Europeanisation: Solution or problem? (*)

Claudio M. Radaelli

European Integration online Papers (EIoP) Vol. 8 (2004) N° 16; http://eiop.or.at/eiop/texte/2004-016a.htm

Date of Publication in : 6.10.2004

| Abstract | Back to homepage | PDF | | This paper's comments page | Send your comment! to this paper |

Contents:

• Introduction
• 1. The specific domain of Europeanisation
• 2. Governance, institutionalisation and discourse
• 3. Making sense of Europeanisation
• 4. How Europeanisation produces domestic change
• 5. What does the evidence tell us?
• 6. Conclusions: Where do we go from here?
• References

1

Introduction

Some years ago, Simon Hix and Klaus Goetz (2001:15) observed that ‘Europeanisation has all the hallmarks of an emergent field of inquiry’. The field has now come of age, and European Integration online Papers has been an important outlet for the academic debate on Europeanisation. Hence this is the best place to take stock of what has been done so far.

Indeed, we may well start from the radical question whether the shift towards Europeanisation reflected by biblio-metric exercises (see the data in Featherstone 2003: 5-6; Giuliani 2004) is really indicative of a new theoretical orientation or simply reflects the faddish popularity of the term. Almost twenty years ago, Kratochwil and Ruggie (1986) put a similar question to the then emergent field of ‘international governance’ by asking whether the shift represented a haphazard sequence of ‘theoretical or topical fads’? To use their language, which draws on Imre Lakatos’ terminology, is Europeanisation rooted in a core concern and ‘set of puzzles’ that provide some coherence to this field? Is Europeanisation a progressive programme? Or does it produce degenerative problem shifts?(1)

It is difficult to operationalise this question, especially in the context of a relatively new field of scientific inquiry. Additionally, how can one pin down precisely what the ‘set of puzzles’ really is – if different approaches lead to different definitions of the ‘set’? Further, it all depends on how demanding the definition of progressive programme is. My view is that it is too early to engage with ‘yes-no’ tests on the ‘progressive programme’ question. And of course Lakatos had in mind large fields of inquiry, whereas Europeanisation is a relatively small sub-set of an interdisciplinary field of research (that is, European Studies). However, we can use Lakatos’ framework to ask the question whether some interesting puzzles are emerging (at least in embryonic form), and whether some novel facts have been uncovered by Europeanisation. This article answers to this re-formulation of the question with a preliminary ‘yes’.

2

The most original work on Europeanisation(2) has contributed to the emergence of new insights, original explanations, and interesting questions on three important issues: the understanding and analysis of ‘domestic impact’ of international politics, how to endogeneise international governance in models of domestic politics (in terms of research design), and the relationship between agency and change. These three issues are prominent in the research agendas of international relations, theoretical policy analysis, and comparative politics. Hence research on Europeanisation has the merit of having brought European Studies into mainstream political science.(3)

One theme throughout the article is that, in order to develop a progressive agenda, Europeanisation should be seen as a problem, not as a solution. Europeanisation does not provide any simple fix to theoretical or empirical problems. Quite the opposite, it can deliver if it is approached as a set of puzzles. A problem in search of explanation – not the explanation itself. At this stage, its potential is greater if one turns to Europeanisation as ‘something to be explained’, not ‘something that explains’.(4)

The article introduces the specific domain of Europeanisation (Section 1) and the different (but not necessarily incompatible) perspectives on Europeanisation (Section 2). Section 3 suggests how to make sense of Europeanisation in terms of research designs. Section 4 presents the main mechanisms of Europeanisation. Section 5 presents five main results in terms of evidence. Finally, Section 6 concludes on what we have learned so far and what we need to learn.

A preliminary remark: this article is eminently on Europeanisation in the context of European integration. It is also focused on the long-term dynamics of European integration, and therefore does not specifically review studies on how European politics and policy affect the new member states. Although some general propositions presented here may travel well (for example in the context of enlargement, see Grabbe 2003), practical considerations suggest this limitation of scope.

1. The specific domain of Europeanisation

Although there are many possible approaches to Europeanisation, depending on the disciplines and the specific questions of different research programmes (Featherstone 2003, Olsen 2002), one simple way to get to grips with definitions and their scope is to make a distinction between background concepts and systematised concepts (Adcock and Collier 2001). As background concept, Europeanisation refers to all the possible meanings we may want to include in an encyclopaedia. Thus, the historian may well look at the evolution of Europeanisation starting from the Renaissance, and link it to the rise of trade and individualism in Europe. It is the task of an encyclopaedia to report on all the major meanings associated with a concept. But in a dictionary used by a community of specialists, there is a need for a systematised concept. This article deals with the latter. It assumes that the readership is made up of political scientists interested in the domestic consequences of the process of European integration. In this connection, Europeanisation provides a shift of focus in relation to theories of European integration, theories of governance, and classic themes in comparative politics.

Let us start by demarcating the difference between Europeanisation and the set of puzzles typical of integration theories. Europeanisation is a set of post-ontological puzzles. This means that we start from the notion that there is a process of European integration under way, and that the European Union (EU) has developed its own institutions and policies over the last fifty years or so. Accordingly, the puzzles do not refer to the nature of the beast (to paraphrase Puchala 1973), i.e., why and how do member states produce European integration, and whether the European Union is more inter-governmental or supra-national.

3

Instead, the theoretical effort in Europeanisation as a research agenda is all about bringing domestic politics back into our understanding of European integration, without assuming that the balance of power between the state and European institutions is being tilted in one direction or another. The question of the balance of power is more important for theories of European integration than for Europeanisation. Börzel’s conclusion (2004) is that integration theories are not well suited to understand Europeanisation as their main puzzle is the explanation of dynamics and outcomes of European integration rather than domestic effects.

Some theories of governance (see Rhodes 1997) go even further than neo-functionalist’s arguments about the balance of power between state and supranational institutions. They predict the hollowing out of the state from above and from below.(5) By contrast, Europeanisation is mostly interested in adaptation to Europe, without making bold predictions. Domestic institutions are supposed to be malleable to different degrees, but they are not withering away. A final element that distinguishes Europeanisation from other conventional approaches to politics in Europe is the lack of assumptions about convergence. Previous work in comparative politics has shown the role of Germany as a model of politics for Europe, specifically because of its semi-sovereign and decentralised governmental system (Katzenstein 1997; Bulmer and Paterson 1987).(6) Research on Europeanisation provides more open answers to convergence and divergence, seems to downplay the role of Germany as a model of governance, and qualifies the results of convergence and divergence by policy areas and historical conditions.(7)

Turning to definitions of Europeanisation, there are various suggestions in the literature. Risse et al. (2001: 3) opt for the following definition:

‘We define Europeanisation as the emergence and the development at the European level of distinct structures of governance’

Although the whole book by Cowles et al. (2001) is clearly dedicated to domestic policy adjustment, this definition seems to shed more light on the creation of European governance than on the domestic consequences of integration. As such, it does not demarcate the specific territory of Europeanisation as a field of inquiry.

Olsen (2002: 944) shows the limitations of definitional exercises by arguing that Europeanisation is a set of model-building puzzles, not a set of definitional puzzles. He then goes on to illustrate the many faces of Europeanisation. His approach covers a very broad spectrum, including Europeanisation as political integration and political unification. This wide spectrum is almost without boundaries and somewhat discouraging. Yet Olsen’s approach has the advantage of re-directing scholars from unidirectional patterns of causation to mutual adaptation and co-evolution between the domestic and the European level – a point well-illustrated by Gualini (2003:6).

