Comparative research on interest group politics in Europe. Project application. Daniel Naurin Department of Political Science University of Gothenburg

Transcription

1 Comparative research on interest group politics in Europe. Project application Daniel Naurin Department of Political Science University of Gothenburg Dear participants at the SWEPSA workshop, This is an application for research funding submitted to the European Science Foundation. We do not yet know whether it will be successful or not, but I would still be interested in discussing the ideas with you. The project includes eight different country contributions in Europe and a sister-project in the USA. I will not force all the application s 80 pages on you, but have kept only the introduction and the section on the country contribution for which I am responsible. Looking forward to meeting you in Göteborg Daniel

2 European Collaborative Research Projects in the Social Sciences (ECRP) ECRP VI (2010) Project title: Comparative research on interest group politics in Europe (INTEREURO) SECTION ONE: SUMMARY OF THE COLLABORATIVE RESEARCH PROJECT 1.1 Main aims of the Collaborative Research Project (Max. 5 aims/150 words) The main purpose of this CRP (INTEREURO) is to promote a more comprehensive theoretical and empirical understanding of the role interest groups play in the European polity. Specifically, we will examine interest group mobilization; organizational maintenance and professionalization; strategies for influencing political decision-making; framing processes; and their impact on policy outcomes. The project will link the different aspects of the role of interest groups in the European policy process in an integrated theoretical framework. Our guiding intellectual tenet is the generation and strategic use of information by organized interests. To confront our theoretical contribution with empirical data, we will develop a comprehensive database that contains information on the individual stages of the influence production process at the national and EU levels. Employing an issue-centred sampling approach we will combine bottom-up with top-down sampling of interest groups and integrate quantitative and qualitative research methods. 1.2 Potential impacts (academic and non-academic) of the Collaborative Research Project (Max. 200 words) By providing information and other input to the policy process, interest groups can improve the quality and legitimacy of EU decision making. Groups also use access to decision-makers to advance their interests. Who are these lobbyists? How do they organize? What are the implications for public policy and democracy in the EU? These are key research questions in the fields of civil society, lobbying and public policy. While research on European interest group politics and civil society has increased substantially during the past 15 years, very little collaborative and systematic comparative work has been conducted. The CRP will develop an integrated theoretical framework and generate a multi-level, multi-country data set. It will be the first study covering all important dimensions of the lobbying process in order to generate a joined-up understanding of interest representation in compound polities. The impact of this research will extend beyond academia, being of direct benefit to policymakers, interest groups, and the media. It will shed much needed light on issues such as transparency and lobbying, the democratic deficit, the themes of biased access and influence, and interest group regulation as a potential remedy of bias. 1.3 Added value of the multinational collaboration (Max. 200 words) This project will provide an intellectually and methodologically robust investigation of interest group behaviour. The primary shared research focus is the EU, but the integrated modular comparison of national and EU systems allows us to account for the role of contextually driven forces in different segments of the influence production process. The collection and analysis of the data at the EU and national levels, necessary for testing the hypotheses that will be developed, cannot be conducted by single country teams or by national teams working discretely. The testing of a comprehensive theory requires systematic collaboration among several teams to build an integrated, large-n data-set. The multinational collaboration will allow us to propose nuanced contextualized explanations pertaining to old and young democracies, older and more recent members of the EU, small and large countries, and politico-economic systems with corporatist as well as pluralist institutional structures and traditions. The research teams include both highly regarded and newly emerging scholars with all the necessary expertise to undertake the research necessary for generating a robust understanding of the European policy process. The project will benefit from cooperation with a sister study to be conducted in the

