Gaby Umbach, The Concept of Europeanisation: Why ‘Brussels’ Matters, Why ‘Back Home’ Matters and Why It Is a Matter of Perspective in:

Gaby Umbach

Intent and Reality of the European Employment Strategy, page 79 - 111

Europeanisation of National Employment Policies and Policy-Making?

1. Edition 2009, ISBN print: 978-3-8329-4128-4, ISBN online: 978-3-8452-1247-0

Series: Studies on the European Union, vol. 1

Bibliographic information
Theoretico-Empirical Frame of Reference 79 2.1.2 The Concept of Europeanisation: Why ‘Brussels’ Matters, Why ‘Back Home’ Matters and Why It Is a Matter of Perspective “Europeanisation is not a new theory. It is an approach that enables us to orchestrate existing 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” (Radaelli 2004a:5; cf. Auel 2005:313; Eising 2003:412; Olsen 2002:944). Given that European integration studies have, for a long time, dedicated lots of “intellectual energy in seeking to understand the ‘nature of the beast’, that is, the nature of European integration, political scientists have .. realised that a EU political system is in place, produces decisions, and impacts on domestic policies in various guises” (Bulmer/Radaelli 2004:3; cf. Radaelli 2004a:2). This turn pays tribute to the fact that the increasing differentiation of the European polity towards a system of EMLG is caused by the growing transfer of competences and resources from the (sub-)national levels to the EU level that itself is accompanied by and impacts on changing modes of governance that, in most different ways, strengthen decisionmaking powers of the EU (cf. chapter Simultaneously, this development led to an increasing influence of European provisions and structures in domestic political arenas of EU member states. With the expansion of political interactions across national borders, that is, with the constant increase of supranational co-operation between EU member states, the impact of the Union on its member states’ polity, politics, and policies became a flourishing field of European integration studies. Therefore, the ever-closer integration of EU member states within an EMLG system, with a tendency towards fusion of national instruments (Wessels 1992, 2000), became increasingly important to the analysis and explanation of political processes within the EU. Within this strand of analysis, the EMLG approach conceptualises “the EU as an independent, autonomous source of influence on Member State politics, either directly via the actions of European-level actors, or indirectly through institutions and policies that alter the distribution of resources and capabilities” (Anderson 2002:799). Consequently and parallel to the EMLG-focused analysis of changes in governance, Europeanisation became a prominent analytical concept, explanative element and “a growth industry covering a broad research agenda” (Jachtenfuchs/Kohler-Koch 2004a:109; cf. Mörth 2003:159; Olsen 2002:921). Based on this general perception, the present sub-chapter introduces the Europeanisation concept as the analytical backbone of the empirical analysis of the impact of the EES on domestic employment policy co-ordination structures and policies in the UK and Germany. As such, it not only provides for an overview on different academic conceptualisations. It also compiles relevant elements for designing the analytical matrix for this book’s empirical study (cf. table 7) and collects insight into relevant intervening factors and variables. By doing so, the sub-chapter lays the ground for conceptualising and measuring the Europeanisation impact of the EES. Theoretico-Empirical Frame of Reference 80 Within its first part, a general overview on the Europeanisation approach, including its analytic-conceptual benefits and shortcomings, is given. The second part deals with central characteristics of the approach and develops the analytical matrix for the latter empirical research. Finally, the third part offers an overview on aspects of domestic change through Europeanisation, including a categorisation of different facets of policy and institutional change. Multiple Definitions and Latent Concept Stretching – An Analytical ‘Wunderkind’ within a Conceptual ‘Tower of Babel’ On a very abstract level, Europeanisation can be defined in three broad conceptual ways. Firstly, given that “Europeanisation is a process of changing understandings of governance in Europe” (Radaelli 2004a:6), it can be understood as the development of governance. As such, it impacts on change and adaptation of common ideas of governance–such as public-private partnership, policy networks, multilevel policy-making (cf. ibid.). In a second broad perspective, Europeanisation can be conceptualised as a process of institutionalisation triggered off by adaptation pressure in “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” (ibid.; cf. Andersen 2004:14; Mörth 2003:159; Olsen 2002:929). Finally, Europeanisation can be viewed as discourse in terms of ideas, language and interaction, able to transform European and national polity, politics and policy dimensions given that “[p]olicy-makers and stakeholders construct Europe through language and discourse” (Radaelli 2004a:8). Although these three conceptualisations focus on a broad variety of aspects related to European integration, Europeanisation “is not a simple synonym for European regional integration or even convergence, though it does overlap with aspects of both” (Featherstone 2003:3; cf. Bulmer/Radaelli 2004:5). However, the proximity of both allows for the (critical) question of “whether the former offers anything new to the analysis of the EU” (Howell 2004a:1; cf. Auel 2005:296; Eising 2003:394). In contrast to classical definitions of European political integration as a process of evolution and development of a new supranational political entity with a distinct institutional set up (cf. inter alia Haas 1958), Europeanisation is perceived as “a set of post-ontological puzzles” (Radaelli 2004a:2, cf. ibid. 2003a:33), focusing on EU polity, politics and policies ‘at work’ and national responses to European integration rather than just on the emergence of the EU polity itself (cf. Auel 2005:294ff.). Therefore and in order to clarify the relation of both, the conceptual lucidity of this study is, in a first step, sought to be increased by siding with the basic differentiation between European integration as the emergence and evolution of the EU’s polity and its institutions and Europeanisation as the impact of European integration on domestic political systems as well as on the EU system (cf. Eising 2003:394ff.). Given these introductory remarks on broader definitions of Europeanisation, it becomes clear that ever since its inception, the Europeanisation approach has been Theoretico-Empirical Frame of Reference 81 “a fashionable but contested concept” (Olsen 2002:921). With it, a rapidly growing acquis académique on the development of the EU’s political system (cf. Featherstone 2003:14; Lawton 1999) as well as on the impact of European integration on the political systems of EU member states came to the fore of EU-related policy analysis and studies.19 The former is closer to the original definition of European integration and particularly incorporates the analysis of the “development at the European level of distinct structures of governance, that is, of political, legal, and social institutions associated with political problem-solving that formalize interactions among the actors, and of policy networks specializing in the creation of authoritative European rules” (Risse/Cowles/Caporaso 2001:3; cf. Auel 2005:293; Börzel 2003:1; Börzel/Risse 2000:1; Bomberg/Peterson 2000:3; Cowles/Risse 2001:218; Mörth 2003:159). Complementary to this focus, the latter perspective especially comprises studies on domestic change caused by European provisions20, including policy change and learning (cf. Börzel/Risse 2003; Jordan 1998; Mazey 1998; Radaelli 1997 and 2003a), institutional change (cf. Cole 2001; Cowles/Caporaso/Risse 2001; Ladrech 1994; Radaelli 2003a:30) as well as adaptation of public administrations (cf. Ágh 1999; Bulmer/Burch 2001; Goetz 2001 and 2000; Lippert/Umbach 2004a, 2004b, 2005; Lippert/Umbach/Wessels 2001; Wessels 1998). In these studies, “Europeanization is conceptualised as the process of downloading European Union (EU) regulations and institutional structures to the domestic level” (Howell 2004a:1). Therefore (and in line with the analytical focus of this study), in most of the cases applied in this understanding, Europeanisation increasingly related to the analysis of the impact of European law, rules, and provisions on national political systems, including adaptation of institutional arrangements and domestic policies21. Regardless of this broad focus of application in EU studies, “this conceptualisation has [yet also] been criticised for its positivistic nature and extended in the literature in terms of up-loading, policy transfer, shared beliefs, fit/misfit etc.” (Howell 2004a:1). In line with this ‘bifocal’ approach in Europeanisation studies, out of Olsen’s five possible concepts of Europeanisation, two are of relevance here. Firstly, the idea that Europeanisation covers the analysis of the further development of distinct formal institutions at supranational level, including “centre-building with a collective action capacity, providing some degree of coordination and coherence” (Olsen 2002:923). Secondly, the definition that it focuses on the central “penetration of national systems of governance … [which] involves 19 Cf. Börzel 2002, 2001b; Börzel/Risse 2000:1; Bulmer/Lesquenes 2002; Bulmer/Radaelli 2003:1; Featherstone 2003:3 and 7ff.; Featherstone/Radaelli 2003; Goetz/Hix 2000; Jachtenfuchs/Kohler-Koch 2003:34; Kohler-Koch 2000; Maurer/Wessels/Mittag 2000; Olsen 2002; Radaelli 2004a; Risse/Cowles/Caporaso 2001; Wessels/Maurer/Mittag 2002. 20 Cf. Börzel/Risse 2003; Buller/Gamble 2002:17; Cowles/Caporaso/Risse 2001; Featherstone 2003:3; Giuliani 2003; Goetz/Hix 2000; Ladrech 1994:69; March/Olsen 1995; Olsen 1996. 21 Cf. Anderson 2002:796; Auel 2005:296; Börzel 2003, 2001 and 1999; Börzel/Risse 2003 and 2000; Bomberg/Peterson 2000:4; Buller/Gamble 2002; Bulmer/Radaelli 2004; Cowles/Curtis 2004:303; Eising 2003:387; Howell 2004; Jachtenfuchs/Kohler-Koch 2004a:109; Ladrech 1994; Radaelli 2004a, 2003a and b, 2000:2; Risse/Cowles/Caporaso 2001:3; Sturm 2005:102. Theoretico-Empirical Frame of Reference 82 the division of responsibilities and powers between different levels of governance” (ibid.). Generally, however, the acquis académique is not a symmetrical body regarding its theoretico-empirical divide: while empirical studies prevail “theoretical work on Europeanization is still in its early days” (Radaelli 2003a:28). Within the political process, Europeanisation is also conceptualised to impact differently on the different phases of “policy formulation (construction); putting policy into practice (institutionalisation); and in a much less structured manner (diffusion), where the EU’s role may be quite limited” (Bulmer/Radaelli 2004:3). Europeanisation studies, therefore, analyse most different, but related aspects emphasising either a bottom-up or a top-down perspective of Europeanisation or a (cyclical) relation between both (cf. chapter Within these multi-directional Europeanisation processes, in which especially the “cognitive dimension of political life matters” (Radaelli 2003a:52; cf. Auel 2005:297), the EU becomes “an increasingly more relevant and important point of political reference for the actors at the level of the member states” (Hanf/Soetendorp 1998:1). As a consequence, “domestic policy areas become increasingly subject to European policy-making” (Börzel 1999:574), which also holds true for socio-economic policy-making and the specific case of employment policy co-ordination analysed in this study. Focusing on the domestic level of EU member states, Europeanisation can, therefore, also be understood as an “incremental process re-orienting the direction and shape of [domestic] politics to the degree that EC political and economic dynamics become part of the organizational logic of national politics and policy-making” (Ladrech 1994:69). In order to “distinguish the simple fact that there is more ‘Europe’ in domestic policies and the more profound impact of the European Union in policy areas which have now become dominated by a European logic of behaviour” (Radaelli 2003a:29, cf. ibid. 2000:2), more detailed definitions of Europeanisation–some drawing heavily on sociological institutionalism and the “logic of appropriateness” (March/Olsen 2004:3ff.)