4

Box 1 provides the definition proposed by Radaelli. It is grounded in an understanding of Europeanisation as interactive process, and not a simple process of uni-directional reaction to ‘Europe’ (Salgado and Woll 2004:4). It is useful to approach Europeanisation well beyond a narrow, linear, top-down notion of ‘impact’ (of the EU on domestic systems). The idea of impact is somewhat static and mechanistic, whilst real-world processes of Europeanisation provide considerable opportunities for creative usages of Europe. Domestic actors can use Europe in many discretionary ways. They may discursively create impacts. They may draw of Europe as resource without specific pressure from Brussels. They may get entrapped in European discourses and socialisation processes that cannot be captured by a narrow notion of impact (Jacquot and Woll 2003; Thatcher 2004). Europeanisation deals with how domestic change is processed, and the patterns of adaptation can be more complex than simple reactions to ‘Brussels’. Indeed, complex adaptation patterns and interactive logics are common terms in new studies on Europeanisation (Megie and Ravinet 2004).

Figure1

This leads to the discussion of what is new in this field. Figure 1 contrasts the focus of European integration studies (first arrow) with two approaches to Europeanisation (roughly speaking, top-down and bottom-up). The top-down approach to Europeanisation, typical of work done in the 1970s and 1980s,(8) but still present in current research projects, tends to rely on the following ‘chain’:

‘pressure’ from Europe on member states à intervening variables à reactions and change at the domestic level.

This is the cluster of studies focusing on uni-directional changes and narrow impacts. Most of the analysis of top-down Europeanisation conducted before the 1990s revolved around the idea of tracking down the implementation of European policies. Alternatively, the thrust was to understand how member states organised their European business. Both ‘pressure’ and ‘intervening variables’ received a rudimentary treatment.

Recent projects informed by this approach are more theoretically robust. An outstanding example is the volume edited by Cowles et al.. Figure 1 shows that the main innovation brought about by top-down Europeanisation research designs consists of a move from the analysis of European integration to the analysis of the domestic impact of the EU. Graphically, there is a shift from the first arrow to the second in figure1.

The bottom-up cluster has a completely different research design. Instead of starting from European policies (or politics) as independent variable and tracking down the consequences for domestic actors, policies, and politics, it starts and finishes at the level of domestic actors. This may seem paradoxical, but actually the third arrow in figure 1 simply exposes the limitations of the language of dependent and independent variables. As Gualini convincingly argues (2003), Europeanisation is not the explanans (i.e., the ‘solution’, to paraphrase the title of this article; or the phenomenon that explains the dependent variables), but the explanandum (i.e., the ‘problem’ that needs to be explained).(9) Essentially, the third arrow shows a bottom-up research design. It starts from actors, problems, resources, style, and discourses at the domestic level. Put differently, the starting point is a system of interaction at the domestic level. By using time and temporal causal sequences, a bottom-up approach checks if, when, and how the EU provides a change in any of the main components of the system of interaction. Finally, ‘bottom-uppers’ try to measure the consequences of all this in terms of change at the domestic level. Of course, one can see this as yet another mechanism of ‘impact’, but with the qualification that the notion of impact goes beyond the ‘reaction’ to Europe and includes creative usages.

5

Let us illustrate figure 1 with an example. In a study on alcohol monopoly in Finland, Norway and Sweden, Ugland (2003) shows that at the beginning of the ‘policy story’ domestic policy actors were coping with alcohol as a problem of health policy. The advent of EU competition policy produced a re-formulation of policy problems. Alcohol policy is now subject to competition policy and to single market regulations. Countries like Finland can no longer maintain monopolies and at the same time export ‘Finlandia’ – a well-known vodka – throughout Europe by using monopolistic protection at home as power base. The very idea of alcohol policy, at the end of Ugland’s story, has been re-categorised as competition policy (as opposed to the previous categorisation in terms of health policy). This change, he argues, cannot be ascribed to any other force than EU policy. As mentioned, the story starts and finishes at the level of the domestic system of interaction.(10)

By doing so, Ugland avoids the pitfall of studies starting from European policies, directives and regulations and tracking down their implementation at the domestic level. The pitfall is one of pre-judging the role of Europeanisation (Dyson 2002). How do we know that Europe is really affecting the logic of interaction at home? All in all, we observe change at the domestic level, but this may well be the result of some other processes.(11) It well may be that Europe is not important at all in terms of logic of interaction, even if directives and policies coming from Brussels have been implemented. With a research design like the third in figure 1, one can see if and when Europe changes the logic, and the analyst is also in a favourable position to observe when major alterations of the logic are produced endogenously at the domestic level or by more global pressures.

There are other interesting properties of the second-generation studies. To begin with, Europeanisation is not approached as a new form of theorisation with its own new vocabulary (like goodness of fit, see below). Rather, Europeanisation is seen as ‘orchestration’ of existing concepts and theories, with major theoretical import from comparative politics and theoretical policy analysis (Featherstone and Radaelli 2003:340). Secondly, whereas previous research considered Europeanisation an end state, recent research has embraced the notion of Europeanisation as a process (Goetz 2002). The question is not one of assessing whether a country has become Europeanised or not. As explanandum or ‘problem’, Europeanisation demands explanation of what goes on inside the process, not a simple black-box design in which one correlates the input ‘EU independent variables’ to the output ‘domestic impact’. Thirdly, Europeanisation covers both vertical processes (from the EU to domestic politics) and horizontal dynamics. The EU may provide the context, the cognitive and normative ‘frame’, the terms of reference, or the opportunities for socialisation of domestic actors who then produce ‘exchanges’ (of ideas, power, policies, and so on) between each other.

Fourthly, although some authors still make the argument that a precise definition of Europeanisation should cover the homogeneising impact of the EU on ‘specific institutions and practices across a wide range of state activities’ (Page 2003: 163), the new approach makes a careful distinction between the definition of Europeanisation and its possible impacts in terms of convergence or divergence. Indeed, as will be shown later on, there is more empirical evidence pointing towards a differential impact of Europe than towards convergence. Thus, it is not surprising that Page has not found much support for Europeanisation as homogeneity. His definition of Europeanisation confuses the concept with its possible outcomes.

To conclude on the insights provided by bottom-up studies: Europeanisation is not a new theory. It is an approach that enables us to orchestrate exiting concepts and to contribute to cumulative research in political science (by drawing systematically on concepts and models produced by comparative politics and policy analysis). It is a process, rather than end-state. It is an explanandum, rather than an explanans.

6

2. Governance, institutionalisation and discourse

There are three aims in this Section. The number of studies on Europeanisation is well in excess of what can possibly be reviewed here. However, they can be reduced to a limited number of ‘deep interrogations’ on the very nature of the process. Accordingly, the first aim is to show how Europeanisation can be seen as governance, as institutionalisation, and as discourse. These are three ideal-types, and most authors find it useful to mix them in order to provide empirically rich accounts. But the distinction sheds light on the nature of Europeanisation. To clarify: the differences in the three approaches to the ‘deep interrogations’ should not be exaggerated; perhaps they simply show the most important sides of the prism. The second aim is to illustrate a two-track strategy of inquiry. One the one hand, fundamental logic suggests that we focus on change (or lack of) and consider Europeanisation in the context of other possible factors of change, without pre-judging its role. On the other, the processual nature of Europeanisation should be acknowledged up-front. The third aim is to discuss the implications of this approach.

Let us commence with Europeanisation as governance. A number of authors (Bache 2003; Buller and Gamble 2002; Goldsmith 2003; Gualini 2003; Kohler-Koch and Eising 1999; Scharpf 1999; Winn and Harris 2003) treat Europeanisation as a process of governance. Bache anchors his analysis to contemporary models of governance, such as the ‘differentiated polity’ model put forward by Rod Rhodes (1997). Goldsmith and, more explicitly, Gualini, draw systematically on multi-level governance. Kohler-Koch and Eising (1999) argue that Europeanisation is a process of changing understandings of governance in Europe. To illustrate, Europeanisation has modified shared notions of governance in EU member-states by establishing the principle of partnership between private and public actors, and by inserting regions into a complex set of layers of governance.