3 US by leading interest group scholars (Baumgartner and Mahoney). 1.4 Data handling aspects (if relevant): quality assurance, storage, access (Max. 200 words) Throughout the project, country teams will collaborate closely to develop a single coding protocol that ensures comparability and equivalence across countries and levels. A steering committee (Beyers, Dür, Eising, Lowery, Maloney) will coordinate and guide this process. Under this common guidance each country team will be responsible for a particular aspect of the data-collection process and maintenance. The existing experience and expertise of the country teams in different data collection methods will provide quality assurance. Country specific data and collaboratively generated data from the shared interview, document, and website coding elements of the project will be collated into a single database and will be made publically available in a downloadable format to all interested parties at the end of the project. This dissemination will become an important public good relevant to several audiences. SECTION TWO: DESCRIPTION OF THE COLLABORATIVE RESEARCH PROJECT The Project Leader should describe the collaboration using the structure below (Country Contributions will be described in detail in Section Three). (Max words, excluding annexes. Entries exceeding 2500 words will not be accepted) 2.1 Short description of the state of knowledge, illustrating the context of the proposal and explaining the originality of the proposed research. The contemporary political-institutional context is characterized by the blurring and transformation of traditional boundaries. Political responsibilities are increasingly fragmented, shared or divided among different political levels (transnational, EU, national and regional). National political authorities have to cope with a hollowing out of their competencies, and with various organized interests bypassing them. Recent years have also witnessed the decline of political parties as the aggregators of societal interests (Mair 2005). In response, there have been calls for the greater involvement of a diverse range of groups in contemporary democratic systems. The EU Commission s White Paper on Governance and the post-lisbon settlement, for example, are focused on reducing the EU s democratic deficit by bringing citizens in via a wide variety of civil society organizations and fora. However, our theoretical understanding and empirical knowledge of the role that interest organizations play in international and European multilevel systems remains limited. Current limited intellectual understandings of interest groups politics are the result of two parallel trends in the literature. First, we face a diverse collection of theoretical perspectives on interest group behaviour that fail to communicate effectively (see Beyers et al. 2008; Eising 2008). Analytic-descriptive approaches that focus on interest group institutionalization and patterns of state-society relations provide us with useful generic typologies, such as the corporatism-pluralism continuum, but these typologies are less useful for developing testable hypotheses on interest group behaviour in multi-level governance systems (Falkner 1998). Other explanatory approaches, such as the exchange (Bouwen 2002), framing (Baumgartner and Mahoney 2008), and venue shopping and agenda-setting perspectives (Princen 2009) have provided testable hypotheses on interest group behaviour, but the interconnections between these approaches have not been investigated. Second, until recently research on interest group behaviour was mainly based on qualitative case study designs. While these studies have generated valuable insights in specific sectors (e.g. Pedler 2002; van Schendelen 1993), they rely heavily on non-comparable research

4 designs in terms of sampling, measurement, and data analysis. This makes it difficult to combine their insights into a comprehensive understanding of group behaviour in the European policy process. More recently, research has developed a more quantitative and explanatory approach, focussing more extensively on a wider range of theoretical issues (Crombez 2002; Eising and Kohler-Koch 2005; Wessels 2004) and using quantitative research designs that facilitate steps toward comparable research strategies needed for the advancement of knowledge (Bernhagen and Mitchell 2009; Beyers 2004; Bouwen 2002; Broscheid and Coen 2007; Mahoney 2008). Although these studies test general propositions about interest group behaviour, they are often based on idiosyncratic research designs that make direct comparison and wider inference difficult. The political reality of fragmented compound polities, the related democratic deficit and limited knowledge accumulation on interest group behaviour mean that a comprehensive analysis of interest group politics in the EU would be intellectually timely. We aim to build upon existing work by developing a novel theoretical framework that connects different aspects of interest group behaviour and aims to construct a comprehensive dataset to test our hypotheses and contextualize accounts of interest group behaviour in the EU. In so doing, our ambition is to reconnect the study of European interest group politics to the broader scholarship on comparative and multi-level governance and to rescue interest group studies from its Cinderella status compared to political party scholarship (Beyers et al., 2008). 2.2 CRP aims and objectives; potential contribution to knowledge and academic and nonacademic impacts. The project s overall objective is to provide a better understanding of the influence production process in the EU. The theoretical framework incorporates different aspects of interest group politics and the research design will generate an integrated dataset enabling us to test implications of the theoretical framework. The project has broader relevance for studying the evolution of European democracy and the emergence of a European public sphere. It will be the first comprehensive attempt at studying the preferences, behaviour and influence of interest groups across the entire range of the European influence production processes. It will allow us to assess the contribution of civil society to the emergence of a European public sphere and the structuring of the European political space in a broad range of policy issues and in different politico-institutional contexts. 2.3 Research design, theoretical framework and methods. Include a work plan with time frame and description of the execution of the proposed collaborative work Our theoretical framework proceeds from the assumption that information is the key currency in policy making interactions. Policy-makers require scientific, technical, legal, economic and social (and implementation) advice and expertise as well as guidance on constituency preferences. Such information can: be seen as a tradable asset that facilitates political access (Bouwen 2002); affects beliefs and expectations about outcomes; and assists policymakers goal attainment priorities (Hall and Deardorff 2006). Interest groups differ in the kind of information they possess, their capacity to control it, and their ability to use that information as a strategic asset to gain political access and influence policy outcomes. The reliability, credibility and the quality of groups information will be influenced by organizational resources, such as money, professional staff, type of membership, and the size and importance of the sector or constituency they claim to represent. We organize our analysis around the stages of the influence production process. The process runs from the initial mobilization of societal groups through organizational maintenance and survival to the influence stage. In this process, many organizations professionalize their operations and apply a highly expert and scientific approach to selecting issues, strategies and frames within specific institutional contexts to influence outcomes. Information plays a