–additionally highlight the impact of Europeanisation on “the development of formal and informal rules, procedures, norms and practices governing politics at the European, national and subnational levels” (Risse/Caporaso/Cowles 2001:3; cf. Börzel 2002:193; Börzel/Risse 2003:66; Radaelli 2000:12). By doing so, they emphasise “less tangible aspects, such as beliefs and values” (Bulmer/Radaelli 2004:3; cf. Eising 2003:401).22 In this perspective, Europeanisation, as a process of mutual relatedness of the supranational and national/subnational level, is viewed to promote the evolution of both “regulative … and constitutive rules” (Mörth 2003:161ff.)23, which results in the “progressive emergence of a bundle of common norms of action, the evolution of which escapes the control of any particular member state and 22 These aspects are perceived to be more resistant to change “given their deep embeddedness in dominant beliefs of domestic actors. Instruments and, even more, settings, by contrast, can be adjusted without necessarily demanding ideational change” (Knill 2005:771). 23 While regulative rules are equivalent to legal rules of the EU including soft law elements, constitutive rules are “rules that in a fundamental way determine how an issue is to be interpreted” (Mörth 2003:162) more equivalent to paradigms, fundamental ideas and frames. Theoretico-Empirical Frame of Reference 83 yet decisively influences the behaviour of public policy actors” (Mény/ Müller/Quermonne 1996:8f.; cf. Cowles/Risse 2001:219; Mörth 2003:160; Olsen 2002:935) within European (administrative) networks (cf. chapter In this context “European policies, norms, and the collective understandings attached to them exert adaptational pressures on domestic-level processes, [especially if] .. they do not resonate well with domestic norms and collective understanding” (Börzel/Risse 2003:58f.). This view strongly “underlines the role of adaptation, learning, and policy change” (Radaelli 2003a:30) for the adaptation of belief systems as basis for domestic public policies. In the context of the present study it is, therefore, especially important to analyse the degree and circumstances under which “Europeanization can change ‘policy core’ beliefs and facilitate learning and non-incremental change” (Radaelli 2003a:49). It has to be kept in mind, however, that such change can only be labelled Europeanisation when the adaptation of norms at supranational level is indeed followed by domestic change (cf. Radaelli 2004a:10). In line with the generally accepted need for further conceptual refinement of the Europeanisation approach, misfit, ‘mismatch’ (Héritier et al. 1996; Héritier 2001b:44; Knill/Lenschow 1998:596) or ‘goodness of fit’ (Bulmer/Radaelli 2004:8; Featherstone 2003:15f.; Risse/Cowles/Caporaso 2001:6) play an important role as categories that presuppose “a clear, vertical, chain-of-command, in which EU policy descends from Brussels into the member state” (Bulmer/Radaelli 2004:9), being best applied to positive integration (cf. ibid.). So, as a precondition for domestic change, Europeanisation is viewed to matter “only if there is divergence, incompatibility, or ‘misfit’ between European-level institutional processes, politics, and policies, and the domestic level” (Radaelli 2003a:44; cf. ibid. 2000:16). Hence, in order to cause change, Europeanisation “must be inconvenient, that is, there must be some degree of ‘misfit’ or incompatibility” (Börzel/Risse 2003:58) between European provisions and national arrangements (cf. Auel 2005:304; Börzel 1999:574; Börzel/Risse 2000:1 and 5; Duina 1999, 1997). Within the Europeanisation approach, misfit is perceived to be an important condition in order to generate adaptation pressure for domestic systems to adapt (cf. Risse/Cowles/Caporaso 2001:2 and 7). The degree of adaptation pressure is high when misfit is high, that is, when domestic structures and policies diverge strongly from EU provisions. Thus, the “lower the compatibility between European and domestic processes, policies, and institutions, the higher the adaptational pressure” (Börzel/Risse 2003:61; cf. Börzel 2002:194; Börzel/Risse 2000:5; Héritier 2001b:44; Treib 2004:49), creating a “curvilinear” (Radaelli 2003a:44) relationship between the two. “If adaptational pressures are very high, European institutions seriously challenge the identity, constitutive principles, core structures, and practices of national institutions” (Risse/Cowles/Caporaso 2001:8). In such cases, adaptation is extremely expensive and can become unrealistic or disfavoured by national political actors. Adaptation pressure is low when misfit is low, hence, when domestic arrangements are in line with the European rules, norms, and laws to be implemented (cf. Bailey 2002:792; Börzel 2003:4; Börzel/Risse 2003:61f.; Knill/Lenschow 1998:596; Radaelli 2003a:44f.; Risse/Cowles/Caporaso 2001:7). Theoretico-Empirical Frame of Reference 84 Some authors assess the impact of European provisions on EU member states to be strongest, if there is a “moderate goodness of fit” (Bulmer/Radaelli 2004:8; cf. Börzel/Risse 2003:70). Misfit can relate to policies and institutional arrangements, leading to either policy or institutional misfit, the latter being more incremental than the former24, with domestic core institutional arrangements largely resistant to change. Both types of misfit increase adaptation pressure and potentially lead to change of either domestic “policy goals, regulatory standards, the instruments or techniques used” (Börzel/Risse 2003:61) or to adaptation of “domestic rules and procedures and the collective understandings attached to them” (ibid.:62). “In any case, policy convergence seems to be more likely than institutional convergence as policy changes are more easily achieved” (ibid.:72; cf. Radaelli 2004a:14 and chapter Yet, policy mismatch might also create adaptation pressure on domestic institutional arrangements (cf. Risse/Cowles/Caporaso 2001:7). With a view to the assessment of the general preconducive relevance of misfit and adaptation pressure for Europeanisation some authors, nevertheless, critically state that Europeanisation can also take place without adaptation pressure and does, therefore, not depend on misfit as a precondition for change (cf. Featherstone 2003:16f.; Sturm 2005:116). In this perspective, Europeanisation–especially related to institutional change–can take place also in cases of negative integration or policy coordination, leading towards adaptation of congruent institutions or even policies in order to enhance the efficiency of domestic arrangements (Sturm 2005:116f.). Additionally, the idea of ‘misfit’ as a precondition for adaptation is also criticised for structurally neglecting political actors’ behaviour and impact (cf. Radaelli 2004a:7; Treib 2004:48 and 60). Besides the positive definition of what Europeanisation is about, it seems promising to add some ‘negative’ definitions to demarcate what Europeanisation is not in order to sharpen the analytical view towards the conceptual margins of the approach. Europeanisation is neither equivalent to convergence nor to harmonisation (cf. Radaelli 2003a:33, 2000:6, 2004:14). As Europeanisation refers to processes of adaptation and change, and given that convergence is a possible but not necessary outcome (as divergence is)–just like “[b]irds sometimes sing and sometimes they don’t” (Lenschow/Liefferink/Veenman 2005:797)–Europeanisation and convergence need to be kept separate in terms of definition (Andersen 2004:7; Börzel/Risse 2003:59; Radaelli 2004a:3; 2003a:33) and further categorised (cf. chapter In this context, also the phenomenon of ‘clustered convergence’ (Börzel 2002) needs to be kept in mind (cf. Risse/Cowles/Caporaso 2001:1). It describes an analytical continuum were minimal “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… . When 24 Börzel/Risse 2003:63 and 72; Börzel/Risse 2000:5; Bomberg/Peterson 2000:36; Cowles/Risse 2001:222f.; Eising 2003:400 and 406; Falkner 2003:2f.; Radaelli 2004a:14; Risse/Cowles/Caporaso 2001:7; Treib 2004:49. Theoretico-Empirical Frame of Reference 85 similar decisions are implemented in a relative uniform way, the degree of convergence increases. Finally, one can imagine the case of convergence in outcomes” (Radaelli 2004a:14; cf. Featherstone 2003:11). Furthermore, Europeanisation does neither impose nor directly aim at harmonisation, since domestic adaptation depends on several intermediating factors that impact on the outcomes of Europeanisation (cf. chapter and 2.2), resulting in “regulatory diversity, intense competition, [or] even distortions of competition” (Radaelli 2003a:33). Concluding from the above presented definitions and demarcations, Europeanisation studies generally focus on the analysis of one or several elements of the interrelation between the supranational and the national/subnational levels, which also build the focus of this study’s empirical analysis: • “adaptation of institutional setting … at different political levels in response to dynamics of integration;… • emergence of new, cross-national policy networks and communities; • nature of policy mimicry and transfer …; • shifts in cognition, discourse, and identity affecting policy in response to European developments; • restructuring of the strategic opportunities available to domestic actors” (Featherstone 2003:19f.). With this widespread conceptual span, Europeanisation covers most different understandings and is analytically “applied within four broad categories: as an historical process [largely related to European (imperial) efforts of exporting cultural norms]; as a matter of cultural diffusion [e.g. diffusion of norms, ideas, behaviour, etc.], as a process of institutional adaptation [of member states in response to European provisions], and as the adaptation of policy and policy processes [by direct or indirect impact of the EU]” (Featherstone 2003:5; cf. Sturm 2005:102). With this span, Europeanisation, “as a meso (middle range) theory” (Howell 2004a:1), covers “the major interests of political scientists, such as political structure, public policy, identities, and the cognitive dimension of politics” (Radaelli 2003a: 30). At the same time, it remains a rather “fuzzy concept” (Jachtenfuchs/Kohler-Koch 2004a:109) and “a loose epithet” (Featherstone 2003:12) without a “single precise or stable meaning” (Olsen 2002:921). It favours extension rather than intension (Radaelli 2003a:32, 2000:4), including the analysis of very different phenomena, elements and research objects based on different and partially vague conceptual assumptions (Jachtenfuchs/Kohler-Koch 2003:34). Furthermore, with a view to measuring Europeanisation, the approach can become victim to “degreeism” (Radaelli 2003a:28) given that “differences in kind are replaced by differences in degree” (ibid.:32), owing to the fact that a categorisation of its impact is a difficult venture as “its structural effects are not necessarily permanent or irreversible”, but are “typically incremental, irregular, and uneven over time and between locations” (Featherstone 2003:4; cf. Börzel/Risse 2003:74). Theoretico-Empirical Frame of Reference 86 With this immense range of application and its inherent risk of ‘degreeism’, the concept is scientifically characterised by the remarkable, albeit disturbing, variety of definitions and conceptual understandings (outlined above). They expose the approach to the dangers of an all-encompassing concept stretching (cf. Auel 2005:295; Eising 2003:389; Howell 2004a:1 and 4; Radaelli 2003a:27 and 32, 2000:1 and 4f.; Treib 2004:63) accompanied by a high level of “empirical complexity and conceptual confusion” (Olsen 2002:923). With this drift, Europeanisation is jeopardised to inter the concept’s inherent analytical ‘Wunderkind’, able to perform in most diverse systemic settings and at different political levels, within a conceptual ‘Tower of Babel’ endangered to collapse due to its own instability. Nevertheless, with “all