector in particular, typically concludes that such entrepreneurship play a key role in explaining the successes of the 1990s (Andersen 1995, Schmidt 1997, Matlary 1997, Eldvik 1999, Nylander 2000). It is generally concluded that processes and outcomes are not consistent with a strict neo-intergovernmentalist model. Most member states are initially against initiatives, they do not dominate the policy-making process and preferences seem to be changing as part of the EU level bargaining process (in contrast to exogenous domestic preference changes assumed by the intergovernmental model).

The conclusion of such studies is that there are integrating pressures at work and that the EU process and the Commission seem to have a certain autonomous influence (Schmidt 1997, Matlary 1998). The sources of this autonomy are existing EU law and obligations which serves as a context for legitimate political entrepreneurship. Schmidt (1997), comparing Europeanisation electricity and telecommunication policies argue that the mix of intergovernmental and supranational factors varies with the distribution of, and relationship between, mobilised interests. It is not always clear how such observations can be reconciled with the idea that the EU is primarily an intergovermental undertaking. However, the assumption seems to be that lack of a positive shared sector interest may be overruled by more general member state commitments to the internal market and to the EU integration process.

Accepting Commission initiatives and adjusting to them may be seen as part of a learning process where preferences are adjusted, as side payments in nested or even secret games. Presumably, the costs of not accepting a certain sector policy, although not in ones interest, would be greater since it could harm a more general and long term interest. If the member states were really against it, they would revoke earlier commitments, the argument goes. Legal rules project state interests and legitimacy of Commission entrepreneurship in the final analysis rests on a cost-benefit analysis.

This line of argument has some merit. From an empirical perspective, however, it raises the problem of how state interest may explain process and outcomes even when the expressed sector interests are the exact opposite, and when `broader underlying' interests may be unobservable except as a post hoc rationalisation. Theoretically, the explanatory value of given interests is problematic when they are changed by external pressures and the momentum of the EU process. On the other hand, such a fundamental intergovernmental model is hard to refute on the basis of specific empirical observations. Although, assumptions about rationality are strict, almost all decision may be rationalised along such lines post hoc. This kind of problem has been discussed by Pierson (1996) in his version of historical institutionalism.

The concern here is to better understand Europeanisation of policy as transformation processes. This points to another class of theoretical models; namely institutional approaches to the study of the EU which lately have become popular (Andersen 2000). Below we will briefly discuss major perspectives and introduce a neo-institutional sociological approach to the study of Europeanisation.

An alternative institutional approach

Since the mid 1980s a growing number of researchers argue that the best starting point for understanding EU integration and Europeanisation of policy-making can be found in various institutional approaches (Andersen 2000). The most common ones are variations of so-called historical institutionalism; where (bounded) rational choice is framed by institutional rules. Mechanisms mediating between context and actor choices may vary. One major type of explanations refers to various institutional rules framing rational action (Bulmer 1994). Another type of explanations emphasises the impact of accumulated unintended consequences of earlier decisions and commitments that frame new choices (Pierson 1996). In both cases changes reflect attempts to accommodate self interests in a local problem-solving context. As such they may still be regarded as variations of rational actor arguments.

Few, if any, have explicitly studied EU energy policy from such perspectives. It would seem that they would face some of the same challenges as an intergovernmental model in that they rely on (bounded) rational action to explain change and transformation. Interest is still the primary integrating mechanism. On the other hand, they would open up for other actors playing a key role, and that the role of member state governments has to be determined empirically. There are less strict assumptions about efficiency gains, particularly among those historical institutionalists that regard interest articulation and action as reflecting a web of institutional rules (Bulmer 1994, Bulmer and Bursch 1998).

The transformation of EU energy policy can be regarded as the parallel re-institutionalisation of policy-paradigms at the EU and at the national level. This transformation is not only a matter of changing policy content, but also part of a wider EU integration process. The question is how such transformations should be conceptualised, and if there are alternative ways to think about integration mechanisms than those found in intergovernmentalist and integration theory:

For intergovernmentalists, as historical institutionalists, integration is mainly based on an accommodation of self-interests and this is reflected in institutional change. In integration theory, in contrast, the concept of spill-over (economic and politic) underline the impact of the institutional context. There is a dynamic relationship between the level of EU integration and national formulations of interests, and the EU elites play a key role in identifying positive gains and new collective formulations of policy. The efficiency of EU level solutions undermines the sub-optimal nation-state arrangements. In both intergovernmentalist and integration theory the external economic environment is an important driving force for integration, and responses reflect local factors. Local solutions are either the aggregate of individual state interests or the systematic response by (entrepreneurial) EU elites.

From the alternative perspective of neo-institutional sociology EU integration may be viewed as part of a wider `modernisation process'. The concept of institution is used here simply to refer to a location of meaning and identity outside the social system under examination (Meyer 2000:3). The argument is that certain forms of organisation and arguments spread from a global rationalised culture; characterised by a set of abstract and universalistic principles, norms and cognitive frames that drives standardisation in various areas. Global models may be more or less developed, more or less contested. Increasingly we find models that reflect international experiences, codified in scientific and legal language.

International organisations, like OECD and GATT in the economic area, play a key role in developing such models which become autorititative frames. Regional and local institution building and change increasingly relate to and reflect such models. In this way global models shape economic and political responses to challenges in the international environment. Arguments about efficiency and functionality will be embedded in a broader cultural context where legitimacy or appropriateness becomes an important element. When facing questions of institutional change involving different types of policy paradigms it is difficult to calculate and evaluate efficiency. Authoritative reasons outweigh means-end calculation.

Dynamics is related to the relationship between global rationalised models, on the one hand, and local and particular models, on the other hand. External legitimacy is not only linked to economic performance pressures, but institutional models heavily vested in cultural forms. Solutions are not local constructions, but variations over global trends and themes that facilitate and shape changes. The argument is that the impact of these particular cultural forms helps account for two aspects of EU transformations. On the one hand it bridges the gap between transformational capacity and a weak organisational core. On the other hand, it helps understand why EU level outcomes seem to reflect global trends, expressed in terms of more universalistic (rather than national particularistic models).