5 key role in all stages of the influence production process: POPULATION ECOLOGY: The study of the population ecology of groups is essential for our informational approach because the density of an interest group population is likely to have an impact on the scope and type of information that groups can offer to decision-makers, and on the strategic value of that information. The denser a population, the more information sources are open to policy-makers and the less exclusive the information provided by a group will be. ORGANIZATIONAL MAINTENANCE AND PROFESSIONALIZATION: Within the analysis of internal interest group dynamics (organizational maintenance and professionalization) and its links to external dynamics (influencing outcomes) it is important to understand the ability of interest groups to exploit information as a strategic asset. We suspect, for instance, that the level of professionalization affects the capacity of groups to select issues, frame them, employ expert and political information in lobbying strategies. Professionalization also affects groups capacity to link citizens more effective to the state and to offer a partial remedy to the (EU) democratic deficit. STRATEGIES: Finite resources and the limited amount of information that can be processed by one organization mean that interest groups can prioritize a small number of issues. Which issues groups prioritize partly depends on the type of information they possess. The type of information available to a group also affects the choice of strategy, because strategies differ in terms of the information that can be transmitted. Insider strategies can be particularly useful for the transmission of detailed operational and technical information, while outsider strategies inform policymakers of the size and the scope of political conflict. FRAMING: Framing processes connect the study of interest representation directly to policy outcomes because the way in which an issue is understood fundamentally influences the outcome of a policy debate. The link with information exists on two levels: framing studies highlight that policy information is socially constructed and not objectively given; and the nature and amount of information (technical or political) that is available to policy advocates in the EU political process shapes the frames actors seek to construct. INFLUENCE: Information is a crucial power source because many lobbying efforts aim to change policymakers perceptions and perspectives by widening the lens, presenting new evidence or concealing unfavourable information (Bernhagen 2007). Policy makers rely heavily on the information provided by interest groups about complex policy areas and grassroots preferences that can be provided by interest groups to attain their policy goals and objectives. Interest groups use their information capabilities technical expertise, data on markets and production costs, and information on citizen preferences to achieve political goals. These stages and their associated activities are connected in complex but yet underspecified ways and we seek to shed some much needed light on the various interconnections. To do so this project will tie the different phases of the influence production process to the informational dimension of EU policy-making. Research design Our research design is based on modular comparisons of national and EU interest group behaviour. The modular comparative approach is designed to test hypotheses on specific aspects of the influence production process. The inclusion of both national-level and EU-level analyses ensures that the design is sensitive to the multilevel dynamics of EU policy-making and the specific contexts in which organized interests are active. This requires variation on

6 key contextual variables such as corporatism versus pluralism; large versus small countries; young versus old democracies, and old versus new EU members. To achieve this, we will study interest group behaviour at the national level in six of the eight collaborating countries: Germany, the Netherlands, Slovenia, Spain, Sweden, and the UK. The data collection will proceed in several steps (see also figure 1). 1. We begin with a sample of 120 legislative proposals introduced by the European Commission from 2008 to 2010 (see annex III). 2. We then identify the population of policy advocates involved in each proposal, both at the EU and the national level in the eight member countries. Each country team is responsible for identifying the interest groups involved in their national issues as well as the identification of the EU-level advocates for a subset (n=15 per country team) of the 120 issues. This will entail extensive document analysis and some snowballing through a limited number of expert interviews (see annex II for the details). 3. Parallel to steps 1 and 2, the Dutch team develops an integrated census of the EU interest system using website coding of organized interests. This bottom up census is important for studying the population dynamics involved in the EU interest system and for weighing the results of the issue studies, as they are conducted on the basis of the top-down sampling of issue populations (see step 1 and 2). In this way, we can generalize from the sample results to the EU population of interests as a whole. 4. The next steps in the research project combine two modes of data-collection, one focusing on document analysis and another on interviewing. Each country team will be responsible for concrete aspects of the data collection (see Annex II and the separate projects): a. Detailed interviews will be conducted with public officials (120 face to face, plus 120 by phone) and interest group representatives (app. 150 face to face, covering an important share of the population of groups active on our sample of legislative proposals). Critically, these interviews will be based on the same questionnaires addressing the full range of substantive topics addressed in the specific projects covered by the country teams. 5. For the analyses of professionalization, framing, and influence, extensive document analysis and/or coding of websites is required (see the separate project descriptions). In addition to the shared EU research, the Spanish, German, UK, Swedish, Slovenian and Dutch teams will examine 20 issues in their national context, thereby providing comparisons between the EU level and the national level and across the member states. The sampling criteria and data collection methods for these national studies will be analogous to those applied to the EU-level data collection and include interviews with interest group officials (40) and key policy makers (20). Figure 1 shows how the different data collection segments are located within the duration of the project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