its limitation, Europeanisation provides a fascinating perspective on how governance is changing” (Radaelli 2004a:16; cf. Risse/Cowles/ Caporaso 2001:1). Therefore and despite of the tendency of concept stretching, the approach provides for a well-fitting analytical tool to scrutinise the impact of the EES on domestic change and adaptation, providing for answers to research questions related to institutional and policy change, policy learning and transfer, and the potential impact of the OMC. Central Characteristics: ‘Up and Down and Cross’ – Scores of Ways to Impact Given this broad range of conceptualisations, it is necessary to clarify the central characteristics and key understanding of Europeanisation as applied in this study in order to ‘un-stretch’ the concept for a focused empirical analysis. First of all, with a view to Featherstone’s four broad analytical categories to which the approach is generally applied (cf. chapter, this study’s impact analysis of the EES on the Europeanisation of national employment policy co-ordination and policies in the UK and Germany focuses on the • (cultural) diffusion of new policy paradigms and patterns of politics caused by the introduction of the EES and the application of the OMC within employment policy co-ordination; • related processes of institutional adaptation and domestic change; • adaptation of policies and policy processes at supranational and national levels. The analysis does not touch upon the historical process of Europeanisation as defined by Featherstone, given that this relates to a historiographic or even anthropological view of Europe (cf. Featherstone 2003:6). Nevertheless, it touches upon the historical process of up-loading, leading to the emergence of employment policy coordination at supranational level (cf. chapter 3.1). In a broad sense, the Europeanisation approach applied in this study analyses an “asymmetric process” (Featherstone 2003:4) “of structural change, variously affecting actors and institutions, ideas and interests” (ibid.:3). In a narrower, albeit multi- Theoretico-Empirical Frame of Reference 87 dimensional perspective, it focuses on three main (sub-)processes (Howell 2004a:5f., 2004b:54ff.): (1) up-loading/bottom-up Europeanisation: transfer of tasks and competences to the European level accompanied by the establishment of European policy areas and policy-making structures; (2) down-loading/top-down Europeanisation: adaptation of national political systems caused by down-loading (that is, transfer and implementation of) European provisions to the national arenas; (3) cross-loading/vertical policy transfer: transfer of policy approaches, ideas, paradigms, and policies between EU member states, originating in EU provisions and based on EU stimulus. These three processes do not represent a special sequential logic of Europeanisation, but are related to different constellations of impact. Moreover, each lays its analytical focus on different areas and scrutinises Europeanisation effects, leading to adaptation and change, at different levels of the EMLG system. Up-loading or bottom-up Europeanisation, as an “effective strategy to maximize the benefits and minimize the costs of European policies” (Börzel 2002:196), highlights the process in which new polity, politics or policy elements are ‘born’ at European level by transfer of national sovereignties. Another analytical focus of this perspective is member states’ influence within this process (cf. Börzel 2003:2) caused and/or followed by an increasing shift of attention of national actors towards the European level and by the creation of supranational elements (Bomberg/Peterson 2000:6; Risse/Cowles/Caporaso 2001; Wessels/Rometsch 1996:328). In this perspective, “Europeanisation is not the explanans …, but the explanandum” (Radaelli 2004a:4). Bottom-up Europeanisation refers to constellations, “in which the dynamics and the outcome of the European institution-building process are the main dependent variable” (Börzel/Risse 2003:57). It analyses the process by which “[n]ational policy models or rules are inserted into EU-level negotiations” (Bulmer/Radaelli 2004:5). This perspective focuses on the “process whereby the European institutional political arena becomes autonomous in respect to the constraints, preferences, and governance habits of the member states” (Giuliani 2003:135), creating a feed-back loop with top-down Europeanisation processes. Yet, the real process of domestic up-loading, that is, the way how member “states responses to Europeanization feed back into EU institutions and policy processes” (Börzel 2003.1) is rarely examined in-depth (cf. Howell 2004a:5). Within processes of bottom-up Europeanisation, member states are key actors, able to ex ante reduce their own costs related to domestic change (cf. Auel 2005:311; Börzel 2003:3f. and 6; 2002:194f.; Bomberg/Peterson 2000:20; Eising 2003:403) and influence the outcome of Europeanisation at the supranational level. Hence, if “member states have lobbied effectively and had their perspective included in EU policy, misfit will be limited and consequent domestic change will be minimal” (Howell 2004a:5). As a consequence, “the more the government and other Theoretico-Empirical Frame of Reference 88 domestic actors are involved in the rule making process at the European Union/ European level, the less effort they have to put into domestic editing” (Mörth 2003:163; cf. Börzel 2003:4).25 Taking the antipodal perspective, down-loading or top-down Europeanisation starts “from European policies (or politics) as independent variable and track[s] down the consequences for domestic actors, policies, and politics” (Radaelli 2004a:4). It examines the impact of EU laws, norms, and provisions on the national political systems, focusing “on the consequences of European integration for the autonomy and authority of the Member States” (Börzel 2003:2; cf. Börzel/Risse 2003:57, 2000:1f.; Giuliani 2003:135; Ladrech 1994; Olsen 2002:932; Radaelli 2004a). Down-loading is assessed to impact through “[p]rocesses of (a) construction, (b) diffusion, and (c) institutionalization of formal and informal rules, procedures, policy paradigms, styles, ‘ways of doing things’, and shared beliefs and norms which are first defined and consolidated on the making of EU public policy and politics and then incorporated in the logics of domestic discourse, identities, political structures, and public policies” (Radaelli 2003a:30, 2000:4).26 Due to the fact that policy-making at supranational and at national level are likely to often be “mutually constitutive” (Mörth 2003:173) and parallel processes (ibid.:163), both, bottom-up and top-down Europeanisation, are no separate but reciprocal processes (cf. Auel 2005:298; Börzel 2003:3; 2002:195; Bomberg/Peterson 2000:6f.; Felder 2001:190; Radaelli 2004a:4; Sturm 2005:103). Blurring the boundaries of dependent and independent variables (Bulmer/Radaelli 2004:3), these two Europeanisation processes are, thus, interlinked through “feedback processes among and between the various levels of European, national, and subnational governance” (Börzel/Risse 2003:58; cf. Howell 2004a:3 and 11; Mörth 2003:160; Treib 2004:69). Therefore, especially the EMLG approach opens an optimal (theoretical) perspective on the interwoven political processes at national and supranational level and their mutual interrelatedness given that the “success regarding up-loading .. will determine the level of change in relation to down-loading” (Howell 2004a:5; cf. Börzel 2002:195; Majone 1998:1; cf. chapter The process of cross-loading includes policy transfer across and among different states (cf. Bomberg/Peterson 2000:10f.; Cowles/Risse 2001:221; Howell 2004a:5ff.; Radaelli 2004a:5 and 2003a:27). It covers two directions: horizontal and vertical policy transfer. While horizontal policy transfer involves no EU stimulus, vertical policy transfer does rely on European stimulus in order to be labelled Europeanisation. Thus “[v]ertical policy transfer emerges through EU policy or European integration processes. Horizontal policy transfer incorporates learning from, and assimilating other member state policies without EU involvement” (Howell 2004a:5). In order to highlight the relevance of EU-imposed adaptation pressure within vertical policy transfer processes, a more precise term, pointing at the relevant stimulusreaction constellation, would, yet, be “EU-ization” (Auel 2005:299; cf. Lodge 25 This process of up-loading is analysed in chapter 3.1. 26 This process of down-loading is analysed in chapter 4.2 and 5. Theoretico-Empirical Frame of Reference 89 2001:15; Mörth 2003:159; Radaelli 2003a:27), “EU-Europeanisation” (Kohler-Koch 2000:12; cf. Eising 2003:393) or even Unionisation (Wallace, H. 2000a:3ff.). In view of the case under analysis in this study, cross-loading/vertical policy transfer is perceived to have indeed “produced far more Europeanization of national policies than actual common policies” (Bomberg/Peterson 2000:32).27 Graph 2: Bottom-up, Top-down, and Cross-loading Processes of Europeanisation European Integration EUROPEAN UNION Europeanization Bottom-up Member States Top-down European policy-making Supranational institution-building Delegation of national competencies European policies, norms, rules, and procedures, political processes (Member) state 1 (Member) state 2 (Member) state n Cross-loading Source: Börzel 2003:3; amended by the process of cross-loading cf. below. Due to distinct domestic preconditions, member states are empowered differently to respond to the demands of Europeanisation (cf. Börzel 2003:3; Giuliani 2003:135; cf. chapter and 2.2). The degree to which they react to Europeanisation pressures can range from being “merely passive takers of European demands for domestic change” to the “proactive shap[ing of] European policies, institutions, and processes to which they have to adapt later” (Börzel 2003:3; cf. Auel 2005:312). Within the literature, three strategies of member states within the process of Europeanisation have been identified (cf. Börzel 2002:196ff.): 27 This process of policy transfer by cross-loading is analysed as a core element of the OMC in chapter 4.2 and 5. Theoretico-Empirical Frame of Reference 90 • Pace-setting: proactively promoting policies at supranational level, including coalition-building, reflecting domestic policy preferences to reduce adaptation and down-loading costs; • Foot-dragging: impeding and hindering policies that are incompatible with national policies and likely to increase implementation costs, trying to get compensation via package deals; • Fence-sitting: rather neutral, policy-dependent strategic coalitionbuilding to push forward national interests without a specific preference for one of the two other strategies given that no significant implementation costs are expected. The choice of one of these three strategies depends on domestic policy preferences and on the state of the national “economic development, which largely influences the degree of domestic regulation and the action capacities of a member states” (Börzel 2002:194, cf. ibid.:196). On this background, the impact of Europeanisation on domestic public policies can take “different forms, such as convergence…, direct and indirect transfer of models from Brussels…, and a profound impact of EU regulation[s]” (Radaelli 2003a:35), depending on the governance mode applied at supranational level. The different existing modes of governance favour the three directions of Europeanisation in different ways (cf. Radaelli 2004a:12, 2000:16ff.). Mechanisms to pressure Europeanisation range from (1) the presence of a European model in the case of positive integration, over (2) the competition between national regulatory systems related to negative integration, to (3) the role of framing, policy learning/transfer and convergence towards commonly established policy paradigms within new modes of governance such as policy co-ordination (cf. Auel 2005:304ff.; Bulmer/Radaelli 2004:4; Featherstone 2003:14; Howell 2004a:4; Knill/Lehmkuhl 1999; Radaelli 2003a:40ff.; Treib 2004:56). Within this context, bargaining and negotiation, as the cradle of every EU policy, favour up-loading processes and lead to bottom-up Europeanisation as “European policy does not emerge from thin air” (Bulmer/Radaelli 