The spread of global culture is often supposed to follow a diffusion pattern. An important modification, which is sometimes stressed, is that the degree of resonance between global and local models is an important precondition; i. diffusion is not only dependent upon interaction patterns. External legitimacy is an independent explanatory mechanism. On the other hand, research within an neo-institutional sociology perspective has also produced a number of studies that emphasise the role of national paradigms in the Western world, as specific operationalisations of a general and rationalised worldview and principles. Such studies stress that local models represents socially constructed rationality, sometimes discussed as policy paradigms. National paradigms may adjust to changes, but their basic assumptions and structure seem suprisingly stable over long periods of time (Andersen 1993, Dobbin 1994). In such studies the internal legitimacy of these paradigms is the key factor which explain continuity. How such stable pattern is reproduced in the face of global pressures is not clear. Presumably, the degree of perceived (in)efficiency and the resonance between global and local models plays a key role.

In its standard formulation neo-institutional sociological theory has little to say about the role of the EU as an independent institutions between the global trends and national adaptation. The mechanisms for change are assumed to work directly on the national level, so that the EU is explained as a variation over global modernisation processes where the nation state is the dominant actor. Here there is a parallel to new intergovernmentalism that views the EU as primarily transnational in character, but where ordinary diplomacy can produce extraordinary results. In neo-institutional sociology the EU is regarded as an `especially intense form of elaborating the global system' (Meyer 2000: 14). At most it would seem to synchronise national changes. This perspective is also reflected in attempts to introduce elements of such thinking in the international regime literature (Finnemore 1996). Collective international actors are only `non-state' actors vested with cognitive and normative authority – `otherhood' reflecting, focusing, elaborating and emitting authoritative `globalised' institutional models.

The idea here is that EU policies reflects external economic and institutional pressures, and that solutions cannot be understood solely as local constructions. Like nation states, the EU to a surprisingly high degree reflects global institutional models and themes. However, the argument presented here differ from the standard presentation of neo-institutional sociological theory in several ways:

First, the EU member countries represent historical traditions consistent with key elements in global rationalised culture and therefore have accumulated experience and competence in the form of embedded political, organisational and legal concepts and rules. Such characteristics make each country perceptive to, and able to, influence external change impulses. At the same time such elements may also serve as the common denominator that makes it possible to find collective solutions in situations characterised by diverse interests and policy paradigms.

Second, the EU is seen as a collective institution that, over time, has acquired characteristics which tend to reinforce elements of a rationalistic global culture. The institutional set-up and political culture emphasise rational and universalistic arguments even more than in international organisations. At the same time co-operation is comprehensive and codified in an accumulated set of legally binding rules. This creates a context of legitimate discourse where legitimacy often becomes a question of legality.

Such factors play a key role when new policy paradigms are developed in the EU. Sometimes the neo-institutionalist perspective is given a rather deterministic and uni-directional interpretation. Here, in contrast, the emphasis is on the interaction between global models and regional and local institutionalisation of policy paradigms. Global models may be more or less developed. They represent an external pressure for change, but how and to what extent depends on a number of factors. They may serve as sources of `concepts of control' (Fligstein 1996), but there is considerable room for modification and innovation during regional and local institutionalisation processes.

Fligstein (1996) discusses the creation of the internal market programme as the construction of a market. He argues that the concept of the internal market served a `concept of control', bringing together a number of ideas and proposals in a way which creates commitment to the wider EU integration project. However, he says little about how this general frame is translated into specific sector legislation. This is the focus here, and we will pursue this though the example of EU natural gas policy. In discussing the transformation that has taken place over the last decade it is useful to distinguish between three aspects of this change:

The first concerns the introduction and legitimisation of a new policy frame, as an alternative way to think about energy issues. This involves the interaction between global, the EU and national level. The second aspect is how the emerging policy paradigm on the EU level influences strategic orientation and action between government, in the court system and the industry. The third has to do with how and what ways the new European policy paradigms, as expressed in a new directive, is implemented in a national context.

Changes in EU energy policy – and the increasing integration of this policy area in the EU system - reflect that new global models of economic organisation emerge. Such emerging models affect both the EU and the individual countries directly. The general hypothesis is that, everything equal, the external global models will spread more easily in situations where there is high degree of resonance with regional and national policy paradigms. However, when looking at particular cases this hypothesis will be modified by a several other factors. Developed Western countries are more likely to be able to uphold their own models, reflecting core principles of rationalised culture, than countries, which lack this tradition. This is more likely for big powerful countries – or influential groups of countries – than for smaller countries. Such factors may influence the willingness of member countries to engage in reform processes. The collective aspect of EU transformation also add two other factor; namely the political weight of `movers' versus `resisters' in the sector and the impact of the wider EU-integration context.

Natural gas - the introduction of a new frame

Within a neo-intergovernmentalist perspective changes may be externally driven, but external forces are mainly linked to competitive pressures or demands on efficiency. In a strict version regional and local changes are down-up processes where state interests drive and control the process, and where institutional models reflect local problem solving efforts. This has not been the case in the energy sector, or in the particular case of natural gas reform. Commission initiatives were taken despite the expressed opposition of almost all member states. These states reacted strongly in the Council of Ministers, finding the Commission initiative unacceptable. This would appear to be in line with softer version of neo-intergovernmentalism which assumes member states' control with the process. However, the initiative was kept alive, despite member states' strong opposition. How can this be interpreted?

The external pressures motivating the Commission proposal were not related to competitive pressures undermining existing national arrangements. The introduction of the proposal must be viewed in the context of the internal market, but the content of the proposal primarily reflected the growing support for global institutional models emphasising market organisation of the energy sectors. The model for a reorganised gas market was inspired by an international normative model, most clearly expressed in the US system with its emphasis on an active competition authority. The initial content of the directive proposal did not reflect local performance problems, but a general model emphasising market functionality and a strong competition authority.

Also, member state opposition was expressed in a way which question the assumption about member states as unitary actors: `(S)o successfully did the Continental European gas industries enlist the support of their governments, that nearly a decade passed before an EU directive dealing with the opening up of gas industries to competition and liberalised access to networks was agreed' (Stern 1998: vii). The confrontation that resulted was characterised by a mismatch. The alternative model was argued from universal principles of market functionality. The counter arguments were based on alternative ideas about appropriateness and the acceptable performance of existing national systems, but also the possible damaging consequences of the proposal for existing national models. In the first period, the confrontation can be seen as mainly being a struggle over the legitimacy of competing models of industrial organisation.