This way of making sense of Europeanisation has the advantage of contributing to cumulative research in political science. Instead of assuming a sui generis ‘nature’ of Europeanisation, one can relate it to our knowledge of governance. Future research informed by ‘Europeanisation as governance’ could tackle the thorny issue of whether Europeanisation produces new governance, or hybrids of existing forms of governance. Another interesting question for this approach relates to normative implications: is Europeanisation producing good and legitimate governance in Europe? Or is it reducing social policy-making capabilities at the domestic level without compensating this with a model of ‘social Europe’ to be delivered by EU institutions (Scharpf 1999, 2001; Mair 2001:47)?

Without denying that Europeanisation is eminently about governance, a second group of authors has put emphasis on Europeanisation as institutionalisation (Cowles et al. 2001, Börzel 2004; Kurzer 2001; Olsen 2002; Radaelli 2003a). The definition in box 1, indeed, refers to Europeanisation as processes through which formal rules and informal ways of doing things are first discovered and experienced in the EU context and then institutionalised inside the logic of behaviour of domestic actors. This is pretty close to the idea of institutionalisation of Europe proposed by Stone Sweet, Sandholtz, and Fligstein (2001). The difference is that the latter make a case for the growth of supra-national governance along revised neo-functionalist lines, whereas the object of Europeanisation is elsewhere, as illustrated above. Cowles et al.’s (2001)(12) approach institutionalisation as the emergence of distinctive structures of governance – the proof that the different visions are often complementary rather than mutually exclusive. Their take on institutionalisation is quite distinct in that it is based on the claim that the exclusive factor producing domestic change is the misfit between the domestic and the EU level. Put differently, in order to create domestic change, ‘Europe’ must be somewhat uncomfortable for domestic institutions.

7

As mentioned, the causal chain of institutionalisation starts with adaptational pressure (created by misfit) and proceeds through two different possible pathways. The former is based on resource re-distribution and the role of actors in an opportunity structure altered by EU variables. The latter is more social constructivist and hinges on processes of socialization. Börzel and Risse (2003) account for specific intervening variables. When pressure operates as new opportunity structure, the intervening variables are the number of veto players and supporting formal institutions. Veto players and formal institutions convert new opportunities into concrete redistribution of resources, and thus contribute to differential empowerment. When pressure is instead channelled through collective understandings and norms, norms entrepreneurs and cooperative informal institutions can facilitate socialisation and social learning. In turn, socialisation and learning contribute to the institutionalisation of norms and to the development of new identities. To conclude, this is a three-step framework based on adaptational pressure, mediating factors, and domestic change.

The major criticisms to Börzel and Risse boil down to the observation that this is not the only way things work. There are examples of Europeanisation without major adaptational pressure. Domestic actors can use ‘Europe’ even in the absence of pressure. They can adapt domestic policy and produce change independently of pressures arising out of institutional misfit (Knill and Lehmkuhl 1999; 2002; Thatcher 2004, Jacquot and Woll 2003). Add to this that the degree of fit is discursively and socially constructed, it is not an objective measure. Much of the debate on welfare state reforms in Southern Europe, for example, is between those who argue that domestic policies are in line with European constraints (such as the stability and growth pact), and those who think that the fit between domestic policies and ‘Europe’ is poor, and hence major reforms are needed. All in all, the ‘goodness of fit’ framework is somewhat excessively structural. There is not enough room for agency. True, actors are not completely neglected, but they act only in response to pressure. Instead, actors can also choose and learn from Europe outside adaptational pressures (Jacquot and Woll 2003:3). Thatcher (2004) shows how domestic reformers have been able to use the European telecommunication policies by transforming minimum pressure into large domestic change through a variety of discursive and bargaining strategies. Structural pressure would have predicted limited change. In order to understand the magnitude and the process of change, one has to enter domestic actors and their creative usages of Europe.

The framework of institutionalisation, however, cannot be reduced to the goodness of fit propositions. Other authors have gone much further. Gualini, for example, blends multi-level governance with institutionalisation to show how policy change can create major institutional transformation – a point that chimes with Giuliani’s observation (2001) that Europeanised policies impact on polity. This makes the analysis of Europeanisation much more dynamic. Instead of separating the domains of policy, politics, and polity (as Börzel and Risse do, 2003), one can investigate the dynamic relations between policy change and macro-institutional structures. Deep down, one can see the research agenda suggested by Theodor Lowi many years ago (‘policies determine politics’) rejuvenated by these recent developments.

Giuliani (2004) links explicitly new institutional theory (specifically, the organisational theory of institutionalisation) and Europeanisation. He argues that ‘Europeanisation as institutionalisation’ means essentially that

1. the EU is specialised and differentiated,
2. there is self-validation of the EU,(13) and
3. EU decisions cannot be derived from the utility functions of one or a stable group of member states.

8

By putting Europeanisation on solid neo-institutional tracks, this suggestion sheds light on the ‘deep nature’ of Europeanisation. It tells us that it is a process through which the EU gains its own autonomous meaning and self-validation within the logics, cognitive frames, and norms of behaviour of domestic actors. We will turn to the implications of this point in a moment. But we need to introduce the third framework first.

The third understanding is Europeanisation as discourse. This claim has been made by Hay and Rosamond (2002) and Kallestrup (2002). Policy-makers and stakeholders construct Europe through language and discourse. In turn, discourse is cast in different forms, from rhetoric (Schimmelfennig 2001) to policy narratives (Radaelli 1999). Hay and Rosamond note that Europeanisation can be the vehicle through which discourses on globalisation are institutionalised in domestic politics. Schmidt and Radaelli (2004; see also Schmidt 2002) look at discourse in its institutional setting. They use the notion of discursive institutionalism. Discourse has a transformative power in EU policy and politics, but in order to understand if and when Europeanisation produces change, one has to situate discourse in its institutional ‘riverbeds’. Turning to figure 1 one more time, discourse can change the preferences of actors, reformulate policy problems, make a style more confrontational or more cooperative, and can also increase or decrease the value of resources (for example by contesting the legitimacy of corporate actors).

Yet discourse is not just language. It is also an interactive process. Indeed, it is a set of ideas and an interactive process. The ideational dimension itself divides into two activities: a cognitive activity which enables actors to make sense of reality (drawing on knowledge, policy analysis, information about problems, actors, and resources) and a more normative activity of assessing and judging reality, which thus refers to the world of norms, values, and principles. The interactive dimension of discourse covers the relations between policy-makers at the stage of policy formulation (how they convey meanings, in which institutional forums, through which acceptable norms of behaviour and expression), and how policies are communicated to the public.

3. Making sense of Europeanisation

It is clear that the three different answers to the deep interrogation about the nature of Europeanisation are not mutually exclusive. Most authors consider more than one framework.(14) Let us try to take stock of these complementary frameworks then. One first element in common to the three approaches is the emphasis on domestic change. This is a convenient point of departure. Change or lack of change may be difficult to measure. Some aspects of change are subjectively (and often discursively) defined. Yet domestic change or lack of it provides a clear focus for the analysis of Europeanisation as process. The question to address is how does one know that change is correlated or caused by Europeanisation, and not by other variables. We have already mentioned the danger of pre-judging the impact of Europeanisation. Other variables, like globalisation and domestic politics, may matter more than Europeanisation (on the ‘globalisation versus Europeanisation’ debate see: Verdier and Breen 2001; Fligstein and Merand 2001). The problem is compounded by the strategy of political leaders to disguise globalisation or domestic politics under a discourse of Europeanisation – either by blame-shifting strategies or by using the appeal of Europe to add legitimacy to choices originating at home.

9

One can try to reduce the probability of making mistakes by using simple devices. The first rule is that Europeanisation, in order to produce domestic change, must precede change. This is not as simple as it seems. Europeanisation often covers slow processes of socialisation of domestic elites into European policy paradigms. These processes may co-evolve with national processes of re-definition of policy paradigms. The result is that it is not really clear whether Europeanisation has overtaken domestic processes or just added to them. Take the case of Italian elites in Economic and Monetary Union. Prior to Maastricht, the experience of the Exchange Rate Mechanism had certainly contributed to the socialisation of Italian elites into a framework of sound finance and policy coordination. But during the same years a slow process of re-definition of monetary policy was under way at home, with the divorce between the Treasury and the Bank of Italy and the rise of monetarist ideas at the level of elites (Dyson and Featherstone 1999). In these circumstances, it is not easy to show whether Europeanisation precedes domestic change or not.