2004:4), but is rather a product of negotiation processes (cf. Auel 2005:310f.; Radaelli 2004a:12). Positive integration, as a type of governance by hierarchy, is followed by down-loading, that is, top-down Europeanisation. Here, the supranational level disposes over the power to impose new policies on EU member states by employing a command-and-control approach (cf. Auel 2005:302; Bulmer/Radaelli 2004:5f.; Featherstone 2003:14; Radaelli 2004a:12, 2000:16f.). Negative integration and policy co-ordination tend towards cross-loading. While “negative integration relates to areas where the removal of national barriers suffices to create a common policy” (Auel 2005:306; cf. Bulmer/Radaelli 2004:6; Cowles/Risse 2001:219; Featherstone 2003:14), policy coordination touches upon areas where member states remain the main actors and decision-makers of the supranational political process (cf. Auel 2005:306; Bulmer/Radaelli 2004:7; Featherstone 2003:14; Radaelli 2004a:12, 2000:17f.; cf. chapter As negative integration constitutes a mixture of vertical EU stimulus and horizontal Europeanisation impact through trans-national competition Theoretico-Empirical Frame of Reference 91 (cf. Auel 2005:306), policy co-ordination as in the case of the EES also impacts horizontally, involving “different forms of framing” (Radaelli 2003a:41), learning processes and mutual socialisation as “there is no [immediate] pressure to conform to EU policy model” (ibid.). Taking up this further categorisation of Europeanisation processes, the present study opts for a focus on top-down Europeanisation, that is, down-loading processes related to the EES. It combines their analysis with the assumptions of cross-loading, be it vertical or horizontal. At the same time, also the bottom-up perspective of Europeanisation is taken into account when looking at the development of the EES and the analysis of the related supranational policy network in order to identify relevant elements for domestic change within the two selected EU member states. Domains of Europeanisation: Domestic Targets of Adaptation Pressure – Increasing Complexity to Enhance Conceptual Lucidity In order to structure its “wider analytical toolbox” (Radaelli 2003a:34), to explain the selection of elements analysed and to put a particular spotlight on the question “What is Europeanized?” (Radaelli 2003a:35; cf. ibid. 2000:14), this study additionally takes up Radaelli’s differentiation of Europeanisation domains. At a very first glance, this seems to increase complexity instead of minimising it. However, this step is especially rewarding as it helps to more closely categorise the targets of analysis in order to enhance conceptual lucidity. The focus of the present analysis is directed towards the following domains of Europeanisation: (1) domestic structures, (2) public policy and (3) cognitive and normative structures (cf. Radaelli 2003a:35; cf. table 6). Domestic structures, as main institutional targets to Europeanisation (cf. Treib 2004:49), comprise formal political, legal, administrative, and institutional structures that root in specific national traditions of domestic political systems of EU member states. Moreover, political parties, interest representations, and cleavage structures can be dealt with under this domain (cf. Radaelli 2003a:35). The analysis of Europeanisation of domestic public policy can touch upon every component of public policies, including political as well as non-state actors, resources, policy instruments, paradigms, frames, narratives, and styles, be they “more or less conflictual, corporatist, or pluralist, or more or less regulative” (Radaelli 2003a:35; cf. Bulmer/Radaelli 2004:15; Radaelli 2000:14; Töller 2004:1). The domain of cognitive and normative structures embraces more informal aspects of national political systems. It might be the domain strongest affected by Europeanisation through the EES and the OMC. It is analytically linked to sociological institutionalist assumptions on the impact of European provisions on the Europeanisation of constitutive rules, values, norms, perceptions, principles, priorities, and discourses within EU member states. Changes in this domain most likely affect the other two domains, leaving broad room for discretion and opening a considerable leeway “for domestic interpretation, editing, and translation” (Mörth Theoretico-Empirical Frame of Reference 92 2003:161f.; cf. Andersen 2004:14; Olsen 2002:936; Radaelli 2003a:35). Given this dual aspect of being ‘Europeanisable’ and exerting Europeanisation pressure on the other two domains as well as for reasons of analytical transparency, this particular domain is conceptually kept separate from the two former (cf. Auel 2005:300). Table 6: Domains of Europeanisation Domain Elements Domestic structures 1. Political structures a) Institutions (e.g. cabinet-assembly relations) b) Public administration c) Intergovernmental relations d) Legal structure 2) Structures of representation and cleavages a) Political parties b) Pressure groups c) Societal-cleavage structures Public policy a) Actors b) Policy problems c) Style d) Instruments e) Resources Cognitive and normative structures a) Discourse b) Norms and values c) Political legitimacy d) Identities e) State traditions–understanding of governance f) Policy paradigms, frames, and narratives Source: Radaelli 2003a:35; bold by the author of the present study to mark the relevant elements related to the empirical analysis conducted. Given that already Radaelli pointed at the difficulties of analysing all these elements within one single study (cf. Radaelli 2003a:37), the present analysis focuses on the most relevant ones in view of its research puzzle. Due to this selection, the analysis concentrates on the Europeanisation of domestic structures and public policies related to the impact of the EES. It, nevertheless, does not neglect cognitive and normative structures as (drawing on historical institutionalism) it also analyses the extent to which state traditions intervene in the Europeanisation of domestic institutions, structures, and policies. Moreover, (owing to sociological institutionalist suppositions) it examines the degree of change of policy paradigms due to the impact of the supranational provision. This integrated approach is, both theoretically and empirically, perceived to have the potential of drawing a more comprehensive picture of the Europeanisation impact of the EES and the OMC. Theoretico-Empirical Frame of Reference 93 Categories of Europeanisation: Domestic Change between Retrenchment and Transformation Following Olsen’s proposition to clarify the “meaning, the phenomena in focus, the simplifying assumptions used, [and] the models of change” (Olsen 2002:921), five categories of Europeanisation, covering “both the magnitude of change and its direction” (Radaelli 2003a:37), serve to measure the outcomes of down-loading, that is, the degree of impact of top-down Europeanisation, in order to avoid the danger of non-reflected ‘degreeism’ outlined above. Following Radaelli (2003a:37), four of these five basic categories of Europeanisation are retrenchment, inertia, absorption, and transformation. These four categories are amended by the fifth category of accommodation or upgrading (cf. Börzel/Risse 2003:70, 2000:10f.; Meyer/Umbach 2007). Graph 3: Categories of Europeanisation Direction of domestic change retrenchment inertia absorption accommodation/ upgrading transformation - 0 + ++ +++ Source: Radaelli 2003a:35; amended by Börzel/Risse 2003:70 and Meyer/Umbach 2007:94f.; ‘policy change’ as referred to in Radaelli’s original table is supplemented by ‘domestic change’ as also institutional change will be measured in these five categories. One side of this conceptual scale is made up of two categories of ‘negative’ or ‘non’-adaptation. The first, retrenchment, describes the phenomenon of a member state becoming “less European” (Radaelli 2003a:38, 2000:15; cf. Auel 2005:300f.) due to Europeanisation pressure that “mobilizes intra- or extra-parliamentarian resistance not only against changes suggested by the EU, but also in favour of initiatives which run counter to those suggestions and demands” (Meyer/Umbach 2007:95; cf. Howell 2004a:3). The second, inertia, is to a certain extent a twofold category. It describes a “lack of change” (Radaelli 2003a:37, 2000:14; cf. Auel 2005:300) due to the lack of political will for change (Howell 2004a:3). This lack of political will for change is caused by too huge a difference between European and national provisions, resulting in high adaptation pressure. So, “bringing domestic arrangements in line would be too costly given a high degree of ‘misfit’” (Meyer/Umbach 2007:95), requiring a “radical break with the past, abandoning old routines, removing outmoded institutions, and switching over to new structures” (Genschel 1995:6). On the other hand, inertia can also be the result of too low adaptation pressure, when EU Theoretico-Empirical Frame of Reference 94 provisions and national arrangements are largely in line with each other. Under these circumstances, no domestic change is needed to adapt in order to meet the requirements of the new supranational provision. As a consequence, “European policy frames which resonate with domestic policy ideas and discourses are unlikely to trigger collective learning processes which could change actors’ interest and identities” (Börzel/Risse 2003:61). Inertia can, therefore, be found in cases of too high as well as too low adaptation pressure, that is, huge or minimal differences between EU and national policies (cf. Radaelli 2003a:45, 2000:14). It “may take the forms of lags, delays in transposition …, and sheer resistance to EU-induced change. … [Moreover,] long periods of inertia should produce crisis and abrupt change” (Radaelli 2003a:37) in case of high adaptation pressure. At the centre of the scale, the ‘positive side’ begins with the category of absorption. It stands for a rather surface, camouflage or façade Europeanisation “as actors and institutions seek to make domestic arrangements work without changing their core organisational principles, routines and discourses or having to meet substantial costs” (Meyer/Umbach 2007:95). With absorption, domestic structures, public policies as well as cognitive and normative structures “provide a mixture of resiliency and flexibility. They can absorb certain non-fundamental changes, but maintain their ‘core’” (Radaelli 2003a:37: cf. Auel 2005:300; Héritier 2001b:54; Zandstra 2004:6). Due to (shallow) “learning some clever strategies as a response to stimulation” (Radaelli 2003a:38) without substantial modification of the original polity, politics and policy dimension of the member state, Europeanisation of this degree remains low (Börzel/Risse 2003:70) and superficial. Between this slightly positive and the most positive degree of Europeanisation lays accommodation or upgrading. This category constitutes a medium degree of Europeanisation and pays tribute to the fact that member “states accommodate Europeanization pressures by adapting existing processes, policies, and institutions without changing their essential features and the underlying collective understandings attached to them” (Börzel/Risse 2003:70). Accommodation/upgrading can, therefore, “be seen as a moderate form of adaptation in so far as some aspects of organisation, practice or ideas are changed in a way that is novel, while other parts remain essentially the same” (Meyer/Umbach 2007:95). This category is inserted into Radaelli’s original scale in order to meet “the need for a further intermediate grade to ensure equidistance between the four values” (ibid.). Finally, transformation means that Europeanisation results in a “[p]aradigmatic change … [of] the fundamental logic of political behaviour” (Radaelli 2003a:38, cf. ibid. 2000:15; Auel 2005:300). “[N]ew institutions are being set up, new policy paradigms are established and implemented, or identity discourses and constructions shift in a way that changes dramatically the behaviour of a given arrangement, actor or system” (Meyer/Umbach 2007:95; cf. Börzel/Risse 2003:70). With transformation, learning provides for change of deeply embedded patterns, norms, values or views, producing a profoundly different way of thinking (Börzel/Risse 2003:59; Radaelli 2003a:38). Theoretico-Empirical Frame of Reference 95 The combination of these categories and the domains of Europeanisation leads to an analytical matrix (cf. table 7) of this study that serves both to measure and locate the research results of the empirical analysis. Table 7: Matrix of Categories and Domains of Europeanisation Domain of Europeanisation Europeanisation impact Category of Europeanisation Domestic Structures Public Policy Cognitive/ Normative Structures Retrenchment (becoming less European) Inertia (absence of change) Absorption (largely superficial form of change) Accommodation / Upgrading (moderate form of adaptation) Missing (adaptation pressure too strong) High (changes of fundamental logics) Transformation (fundamental change) Source: Own design based on the contents of Radaelli 2003a:35, Börzel/Risse 2003:70 and Meyer/Umbach 2007:94f.; partially inspired by the design of Mittag/Wessels 2003:414. Europeanisation and Domestic Change: Cause and Effect of Policy Change and Institutional Isomorphism – Different Sides of a Multi-Dimensional Coin? “The domestic effect of Europeanization can be conceptualized as a process of change at the domestic level in which the member states adapt their processes, policies, and institutions to new practices, norms, rules, and procedures that emanate from the emerging European system of governance” (Börzel/Risse 2003:63). In order to theoretically embed the analysis of the Europeanisation impact of the EES and to show, if and how this European provision provokes domestic adaptation even without being legally binding, a closer conceptual look at possible domestic Theoretico-Empirical Frame of Reference 96 effects of Europeanisation, including policy and institutional change, seems to be valuable and necessary. As throughout the study, also this close-up look applies “the distinction between policies, politics, and polity to identify three dimensions along which the domestic impact of Europeanization can be analysed and processes of domestic change can be traced back” (Börzel/Risse 2003:60; cf. Bomberg/Peterson 2000:5; Radaelli 2004a:14). Doing so promises to improve the structure of the latter empirical analysis given that top-down Europeanisation is supposed to exert different adaptation pressures on different dimensions of the distinct domestic political systems of EU member states (cf. Börzel 2003:3; Bulmer/Radaelli 2004:1; Giuliani 2003:134; Risse/Cowles/Caporaso 2001:11). As to its domains, that is the question of ‘what is Europeanised’ (cf. chapter, Europeanisation can generally influence changes of domestic policy standards, paradigms, styles, logics, narratives, discourses, instruments and problemsolving approaches within the domestic policy dimension. Related to domestic politics, Europeanisation is perceived to impact on different processes, ranging from interest formation over aggregation towards representation of interests, including public discourses. Finally, with a view to the domestic polity dimension, Europeanisation–in case of misfit–can exert adaptation pressure on political institutions, intergovernmental relations, judicial structures, public administration, state traditions, economic institutions, state-society relation, and collective identities (cf. Börzel/Risse 2003:60; Börzel/Risse 2000:3f.). Europeanisation, thus, touches upon the entire domestic political system (cf. Bulmer/Radaelli 2004:11; March/Olsen 2004:13; Risse/Cowles/Caporaso 2001:12). Broadly speaking, the phenomena of policy and institutional learning propel the transformation of public policies, changes in established modes of public-private interactions, the adaptation of legal aspects as well as administrative or institutional adjustment (Treib 2004:72) in “a culturally and institutionally pre-structured process” (Lenschow/Liefferink/Veenman 2005:800). Both processes of policy and institutional learning include different aspects such as to “destabilize existing understandings; bring together people with diverse viewpoints in settings that require sustained deliberation about problem-solving; … reconfigure policy networks; encourage decentralized experimentation; produce information on innovation; require sharing of best practice and experimental results; encourage actors to compare their results with those of the best performers in any area; and oblige actors collectively to redefine objectives and policies” (Trubek 2002:4). With a view to the intensity of domestic change, that is, the five categories of Europeanisation outlined above (cf. chapter, it needs to be “distinguish[ed] between instances in which actors merely adjust means and strategies to achieve their given goals and preferences (‘single-loop learning’…[or ‘thin learning’ (Radaelli 2004b:11)]) and situations that lead actors to change these goals and preferences themselves (‘double-loop learning’ or ‘complex learning’)” (Risse/Cowles/Caporaso 2001:12). Moments of double-loop or “thick learning” (Radaelli 2004b:11), nevertheless, are very seldom and often follow crises or moments of high political ambiguity (cf. Börzel/Risse 2003:67). Theoretico-Empirical Frame of Reference 97 Domestic change “may be a result of rule-following and the application of standard operating procedures … [,] an outcome of problem-solving and calculating expected consequences, or of conflict resolution and confrontation … [, or] be produced through experimental learning or competitive selection” (Olsen 2002:924). In the case of the EES, the last two options seem to be most relevant. Within the process of Europeanisation, policies are forced to change and to adapt to newly created institutional and ideational arrangements (Mörth 2003:159). As a consequence, domestic change first affects various elements of public policies, ranging from underlying policy ideas and paradigms, instruments applied, over actors’ involvement towards the distribution of resources (Bulmer/Radaelli 2004:15). The degree of “impact of EU public policy is contingent on whether a country is already involved in a process of reform or not” (Mörth 2003:173, cf. above). At the same time, “European pressure to act may create a political space for reform, a window of opportunity … that may be exploited by the strategic action of political elites in the specific institutional context of a domestic political system” (Héritier 2001b:53; cf. Bulmer/Radaelli 2004:12). Policy change, moreover, “requires robust networks of stakeholders that facilitate the adoption of new policies at home, a strong civil society, and administrativepolitical capability the consciously modify, edit, and adapt foreign experience to national circumstances” (Radaelli 2004a: 13; cf. Bomberg/Peterson 2000:23; Zandstra 2004:12). Europeanisation of Public Policies: Means and Results of Policy Change in-between Policy Diffusion, Transfer, and Convergence Policy change can be stimulated by different processes (‘means’) and achieve different effects (‘results’) (cf. table 8). The means range from policy diffusion to policy transfer that potentially aim at policy convergence in view of commonly defined goals, that again impacts on institutional and/or procedural isomorphism (cf. Bomberg/ Peterson 2000:34f.). Each of these means and/or results focuses on different aspects of policy (and institutional) change, constituting different sides of the multi-dimensional coin of domestic change. Policy diffusion and policy transfer focus on the procedural aspects and, hence, the ways towards domestic public policy change (cf. Zandstra 2004:5). Policy convergence analytically relates to the effects, that is, the results of policy change through Europeanisation concerning policy characteristics. Isomorphism deals with institutional and organisational change in terms of results (cf. Holzinger/Knill 2005:779), that is often driven by policy change given that policy misfit is capable of also exerting adaptation pressure on domestic institutional arrangements. Theoretico-Empirical Frame of Reference 98 Table 8: Means and Results of Policy Change Policy convergence Isomorphism Policy transfer Policy diffusion Analytical focus Effects Effects Process Process Empirical focus Policy characteristics Organizational structures Policy characteristics Policy characteristics Dependent variable Similarity change Similarity change Transfer content transfer process Adoption pattern Source: Knill 2005:768. Policy diffusion and policy transfer conceptualise the process of policy change, which can lead to policy convergence (Heichel/Pape/Sommerer 2005:818). Both processes focus on the way main policy characteristics (that are ideas, principles and goals, objectives, instruments, modes of governance) change, while analysing different dependent variables (cf. Bache 2000:3; Holzinger/Knill 2005:779). Both hold that exchange on policy choices among governments based on “common affiliations, negotiations and institutional membership” (Knill 2005:767) is a crucial precondition for policy learning and policy change, which involve adaptation of “the principle goals that guide policy …; the techniques of policy instruments used to attain these goals; the institutional setting in which policy making takes place” (Zandstra 2004:4). Policy diffusion describes processes of policy change that potentially lead to the enhancement of cross-country public policy proximity. It focuses “more on the spatial, structural and socioeconomic reasons for particular adoption patterns rather than on the reasons for individual adoptions as such” (Knill 2005:767; cf. Bache 2000:3). It describes the process that changes policy characteristics and especially analyses related implementation patterns of the cross-national dispersal of public policies (cf. Zandstra 2004:5), including instruments like peer review, multilateral surveillance, scoreboards or bench-marks (Radaelli 2004b:13). With this focus, policy diffusion is “defined as the socially mediated spread of policies across and within political systems, including communication and influence processes which operate both on and within populations of adopters” (Knill 2005:766). Policy change induced by policy diffusion might lead to policy convergence across countries, even though this effect is not the only possible result. Conceptually, two possible patterns of policy diffusion can be identified. Firstly, policy diffusion can impact “as a distinctive causal factor that drives international policy convergence” (ibid.) by means of voluntary policy transfer within the international environment. Secondly, policy diffusion can spread “policies across countries with the possible result of cross-national policy convergence” (ibid.:766f.) not considering causative elements. In the second case, it can impact through more coercive mechanisms such as “legally binding harmonization requirements defined in international agreements or supranational regulations, [or] .. the imposition of policies on other countries Theoretico-Empirical Frame of Reference 99 through external actors” (ibid.:766). Policy transfer describes both “the underlying causes and content of singular processes of bilateral policy exchange” (ibid.:767), taking into account also the role of agencies (cf. Bache 2000:3). It is characterised as an either voluntary or coercive, but in any case intentional process (cf. Bomberg/Peterson 2000:11), “by which knowledge about policies, administrative arrangements, institutions and ideas in one political system (past or present) is used in the development of policies, administrative arrangements, institutions and ideas in another political system” (Dolowitz/Marsh 2000:5; cf. Bomberg/Peterson 2000:10). While voluntary policy transfer by embracing other countries’ best practices usually emerges from policy crisis, coercive policy transfer is induced by external stimuli such as inter-/supranational provisions (cf. Bache 2000:2; Bomberg/Peterson 2000:10; Bulmer/Padgett 2005:104ff.; Zandstra 2004:5f.). Most recently, the EU has developed its own tool kit of policy transfer, including elements such as the “agreement on general goals and standards …; institutionalised peer review and the identification of best practices …; the construction of league tables ranking national policies” (Bomberg/Peterson 2000:19) as applied also with the EES. With this