For a long time the energy sector was even in the Western world influenced by strong state regulation and hierarchy, in the planned economies it represented the core. Sometime during the 1980s the dominant global model became contested. This happened partly through the general change towards a market model. This led to far-reaching changes in the US, but also other countries, most notable the UK. In both the US and in the UK, energy had been heavily regulated for decades (Tugwell 1988), and liberalisation of the energy sectors followed a political break through for `market ideology'. The sources of this change are outside the scope of this paper (but see Hall 1989, Haas 1992). However, as a result an international model of energy policy has emerged emphasising a more active use of markets, with less political interference.

The change towards a new global economic policy model for energy reflected the general political climate, but also changes in the international supply situation which had been `normalised' after more than a decade of turbulence. The change in the dominant political-economic perspective on energy was partly formalised by the Energy Charter Treaty during the early 1990s. This treaty brought together the former Soviet and its satellites and the OECD countries. In the EU such changes were first reflected in the UK during the mid 1980s. However, reforms, most importantly as privatisations, were not primarily driven by new theorising about market functionality. They were motivated and legitimated by a revival of liberal ideology in economic policy (`Thatcherism') rather than the failures of the old system (Andersen 1993). At the EU level pressures for change were picked up and reinforced by the Commission and the British government, as part of the internal market programme. In other words the changes in the energy sectors mimicked the general changes towards market based solutions, as part of a broader trend.

There were, however, some special characteristics of the energy sectors in EU countries that complicated the further transformation. They represented historical experiences and perceptions of market failures as well as political objectives:

There were real reasons why the energy sector had been organised with clear elements of state control and hierarchy. Important energy resources were concentrated in a few countries outside the EU and this had led to concern for supply and price stability. Also, there was a strong tendency in European countries to view the energy sector as more than just an economic sector.

Another reason was that state interests were heavily biased by the fact that energy policy in every sector was dominated by monopolistic companies (often state owned) and regulatory authorities which could articulate clear and strong interests within established national paradigms. The preceding decade of turbulence in the oil market has reinforced this.

Estrada et al (1988) provided an authoritative overview over the West-European gas market by the late 1980s. Their conclusion was that the impressive `existing structure.. is well founded in a complex network of economic and political interests' (pp. 254). Discussing the potential for transformation in the 1990s they conclude that: `It does take a lot of imagination to foresee the present structure being undermined' (i.e. liberalised). Political `undercurrents have to grow to a wave of political determination and a forceful coalition of economic interests to become a serious challenge for the status quo' (pp. 259).

For such, and other, reasons it would take some time before EU's internal market philosophy could be transformed into an operational policy paradigm on the EU level. The energy sector was deliberately kept out of the first round of the internal market `package'. However, it was never the Commission's intention to make a permanent exception. That would violate the general pattern of economic reform, and besides the Commission's general argument that increased EU-wide competition would improve global competitiveness was assumed to hold also for energy. When energy was introduced a few years later it was thought that it would be possible to achieve success despite strong national resistance, precisely because of the momentum of the wider integration process.

The natural gas sector was relatively new and it had expanded during the 1980s, almost as a `counter culture' to the internal market programme, characterised by high degree of regulation and oligopolitics sector organisation. The introduction of a new policy frame, as a directive proposal on the liberalisation of EU natural gas markets, challenged key elements the existing system. The transmission companies, linking major sellers and buyers had an especially strong position. They would use their monopoly to serve as an intermediary; buying from a few producers and selling to national monopolies, with high profits. As a consequence there was no gas-to-gas competition. Natural gas was priced according to competing energy, most often oil. The benefit of the system was that long-term contracts made it possible to secure the financing of gas field and transportation development, and thereby security of supplies.

The so-called gas directive proposal was part of a package. Other directives concerned price transparency for electricity and gas contracts and transit rights for electricity and gas. The so-called gas directive proposal aimed at liberalising the gas market by attacking the present structure. The proposal covered four major areas:

One major issue concerned market opening; i.e. how much of the gas market should be opened for competition, and who should have the right to send gas through transmission lines. The latter had so far been the exclusive right of the pipeline owners; the proposal introduced the concept of regulated third party access with a EU-level regulatory authority. This became the major source of controversy. The second major issue concerned the impact of liberalisation on long term contrasts (take-or pay) which traditionally had been a key instrument to secure investments and long term supplies. The third issue was whether it was possible to impose special obligations (equal price, supply security, environmental protection etc.) on suppliers all over Europe. The fourth issue concerned the need to establish separate markets around various functions in the gas market which had before been `bundled' by monopolies.

The Commission proposal reflected a strategy for deregulation and liberalisation. However, with the exception of the UK, no EU member countries had any clear recognition of the crucial role to be played by regulatory authorities in the implementation and policing of competition and liberalisation (Stern 1998: xxii). The introduction of the directive proposal led to strong confrontation and polarisation, where a clear majority (11 of 12 members) rejected the Commission initiative, and so did the industry. The Commission's attempt to involve the industry through committees, to achieve dialogue and find common ground, broke down. Why was the initiative not dropped? Part of the answer was that the existing national industrial models in the majority of member countries did not have the proper form in the new international and EU context where market models were on the offensive. Representatives of traditional models were used to argue that `energy was too important to leave to the market', they were unprepared for the new discourses focusing mainly on market functionality.

The degree of resonance and conflict between the emerging model and pre-existing national paradigms varied. These paradigms partly reflected traditions of state – economy relations in various countries, and partly the countries' position with respect to natural gas resources, affecting interests related to an emerging European market:

Most countries were only, or primarily, consumers and heavily dependent on import. For this reason the Commission initiatives found it politically feasible to start its reform effort with an up stream focus (production and transmission). The expectation was that consumer prices would be lower in a (more) deregulated market. This would also lead to expansion of local market, pushing out other types of energy that were seen as harmful to the environment (coal, nuclear). A major concern for most consumer countries was related to security of supplies, traditionally dealt with through diversification of supply countries and long term contracts.