In any case, the limitations of the post-hoc, ergo propter hoc prediction are well-known. A process can take place before another without necessarily being correlated with it, let alone be the cause. This suggests a second control, based on mental counterfactuals. Would change have taken place anyway without Europeanisation? Take the example of alcohol monopoly mentioned above. Would Finland and Sweden have re-defined alcohol monopolies as a problem of competition policy without the activism of the European Commission and the treaty obligations on state aids and competition? Now take the example of Italy and EMU. Would Italy have moved to a sound finance paradigm and limited budget deficits without the role played by the European ‘constraint’? Would this country have found its own way to macro-economic stability without the fear of being left out of the Euro-zone?

Counterfactual analysis is a relatively common device in international relations and comparative politics. Yet in order to perform it correctly one has to specify a number of ‘alternative mechanisms’. This is where the third device to control hypotheses enters the scene. It refers to the explicit formulation of rival alternative hypotheses. All too often research on Europeanisation is not clear enough on how globalisation or domestic politics (to mention the classic rival alternative hypotheses) could have produced change. Research designs based on bottom-up approaches provide powerful controls for alternative hypotheses, but only if the alternatives are clearly formulated at the outset. Otherwise there is the risk of examining the evolution of policy and politics at the domestic level without awareness of what else (apart from Europeanisation) can affect the system of interaction. Put bluntly, the risk is one of not ‘seeing’ rival mechanisms at work.

The three views of Europeanisation sit comfortably with these controls. Perhaps some ‘Europeanisation as discourse’ advocates would reject the positivist bias of this methodology. But to be frank, most people looking at Europeanisation as discourse are outside the post-modernist camp and still operate by dint of the logic of scientific inquiry and empirical analysis. Another possible exception comes from the ‘goodness of fit’ school. In this school, the issue of whether ‘Europe’ precedes ‘domestic change’ is not problematic: the study of Europeanisation – they would argue – exists only insofar as there is a clear ‘EU origin’ (policy or politics). Starting from this origin, the analyst tracks down adaptational pressure and investigates lack or presence of change. But most of the studies seem to point towards three important elements:

10

1. There is Europeanisation when the logic of domestic political actors changes. This happens when elements of EU policy-making become a cognitive and normative ‘frame of reference’ (Muller 1995; Surel 2000) and both the logic of action and the logic of meaning are guided by Europe. Think of Europe as the ‘grammar’ of domestic political action.(15)
2. Europeanisation is change both in the sense of responses to EU pressures and in the sense of other usages of Europe which do not presuppose pressure.
3. Europeanisation is a process consisting of complex sequences and time patterns.

The reference to time and sequences is not trivial. One implication of getting the time dimension right is that it can avoid classic pitfalls in causal analysis. There are areas like justice and higher education where many initiatives of cooperation have not taken place in the context of EU processes (Megie and Ravinet 2004). Yet there may be cases in which these initiatives have been facilitated by tentative Commission’s white papers and socialisation processes orchestrated by the European Commission. One may well find that that the Commission failed to create consensus on ‘genuine EU measures’, and cooperation among certain countries proceeded along non-EU tracks. But these tracks would not have been discovered if the Commission had not contributed to agenda setting and socialisation.

The presence of fully-fledged European policies in a certain domain is not a pre-condition for Europeanisation. Irondelle (2003) shows how French military policy has become Europeanised even if EU policies in this area have never taken off. Despite the absence of EU policies, the cognitive and normative policy frames of French policy have been gradually and increasingly re-defined along European lines. Irondelle does not argue exclusively in terms of socialisation. He demonstrates how some pressure and rational calculations have been part of the picture. His explanation is compatible both with rational choice and social constructivism. Another implication is that Irondelle reaches his result only by getting the time dimension and the sequences right, as shown by his process-tracing design (Irondelle 2003: 214). To conclude, Europeanisation does not require the formulation of EU policies, but processes of socialisation are not a sufficient condition for Europeanisation. There may be considerable socialisation without policy change at home. It is only when socialisation to Europe is followed by domestic change than one can speak of Europeanisation.

Socialisation is not even necessary in processes of Europeanisation without EU policies. One can imagine a process of negotiation of EU directives for, say, liberalisation, that remains stymied and controversial for several years. Decisions at the EU level do not materialise, and beliefs remain polarised. Yet policy-makers from countries adverse to liberalisation may feel that they will loose the battle one day, and in order to suffer less later decide to launch a limited process of liberalisation now. Controlled liberalisation ex ante is better than imposed liberalisation ex post. In this case, Europeanisation is not the product of socialisation, but of more rational mechanisms of anticipated reactions.

Another advantage of this approach is that it gets rid of the confusion between up-loading and down-loading. There has been some discussion on whether Europeanisation refers only to the process of downloading policies, or should also include the activities of up-loading preferred policies in the context of EU policy formulation and EU decisions (see Börzel 2002 on up-loading and downlaoding). An understanding of Europeanisation based on the three elements mentioned above (Europe as logic and frame, change, and time-sensitive analysis) does not need to postulate a simplistic life-cycle of policies.

11

Failed up-loading may nevertheless lead to some repercussions and re-orientation at home. Down-loading can make a marginal impact on domestic policies, but there may be more horizontal processes of benchmarking and cognitive re-orientation in the same policy area eventually conducive to change. In this case again, what matters is change at home, of the type that can be captured by bottom-up research designs and careful process-tracing.(16) Box 2 sums up the results of our discussion.

4. How Europeanisation produces domestic change

In order to analyse and measure the power of Europeanisation, one has to think of mechanisms. We have already touched upon one of these mechanisms, the goodness of fit. But now it is useful to proceed more systematically. A number of building blocks for the analysis of mechanisms have already emerged. Europeanisation is about governance and processes. It covers both cases in which EU policies exist, and other cases in which EU-level discussion does not end up with policies, yet domestic actors re-orient their behaviour because ‘Europe’ has become the common grammar. This re-orientation can be explained by preference change within socialisation, but can also be the result of more rational calculation and anticipated reactions. Finally, Europeanisation can be both ‘vertical’ (as shown by the discussion on up-loading and downloading) and ‘horizontal’.

Table 1

12

Table 1 (adapted from Bulmer and Radaelli 2004) introduces the different mechanisms and links them to governance. There are at least three different modes of governance in the EU policy process. They can be described as bargaining, hierarchy, and facilitated coordination. Bargaining and negotiations are not exclusive of policy formulation. There is a lot of bargaining at the level of domestic actors and between governments and the EU when directives are transposed into national legislation and implemented. Compliance can also be seen as negotiation. Actors can bargain over sanctions, but also in terms of interpretation of rules (Tallberg 2002). When actors bargain, the main mechanism of Europeanisation is adaptation as a result of anticipated reaction. Domestic actors formulate expectations of what the result of negotiation may be. They may calculate that it is better for them to change before the process of bargaining is concluded in order to gain some credits at home, or to limit the negative consequences of future decisions. Alternatively, one can see bargaining as a sui generis process of socialisation, and hypothesise that preferences may gradually change in the context of long processes of negotiation. Although the default explanation of change at home is in terms of opportunity structure (last column in table 1), there is some limited mileage for learning.

The second mode of governance is hierarchy. The EU governs hierarchically in two different ways. It can produce policies of positive integration that ‘correct’ the results of the market. Or it can engage in negative integration, by striking down the barriers to the market (Pinder 1968; Scharpf 1999). Whether the EU does this via Commission’s direct activity (competition policy), Council’s legislation, or the jurisprudence of the European Court of Justice is not relevant for our discussion. They are all forms of governance by hierarchy. The key difference between positive and negative integration is the presence or absence of EU models. In positive integration, more often than not the EU produces its own model of how policies should look like. As there is a model, there is some form of adaptational pressure. Not the same pressure for every member state, of course. If pressure if very high, member states will not have the capability to change. The model is far too distant from their policies. One can think of the difficulty of countries like France to follow the EU templates for the liberalisation of electricity. At the opposite pole of the spectrum, if adaptational pressure is very low, because EU and domestic models are compatible, there is no need to change things at home. The result is that major change is predicted in cases of manageable misfit. Overall, this is a vertical mechanism.