definition, policy transfer overlaps with Europeanisation ‘á la’ crossloading, that is, vertical and/or horizontal policy transfer (cf. chapter and can lead to different results, including emulation, synthesis and influence (Bulmer/Padgett 2005:106ff.). Table 9: Forms and Effects of Policy Transfer Form Effect Emulation Exchange of domestic policy solutions by entirely taking over external policy models Synthesis Combination of aspects of different policies of different proveniences Influence Mere inspiration of own domestic policy solutions by external policy examples Source: Own compilation based on Bulmer/Padgett 2005:106ff. The result of policy transfer might be convergence among public policies in different dimensions, including policy “goals, structure and content; policy instruments or administrative techniques; institutions; ideology; ideas; attitudes and concepts” (Bache 2000:3). Nevertheless, this need not be the case as national adaptation processes are influenced by diverse domestic intervening variables and national traditions (cf. chapter 2.2). Policy transfer can also result in the amalgamation of transferred policy ideas and arrangements with domestic styles or routines, producing distinct national solutions not converging with the emulated blueprint (cf. Bomberg/Peterson 2000:11; Holzinger/Knill 2005:782f.; Knill 2005:766; Lenschow/Liefferink/Veenman 2005:799). In this case, Europeanisation would most likely result in accommodation or upgrading (cf. chapter Theoretico-Empirical Frame of Reference 100 Policy convergence focuses on the effects of policy change on policy characteristics. This concept is strongly influenced by “the tendency of societies to grow more alike, to develop similarities in structures, process, and performances” (Kerr 1983:3; quoted in Knill 2005:765), favouring an increase of proximity of public policies (Holzinger/Knill 2005:776) among states. Policy convergence, thus, categorises the results of policy change and “explain[s] changes in policy similarity over time. By contrast, transfer studies investigate the content and process of policy transfer as the dependent variable, while the focus of diffusion research is on the explanation of adoption patterns over time” (Knill 2005:767f.). Policy convergence can be supported by some assisting factors, that are, common problem perception and/or pressure, proximity of national traditions and system variables, policy imposition, harmonisation through law (be it supra- or international), competition induced by globalisation and economic co-operation as well as cross-national communication, inspiring policy learning, lesson drawing, network-building and problem-solving among nations (cf. Holzinger/Knill 2005:779ff.). In line with historical institutionalist assumptions, it can be supported or suppressed by intervening variables such as cultural, socio-economic, procedural, institutional or administrative distance/proximity of countries and policy characteristics leading to a low probability of convergence in the case of highly redistributive policies due to expectable distribution conflicts (Knill 2005:770f.; cf. Holzinger/Knill 2005:779ff. and 792). Table 10: Causal/Facilitating Factors and Mechanisms of Policy Change and Convergence Factor Aspect / Mechanism Stimulus Response Causal mechanisms Independent problem-solving Parallel problem pressure Independent similar response Imposition Political demand or pressure Submission International harmonization Legal obligation through international law Compliance Regulatory Competition Competitive pressure Mutual adjustment Transnational communication Lessondrawing Problem pressure Transfer of model found elsewhere Transnational problem-solving Parallel problem pressure Adoption of commonly developed model Emulation Desire for conformity Copying a widely used model Theoretico-Empirical Frame of Reference 101 Factor Aspect / Mechanism Stimulus Response International policy promotion Legitimacy pressure Adoption of recommended model Facilitating factors Cultural similarity Country-group related Institutional similarity Socioeconomic similarity Policy type Policy-related Policy dimension Source: Combination of Knill 2005:771 and Holzinger/Knill 2005:780. Borrowing from economic literature on convergence, different types of policy convergence can be distinguished. The most widespread type of convergence is socalled ‘?-convergence’ that marks “a decrease in variation of policies among the countries under consideration” (Knill 2005:769; cf. Holzinger/Knill 2005:776). Conceptually aiming at categorising the increasing similarity among domestic solutions, this sigma-convergence “is found in areas where forces of economic globalization seem to play a dominant role” (Heichel/Pape/Sommerer 2005:831). The next type, ‘-convergence’, “occurs when laggard countries catch up with leader countries over time” (Knill 2005:769). The underlying principle of (economic) growth provides this type of convergence with its label alongside the economic growth coefficient (Heichel/Pape/Sommerer 2005:832). “Beta-convergence often goes along with sigma-convergence, as ‘growing together’ presupposes a process of catchingup by beta-convergence” (ibid.). As a possible consequence of beta-convergence, ‘convergence’, as a third type, “is measured by changes of country rankings [over time] with respect to a certain policy” (Knill 2005:769). “A low degree of similarity between rankings indicates a high mobility of countries over time” (Heichel/Pape/Sommerer 2005:833). Moreover, if rankings do not change significantly, this indicates at the quality of both ?-convergence and ?-convergence (ibid.). Finally, ‘?-convergence’ is to be found “when similarity change is operationalized by comparing countries’ distance changes to an exemplary model” (Knill 2005:769), such as a supra- or international one. This last conceptualisation of convergence is closest to the assumed impact of Europeanisation on domestic political systems. Moreover, “sigma- and delta-convergence often occur simultaneously. If countries reach total similarity relative to a policy model, variance between them is obviously reduced” (Heichel/Pape/Sommerer 2005:833). Theoretico-Empirical Frame of Reference 102 Table 11: Types of Policy Convergence Type of Convergence Characteristic ?-convergence Increasing policy similarity -convergence Worst performers pull alongside best performers -convergence Alteration of policy-related country ranking across time ?-convergence Bench-marking in order to measure change of similarity among countries towards a common model Source: Own compilation based on Knill 2005:769. Nevertheless, besides its benefit for theoretically conceptualising policy change more in detail, critical voices–inspired by historical institutionalist ideas–comment on the empirical benefit of the concept of policy convergence by emphasising that “important differences in national institutions and opportunity structures for domestic actors” (Knill 2005:764) are main reasons for remaining domestic public policy divergence, hampering such a thing as cross-national convergence through EU provisions. As a result of this criticism, the fact that Europeanisation does not exclusively trigger convergence, but equally impacts on national divergence to persist is broadly discussed within the academic literature (cf. Bomberg/Peterson 2000:34). Given the overall empirical focus of the present study, it seems not only acceptable–although perhaps slightly confusing at first glance, while, however, analytically convincing in the end–to introduce these different degrees of policy convergence in order to measure change of national employment policies and policy co-ordination influenced by the underlying ideas, paradigms and priorities of the EES, that is, to measure the potential Europeanisation impact of the EES on domestic public policy change in terms of convergence towards its underlying ideas (cf. chapter 3.4.3, table 50). Europeanisation and Institutional Change: Institutional Isomorphism on the Winning Track? Institutional isomorphism empirically concentrates on the analysis of change in institutional and organisational structures in terms of decreasing divergence and increasing similarities. Beyond and by public policy change, Europeanisation can impact on the domain of domestic structures, effecting domestic institutional change (cf. chapter as “Europeanised policies change state-society relations…, empower technical bureaucracies, [and] change the institutions of economic policy” (Radaelli 2004a:14). Changes in formal and informal institutions are generally inspired by political interaction in decision-making processes (cf. March/Olsen 2005:5; Peters 2000:11). They stand for institutional change, which is more likely to occur “if the paradigms Theoretico-Empirical Frame of Reference 103 and norms of an institution are transforming and patterns of behaviour are changing” (Knodt 2004:704) and if new institutional provisions correspond to existing institutional logics, that is, in case of low or medium adaptation pressure (Knodt 2005:41 and 44; cf. March/Olsen 1989). These changes can affect the institution as a whole or be related to the cognitive/normative dimensions of enshrined ideas, values and norms (cf. March/Olsen 2005:15; Peter 2000:7). While the opportunity of institutional adaptation and structural change are accentuated (cf. Benz 2004a:27; Knodt 2005:18 and 40), historical institutionalism, strongly “emphasizes the ‘stickiness’ of identities and the ‘path-dependency’ of institutions” (Schmitter 2004:49)28, focusing “on how the element of willed change is influenced and constrained by existing institutional arrangements” (Olsen 2002:930; cf. chapter So, even if institutions are perceived to be non-static, in constant flux and evolution (March/Olsen 2005:8 and 13; Stone Sweet/Sandholtz 1998:16), the range of institutional change is viewed to be rather limited and dependent on institutional traditions, cultures and past paths of development (cf. Bulmer 1997:8; Genschel 1995:5; March/Olsen 2005:9; Risse/Cowles/Caporaso 2001:2f.) given that institutions “are relatively stable elements of political life” (Olsen 2002:925). This assumption is based on the idea that institutions provide for a commonly acknowledged framework of rules and procedures that structures interaction among political actors (March/Olsen 2005:8). As a result–following the assumptions of sociological institutionalism–“individuals adhere to these patterns of behaviour because deviation will make the individual worse off than will adherence” (Hall/Taylor 1996:8). Nevertheless, once changed, patterns of interaction and behaviour are adapted to new institutions (cf. Stone Sweet/Sandholtz 1998:17) as they–be they national or supranational–impact on the change of decision-making, political interaction, preference building, and strategic planning at all political levels of the EU and national political systems (ibid.:19; cf. Genschel 1995:5). Intentional institutional change, however, is perceived to be hardly steerable by ex-ante intentional institutional design (cf. March/Olsen 1989:5f. and 2005:9, 13f. and 19; Peters 2000:6; Pierson 1998:30). Institutional change can be explained by the fact that it increases the overall legitimacy of institutions and/or organisations and very much depends on their existing patterns (Fligstein/Stone Sweet 2002:1211; Hall/Taylor 1996:16 and 20). Thus, institutional change often follows the adaptation logic of ‘selective mimesis’ and ‘lending’. Generally, different mechanisms of institutional change can be identified. 