Most directly affected were countries with producer/ exporter interests, but also here interests differed. British gas production was primarily for domestic consumption. The UK was the first to privatise its state monopoly. The country initiated the EU reform, but knew that it would take time, and in the meantime the national market could be protected and developed. In the wider picture the active British stand in this area could balance some of the blame for its scepticism towards EU integration in other areas. The Netherlands, the other major EU producer and exporter, had a production and sales monopoly based on a partnership between the government and private international companies. It has played a key role in developing the European gas industry as major exporter and there was initially no interest in reforms.

The gas directive was motivated by a wish to increase gas-to-gas competition and lower prices. It would also make it difficult to keep up traditional statist structures (strong regulation, direct ownership and state companies) and national (state-dominated) sales monopoly. This was a particular challenge for Norway as the major Western country supplying the EU with natural gas. As an EEA member, Norway would be affected by the new directive.

The European gas industry was strongly against the reform initiatives. The national gas industries, serving national markets, were not experiencing problems in terms of technical efficiency or earnings. They had vested interests in national infrastructure – for example pipelines - which the Commission wanted to open up for competitive use. The only industrial interests that supported the Commission were large industrial customers (like the chemical industry) that hoped for lower prices.

The first years after the introduction of the Commission directive proposal was characterised by strong opposition, periodic stale mate and uncertainty about the feasibility of the project. The opposition in the natural gas sector was to a certain extent legitimised by the general scepticism towards further EU integration which characterised this period, symbolised by reactions to the Maastricht treaty. The discussion on the natural gas directive was kept alive, mainly due to the Commission. However, the legitimacy of Commission actions rested not only on formal mandates and earlier state commitments which was weighed against the costs of accepting new sector arrangements. The legitimacy of the alternative model as an appropriate form for organising any economic activity was an important factor, and this was expressed in member states' lacking ability to come up with legitimate reasons for preserving the existing sector arrangements. Although opposition to parts of the proposal was still strong, it gradually became clear that natural gas issues had to be addressed in terms of new concepts and perspectives.

Frame elaboration and strategic re-orientations

By the mid 1990s, the fate of the natural gas reform, seemed quite uncertain. In 1994 the original Commission proposal was formally rejected. Over the next years the EU Commission worked on an alternative, some would say watered out, version of the original directive. Key concepts and formulations were altered, but the general perspective and normative orientation remained the same. The observed development contradicts an intergovernmentalist model of explanation, as pointed out by several students of EU energy, and more specifically natural gas, policy (Matlary 1997, Eldevik 1999). Member states are not the privileged actors; changes involve not only interests, but also perspectives and models and such changes cannot realistically be seen as independent domestic processes. The EU is not just a neutral arena for given interests (Eldevik 1999:82-83).

During this stage the nature of the process changed in several ways. A key element in the construction of the new paradigm had been to develop implication of market theory, emphasising market functionality, with political concerns related to security of supply, public service obligation. Increasingly this had made it possible to rethink a number of issues. Another key element was that the legitimacy of existing national paradigms had been undermined along the way. This was not primarily due to failure of performance, but the direct impact of market ideology in individual countries, sometimes pushed by privatisation schemes to finance public deficits.

Increasingly we find that arguments and the nature of confrontations change. This happened partly in political negotiations, partly through long drawn expert driven arguments. Along the way we find a more active role of court decisions, and also new strategic initiatives from parts of the industry. This trend reflected that the new frame had been legitimated. It became clear that alternative member state and industry views could not be sustained over time. The implementation of the electricity directive served as a model, both in term of providing legitimacy and in clarifying the strategic implications of a new perspective. The new general EU dynamics, related to the Amsterdam treaty and the introduction of the Euro reinforced such processes. Increasingly the question was not if a new directive was to be passed, but when and with what provisions. Gradually arguments reflected calculations of future effectiveness.

Stern (1992:ix), a major authority on the EU gas market, was one of the first to argue that the debate on the liberalisation of the natural gas sector had far too much stress on the European Commission initiatives and too little on the actions of powerful market players. His book Third party access in European gas industries (1992) has the subtitle `Regulation driven or market led'? This book does not, however, challenge the role of the Commission as the promoter of an alternative new frame for European natural gas policy. The argument is, rather, that the content of major provisions, like third party access, most likely will materialise as a result actions taken by major players. The alternative would be more clear-cut EU regulations to be imposed by a powerful EU level regulatory authority. In later work Stern (1998) also describes how market strategies as well as a more active court system has been instrumental in bringing about changes. His analysis provides a good account of how a contested new frame is specified in political process that runs parallel to processes in the court and market system.

Changes have taken place with respect to ambitions and formulations affecting all four controversial areas mentioned above:

The concept of EU level regulated third party access to pipelines was inspired by a so-called common carrier concept in the US, but this was effectively blocked. A key controversy over the whole decade was how to secure so-called voluntary or negotiated third party access to pipelines that controlled the flow of natural gas. Related to this was the question of how much of the traded volume should be covered by such arrangements. Another key controversy concerned the problems of securing long term investment, and thereby supplies, in a system which undermined traditional long-term contracts. Gradually a new concept of security developed, emphasising the mutual interdependence between exporters and importers and the reduced operational risk in an integrated European system. The controversy over the nature and level of public obligations to be met by companies has been resolved, partly through the use of the court system. The controversy over unbundling of commercial functions has led to a general accept for the principle, in line with competition law.

For Britain it was important to achieve explicit rules for third party access and a high degree of market opening as soon as possible. The proposed 28 per cent minimum that became the result was considered too low. Also, they would like to see regulated access, where companies' access to pipelines was regulated through a tariff. Most countries had reservations about even the modified liberalisation directive. For some time the Netherlands was working against reform. However, as it turned out it was possible for the Dutch interests to reposition themselves under a new regime, to see advantages that might outweigh costs. After resisting the new regime for several years, the Dutch were among the first to change their strategy (1995). Also, Spain is characterised by a pronounced change in attitude to the new directive. In both countries governments decided to move towards opening up market more quickly than dictated by the directive.

Other countries remained reluctant throughout the whole process. In France, part of the defence for the national model was pursued through the court systems. The objective was to limit the role of competition to preserve the public service obligation. In Germany the dominant position of Ruhrgas prevented rapid political changes. However, as will be discussed below, important changes were introduced through the market. Norway, as one of three major exporters to the EU (Russia, Algeria) has waged a considerable lobby campaign against the new natural gas directive. Under the old regime the three major exporters were involved in a strategic game with a few European buyers. In Norway natural gas was produced by several companies and has to be regulated through one central negotiating authority (GFU). The present challenge is to find a way to preserve a legitimate form of regulation.