When the EU engages with negative integration, there is no major model or EU template. All the EU imposes hierarchically is the ‘level playing field’. Lack of discrimination between residents and non-residents is a typical example. ECJ jurisprudence striking down barriers to trade and freedom of establishment is another. Negative integration is imposed hierarchically on member states, but the final outcome depends on the propensity and capacity of governments to engage in regulatory competition. As such, the mechanism is more horizontal than vertical. It is not at all a matter of fit or misfit. It is a matter of willingness and capability to change domestic policies in order to attract capital and highly-skilled labour.

The problem is that we still do not know much about regulatory competition.(17) The idea of races (to the bottom or to the top) does not help much. Policies and legal systems of different countries do not compete the same way two companies compete in terms of prices. For example, a country may well reduce a tax rate in order to attract corporations, but firms will change their location on the basis of a number of variables, including the quality of the labour force, the presence of major risks such as organised crime, and transport costs. A tax rate does not provide information on how taxes are enforced. A recent review of regulatory competition in Europe (Radaelli 2004) did not find much evidence of this mechanism in Europe, probably because aggressive competition necessitates a transparent and fundamentally open market, and many EU markets are still somewhat opaque and protected.

13

The final mode of governance is facilitated coordination. When the EU uses this mode, it operates like a forum for discussion and a platform for policy transfer. Some forms of facilitated coordination have been around for a while, and not only in the EU, as shown by the literature on soft-law (Schäfer 2004; Snyder 1994). However, facilitated coordination has been re-discovered and improved in connection with the Lisbon agenda to make the EU ‘the most competitive knowledge-based society in the world’ (Joao Rodrigues 2004).

Under facilitated cooperation, the EU organises cooperation among member states, but does not produces European legislation. It produces opportunities for learning – the default explanation of Europeanisation for this mode (table 1, last column).

This mode of governance is typical of benchmarking exercises. For example, the directors of better regulation programmes of the EU meet routinely to discuss good practice in public management reforms and to assess the quality of regulation at the domestic and EU level. The aim is to facilitate learning from each other and diffusion of good practice. With the open method of coordination, governance by learning, socialisation, and policy transfer have been codified and somewhat institutionalised. There is a lively debate on whether the open method of coordination can deliver or not (Borras and Greve 2004; Ferrera et al. 2002, Radaelli 2003b, Scharpf 2002, Wincott 2003). In terms of Europeanisation, it is difficult to show that some domestic policy changes are a result of the open method of coordination rather than the consequences of domestic politics. It is not easy to measure the power of soft law, learning, and peer pressure. There is more research at the theoretical level than systematic empirical evidence. When available (Borras and Greve 2004; De la Porte and Pochet 2002; Bonoli and Bertozzi 2002, Radaelli 2003c), empirical research shows that in some policies Europeanisation by facilitated coordination proceeds by ideational convergence, that is, changes of the cognitive and normative frames used by domestic policy-makers. This does not mean convergence of domestic policies, of course. Indeed, one sceptical line in the debate is that the open method of coordination has created a community of discourse, but no real change in labour market, pensions, and social inclusion – unless these changes are pushed by more concrete domestic processes of politics and policy-making.

Indeed, the literature on policy transfer tells us that effective transfer requires more than governance architectures like the open method of coordination. It also requires robust networks of stakeholders that facilitate the adoption of new policies at home, a strong civil society, and administrative-political capability to consciously modify, edit, and adapt foreign experience to national circumstances (Jacoby 2000, Rose 2002). Although the open method of coordination has produced momentum for the discussion of political sensitive reforms of the welfare state, its potential in terms of Europeanisation is limited when the domestic coalitions for reforms are weak, or the stakeholders do not engage creatively with the imported institutional models.

5. What does the evidence tell us?

To report on the empirical results achieved by the literature is not an easy task. The results of the major projects completed by teams of scholars are of course available (Cowles et al. 2001, Héritier et al. 2001, Featherstone and Radaelli 2003; Hix and Goetz 2000; Knill 2001; for Germany see Dyson and Goetz 2003). But they vary depending on the different frameworks employed by individual authors. As mentioned above, Europeanisation can be approached as governance, institutionalisation, and discourse.

14

Further, the notion of impact embraced by authors like Risse and Börzel refers to vertical mechanisms of adaptation. For them, evidence should point out to change as result of adaptational pressure. They approach Europeanisation as explanans rather than explanandum. By contrast, authors interested in facilitated coordination and ‘framing mechansims’ (Knill and Lehmkuhl 2002) search for evidence that does not necessarily arise out of adaptational pressure. In more discursive approaches, Europeanisation is investigated as a problem (that is, an explanandum). Mörth (2003) and Kallestrup (2002), for example, are interested in the process of editing and translating EU ideas and paradigms into national policy.

Five results stand out, however. Firstly, the evidence on the Europeanisation of public policy is more robust than the evidence on the Europeanisation of political competition, state structures, and the polity domain. Some policies are more impermeable than others to ‘Europe’ (Bulmer and Radaelli 2004), as shown by the difference between competition and environmental policy, on the one hand, and higher education and justice, on the other.

One limitation of the current literature is that it works in a ‘compartimentalised’ manner by considering only one dimension at the time and ignoring the others. The most exciting projects, however, have shown that the three dimensions (politics, policy, and politics) interact, often in subtle and indirect ways (Giuliani 2001; Gualini 2003; Ladrech 2002). This is the second result. Europeanised policies change state-society relations (see Cowles et al. on pressure groups), empower technical bureaucracies, change the institutions of economic policy (Dyson 2002), transform the cultural and organizational ‘governance software’ of departments (Jordan 2003), transform the operating environment for party politics (Ladrech 2002), and enable domestic actors to tilt the balance of power between regions and central government (Bull and Baudner 2004). The adage that ‘policies change but politics and polity do not’ is obsolete. Europeanisation has therefore an important message for those who are interested in how governance is changing in Europe.

Thirdly, and (to some extent) consequently, the linkages between policies and the political-polity dimensions lead to an increasing stickiness of the national (and often regional) and European levels. There is no fusion of the various levels, but if Europe becomes the grammar of domestic political action and if the external environment becomes endogenous, the different levels may collapse in one dimension of Europeanised political action.

However, Europeanisation is not convergence. Even in cases of convergence, the pattern is more one of clustered convergence (Börzel 2002; Goetz 2002) than of uniform convergence across Europe. The idea is that countries with the same structural characteristics respond with similar strategies to the opportunities and constraints provided by Europeanisation. However, it is difficult to generalise across policy areas, as the same country often responds in dramatically different ways to Europeanisation, depending on the constellation of actors and resources available in policy areas.

Convergence should be measured along a continuum. At a minimum, convergence means that domestic policy-makers share ‘European’ vocabularies. If Europeanisation produces a convergence of paradigms and ideas of good practice, one can also speak of ideational convergence (Radaelli 2003c). The next stage is convergence at the level of decisions. When similar decisions are implemented in a relatively uniform way, the degree of convergence increases. Finally, one can imagine the case of convergence in outcomes. Future research should discuss more systematically types of convergence. More importantly still, it should expose the implicit theories of implementation contained in claims about convergence. For example, empirical studies routinely confirm that even hierarchical modes of governance in the EU do not produce policy convergence.

15

But to imagine convergence of outcomes in the EU requires a theory of implementation close to the ideal-type of perfect, rational, chain-of-command administration. This is tantamount to ignoring the last thirty years of implementation research. Another example of the misunderstandings about convergence is provided by the discussion on the limitations of facilitated coordination. If one argues that the open method of coordination is failing because it is not delivering convergence, one is neglecting the basic fact that facilitated coordination acknowledges diversity and different models of capitalism up-front. More generally, the whole issue of convergence in the EU should be framed in the context of the principle of subsidiarity.