28 Cf. Bulmer 1997:7; Eising 2003:398f.; Hall/Taylor 1996:8f.; March/Olsen 2005:14, 2004:12; Olsen 2002:925; Peter 2000:3 and 5; Pierson 1996, 1998:34, 44 and 46, 2000; Stone Sweet/Sandholtz 1998:19. Theoretico-Empirical Frame of Reference 104 Table 12: Mechanisms of Institutional Change Mechanism Impact imposed • Revision of constitution • Treaty revision • Formal intergovernmental or ‘inter-organ’ arrangements via integration • Participation in policy-shaping within new networks and arenas and initiation of policy-learning by this means • Joint development of new and shared perceptions of the definition of problem-solving and patterns of problemsolving • Confrontation with new demands by offer • Active propagation of desirable principles of order • Offering of access points/options for the use of new resources–“windows of opportunities” Source: Translated version of Knodt 2005:47. With a view to Europeanisation through the OMC, a main source of institutional change through “experimental learning and competitive selection” (Olsen 2002:932)–even if not undisputed in its real impact (cf. Giuliani 2003:152)–is “institutional isomorphism, suggesting that institutions which frequently interact, are exposed to each other, or are located in a similar environment develop similarities over time in formal organizational structures, principles of resource allocation, practices, meaning structures, and reform patterns” (Börzel/Risse 2003:66; cf. ibid.:72; Cowles/Risse 2001:224; Knill 2005:768; Lodge 2001:2; March/Olsen 2004:13; Risse/Cowles/Caporaso 2001:16f.; Sturm 2005:118) and move towards institutional convergence. Institutional isomorphism, moreover, also relates “to the mechanisms through which organizations become more similar over time” (Knill 2005:768). It can be inspired by the three sources “coercive means via resource dependence on a higher level of government, mimetic sources in which actors in a state of uncertainty search for models which they perceive to be legitimate and appropriate, and normative pressures which result from an increasing professionalisation” (Howell 2004a:11; cf. Risse/Cowles/Caporaso 2001:17). Alongside sociological institutionalist logics that accompany the analysis of institutional isomorphism and especially due to the fact “that the European Union itself … displays the typical features of a consensual system” (Giuliani 2003:136), consensus-oriented system are assumed to better perform at EU level than majoritarian system. Institutional isomorphism is, hence, a result of a combination of systemic similarities and increasing interactions of actors of different levels of the EMLG systems within a growing number of policy areas, applying most different new policy instruments and modes of governance based on learning and ideational convergence, stimulated by peer review or bench-marking, rather than on coercion (Knill 2005:768; Lodge 2001:11; Mörth 2003:174; Radaelli 2004b:18). Theoretico-Empirical Frame of Reference 105 Another source for institutional change is misfit between European and national institutional arrangements (cf. Eising 2003:399; chapter “The more European norms, ideas, structures of meaning, or practices resonate (fit) with those at the domestic level, the more likely it is that they will be incorporated into existing domestic institutions … , and the less likely it is that European norms will lead to domestic change” (Börzel/Risse 2003:67; cf. Giuliani 2003:136). As a consequence, the more misfit is to be found, the more adaptation pressure will be exerted the less likely mere incorporation in present national institutions becomes and the more likely institutional change occurs at national level as a consequence. “The growth and decay of institutions, roles, and identities, with their different logics of action are therefore key indicators of political change” (March/Olsen 2004:11). Institutional isomorphism, convergence, and “homogenization of organizational structures” (Börzel/Risse 2003:66, cf. ibid. 2000:12) are possible, but not necessary outcomes of the Europeanisation of domestic institutions exposed to change emanating from European provisions. The direction and degree of Europeanisation strongly depends on ‘national editing’ and certain intervening variables (cf. Mörth 2003:162f.; cf. chapter and 2.2), such as the institutional capacity to support change, policy structures conducive to Europeanisation or–following an historical institutionalist logic–state traditions diversifying the outcomes of Europeanisation among member states (cf. Andersen 2004:8; Börzel 1999:575; Börzel/Risse 2003:66; Cowles/Risse 2001:232). So, there “is no evidence that domestic institutional change mean[s] the comprehensive rejection of national administrative styles, legal cultures, societal relationships, and/or collective identities” (Börzel/Risse 2003:72). The ‘Whole Picture’ of Europeanisation: Key Elements Boosting or Blocking Domestic Change Generally and in view of analysing the impact of Europeanisation on domestic change, a frequently articulated “problem is that to cast the discussion on Europeanization exclusively in terms of its effects means to assume that … there are EUinduced effects!” (Radaelli 2003a:50). Therefore, the analysis of Europeanisation needs to be contextualised, taking into account that the presence of other elements enhances or “constrains the institutional capacity to produce change” (Radaelli 2003a:46). Hence, the degree of Europeanisation, that is, its category (cf. chapter, is not influenced by the degree of misfit alone (cf. Anderson 2002:798; Auel 2005:304; Bulmer/Radaelli 2004:9; Börzel/Risse 2003:58; Eising 2003:407; Radaelli 2004a:7 and 16; Treib 2004:52ff.). This amendment of the theoreticoempirical frame of reference helps to avoid one of the pitfalls of Europeanisation, that is, a theoretical over-generalisation (cf. Bomberg/Peterson 2000:5; Treib 2004:55), “prejudging the role [and impact] of Europeanisation” (Radaelli 2004a:5). Within the academic literature, elements boosting or blocking Europeanisation are generally labelled ‘intervening variables’, “scope conditions” (March/Olsen Theoretico-Empirical Frame of Reference 106 2004:17; cf. Treib 2004:55) or “facilitating factors” (Börzel/Risse 2003:63). They are largely related to institutional, cultural, and economic peculiarities of national political systems (Lenschow/Liefferink/Veenman 2005:798 and 801f.). Their integration into the research design underlines their influence on “the likelihood and direction of Europeanisation” (Radaelli 2003a:47; cf. Börzel 2002:208; Börzel/Risse 2000:1; Bulmer/Radaelli 2004:14; Howell 2004a:10; Risse/Cowles/Caporaso 2001:2). In a first definition–based on rational choice and sociological institutionalist assumptions (cf. chapter–four main sets of intervening variables can be identified. Based on rational choice institutionalism, the first two are related to a possible influence on a re-shaping of national power relations and political opportunity structures, differently empowering political actors. In line with sociological institutionalism, the latter two sets are viewed to influence the Europeanisation of national belief systems, policy paradigms, norms, and values. The four sets of main intervening variables are: (1) existence of multiple veto points strengthening actors’ capacities to block Europeanisation, (2) formal institutions facilitating change by offering conducive resources, (3) norm entrepreneurs in favour of Europeanisation as well as a (4) political culture supportive to consensus-building29. Graph 4: Factors Facilitating Change Policy / Institutional Misfit Pressure for adaptation New opportunities and constraints New norms, ideas an d collective understandings Factors facilitating change Low number of veto points Supporting formal institutions Factors facilitating change Norm entrepreneurs Cooperative informal institutions Redistribution of resources Socialization and social learning Differential empowerment Norm internalization Development of new identities Domestic change Source: Börzel/Risse 2003:69. 29 Cf. Börzel/Risse 2003:58f. and 64ff., 2000:1ff.; Featherstone 2003:16; Lenschow/Liefferink/Veenman 2005:798; Olsen 2002:926; Risse/Cowles/Caporaso 2001:9; Treib 2004:63ff.. Theoretico-Empirical Frame of Reference 107 These four sets of intervening variables can be subsumed under three wider fields alongside the conceptual differentiation of the three Europeanisation domains (cf. chapter These three wider fields are the domestic institutional capacity, the timing of EU policies and the policy structure under examination (cf. Risse/Cowles/Caporaso 2001:9ff.; cf. table 13). Concerning the domestic institutional capacity for change (domestic structures), several particularities of national political systems influence the degree and direction of Europeanisation (cf. Börzel 1999:577; Börzel 2003:6; Bulmer/Radaelli 2004:9; Eising 2003:396; Giuliani 2003:136; Lodge 2001:1; Olsen 2002:925; Risse/Cowles/Caporaso 2001:10). Given that political leadership “can be integrated or, at the other extreme, fragmented, short-lived, and conflict-ridden” (Radaelli 2003a:47), the number of veto players (cf. Tsebelis 1999 and 2002)–“defined as those actors whose consent is needed in order to produce any political change” (Giuliani 2003:136)–within domestic arenas varies and impacts on domestic change. Integrated leadership, as in the case of the UK (cf. chapter, theoretically minimises the number of veto players and maximises the impact of Europeanisation. Fragmented leadership, as in Germany (cf. chapter, theoretically maximises the number of formal and informal veto players30 and, thereby, tends to decrease the potential impact of top-down Europeanisation31. So, the “more power is dispersed across the political system, and the more actors have a say in political decisionmaking, the more difficult it is to foster the domestic consensus or ‘winning coalition’ necessary to introduce changes in response to Europeanization pressures” (Börzel/Risse 2003:64; cf. Börzel 2003:6; March/Olsen 2004:15; Radaelli 2000:22). As a consequence, the more veto points or veto players are to be found, the more difficult Europeanisation becomes at the domestic level (cf. Giuliani 2003:136f. and 143f.). With a view to bottom-up Europeanisation the opposite effect can be witnessed. Here, a “high number of domestic veto players are likely to increase the capacity of Member States to shape EU policy outcomes because they can ‘tie their hands’ to the preferences of their constituencies” (Börzel 2003:7; cf. Börzel 1999:577; Risse/Cowles/Caporaso 2001:9). Compared to other facilitating factors, the element of veto points/veto players is, hence, assessed to be of pivotal impact on the domestic effects of Europeanisation (Giuliani 2003:150), conceptually adding a more actor-centred perspective to the misfit assumption. This conceptual amendment responds to the criticism of institution-based explanatory elements not being in the position to elucidate change and adaptation in case of medium adaptation pressure (cf. Treib 2004:52). 30 Informal veto players are for instance “trade unions, entrepreneurial associations, regional governments, and courts, whose preferences and traditional rules of behaviour may conflict with the process of Europeanization and aggravate the political misfit of a member state” (Giuliani 2003:152). 