Part of the change in national positions was motivated by threats from the EU Commission to apply competition policy. The Commission threatened both France and other countries with court cases for violating competition law. In the context of growing support for the emerging new paradigm, and with reorientations going on in other countries, this created incentives for reform under domestic control. In other countries changes were motivated by new company initiatives. The announcement by the German gas company Wintershall to build new pipelines in 1989 (to further the consumer interests of the parent company BASF) and in 1990 (in partner ship with the Russian Gazprom) introduced competition for the dominant Ruhrgas.

The natural gas directive was passed in 1998. It can be considered both a success and failure for attempts to achieve liberalisation. On the one hand, it has opened up for an alternative to the pre-existing monopolised system. On the other hand, the direct and immediate requirements of the directive are limited. The directive requires that a minimum of 28 per cent of traded volume is open for competition when the directive is implemented (fall 2000). This number is going to be increased over time, but even in 2008 member countries can refuse small customer the choice of supplier as long as 43 per cent is covered by liberalised system. Systems for regulating third party to pipelines were to be developed on the national level. In addition, companies may apply for exception from the third party access article if serious economic and financial problems should arise as a consequence.

The directive contains very few supernational regulations, and the member countries have achieved considerable freedom to regulate national markets in the future. Two of the most important battles, determination of tariff methodologies and regulatory provisions for enforcing competition and third party access at the EU level, were defeated already before the first attempts of finding an agreement (in 1994). In the words of Stern (1998:xviii):

`The value of the directive, therefore, lies not so much in its specific provisions, which are likely to be rapidly overtaken by events, but rather in the fact that it established both the principle of access to (pipeline) networks, and the assurance that opponents of competition and liberalisation cannot indefinitely procrastinate in the opening up of their gas markets'

Increasingly national regulatory efforts and new market strategies become important in providing the content and pace of market liberalisation. However, the emerging EU level policy paradigm would not seem possible without the persistent role of the Commission in pushing for an alternative frame. The legitimacy of this frame is a precondition for the dynamics to be observed over the last years. In this process there is ample room for the interests and strategies of member states, industry interests and courts. However, the legitimacy of an alternative normative frame is preceding and facilitating such strategies, rather than emerge from them.

To summarise, the energy policy initiatives were introduced at a time when the collective discipline and commitment to the internal market were at the highest, around 1990. [2] However, the following process was influenced by the reduced enthusiasm and even widespread scepticism towards further EU integration around the ratification of the Maastricht treaty (1993). The most controversial issue was the directive aiming at the deregulation of the natural gas market. Conflicts led to delays, and for a long time it seemed uncertain if the initiative could succeed. It was kept alive by the Commission, and more and more it became clear that changes were taking place both at the EU and at the national level which paved the way for a new directive by 1998, after about 10 years of continuous struggle. The next stage was the implementation of the directive, with special emphasis on subsidiarity.

National implementation - convergence and divergence

The EU gas directive was passed in June 1998, and it entered into force in August the same year. Member states were committed to bring into force the laws, regulations and administrative provisions necessary to comply with the directive no later than two year after this. Compared to the original draft, the outcome was rather modest, but crucial in changing the basic rules of the game. On the other hand, the directive did not go very far in specifying how liberalisation should be achieved in practice. As Stern (1998:viii) puts it: `Not only are the most crucial commitments and provisions defined in rather general terms, but the options to adopt `negotiated access' (rather than `regulated access') could give many transmission companies an ideal opportunity to indulge in protracted and inconclusive `negotiations'.

The limitations of the gas directive, as a cohersive and specific EU level policy paradigm, reflect the opposition of member states over a decade. The modifications and limitation of the directive cannot be explained without the active role of the member states. However, the new general model reflected in the directive cannot be derived from member states' preferences and perspectives. The emphasis on subsidiarity reflects member states' control over the further development. An interesting question is to what extent states can be said to control outcomes. As discussed earlier, this is hard to measure in an exact way. The focus here is on the transformative aspect of the reform process.

Some of the formulations in the directive are deliberately not clear with respect to several important issues. A further clarification of what is meant by different concepts and provisions in the directive was left to the implementation process. The emphasis on subsidiarity, as a political necessity, opened up for different national approaches with respect to how various provisions should be secured on the national level. This meant that the further elaboration of the new policy paradigm shifted from the EU level to member state level. To ensure that implementation took place in time, and in accordance with the directive two follow-up committees were created by the Commission. One consisted of country representatives (including EEA countries), the other of industry representatives. These Committees partly reflect the Commission's responsibility to secure implementation, but they are based on voluntary co-operation.

The implementation process faces the challenge of balancing three important aspects: To clarify the implications of the directive in relation to various national setting, allowing for a certain degree of flexibility with respect to the speed of reform and specific organisational solutions, and at the same time making sure that solutions are consistent with the directive and general EU competition law. In other words, the problems is to secure convergence around one basic model of a liberalised gas market, while at the same time allowing for a certain degree of divergence with respect to how this model is operationalised in national context, even over a longer time period.

For many years the discussion of whether and how much EU integration would lead to convergence or divergence has been central within EU research (Steunenberg and Dimitrova 1999). However, ideas of convergence and divergence are difficult to pursue empirically without a clear theoretic definition of what we can expect to find. In neo-sociological theory Strang and Meyer (1993) argue that convergence is tied to local implementation of idealised models. This also applies to the imitation of specific organisations and institutions. What is imitated is the abstract underlying basic models that certain arguments and structures are assumed to be (more or less complete) representations of (Nee and Strang 1998). Isomorphic tendencies should not be regarded as equivalent with the uniformity of social processes and structures.

To what extent is this consistent with what we find in the natural gas sector? The report on the state of implementation the EU gas directive [3] summarises the situation six months before the process is to be completed. At this stage four countries had already passed legislation which to a large extent implements the directive. Nine countries were well advanced in the implementation process. Two countries were lagging behind. All were committed to fulfil requirements before August 2000, but the countries differed somewhat with respect to how objectives are to be realised.