7 2.4 Describe the integration of the Country Contributions in the CRP and the added value of the multinational collaboration1 Much of the recent progress in the study of interest group politics has grown out of the EU funded CONNEX network, an ESRC funded interest group seminar series (Organized Interests: Democratic and Governance Issues), and a recently established ECPR Standing Group on Interest Groups. This proposal capitalizes on these networks and attempts to move towards an even more integrated research group including a mix of young and established scholars. In each of the countries, we have identified and secured the cooperation of experts in at least one of the constituent elements of the influence production process. Each country team will focus on a specific aspect of the influence production process, on how information plays a key role in that particular stage and its interconnectedness to the other stages. The first module led by Lowery (the Netherlands) will focus on population ecology, while a second module by Maloney (UK) will tackle the organizational development of groups. A further two modules (and an associated project) will examine interest groups involvement in the influence production process at the EU level focusing on political strategies (Beyers, Belgium) and issue framing (Naurin, Sweden; and Baumgartner and Mahoney, US). A fifth module will deal with the measurement of influence over public policy (Dür, Austria). The sixth module, finally, will focuses on national level interest group behaviour (Eising, Germany; Chaqués, Spain; and Fink-Hafner, Slovenia). Annex I: References Baumgartner, F. R. and C. Mahoney (2008). "The Two Faces of Framing. Individual-Level Framing and Collective Issue Definition in the European Union " European Union Politics 9(3): Bernhagen, P. (2007). The Political Power of Business. Structure and Information in Public Policy-making. New York: Routledge. Bernhagen, P. and N.J. Mitchell (2009). The Determinants of Direct Corporate Lobbying in the European Union. European Union Politics 10(2): Beyers, J. (2004). "Voice and access: Political practices of European interest associations." European Union Politics 5(2): Beyers, J., R. Eising and W. Maloney (2008). Conclusion: Embedding Interest Group Research. West European Politics 31(6): Bouwen, P. (2002). Corporate Lobbying in the European Union: The Logic of Access. Journal of European Public Policy 9(3): Broscheid, A. and D. Coen (2007). "Lobbying Activity and Fora Creation in the EU: Empirically Exploring the Nature of the Policy Good." Journal of European Public Policy. Crombez, C. (2002) Information, lobbying, and the legislative process in the European Union, European Union Politics, 3(1): Eising, R. (2008). Interest groups in EU policymaking. Living Reviews in European Governance 3(4). (accessed Feb. 2010). Eising, R. and B. Kohler-Koch (2005). Einleitung: Interessenpolitik im europäischen Mehrebenensystem. Interessenpolitik in Europa. Rainer Eising and Beate Kohler-Koch. Baden-Baden, Nomos: Falkner, G. (1998) EU Social Policy in the 1990s: Towards a corporatist policy community. London: Routledge. Hall, R. L. and A. V. Deardorff (2006). "Lobbying as a Legislative Subsidy." American Political Science Review 100(1): Mahoney, C. (2008), Brussels vs. the Beltway: Advocacy in the United States and the European Union. Washington DC. Georgetown UP.