To conclude the discussion of convergence, this is not a top-down process. In his analysis of competition policy in the UK and the EU, Zahariadis (2004) finds that even pace-setters like the UK may not be interested in up-loading their model (i.e., they may not be interested in contributing to the development of EU policy). The resulting incongruence between the EU and domestic model does not necessarily produce change (much less convergence) at the domestic level (Zahariadis 2004:69). The EU and the UK models of competition policy have not converged for years. When they did converge, this was not the result of the pressure from above, but the outcome of domestic political processes. So, although the EU may provide an ‘activating stimulus’ for convergence, the actual process is driven by domestic politics (Zahariadis 2004:70)..

Fifthly, empirical research has begun to examine the impact of Europeanisation in the context of other possible pressures for convergence, such as globalisation and international policy diffusion. Levi-Faur (2004) has argued that the impact of Europeanisation in electricity and telecommunications decreases when one performs these controls, for example by comparing patterns of privatisation in the EU with similar patterns in Latin America. However, although the overall direction of policy change may not be predicted by Europeanisation, the latter may explain the scope of change and final outcomes – a point on which the discussion is fully open.

6. Conclusions: Where do we go from here?

Let us go back to the initial question about progressive problem shifts and set of puzzles. What is the potential of Europeanisation as research agenda? It provides a valuable shift of focus by generating a set of questions for the analysis of the interplay between different levels of governance. Indeed, Europeanisation may contribute to our understanding of the changing nature of governance and the state by endogeneising international governance in the models of domestic politics and policy. However, Europeanisation is not a new theory. It is a way to organise our concepts and to contribute to the normalisation of political science (Hassenteufel and Surel 2000) by drawing systematically on comparative politics, international political economy, international relations, and policy analysis. Europeanisation is compatible with the different angles suggested by the disciplines of politics and international relations (Featherstone and Radaelli 2003). Accordingly, it does not need to create its own ad hoc theoretical models. This is an advantage, as European Studies are often accused of having abused of ad hoc theory, rather than drawing on comparative politics (Hix 1999).

Having clarified in what sense Europeanisation is a research agenda, this article has shown the set of puzzles posed by research in this field. Methodologically, the most promising puzzles revolve around the notion of impacts, how to control for rival alternative hypotheses, and how to integrate agency in the explanation of change. The preliminary responses given by the literature point towards complex notions of causality (Elster 1989)(18) and comprehensive sociology (Jacquot and Woll 2003). One challenge for future research is how to embrace these complex notions and be ‘clear enough to be wrong’ (Sabatier 2000).

16

To illustrate, if Europeanisation is a frame of reference, what are the empirical indicators that show its existence or absence? This raises the issue of operationalisation. So far, the literature has produced more metaphors than systematic attempts to operationalise variables(19) – and without operationalisation ‘large n’ comparisons are impossible.

Turning to substantive results, most of the current focus is on the mechanisms of Europeanisation and whether they are producing substantive change, where and how. Future research will have to get to grips with the limitations of each and every mechanism. The goodness of fit has to account for agency and the transformative power of discourse to gain more precision. The mechanism of regulatory competition is still un-explored, and perhaps too obsessed with races and their direction, rather than being concerned with how to explain real-world processes of competition. More interdisciplinary research is needed in this field. Finally, mechanisms of facilitated coordination should clarify the conditions for learning and how the latter produces change. The emphasis on learning should not obfuscate the fact that political learning can be conflictual and asymmetric. Additionally, more research is needed on how soft and hard mechanisms (for example, governance by hierarchy and facilitated coordination) interact in the same policy area. Convergence and compliance are still relatively unexplored concepts, and new research shows how the problem-shifts suggested by Europeanisation may help (see Zahariadis 2004 on bottom-up convergence and Tallberg 2002 on compliance).

Finally, political scientists need to bring politics back into the analysis of Europeanisation – a point made by Peter Mair (2004). By focusing on cleavages, dimensions of conflict, and patterns of contestation, Europeanisation can inform on the nature of the EU and its member states as a political system, with its own processes of boundary building, representation, and political structures. Arguably, this is the direction in which Europeanisation can penetrate the nature of EU politics much better than the major theories of European integration.

With all its limitations, Europeanisation provides a fascinating perspective on how governance is changing. As such, it is a valuable tool for political science research. It is not a ‘solution’ (this time in the sense of providing off-the-shelf explanations). It is a challenging, exciting ‘problem’.

References

Adcock, R. and D. Collier (2001) ‘Connecting ideas with facts: The validity of measurement’, American Political Science Review 95(3), September: 529-546.

Bache, I. (2003) ‘Europeanisation: A governance approach’ Paper to be presented at the EUSA 8th International Biennial Conference, Nashville, March 27-29, 2003.

Borras, S. and B. Greve (2004) (Eds.) ‘The open method of coordination in the European Union’, Special issue of Journal of European Public Policy, 11(2).

Börzel, T. A. (2002) ‘Pace-Setting, Foot-Dragging, and Fence-Sitting: Member State Responses to Europeanization’, Journal of Common Market Studies 40 (2): 193-214.

Börzel, T. A. (2004) ‘How the European Union interacts with its member states’ in C. Lequesne and S. Bulmer (Eds.) Members States and the European Union, Oxford, Oxford University Press.

Börzel, T. and T. Risse (2003) ‘Conceptualising the domestic impact of Europe’ in K. Featherstone and C.M. Radaelli (Eds.) The Politics of Europeanisation Oxford: Oxford University Press, 57-80.

Bull, M. and J. Baudner (2004) ‘Europeanisation and Italian policy for the Mezzogiorno’, Journal of European Public Policy, 11, Forthcoming.

Buller, J. and A. Gamble (2002) ‘Conceptualising Europeanisation’, Public Policy and Administration, 17(2) Summer: 4-24.

Bulmer, S. and W. Paterson (1987) The Federal Republic of Germany and the European Community, HarperCollins.

Bulmer, S. and C. M. Radaelli (2004, forthcoming) ‘The Europeanisation of public policy’ in C. Lequesne and S. Bulmer (Eds.) Members States and the European Union, Oxford, Oxford University Press.

Busch, A. (2004) ‘National filters: Europeanisation, institutions, and discourse in the case of banking regulation’, West European Politics, 27(2): 310-333.

Braun, M. (2003) Review of Transforming Europe and The Europeanisation of National Administrations, Perspectives on Politics, 1(1) March: 206-207.

Cowles, M.G., Caporaso, J. and T. Risse (2001) (Eds.) Transforming Europe: Europeanisation and Domestic Change, Ithaca and London: Cornell University Press.

De la Porte, C. and P. Pochet (2002) (Eds.) Building Social Europe through the Open Method of Coordination, Peter Lang, Brussels.

Dyson, K. and K. Featherstone (1999) The Road to Maastricht: Negotiating Economic and Monetary Union, Oxford, Oxford University Press.

Dyson, K. (2002) (ed.) The European State and the Euro, Oxford, Oxford University Press.

Dyson, K. and K. Goetz (2003) (Eds.) Germany, Europe and the Politics of Constraint, Oxford, Oxford University Press.

Elster, J. (1989) Nuts and Bolts for the Social Sciences, Cambridge, Cambridge University Press.

Falkner, G. (2003) Comparing Europeanisation effects: From metaphor to operationalisation, European Integration online Papers, Vol. 7 No. 13 http://eiop.or.at/eiop/texte/2003-013a.htm.

Featherstone, K. (2003) ‘In the name of Europe’, in K. Featherstone and C.M. Radaelli (eds.) The Politics of Europeanization, Oxford, Oxford University Press, 3-26.

Featherstone, K. and C. Radaelli (2003) (Eds.) The Politics of Europeanisation, Oxford, Oxford University Press.