When it comes to the third party access to pipe-lines, one of the key issues, countries may choose either negotiated access (based on publications of the main commercial conditions applicable) or (nationally) regulated access based on published tariffs, or a combination of these two systems. Eight countries will go for regulated third party access, and two countries a combination of the two models. Only three countries have chosen negotiated access, but in a restricted form. The directive requires countries to achieve a minimum level of market opening, 20 per cent by August 2000 and 33 per cent by 20008. However, most countries will go further and faster than this. The same goes for the required expansion of eligible customers, and so-called unbundling of commercial functions.

According to the directive (article 21) member states shall designate a `competent authority, which must be independent of the parties, to settle expeditiously disputes'. Member states shall also (article 22) `create appropriate and efficient mechanisms for regulation, control and transparency so as to avoid abuse of a dominant position'. Here we find that member states are choosing different regulatory approaches. Few member states plan to establish independent regulators for gas, and a few want sector regulators for natural gas as such. The majority is likely to leave regulatory responsibilities in the hands of general energy or common gas/ electricity regulators. In some countries the ministry responsible for energy matters will play a key role in the regulatory system.

Similar patterns are found with respect to other articles of the directive. The point here, however, is that national solutions may vary considerable and still be found to be within the general model laid down in the directive. An important aspect of the implementation process is to make sure that various solutions are consistent with demands of equal treatment and general provisions of EU competition law. This also means that legitimacy of national structures depends on consistency with the underlying model, and that this legitimacy in many ways becomes equivalent to the legality of particular arrangements.

As Stern (1998) has pointed out the further dynamics of EU gas market liberalisation has been moved from the EU level to the national level. The elaboration of the EU model rests heavily on national processes, but in the future industry strategies and court decisions will play a key role in the further maturing of the system. To what extent can we talk about a real transformation? As Fligstein (1996) pointed out in his analysis of the internal market programme, states were more willing to accept changes affecting transaction rules, than rules which involved areas which were deeply embedded in member state structure (like property rights). To some extent we find the same in the gas sector. There is reluctance to loose control over national reforms and regulatory authority.

However, preservation of national divergence with respect to speed and forms of market liberalisation does not imply that changes are mainly a matter of form; new ways to legitimate traditional policies (Brunsson 1989). The gas directive have changed the basic logic and unleashed dynamic forces of liberalisation. The result is not only that national sectors are impacted by EU decision, it seems even more important that the relative weight of the market in relation to politics has been altered in a radical way. The next step is probably going to be a structural transformation of the market; with the introduction of new technology, new companies operating across energy sectors, new types of actors (traders) and price formation through the spot market. [4]

To summarise, the further elaboration of a European gas policy paradigm will most likely depend on developments on the national level for several years to come. This process is likely to create pressures towards convergence, but variations in national arrangements may persists as long as they operate in a way which is consistent with the directive. If companies go to court with complaints that certain arrangements do not function properly, there will be pressures for change. This may be regarded as the continuation of a transformation, which started, with the legitimisation of an alternative and controversial frame and which was elaborated on the EU level to become a new directive. The present situation is characterised by both convergence and divergence, where the basic rules of the game have changed and this is the point of departure for national adaptation.

Understanding complex transformation

Changes in EU energy policy are indirectly driven by global economic competition (wish for lower energy prices), but even more by clearly political changes which reflect and reinforce a paradigmatic shift in global thinking about energy markets. The construction of a new EU level paradigm is not, however, simply a copying process, it modifies and elaborates on the global trend. Such modifications have the underpinning of rational arguments about market functionality and market failure in the face of welfare capitalism. At the same time policy changes reflect and reinforce EU integration; i.e. increased supranational authority vested in law, decision-making procedures, as well as a normative model of EU integration. On the other hand, the principle of subsidiarity opens up for `flexible' implementation under EU competition law.

The changes that have taken place in the EU natural gas policy, legislation and market over the last decade can best be characterised as a radical and complex transformation. It is not only a result of states finding new ways to secure given self-interests or due to documented superior effectiveness of an alternative form of industrial organisation. On the contrary, the construction of a new policy paradigm was only to a limited extent based on calculations, but rather legitimisation and elaboration of a market model. Authoritative reasons seemed to have outweighed means-end calculation of benefits. The emergence of a new paradigm led to changes in political influence between states, it opened for new strategies and new types of calculations. The existence of a policy paradigm is an important precondition for actors being able to act under a norm of rationality. In a rational model, in contrast, the relationship between these factors should generally be reversed, and rationally rests in the actors.

The transformation of the EU energy policy paradigm weakened the role of the government, and strengthened the role of law and the energy companies. These changes suggest a movement towards a more `minimalist' type of regulation, often seen as a characteristic of the EU. However, the process proceeding this market transformation involves considerable state-like authority and capability for keeping initiatives on the track, securing an outcome, and implementing decisions – while preserving the general direction of reform. This ability to carry out radical reforms, which transforms `almost unchangeable' national policy paradigms seems to contradict the image of the EU as a organisationally weak system, with a weak core, unclear institutional boundaries (Kohler-Koch and Eising 1999) and with a high degree of freedom for political entrepreneurship and open access for a wide variety of actors (Andersen and Eliassen 1996).

What keeps such processes from breaking down? What provides direction? What is suggested here is that this gap between the weak organisational core, on the one hand, and EU's ability to transform entrenched national paradigms, on the other hand, should also be understood in he context of the cultural materials institutionalised in this structure. The culture involved is the (originally Western European) culture of rationalisation and something of the same nature is found on the global level. This would imply that the EU is not only an especially dense form of an elaborating global system (Meyer 2000:14). Over time the EU institutional set up has acquired characteristics that reinforce the kind of universalistic discourses implied by this cultural frame.

The EU member states represent historical traditions which make them more able to sustain their models of nation-state hood than countries that lack this tradition. However, the member countries also represent stores of accumulated insight, embedded and operational knowledge reflecting the basic principles of such a culture, as a common core. This common denominator facilitates the construction of new paradigms and it also opens up for the further elaboration and modifications of global models. This is reflected in the basic discourse rules of central EU decision-making processes guiding the search for authoritative and legitimate arguments. And also the growing pool of legitimate concepts and legislation reflecting the wider EU integration.