8 Mair, P. (2005). Democracy Beyond Parties. Center for the Study of Democracy. Paper Pedler, R., ed. (2002). European Union lobbying. Changes in the arena. New York: Palgrave Macmillan. Princen, S. (2009). Agenda-Setting in the European Union. (New York: Palgrave). van Schendelen, M. P.C.M. ed. (1993). National Public and Private EC Lobbying. Ashgate: Dartmouth. Wessels, B. (2004). Contestation Potential of Interest Groups in the EU: Emergence, Structure, and Political Alliances. European Integration and Political Conflict. Gary Marks and Marco R. Steenbergen (eds.). Cambridge: Cambridge University Press: Annex II: Overview of specific research activities associated with each module The overview below presents the research modules run by the different country teams and the data collection modes they rely on. Note that issue and actor sampling precede establishing the issue advocacy population (see project outline above). Each team covers part of the work with regard to the identification of the interest groups involved in about 20 issues. The coordination of the management of this data will be done by the steering committee and this work needs to be done during the first sixth month of the project. Interviewing and surveying will start after some documentary-analysis as this enables us to take into consideration a lot of substantive information regarding the policy issues. In the bottom row of the table we list the members of the steering committee. Each member of the steering committee coordinates the research activities for a specific stage in the influence production process. The X-signs represent the responsibility of each national team. The work for the EU-level will be conducted by the Austrian, the Belgian, and the Dutch teams whereby they have the following division of labour:! Austrian team: taking the lead in building a measurement instrument for document analysis, detailed document analysis for measuring goal attainment, organizing a training session for coders; coordinating this work with the framing study that is conducted by the US-based associated project; coordinating and conducting the fieldwork on strategies in Brussels with the Belgian team;! Belgian team: taking the lead in designing an interview protocol on strategies; in charge of the bulk of the Brussels interviews; organizing a training session for interviewers; coordinating the conduct of the remaining interviews with the national teams;! Dutch team: conducting the bottom-up census of the EU interest group population and coordinating the bottom-up identification of issues and issue-populations. The work for the national level will be carried out in parallel by six teams, all of which concentrate on organizational adaptation and development, the domestic framing of issues as well as the strategies domestic interest groups deploy with regard to EU legislative issues. Each national team adopts the same research strategy which is equivalent and comparable to the EU-level study.

9 1%&#?FD-&(&-#/16579# # # Stages of the influence production process # P'.6-21*'0# Q);20*L21*'02-# R)24*0;# N1)21&;*&/# C08-6&03&# 5&(&-'.4&01#?F#<G$#JS)H#I?T# E# E# E# E# E# I&7&)/H#:T# T'+&)7>#?NP#<U%2O6V/># D# E# E# E# D# W?X#<?*/*0;># D# E# E# E# D# :T#<T'+&)7># D# E# E# E# D# N-'(&0*2#<R*0@D D# E# E# E# D# Y280&)># NZ?#<:26)*0># D# E# E# E# D# F[#<\2-'0&7># D# E# E# E# D# N1&&)*0;# U'44*11&&# T'+&)7# \2-'0&7#?*/*0;# I&7&)/# JS)# Annex III: Sampling policy proposals & identifying policy advocates Our project requires a stratified random sample of 120 legislative proposals introduced by the European Commission between 2008 and To arrive at this sample, we will do the following: 1. In a first step, we will rely on the European Commission s Prelex database ( /prelex/rech_avancee.cfm?cl=en) to create a list of legislative proposals introduced by the European Commission between 01/01/2008 and 31/12/2010. We will exclude proposals for decisions and recommendations from this list as the former tend to be directed at very specific recipients and the latter are not legally binding. We will, however, include green and white books and use them to establish a sample of issues that are not yet formally on the agenda. This allows us to study lobbying at the agenda setting stage. Limiting the list to proposals that have been introduced in 2008, 2009, and 2010 will allow respondents to recollect the details of the decision-making process. 2. We will then create separate lists of proposals for regulations, proposals for directives, and green and white books. This will allow us to create a stratified random sample, which will deliberately over-represent proposals for directives and green and white books. To illustrate, in 2008 the European Commission introduced 177 proposals for regulations, 82 proposals for directives, and 10 green and white books. A random sample including all proposals would most likely include only few of the latter two types of documents, which would be problematic for our project. We consider directives important because they need to be transposed and implemented within the member-states, which makes them theoretically relevant from a multi-level governance perspective. Green and white books are interesting because they allow for a study of the agenda-setting stage of the policyprocess. 3. We set apart legislative proposals that did not generate public attention at a moment equivalent for each proposal. These are likely more technical issues that do not motivate a public response. We will choose publicly salient proposals by checking whether it has been given substantive attention in Agence Europe, the Financial Times, Le Monde, and/or the Neue Zürcher Zeitung. 4. Drawing on our randomly sorted lists of proposals, we will repeat step 3 until we arrive at a random sample of 100 proposals ensuring a sample encompassing both highly and moderately visible issues. The stratified sample will include: 50 proposals for regulations, 40 proposals for directives, and 10 green and white books. We will oversample both proposals for directives and green and white books relative to their frequency in the Prelex universe of legislative proposals for the reasons noted above. 5. We randomly select another 20 proposals from those set aside in step 3 as a control. This sample of proposals that did not garner public attention will be stratified using the same

10 proportions as for the publicly-salient proposals: 10 for regulations, 8 for directives, and 2 green and white books. 6. Of the final list of 120 proposals, we will select 20 proposals based on the type of proposal, salience, level of controversy, and dimensionality that will be analysed in the national-level studies. We will carry out this work in late 2010 and early 2011 to allow for a start of the project in March of that year. The steering committee will develop a detailed protocol in the course of With the list of 120 proposals, we will identify the population of policy advocates involved in the policy-making process for each. The following considerations are important in this context: 1. Our definition of involvement is rather broad and could include many different aspects ranging from giving expert advice within a committee or advertising policy positions in the media. 2. We will concentrate on actors that are identifiable as separate organizations; this may include government agencies, party officials, interest groups, firms, trade unions, universities, cultural institutions, and so on.