Ferrera, M., Matsaganis, M. and S. Sacchi (2002) ‘Open coordination against poverty: The new EU ‘social inclusion process’ Journal of European Social Policy 12(3): 227-239.

Fligstein, N. and F. Merand (2001) ‘Globalization or Europeanization? Evidence on the European Economy Since 1980’ Working paper wps-2001-02, Center for Culture, Organizations and Politics, University of California, Berkeley. http://repositories.cdlib.org/cgi/viewcontent.cgi?article=1001&context=iir/ccop.

Goetz, K. (2002) ‘Four worlds of Europeanisation’, Paper prepared for the ECPR Joint Sessions of Workshops, Turin, Italy, 22-27 March 2002.

Giuliani, M. (2001) ‘Europeanization and Italy: A bottom-up process?’ in K. Featherstone and G. Kazamias (Eds.) Europeanization and the Southern Periphery, London, Frank Cass: 47-72.

Giuliani, M. (2003) ‘Europeanisation in comparative perspective: Institutional fit and national adaptation’ in K. Featherstone and C.M. Radaelli (Eds.) The Politics of Europeanisation, Oxford, Oxford University Press, 134-155.

Giuliani, M. (2004) ‘Europeizzazione come istituzionalizzazione: questioni definitorie e di metodo’, Rivista Italiana di Politiche Pubbliche, 1: 141-161. Pre-Print version at http://www.urge.it/papers/WP_4_2004.pdf.

Grabbe, H. (2003) ‘Europeanization goes East’, in K.Featherstone and C.M. Radaelli (Eds.) The Politics of Europeanization, Oxford, Oxford University Press, 303-327.

Gualini, E. (2003) Multi-level Governance and Institutional Change. The Europeanization of Regional Policy in Italy, Aldershot, Ashgate.

Hassenteufel, P. et Surel, Y. (2000) ‘Des politiques publiques comme les autres? Construction de l'objet et outils d'analyse des politiques européennes’, Politique Européenne, 1: 8-24.

Hay, C. and B. Rosamond (2002) ‘Globalisation, European integration and the discursive construction of economic imperatives’, Journal of European Public Policy 9(2): 147-167.

Héritier, A., Kerwer, D, Knill, C., Lehmkuhl, D., Teutsch,M. and Douillet, A-C. (2001) Differential Europe. The European Union Impact on National Policymaking Lanham, MD: Rowman and Littlefield.

Hix S. (1999) The Political System of the European Union, Macmillan, Basingstoke.

Hix, S. and K. Goetz (2000) (Eds.) Europeanised Politics? European Integration and National Political Systems, London, Frank Cass.

Irondelle, B. (2003) ‘Europeanization without the European Union? French military reforms 1991-1996’, Journal of European Public Policy, 10(2) April: 208-226.

Jacoby, W. (2000) Imitation and Politics: Redesigning Modern Germany, Ithaca: Cornell Univ. Press.

Jacquot, S. and C. Woll (2003) ‘Usage of European integration: Europeanisation from a sociological perspective, European Integration online Papers, Vol. 7 No.12 http://eiop.or.at/ eiop/texte/2003-012a.htm

Joao Rodrigues, M. (2004) European Policies for a Knowledge Economy, Cheltenham, Edward Elgar.

Jordan, A. (2003) The Europeanization of British Environmental Policy, Basingstoke: Palgrave.

Kallestrup, M. (2002) ‘Europeanisation as a discourse: Domestic policy legitimization through the articulation of a need for adaptation’, Public Policy and Administration, 17(2) Summer: 110-124.

Katzenstein, P. (1997) (ed.) Tamed Power: Germany in Europe, Ithaca and London, Cornell University Press.

Knill, Christoph (2001) The Transformation of National Administrations in Europe. Patterns of Change and Persistence Cambridge, Cambridge University Press.

Knill, C. and D. Lehmkuhl (1999) How European matters: different mechanisms of Europeanization, European Integration online Papers Vol. 3 No. 7. http://eiop.or.at/eiop/texte/1999-007.htm

Knill, C. and D. Lehmkuhl (2002) ‘The national impact of European Union regulatory policy: Three Europeanization mechanisms’ European Journal of Political Research, 41(2) March: 255-280.

Kohler-Koch, B. and R. Eising (1999) (Eds) The Transformation of Governance in Europe, London: Routledge.

Kurzer, P. (2001) Markets and Moral Regulation. Cultural Change in the European Union, Cambridge, Cambridge University Press.

Ladrech, R. (2002) ‘ Europeanization and political parties’, Party Politics, 8(4): 389-403.

Lakatos, I. (1978) The Methodology of Scientific Research Programmes: Philosophical Papers, Volume 1 (edited by John Worrall and Gregory Currie), Cambridge, Cambridge University Press.

Levi-Faur, D. (2003) ‘On the net impact of Europeanization: The EU’s telecoms and electricity regimes between the global and the national’, Comparative Political Studies, 37(1) February: 3-29.

Mair, P. (2001) ‘The limited impact of Europe on national party systems’, West European Politics, 23(4 )27-51.

Mair, P. (2004) ‘The Europeanisation dimension’, Journal of European Public Policy, 11(2) : 337-348.

Megie, A. and P. Ravinet (2004) Contrainte de coopération intergouvernementale et processus d’européanisation, La construction des espaces européens de l’enseignement supérieur et de la justice, Paper delivered to the conference on Europeanisation of Public Policies and European Integration, IEP- Paris, 13 February 2004.

Mörth, U. (2003) ‘Europeanization as interpretation, translation, and editing of public policies’, in K. Featherstone and C.M. Radaelli (Eds.) The Politics of Europeanization, Oxford, Oxford University Press: 159-178.

Muller, Pierre (1995) ‘Les politiques publiques comme construction d’un rapport au monde’, in Faure, Alain, Pollet, Gilles et Warin Philippe (Ed.) La construction du sens dans les politiques publiques, débats autour de la notion de référentiel, Paris, L’Harmattan : 153-179.

Muller, P. (2000) ‘L’analyse cognitive des politiques publiques : vers une sociologie politique de l’action publique’, Revue Française de Science Politique, 50 (2), April: 189-207.

Muller P., and S. Rouault (1997) "’La mise en œuvre de l'initiative emploi, une grammaire européenne de l'expérimentation sociale’, Cultures et conflits.

Murphy, D.D. (2004) The Structure of Regulatory Competition, Oxford, Oxford University Press.

Olsen, J.P. (2002) ‘The many faces of Europeanization’ Journal of Common Market Studies 40(5) 921-952.

Page, E.C. (2003) ‘Europeanization and the persistence of administrative systems’, in J. Hayward and A. Menon (Eds.) Governing Europe, Oxford, Oxford University Press, 162-176.

Pinder, J. (1968), ‘Positive integration and negative integration: some problems of economic union in the EEC’, The World Today 24(3): 89-110.

Puchala, D. (1972) ‘Of blind men, elephants, and European integration’, Journal of Common Market Studies 10(2): 267-284.

Radaelli, C.M. (1999) Harmful tax competition in the European Union: Policy narratives and advocacy coalitions, Journal of Common Market Studies, 37(4), December: 661-82.

Radaelli, C.M. (2003a) ‘The Europeanization of public policy’, in K. Featherstone and C.M. Radaelli (Eds.) The Politics of Europeanization, Oxford, Oxford University Press: 27-56.

Radaelli, C.M. (2003b) The open method of coordination: a new governance architecture for the European Union?, Stockholm, Swedish Institute of European Policy Studies. http://www.epin.org/pdf/RadaelliSIEPS.pdf.

Radaelli, C.M. (2003c) ‘The code of conduct in business taxation: open method of coordination in disguise’, Public Administration, 81(3): 513-531.

Radaelli, C.M. (2004) (ed.) ‘The puzzle of regulatory competition’, Journal of Public Policy, 24(1): 1-23.