The institutionalisation of new policy paradigms are reflexive processes where actors changes perceptions about interests (energy as commodity versus part of economic/ social policy), about the proper way to organise and control activities (`some sort of market' which will unleash a new dynamics), and even self image and identities (what is the meaning of sovereignty). This does not in any way imply a romantic notion of conflicts being transcended in a collective heaven of EU co-operation. Conflicts persist, but they are framed in other ways, and as in all political processes some gain more than others do, and some lose.

The persistence of interests and conflicts is an important reason why some have suggested that also complex transformation and learning should be explained in terms of rational actors influenced by the reinforcement of the `stick and `carrot'. Such a model seem to make more sense when actors face non-paradigmatic changes which facilitate `single loop' learning, or when powerful actors may shape or control the process. In situations of radical and complex transformation the ability to calculate interests in a future state is limited, and reinforcement often has to rely on arguments about appropriate form. The process around the natural gas directive illustrates the latter.

How are good and authoritative arguments constructed? What makes old arguments lose their power? This study suggests a process where the authorisation of arguments in relation to a wider cultural frame play a key role as a source of legitimacy, but arguments are not developed from scratch. Concepts and ideas are downloaded from various institutional settings; often from international experience codified in scientific knowledge or legal concepts and rules (OECD, GATT). In this case also from the wider EU context; as established competition law. As integration increases the available repertoire of embedded concepts and rules for defining issues will increase. It will be easier to find elements to build upon in the institutionalisation of new areas. This also means that certain elements become de-legitimated.

The above interpretation seems most consistent with the type of paradigm construction that can be observed in the development of EU energy policy. The long drawn process of confrontation, stalemate and compromise was characterised by the parallel de-legitimisation of traditional statist models, on the one hand, and the legitimisation of new market oriented model. This process took place at both the central EU and at the national level, and the relative weight of central policy, court decisions and industrial initiatives varied over time. Because of the special tensions and conflicts between the EU and national level a major element of the directive was that implementation should take place under the principle of subsidiarity. In other words the directive represented a new policy paradigm, which in principle represented a new normative model that every country, as well as the industry, accepted. On the other hand, there were some `grey areas' as to what some elements of the directive actually implied. The principle of subsidiarity should be an instrument to handle this, but the problem was how to make sure that country specific interpretations did not violate common competition law, linking legitimacy and legality. For this reason two implementation committees were established to handle implementation under subsidiarity. One for the national regulator and one for the industry. Such committees are voluntary, but also linked to the Commission's mandate to overlook implementation. (Norway has participated in the Committees, despite the fact that the directive is still not passed in the EEA).

To summarise: Reforms were originally inspired by US reform and experiences from the UK. However, reforms in these two countries had emphasised different aspects. In the US market functionality had been the key, while in the UK privatisation had dominated with less concern about how and under what condition the market would work. After the initial refusal from a majority of countries there was a stalemate followed by the pragmatic search for acceptable ways to construct a new normative model, but taking into account special needs for modifications due to the nature of the industry: the maturity of the market, import dependence, the public service obligation of the industry (in some countries), national control over reform process and some freedom of choice over the actual organisation of national industry structures and regulatory roles. These factors reflected the special situation in the EU natural gas sector, the dominance of consumer country interests and the tension between supernational and national authority.

Concluding remarks

This paper does not allow for a detailed confrontation between theoretical ideas and data; a reconstruction of how and through what mechanisms changes have taken place in the EU energy policy. Broad social changes of this nature take place in many ways, at many levels and sometimes in rather diffuse ways. Various aspect of this change may be analysed in differently. However, the discussion suggests that the transformations that have taken place in the energy sector cannot be well accounted for a) by models assuming that processes are driven by, or at least accepted by, rational states with given interests determined exogenously to the EU level process, b) by models assuming that EU level solutions drive out less effective national solutions, or c) by models of historical institutionalism which emphasise continuity rather than change or the accumulated pressures of historical decisions.

Experiences from the EU energy sector experience illustrate some general point within a neo-institutional sociological perspective:

  1. EU reforms do not make sense if not seen as part of a broader, global change strengthening the legitimacy of market models. However, Europeanisation of policy is not only a copying process, it also represents a modification that feed back on the wider change process.
  2. The EU system has a surprising ability to carry out radical reforms – even against the will of a majority of states - despite a weak organisational core. This suggests that the culture vested in this loose structure plays a key role in keeping such processes on track and provides direction. The legitimisation of a new paradigm, over time, opens up for calculation of strategic opportunities and efficiencies, and not the other way around.
  3. The Europeanisation of the nation state does not imply uniformity of structure. National structures may reflect the general European model to different degrees, not only as matter of transition. However, legitimacy of form increasingly becomes a question of legality.


Andersen, S.S 1993: The Struggle Over North Sea Oil and Gas. Oslo: Scandinavian University Press.

Andersen, S.S 1995: Europeisering av politikk. Indre marked, petroleum og milj�. Oslo: Fagbokforlaget.

Andersen, S.S. 2000: “How is the EU possible? A neo-institutional sociological perspective”. Paper presented at the ARENA workshop on Institutional approaches to the study of the EU, March 2000.

Andersen, S.S. and K.A. Eliassen (eds.) 1996: The European Union: How Democratic Is It?. London: Sage.

Brunsson, N. 1989: The Organization of Hypocrisy – Talk, Decisions and Actions in Organizations. Chichester: John Wiley and Sons.

Bulmer, S. 1994: “The Governance of the European Union: A New Institutionalist Approach”, Journal of Public Policy , 13:351-80.

Bulmer, S. and M. Burch 1998: “Organising for Europe: Whitehall, the British state and European Union”, Public Administration, 76: 601-28.

Cameron, E. 1991: The European Reformation. Oxford: Clarendon Press.

Dobbin, F. 1994: Forging Industrial Policy. The United States, Britain and France in the Railway Age. Cambridge: Cambridge university Press.

Dore, J. & R. De Bauw, 1995: The European Energy Charter. Origins, Aims and Prospects. London: The Royal Institute of International Affairs.

Eldevik, F. 1999: “Liberaliserng av gassmarkedet i Europa. EUs gassdirektiv av 1998”. ARENA Report 4/99. Oslo: ARENA.

Estrada, J., H.O. Bergesen, A. More & A.K Sydnes (red.) 1988: Natural Gas In Europe. London: Pinter Publishers.