11 COUNTRY CONTRIBUTION Principal Investigator: Country: ECRP Funding Organisation: Daniel Naurin Sweden Vetenskapsrådet 3.2 Description of Country Contribution Describe the specific competence and expertise of your country team with regard to the collaboration. Daniel Naurin is Associate Professor at The Centre for European Studies and the Department of Political Science at Gothenburg University. He has held positions as Marie Curie Fellow at the EUI in Florence, and visiting scholar at Sussex European Institute at Sussex University, and at the Swedish Institute of International Affairs in Stockholm. His research has focused in particular on lobbying and European Union politics, which means that he is well suited for the theme of the project. He is the author of Deliberation Behind Closed Doors: Transparency and Lobbying in the European Union (ECPR Press 2007). Other recent publications include Least Common When Most Important. Deliberation in the Council of the European Union, British Journal of Political Science (2010), Backstage Behaviour? Lobbyists in Private and Public Settings, Comparative Politics (2007), and the co-edited volume (together with Helen Wallace) Unveiling the Council of the European Union (Palgrave 2008) Detail your country team s contribution to the overall work plan. Aims and Objectives A theoretical basis for the overall CRP is the assumption that information is the key currency for interest groups trying to influence the policy-making process. The generation and strategic use of information includes assembling facts about the world and transforming these into persuasive arguments which can be presented to the decision-makers. It also, crucially, includes establishing which information is relevant for the issue at hand. An important part of lobbying therefore concerns defining what the issue really is about. Framing involves selecting and highlighting some features of reality while omitting others (Entman 1991, p. 53). How an issue is understood may fundamentally influence the outcome of a policy debate. Issuedefinition can determine what interests mobilize, how many mobilize, whether the governing party supports or opposes, and if there is an all-out battle or a quiet compromise. Getting everyone to debate an issue on your terms can dramatically improve one s chances of reaching one s policy objectives (Schattschneider 1960, Baumgartner and Jones 1993). Interest groups and policymakers often try to shift attention to the dimension or frame that strengthens their position on an issue, but they cannot single-handedly redefine an issue since all other interest groups and policymakers are simultaneously trying to do the same (Baumgartner et. al. 2009). So how is it that one dimension or frame, or a few, come to dominate on any given issue, even though most issues have many dimensions and could be discussed in countless ways? What is the process by which individual level framing attempts aggregate and a single dimension or frame comes to dominate the debate? Until now, it has been difficult empirically to investigate these questions but the accessibility of large stores of issue and position documentation and the development of new computer assisted text analysis allows us to map the process by which hundreds of individual discussions of an issue