Rhodes, R.A.W. (1997) Understanding Governance. Policy Networks, Governance, Reflexivity and Accountability. Buckingham and Philadelphia: Open University Press.

Risse, Thomas, James Caporaso, and Maria Green Cowles (2001) ‘Europeanization and Domestic Change. Introduction’, in M. Cowles, J. Caporaso and T. Risse (Eds.) Transforming Europe: Europeanization and Domestic Change, Ithaca NY: Cornell University Press, 1-20.

Rometsch, D. and W. Wessels (eds) The EC and Its Member States, Manchester University Press.

Rose, R. (2002) Ten steps in lesson-drawing: A guide to learning from abroad, RSCAS working paper 2002-05, EUI, Florence.

Sabatier, P. (2000) ‘Clear enough to be wrong’, Journal of European Public Policy, 7(1): 135-140.

Salgado Sanchez, R. and C. Woll (2004) L’Européanisation et les acteurs non-Etatiques, Paper delivered to the conference on Europeanisation of Public Policies and European Integration, IEP- Paris, 13 February 2004. http://www.sciences-po.fr/recherche/forum_europeen/prepublications/Papie r_2_Woll_Salgado.pdf.

Schäfer, A. (2004) ‘A new and effective form of governance? Comparing the OMC to mulit-lateral surveillance by the IMF and the OECD’, Paper prepared for the 2004 Conference of Europeanists, Chicago, 11-13 March 2004. http://www.europanet.org/conference2004/papers/D6_Schaefer.pdf.

Scharpf, F.W. (1999) Governing in Europe: Effective and Democratic?, Oxford: Oxford University Press.

Scharpf, F.W. (2001) ‘What have we learned? Problem-solving capacity of the multilevel European polity’, Max Planck Institute Working Papers Series, Working Paper 01/4, Cologne.

Scharpf, F.W. (2002) ‘The European social model: coping with the challenges of diversity’, Journal of Common Market Studies 40(4): 645-670.

Schimmelfennig, Frank (2001). “The Community Trap: Liberal Norms, Rhetorical Action, and the Eastern Enlragement of the European Union, International Organization 55(1), Winter: 47-80.

Schmidt, V.A. (2002) ‘Does discourse matter in the politics of welfare adjustment?’, Comparative Political Studies, 35(2) March: 168-193.

Schmidt, V.A., and C.M. Radaelli (2004) ‘Policy change and discourse in Europe: Conceptual and methodological issues’, West European Politics, 27(2): 183-210.

Snyder, F. (1994) ‘Soft Law and Institutional Practice in the European Community’ in Stephen Martin (Ed.). The Construction of Europe: Essays in Honour of Emile Noël, Dordrecht: Kluwer Academic Publishers.

Stone Sweet, A., Sandholtz, W., and N. Fligstein (2001) (Eds.) The Institutionalization of Europe, Oxford: Oxford University Press.

Surel, Y. (2000) 'The role of cognitive and normative frames in policy-making', Journal of European Public Policy, 7(4) October: 495-512.

Tallberg, J. (2002) ‘Paths to compliance: Enforcement, management, and the European Union’, International Organization, 56(3) Summer: 609-643.

Thatcher, M. (2004) ‘Winners and losers in Europeanisation: Reforming the national regulation of telecommunications’, West European Politics, 27(2): 284-309.

Ugland, T. (2003) ‘Adaptation and integration through policy re-categorization’ Journal of Public Policy, 23(2): 157-170.

Verdier, D and R. Breen (2001), Europeanization and globalization: Politics against markets’ Comparative Political Studies 34(3): 227-262.

Vogel, D. (1995) Trading up. Consumer and Environmental Regulation in a Global Economy, Harvard University Press, Cambridge.

Wincott, D. (2003) ‘The idea of the European social model: limits and paradoxes of Europeanisation’, in K. Featherstone and C.M. Radaelli (Eds.) The Politics of Europeanisation, Oxford, Oxford University Press, 279-302.

Winn, N. and E. Harris (2003) ‘Conclusion: The story of a death foretold?’, Perspectives on European Politics and Society, 4(1): 167-170.

Zahariadis, N. (2004) ‘European markets and national regulation: Conflict and cooperation in British competition policy’, Journal of Public Policy, 24(1): 49-73.

Endnotes

(*)I am grateful to Michael Nentwich and two anonymous reviewers for their comments and suggestions. Michelle Cini and Angela Bourne kindly provided comments on the first draft of this paper. The usual disclaimer applies. This paper was written in the context of my contribution to the Palgrave Guide to European Studies edited by Michelle Cini and Angela Bourne (Forthcoming, 2005).

(1) For Lakatos, in degenerative programmes theories are fabricated only to accommodate known facts. By contrast, progressive research programmes predict novel facts. For a simple introduction to these concepts see the famous talk on science and pseudo-science originally broadcast on 30 June 1973 (text accessible at http://www.lse.ac.uk/collections/lakatos/scienceAndPseudoscienceTranscript.htm). The text also appears in the introduction to Lakatos (1978).

(2) I have come across sceptical readers arguing that Europeanisation is just a waste of time. They point to some ‘Europeanisation’ articles and papers which either re-brand honest empirical research on EU policies as ‘Europeanisation’ or (worse) literally regurgitate classic concepts and ideas about political integration and implementation originally formulated by lawyers, policy analysts, and political scientists. In this article I look at the added value- net of bandwagon effects. As shown by Mair (2004: 337) the number of articles on this topic is increasing. A consequence is that there are some bandwagon effects common to all topics when they gain popularity. Some articles re-brand well-known features of European integration as ‘Europeanisation’. It takes a decade or so for a new field of research to ‘settle down’ and clarify concepts and research agenda. Thus, although bandwagon effects exist, it would be unfair to argue that there is nothing new in Europeanisation. At least some work in this area has produced challenging propositions and fresh ideas. Mair (2004: 346), who certainly cannot be accused of following the Europeanisation bandwagon, contrasts the innovative work on Europeanisation and European integration with his past experience of reading ‘dull descriptive, and a-theoretical’ papers on European politics and policy.

(3) The claim was originally made by Hassenteufel and Surel (2000).

(4) Throughout the article, I will borrow on Gualini’s idea of Europeanisation as explanandum (Gualini 2003).

(5) With the qualification that hollowing out from above can also be the result of processes of regionalisation induced by European integration.

(6) For evidence of a Copernican revolution on how political scientists assess the role of Germany as European model of governance see Dyson and Goetz (2003).

(7) This point is made by Michael Braun in his review of two studies on Europeanisation (Braun 2003).

(8) The classic example is the series of volumes published by Pinder in the early 1990s on the assessment of member states’ membership of the European Community. See also Rometsch and Wessels (1996).

(9) The identification of explanans and explanandum as ‘solution’ and ‘problem’ is quite common. See http://www.iscid.org/encyclopedia.

(10) Other examples of bottom-up research designs include Bull and Baudner (2004), Busch (2004), Thatcher (2004), and Zahariadis (2004).

(11) In this article I do not deal with the thorny issue of how to measure change. Change is often in the eye of the beholder, and what looks like inertia at time n may well become deep transformation at time t+n. My views on this topic are in Radaelli (2003a: 37-40).

(12) See Börzel and Risse (2003) for a systematic exploration of this approach.

(13) Self-validation is a classic feature of institutionalisation processes in complex organisation. It means that an organisation is valued (positively or negatively) per se, beyond the implications for individual members.

(14) Schmidt and Radaelli (2004) even use hybrid terminology like ‘discursive institutionalism’.

(15) I borrow this expression from my conversations with Pierre Muller. See Muller (2000) and Muller and Roualt 1997).

(16) It should be observed that bottom-up research designs have nothing to do with the up-loading vs. down-loading discussion.

(17) On international regulatory competition see Vogel (1995) and Murphy (2004).

(18) In turn, these complex notions expose the limitations of research designs oriented towards tests of simple, linear causal hypotheses.

(19) For recent work on operationalisation see Falkner (2003) and Giuliani (2003).