31 Cf. Bailey 2002:792; Börzel 2003:6f.; Börzel/Risse 2003:64; Börzel/Risse 2000:7; Bulmer/Radaelli 2004:9; Cowles/Risse 2001:226; Giuliani 2003:152; Héritier 2001b:44 and 53f.; Lenschow/Liefferink/Veenman 2005:803; Radaelli 2003a:47, 2000:21f.; Risse/Cowles/ Caporaso 2001:9; Treib 2004:64. Theoretico-Empirical Frame of Reference 108 Also within the wider field of domestic institutional capacity, national institutional settings play an important role as they influence “the distribution of resources among domestic actors affected by Europeanization. The result is that the impact of Europeanization is contingent on institutional factors. A corollary is that Europeanization will produce diversity rather than convergence, because domestic institutions differ widely” (Radaelli 2003a:45, 2000:20; cf. Giuliani 2003:143 and 152). The last point strongly underlines the relevance of historical institutionalist assumptions with a view to the potential and degree of Europeanisation of domestic structures. Thus, domestic administrative capacity and traditions, intergovernmental relations as well as co-ordination mechanisms impact on Europeanisation (cf. Börzel 2003:7f.; Héritier 2001b:44; Radaelli 2000:8; Sturm/Pehle 2005:29; Treib 2004:49 and 60). They “can provide actors with material and ideational resources necessary to exploit European opportunities and to promote domestic adaptation” (Börzel/Risse 2003:65; 2000:7). Among them party system constellations (cf. Treib 2004:64), majority or consensus model democracy (cf. Giuliani 2003:136), administrative traditions, the integration of private actors as well as public interest groups (cf. Eising 2003:403; Peterson 2003:5; Sturm 2005:120; Treib 2004:63). Nevertheless, in line with the core historical institutionalist assumption of path-dependence, national institutional arrangements can also block Europeanisation given that “national traditions of regulation are so embedded in their domestic context that adaptation may not give rise to convergence: whether in response to EU or global stimuli” (Bulmer/Radaelli 2004:10; cf. Sturm 2005:122; Treib 2004:50). With a view to the second wider field of intervening variables (cf. table 13), that is, the timing of EU policies, political background variables such as political party constellations, national elections or domestic reform processes are to be taken into account (cf. Eising 2003:406). “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” (Radaelli 2004a:9). Therefore, if the national ground for reform is already laid, Europeanisation is more likely to impact than otherwise (cf. Radaelli 2000:22). At the same time, in case of temporal parallelism, it is more difficult to relate domestic change clearly back to Europeanisation as domestic political circumstances impact on the timing of transposition, implementation. They, thus, influence the national pace of Europeanisation. “For example, certain policies of liberalization in the European Union have caught some countries unprepared whereas others, most notably the United Kingdom, were already on their way to deregulation and privatization” (Radaelli 2003a: 47f.). In order to grasp this parallelisms in ‘time, timing and tempo’ (cf. Goetz 2000:223), “[c]areful process-tracing and attention to the time sequences between EU policies and domestic changes allow .. to distinguish between globalization effects, on the one hand, and the impact of Europeanization” (Risse/Cowles/Caporaso 2001:4; cf. Treib 2004:54) on the other. This aspect of ‘time’, thus, provides for explanative elements in view of the constellation of intervening variables at the time under examination, while the ‘timing’ of EU policies Theoretico-Empirical Frame of Reference 109 sheds light on the sequence of stimuli, and the ‘tempo’ refers to process of and capacity for domestic reforms (Goetz 2000:223ff.). The final wider field of intervening variables (cf. table 13)–policy structures and advocacy coalitions (public policy and cognitive/normative structures)–combines the domestic policy level with the structural level of policy-making within policy networks. It emphasises the role of administrative and technocratic elites–strengthened by Europeanisation–as central actors for running certain policies, while policy network constellations in need for wider participation are assumed to diminish the likelihood of Europeanisation (cf. Radaelli 2003a:48). These actors are perceived to be “guided by collective understandings of what constitutes proper, that is, socially accepted behaviour in a given rule structure” (Börzel/Risse 2003:65). In the Europeanisation process, political activities of certain norm entrepreneurs boosting Europeanisation of domestic policies can lead to the change of domestic policy paradigms, ideas and discourses that are challenged by European provisions (cf. Risse/Cowles/Caporaso 2001:5). Moreover, the existence of such norm entrepreneurs (that is, epistemic communities or advocacy coalitions, cf. chapter, pressuring “policy-makers to initiate change by increasing the costs of certain strategic options” (Börzel/Risse 2003:67, cf. 2000:9), is to be analysed within this wider field. With this group formation, means of arguing and persuasion as well as a political culture supportive to consensus-building provide “new opportunities for some actors ... severely constraining other actors’ freedom of manoeuvre” (Börzel/Risse 2003:65; 2000:9; cf. Eising 2003:399). This third field of intervening variables, moreover, includes core policy characteristics/paradigms and, with a view to national employment policies, also underlying welfare state traditions as well as socioeconomic preconditions of domestic systems (Featherstone 2003:15; Lenschow/Liefferink/Veenman 2005:800; cf. chapter 2.2.2). Adding to the influence of these intervening variables are also “differences in the degree to which domestic norms and institutions change in response to international institutional arrangements” (Börzel/Risse 2003:66), that is, not only to European, but also to international adaptation pressure. It is, thus, a major obligation to separate domestic change caused by Europeanisation from change inspired by other stimuli such as globalisation or international economic interactions (Cowles/Risse 2001:221; Olsen 2002:926 and 937; Risse/Cowles/Caporaso 2001:3f.). In order to do so, the above three wider fields of intervening variables are amended a fourth wider field comprising supra- and international constellations (cf. table 13). A starting point for this fourth field is provided by the so-called ‘second-image reversed perspective’ of International Relations (Gourevitch 1978 and 1986; cf. Börzel/Risse 2003:66; Lenschow/Liefferink/Veenman 2005:798f.; Mörth 2003:159; Radaelli 2000:6; Risse/Cowles/Caporaso 2001:5; Zandstra 2004:6). This perspective especially highlights the influence of supra-/international constellations on national policy change. It defines them as important factors not to be neglected when focusing on “the original second image, that is, domestic factors shaping the process of policy change and influencing patterns of convergence” (Lenschow/Liefferink/Veenman2005:799). The ‘second-image reversed perspective’, thus, offers additional Theoretico-Empirical Frame of Reference 110 explanative potential for the analysis of Europeanisation. It, yet, has to be extended by the assessment that “states are porous in a globalized world and that borders between the internal and the external processes of the state are diffuse” (Mörth 2003:160). So, given that Europeanisation of national political systems within the EMLG system can be supported both in content and timing by parallel events at the supra- and international level (cf. chapter, a strict separation between the domestic, supra- and international political level, as proposed in the original secondimage reversed perspective, does not seem feasible in the present study. Due to the described sets of intervening variables, rather than a consistent impact of Europeanisation, a “differential impact of Europe”32 is most probably to be found across national borders and policy areas, even if Europeanisation is exerting some sort of uniform adaptation pressure on national political systems to adapt.33 Table 13: Categories of Key Intervening Variables Explaining Europeanisation 1. Institutional capacity to produce change (Europeanisation Domain: Domestic structures) • Veto players in the political system • Scope and type of executive leadership • Formal institutions providing conducive resources • Administrative capacity and co-ordination mechanisms 2. Timing of European policies 3. Policy structure and advocacy coalitions (Europeanisation Domain: Public policy and cognitive/normative structures) • Technocratic capture potential • Adoption-implementation balance • Presence of a legitimising policy discourse • Impact of EU policy on domestic advocacy coalitions • Norm entrepreneurs • Political culture supportive to consensus • Policy characteristics • Socio-economic conditions 4. Supra- and international constellations • Provisions of supra-/international organisations • International framework conditions of the policy field Source: Integrated and amended version of Radaelli 2003a:47, Börzel 2003:8 and Börzel/Risse 2003:58f.; italics amended by the author of this book. 32 Börzel/Risse 2003:66; cf. Anderson 2002:797; Auel 2005:303; Börzel 2003:3; Eising 2003:395; Héritier 2001b:44; Mörth 2003:163; Olsen 2002:933; Radaelli 2004a:5; Risse/Cowles/Caporaso 2001:7; Sturm 2005:104. 33 Cf. Börzel 2003:3; Bomberg/Peterson 2000:7; Featherstone 2003:15; Héritier 2001b:53; Ladrech 2004:63; Olsen 2002:932; Radaelli 2004a:3, 2000:8; Risse/Cowles/Caporaso 2001:8. Theoretico-Empirical Frame of Reference 111 In view of this study’s analysis, the following intervening variables are identified as potentially relevant for domestic adaptation and will be analysed more in-depth in the next sub-chapter: • Institutional capacity (Europeanisation Domain: Domestic structures): o Existence and number of veto points within national political systems; o National political institutions offering or suppressing supportive resources. • Policy structure and advocacy coalitions (Europeanisation Domains: Public policy and cognitive/normative structures): o Domestic socio-economic conditions and welfare state traditions; o National employment policy characteristics; o National cleavages and interest representation related to employment policy. • Timing of reforms: o National political and economic background variables with a view to the period under analysis. • Supra- and international constellations: o Supra- and international influences as sources of adaptation pressures. 2.2 Intervening Variables Boosting or Blocking Domestic Change caused by Europeanisation The present chapter enters the empirical part of the theoretico-empirical frame of reference. It deals with key intervening variables of both the chosen policy area and the political systems of the UK and Germany. Its first sub-chapter examines the domestic institutional capacity to boost or block Europeanisation. It especially highlights central veto points of the national political systems and main institutional arrangements that offer or restrict resources and access, such as the central state executive (including the head of state and the government), EU-related policy-making and co-ordination structures, the parliament (including patterns of the respective domestic political party system) as well as national interest organisations. Moreover, as the EES requires and strongly promotes the involvement of social partners into the process of setting up the annual National Action Plans (NAPs) (cf. chapter 3.2), a view at relevant social partnership arrangements, domestic structures of (tripartite) social dialogue and related traditions within the UK and Germany is of special interest–also in order to define domestic advocacy coalitions/epistemic communities. So, the first sub-chapter goes, moreover, into detail concerning key national cleavages and interest representation structures related to employment policy, identifying potential norm entrepreneurs in the policy field. The second sub-chapter analyses the underlying socio-economic conditions in the UK and in Germany, such as central welfare state traditions and employment policy characteristics. The third sub-chapter

Chapter Preview



Mit ihren spezifischen Merkmalen als neues Politikinstrument – wie etwa ihrem rechtlich nicht bindenden Charakter, dem Ziel des gegenseitigen Politiklernens durch Austausch bester Praktiken oder gemeinsamen Evaluierungsprozessen – stellt die Europäische Beschäftigungsstrategie (EBS) und die mit ihr Anwendung findende Offene Methode der Koordinierung (OMK) beschäftigungspolitische Akteure in der EU vor die neuen Herausforderungen von Politik-Koordinierung, die die Politikgestaltung im europäischen Mehrebenensystem neu prägen.

Das vorliegende Buch beschäftigt sich intensiv mit diesen unterschiedlichen Facetten der EBS und ihrer Wirkung. Es geht dabei über bisherige Einzelstudien zur EBS hinaus und befasst sich nicht nur mit deren Entstehung, Entwicklung und Merkmalen. Es kontrastiert vielmehr den eigenen Anspruch der EBS mit ihrer politischen Realität und untersucht theoretisch hoch reflektiert deren Einfluss auf Politik-Koordinierungsstrukturen, Beschäftigungspolitiken und zugrunde liegenden Ideen sowie deren Zusammenspiel mit anderen wirtschaftspolitischen Bereichen. Neben der EU-Ebene dienen Großbritannien und Deutschland als Fallbeispiele für mitgliedstaatliche Anpassungsprozesse. Das Buch verankert seine Wirkungsanalyse sehr fundiert in der wissenschaftstheoretischen Debatte um Europäisierung und Politikkonvergenz, um deren Anwendbarkeit im Falle der EBS kritisch zu analysieren. Es komplettiert damit Europäisierungsstudien zu regulativer Politik durch die Analyse des Einflusses weicher Politik-Koordinierung im europäischen Mehrebenensystem.