12 aggregate to produce collectively dominant frames. This proposal lays out a strategy for collecting, coding and mapping the process by which ideas tip in the EU. Studying the EU requires attention to the multi-level character of the system. How are frames and framing strategies at the national and the supranational level connected? Are different framing games going on in different member states, and are they in turn different from the EU-level? A classic question in European integration research concerns the role of national vs. supranational factors for forming member states preferences and positions. We will approach this question in a new way by looking at how member state representatives to Brussels are affected by cross-pressure from different supranational and national frames. Outline This project is one section of a broader collaborative project on civil society organizations in the European Union. Scholars in nine countries are working together to study different aspects of organizations and their advocacy activity across 120 EU policy issues. The collaborative nature of the broader project will allow us not only to consider framing processes as a dependent variable but also as an independent variable influencing lobbying success and policy outcomes. To understand the variation in and effects of framing across member states and national and supranational levels we proceed in three stages. First, we compile publicly available and salient information (governmental, media, and interest group policy papers), on the 120 issues at the EU level and on 20 selected issues in six member states: Germany, the Netherlands, Slovenia, Spain, Sweden, and the UK. By studying these particular states we assure variation with respect to member state size and state-interest group model (statist, pluralist, corporatist) as well as number of years of democracy. The documentary data will be analysed with the purpose of defining which frames that are active in the issues at hand, i.e. in which different ways the issues are perceived and talked about. Secondly, we conduct interviews with the government representatives in Brussels, from the six selected member states, who were responsible for negotiating those 20 issues in the Council of the EU on which we also analyse national frames (120 interviews in total). The purpose is to establish if and how cross-pressure from different frames at the national and supranational level influence the understanding of the issues that these key intermediary actors in the EU multilevel system have. Third, a sub-sample of issues will be selected (5-6) for even more in-depth case studies. The Swedish country contribution outlined here takes the lead responsibility for the interview study in the second step. Furthermore, it is also responsible for the collection and content analyses of documentary data on the sub-sample of 20 issues in Sweden. It will also contribute to the overall project by identifying actor populations on the EU-level for 15 of the 120 issues, as well as conducting 60 interviews (40 interest group representatives, 20 public officials) in Sweden for the sub-sample of 20 issues. The affiliated US project will compile and analyse the documentary data on the 120 issues at the EU-level, while the German, Dutch, Slovenian, Spanish and British projects will take responsibility for the data collection and content analysis on the sub-sample of 20 issues in their respective domestic contexts. Finally, the German, Spanish and Slovenian contributions take responsibility for the in-depth case studies (which hence are described in that section of the application). Methods and Workplan First step: Publicly available documentation EU-level and Member State-level For every issue in our random sample of 120 cases we will analyze public position taking at the supranational level. Furthermore, for a sub-sample of 20 issues we will do the same in six member states. The selection of the 20 issues will be based on the type of document, level of controversy, salience and dimensionality. The analysis will be based on official statements from the EU and national government institutions and policymakers, consultation submissions and interest group position papers. The list of active participants on an issue will be constructed iteratively by triangulating information on active groups in the different sources of documents. The study of argumentation and framing at the elite level is presently being revolutionized by new developments in the systematic study of text as data (please refer to the affiliated US project for references and details). In order to identify frames and to assess the dimensionality of the debates in

13 the 120 issues at the EU level, and 20 issues at the national level in six countries, we will use a fully automated content analysis technique drawing on a combination of cluster and correspondence analysis. This technique has previously been successfully used to analyze framing and dimensionality of parliamentary speeches, speeches of presidential candidates and transcripts of the Federal Reserve s Federal Open Market Committee. Three software packages are available with the necessary functions: ALCESTE, T-LAB and Wordstat. We will use either of these programs depending on functionality and pricing. The process for deriving and analysing frames in the documentary data is described in the affiliated US project. The compilation of publicly available information will not only allow us to determine the different types of frames and their relative importance but also to establish the universe of actors active on each issue at the national (in the six member states) and EU level. In this way we can compare mobilization at the national and supranational level. Second step: PermRep interviews In addition to the content analysis of publicly available documents, we will conduct interviews with permanent representatives in Brussels (PermReps) of the six selected member states. The purpose is to analyse the effect of varying national and supranational frames and framing activities on these key decision makers. The PermReps are national actors embedded in the supranational context, and therefore potentially exposed to both national and supranational frames (Naurin and Wallace 2008, Hayes-Renshaw and Wallace 2006). Formally, these civil servants shall represent the positions of their member states governments in the Council of the EU, the main decision-making institution in the EU system. Negotiations take place in higher (including the ambassadors committee Coreper) and lower level committees and working groups within the Council, preparing the ministers meetings on which the formal decisions are taken. About 90 percent of the issues are in practice decided by the PermReps, only the most politically contentious are left for the ministers (Hagemann 2008). The PermReps thus constitute the core of the EU s negotiation machine. A contentious issue is to what extent the PermReps are socialised into thinking about issues in ways that dominate at the European level, rather than in their home member states (Lewis 2008, 1998, Beyers 2005, Checkel 2005). This question connects in turn to the classic question in European integration research of the degree of supranational influence on member states preferences and EU decisions (Moravcsik 1998, Sandholtz and Stone Sweet 1998, Haas 1958). Capturing such perception drift empirically is difficult, however. Anecdotal evidence in the literature states that national ministries sometimes view their Brussels representatives with some suspicion. Part of the PermReps job is to compromise on national positions in order for the Council to be able to take decisions even when member states disagree, which may lead to disappointment with the representatives in the capitals. This project proposes a novel way of capturing the potential supranational influence on the actual positions defended by the member states in the Council, by studying the frames active in the minds of the PermReps, and comparing them to existing frames at different levels. This makes it possible to analyse to what extent the PermReps are captured by the Brussels environment to adopt frames (and potentially, eventually, positions) that are different from those at the domestic level. Doing so will contribute not only to research on framing in multi-level systems, and the ability of national and EU level lobbyists to plant their favoured frames in the Council, but also to the broader European integration literature. Fundamentally, this study concerns the question of what kind of organization/political system the EU is: An organization for two-level nation state diplomacy, or a (multilevel governance) system for European politics (Hix 2008). Since there will be little if any public documentation which could distinguish PermReps frames from national governments frames (which are formally supposed to be the same) interviews are required for this part of the project. For the sub-sample of 20 issues we will interview those persons in the permanent representations from the six countries who belonged to the working group or committee which dealt most extensively with the issue at hand (120 interviews in total). This means that we will normally not interview the chief ambassadors (Coreper II) who are generalists dealing with issues that are left unsolved by the working groups. The interviews will be made over telephone, a method which has been used before by the principal investigator in studies of the Council of the EU (Naurin and Lindahl 2008 and 2009, Naurin 2010). Naurin and Lindahl interviewed 618 PermReps, and had a response rate of over 80 percent. The interviews will be semi-structured and developed on the basis of the findings of the documentary