Directive 98/30/EC of the European Parliament and of the Council 1998: Concerning Common Rules for the Internal Market for Natural Gas. June 22. Brussels.

Finnemore, M. 1996: “Norms, Culture and World politics: Insights From Sociology's Institutionalism”, International Organization, 50(2):325-47.

Fligstein, N. 1996: “Markets as Politics: A Political-Cultural Approach to Market Institutions”, American Sociological Review, 61:651-73.

George, S. 1985: Politics and Policy in the European Community. Oxford: Oxford University Press.

Haas, Peter 1992: ”Introduction: Epistemic Communities and International

Policy Co-ordination” in International Organization, 46:1-35.

Hall, Peter (ed.) 1989: The Political Power of Economic Ideas: Keynesianism

across Nations. New Jersey: Princeton University Press.

Hancher, L. 1998: “EU Energy Policy”. Report 4/1998. BI: Senter for elektrisitetsstudier.

Kohler-Koch, B. 1996: `Catching Up With Change. The Transformation of Governance in the European Union', Journal of European Public Policy , 3:359-80.

Kohler, B. and R. Eising 1999: The Transformation of Governance in the European Union. London: Routledge.

Lyons, P. 1992: EC Energy Policy. A Detailed Guide to the Community’s Impact on the Energy Sector. London: Financial Times Business Information.

Lyons, P. 1994: Energy Policies of the European Union. London: EC Inform.

Matlary J.H. 1991: “From the Internal Energy Market to a Community Energy Policy? An Analysis of Decision-making With Emphasis on Natural Gas.” FNI-Report 005. Lysaker: FNI.

Matlary, J.H. 1996: “Internal Market Regime or New Polity Model: Whither the European Union?” ARENA Working Paper no. 26. Oslo: ARENA.

Matlary, J.H. 1997: Energy Policy in the European Union. London: McMillian Press

Meyer, J.W. 1994: “Rationalized Environments”. In W.R. Scott and J.W. Meyer (eds.): Institutional Environments and Organizations. Thousand Oaks, Ca.: Sage.

Meyer, J.W. 2000: “The European Union and globalisation of culture”. Paper presented at the Workshop on Institutional Approaches ti the Study of the European Union, Oslo, March.2000

Meyer, J.W., J. Boli, and G.M. Thomas 1987: “Ontology and Rationalization in the Western Cultural Account”. I G.M. Thomas, J.W. Meyer, F.O. Ramirez, and J. Boli (eds.): Constituting State, Society, and the Individual . Newbury Park, Ca.: Sage.

Meyer, J.W., J. Boli, G.M. Thomas, and F.O. Ramirez 1997: “World Society and the Nation-state”. American Journal of Sociology , 103(1).

Moravcsik, A. 1993: “Preferences and Power in the European Community: A Liberal Intergovernmentalist Approach”, i Journal of Common Market Studies , 31:473-524.

Moravcsik, A. 1998: The Choice For Europe. Social Purpose and State Power From Messina to Maastricht. London: UCL Press.

Morgenthau, H.J. 1973: Politics Among Nations: The Struggle for Power and Peace. New York: Knopf.

Nee, V. and D. Strang 1998: '”The Emergence and Diffusion of Institutional Forms”. Journal of Institutional and Theoretical Economics, 154(4):706-15.

Nylander, J. 2000: The Power of Framing. A New-Institutional Approach to Interest Group Participation in the European Union. Doctoral Thesis, University of Uppsala.

Padgett, S. 1992: “The Single European Market: The Politics of Realization”, Journal of Common Market Studies, (1).

Peterson, J. 1995: “Decision-making in the European Union: Towards a Framework for Analysis”.Journal of European Public Policy 2:69-93.

Peterson, J. and E. Bomberg 1999: Decision-making in the European Union. Houndsmills: MacMillan.

Pierson, P. 1996: “The Path to the European Integration - A Historical Institutionalist Analysis”. Comparative Political Studies, 29(2):123-163.

Scharpf, F. 1988: “The Joint Decision-trap: Lessons From German Federalism and European Integration”, Public Administration, 66:239-78.

Schmitter, Ph. 1996: “Imagining the Future of the Euro-polity With the Help of New Conceots”. I G. Marks, F. Scharpf, Ph. Schmitter, and W. Streeck (eds.): Goveranance in the European Union. London: Sage.

Scott, W.R. 1994: “Institutions and Organizations: Toward a Theoretical Synthesis”. In W.R. Scott and J.W. Meyer (eds.): Institutional Environments and Organizations. Thousand Oaks, Ca.: Sage.

Schmidt, S. 1997: “Sterile Debates and Dubious Generalisations: European Integration Theory Tested by Telecommunications and Electricity”, Journal of Puvlic Policy 16 (3) pp. 233-71.

Stern, J.P. 1992: Third Party Access in European Gas Industries: Regulation-driven or Market-led? London: The Royal Institutte of International Affairs.

Stern, J.P. 1998: Competition and Liberalization in the European Gas Market. A Diversity Model. London: The Royal Institutte of International Affairs.

Steunenberg, B. and A. Dimitrova 1999: The Search for Convergence of National Policies in the European Union: An Impossible Quest? ARENA Working Paper no. 16. Oslo: ARENA

Strang, D. and J.W. Meyer 1993: “Institutional Conditions for Diffusion”. Theory and Society, 22:487-511.

Tallberg, J. 2000: Making States Comply. Lund: Department of Political Science.

Tugwell, F. 1988: The energy crisis and the American political economy. Politics and markets in the management of natural resources. Sstanford: Stanford university Press.

Usherwood, S. 1998: “Energy policy”. In Gl�cker, Junius, Usherwood og Vassallo, Guide to EU policies. London: Blackstone Press.

Waltz, K. 1979: Theory of International Politics. New York: McGraw-Hill.


[1] Thanks to Johan P. Olsen, and also Dag H. Claes og Ole G. Austvik, for helpful comments.

[2] A white paper on the energy sector(s), and its relation to the internal market programme, was published by the Commission in 1988.

[3] State of implementation of the EU Gas Directive (98/30/EC). Summary report. Prepared for the 3. meeting of the Follow-up Group held 24 February 2000.

[4] As a centrally located informant put it: 'We don't know exactly what is going to happen, but we have confidence in the future'.

[Date of publication in the ARENA Working Paper series: 15.08.2000]