14 content analyses. They will contain two main parts. First, to capture the framing of the issue which comes first to the mind of the officials we will ask an open question (approx.: "In your view, what is this issue about...?"). Thereafter, we will focus more specifically on the different frames found in the document analysis, by giving the respondents some alternatives (approx.: "In your view, is this issue mainly a question of (dimension) X, Y, Z..."). The same questions will be asked in the interviews of public officials and interest group representatives in the six national studies collecting data on the same 20 issues. By comparing frames and framing activities of lobbyists at the national and EU level we will be able to study if and how cross-pressure from different frames defended by different actors affect government representatives perceptions of the issues. Third step: Case studies For lack of space we refer to the German contribution for details on the third step in the framing module. Planned Output & Website Infrastructure As described in the affiliated US contribution we will develop a website where we will make all of the documentation publicly available for other scholars; detail our data collection process; and our coding techniques so that interested scholars could build on the work we are doing Annex I: References Baumgartner, Frank R. and Jones Bryan D Agendas and Instability in American Politics. Chicago: University of Chicago Press. Baumgartner, Frank R., Jeffrey M. Berry, Marie Hojnacki, David C. Kimball and Beth L. Leech Lobbying and Policy Change. Who Wins, Who Loses, and Why. Chicago: University of Chicago Press. Beyers, Jan Multiple embeddedness and socialization in Europe: The case of council officials. International Organization 59 (4): Checkel, Jeffrey T International institutions and socialization in Europe: Introduction and framework. International Organization 59 (4): Entman, Robert M Framing US coverage of international news: Contrasts in narratives of the KAL and Iran air incidents, in Journal of Communication, 43 (4): Haas, Ernst B The Uniting of Europe. Stanford: Stanford University Press. Hagemann, Sara Voting, Statements and Coalition-Building in the Council. Party Politics and National Interests. In Unveiling the Council of the EU. Games Governments Play in Brussels., edited by D. Naurin and H. Wallace. Basingstoke: Palgrave Macmillan. Hayes-Renshaw, Fiona, and Helen Wallace The Council of Ministers. 2. ed, The European Union series. Basingstoke: Macmillan. Hix, Simon What's wrong with the European Union and how to fix it. Cambridge: Polity. Lewis, Jeffrey Strategic Bargaining, Norms, and Deliberation: The Currencies of Negotiation in the Council of the European Union, in Unveiling the Council of the EU: Games Governments Play in Brussels, edited by D. Naurin and H. Wallace. Basingstoke: Palgrave Macmillan. Lewis, Jeffrey Is the 'hard bargaining' image of the Council misleading? The Committee of Permanent Representatives and the local elections directive. Journal of Common Market Studies 36 (4): Moravcsik, Andrew The Choice for Europe. Ithaca, NY: Cornell University Press. Naurin, Daniel Least Common When Most Important. Deliberation in the Council of the EU, British Journal of Political Science. 40: Naurin, Daniel and Rutger Lindahl East-North-South. Coalition-Building in the Council Before and After Enlargement. In Unveiling the Council of the EU: Games Governments Play in Brussels, edited by D. Naurin and H. Wallace. Basingstoke: Palgrave. Naurin, Daniel and Rutger Lindahl Out in the Cold? Flexible Integration and the Political Status of Euro-Outsiders, SIEPS European Policy Analysis,

15 Naurin, Daniel and Helen Wallace, eds Unveiling the Council of the EU. Games Governments Play in Brussels. Basingstoke: Palgrave Macmillan. Sandholtz, Wayne and Alec Stone Sweet European Integration and Supranational Governance, Oxford: Oxford University Press. Schattschneider. E.E The Semi-Sovereign People. Fort Worth: Harcourt Brace Jovanovich Collage Publishers.!