TEORIJA IN PRAKSA let. 53, 3/2016 605 Mitja HAFNER-FINK, Meta NOVAK, Danica FINK-HAFNER, Rainer EISING, Daniel RASCH, Patrycja ROZBICKA* GIANTS AND DWARFS: THE MULTILEVEL LOBBYING STRATEGIES OF NATIONAL INTEREST ORGANISATIONS1 Abstract. The article addresses the bias in interest rep- resentation within the EU by examining the lobbying strategies of national interest organisations within the EU’s multilevel political system. Both our theoreti- cal framework, which includes the determinants of a national interest organisation’s decision to act at the EU level, and the data analysis from the INTEREURO Multi-Level Governance Module (MLG) (www.intere- uro.eu) reveal three main findings. Firstly, the great- est differentiation among interest organisations (IOs) appears to be between those IOs from the older member states (Germany, the UK and the Netherlands), which exhibit above-average levels of activity, and those from the newer EU member states (Sweden, Slovenia), which exhibit below-average levels of activity. Secondly, the variations in IO activity levels are much greater from country to country than from one policy field to anoth- er. Thirdly, although the IOs from all five countries in our study are more likely to employ media and publish- ing strategies (information politics) than to mobilise their members and supporters (protest politics), we can still observe national patterns in their selection of strat- egies and in the intensity of their instrumentalisation. Key words: EU, interest groups, interest organisations, lobbying, strategies 1 This paper was co-financed by the Slovenian Research Agency Research Programme Political Science Research and project N5-0014 INTEREURO as well as by the German Science Foundation (grant number EI 461/6-1) and the European Science Foundation project 10-ECRP-008: Comparative Research on Interest Group Politics in Europe (INTEREURO). * Mitja Hafner-Fink, PhD., Associate Professor, Faculty of Social Sciences; Meta Novak, PhD., Research Assistant, Faculty of Social Sciences; Danica Fink-Hafner, PhD., Professor, Faculty of Social Sciences, University of Ljubljana; Rainer Eising, PhD., Professor, Ruhr University Bochum; Daniel Rasch, PhD. Research Student, Ruhr University Bochum; Patrycja Rozbicka, PhD., Assistant Professor, Aston University, Birmingham. TEORIJA IN PRAKSA let. 53, 3/2016 606 M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA Introduction The representation of interests in the EU political system is highly com- plex – both in terms of the resources that interest organisations2 (IOs) can bring to bear on EU-level policy processes as well as in the interplay between the bottom-up, top-down and horizontal inputs that interest organisations effect in EU policymaking. Within this framework, analysts evaluating policy- making at the EU level have often observed that there are certain biases in the participation of IOs. We are particularly interested in biases that potentially arise from the increasing variety of resources available to national interest organisations as well as the national contexts from which these organisations operate in an enlarged European Union. This aspect of interest representa- tion in the EU remains largely unresearched (Beyers et al., 2008; Eising, 2008) despite the fact that the EU constantly faces a question of legitimacy (Eriksen, 2014), particularly with the public revelations that certain interests enjoy bet- ter access to policymakers3 and the scandal of closed-door deals between policymakers and certain interest organisations – namely, big corporations.4 Interest representativeness has not only gained additional importance as part of the EU legitimacy in the context of the international financial cri- sis (Kröger, 2014; Fossum, 2016), but has also become problematised due to the recent wave of enlargement. A number of empirical studies from Western EU member states have revealed that the decision of a national interest group to Europeanise its lobbying strategy will depend on both its resources and its domestic national institutional context (Klüver, 2010; Beyers and Kerremans, 2012). We might expect to see an extensive gap in interest representation at the EU level due to the fact that the representa- tion of interests from post-socialist EU member states tends to be weaker, since interest organisations (particularly NGOs) from these countries tend to have weaker resources (see the NGO Sustainability Index5). 2 For the argumentation in favour of using this term, see Beyers, Eising, Maloney (2008). 3 See information published by the Corporate Europe Observatory (http://corporateeurope.org/), Transparency International – EU integrity watch (http://www.integritywatch.eu/), Worst EU Lobby Awards (www.facebook.com/worstlobby/ and https://www.youtube.com/watch?v=k0Tq4bTK1CQ). 4 For example, an EU investment proposal will not prevent corporate challenges to health and envi- ronment (http://corporateeurope.org/international-trade/2016/04/eu-investment-proposal-won-t-prevent- corporate-attacks-health-and); Trade Secrets: opponents call on national governments to reject the text (at http://corporateeurope.org/pressreleases/2016/04/trade-secrets-opponents-call-national-governments- reject-text; corporateeurope.org/international-trade/2016/01/dangerous-regulatory-duet); Trade Secrets: who voted what (http://corporateeurope.org/power-lobbies/2016/04/trade-secrets-who-voted-what); a leak reveals the Commission gave inside information to the automobile lobby on new emissions tests (http:// corporateeurope.org/power-lobbies/2016/03/leak-shows-commission-giving-inside-information-car-lobby- new-emissions-tests). 5 The 2014 report is available at https://www.usaid.gov/sites/default/files/documents/1863/ EuropeEurasia_FY2014_CSOSI_Report.pdf. M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 607 The main contribution of our article is to shed light on lobbying by national interest organisations acting at the national and European levels in an enlarged European Union. We will also question how the broad varia- tions in an enlarged EU may contribute to biases in interest representation at the EU level. As a piece of original comparative research that examines the differences in both western and eastern EU member states as well as the differences in various enlargement waves, our study is first of all a descrip- tive comparative view of basic national patterns of interest representation within the EU’s multi-level political system. Our study focuses on the involvement of IOs in the national and Euro- pean level decision-making on EU directives. More precisely, these pro- cesses are investigated from the perspective of the bottom-up input of national interest organisations into EU-level policymaking. We pay par- ticular attention to two sets of questions on strategies adopted by interest organisations: (1) which strategies they adopt in terms of their selection of which level of government to lobby, and their choice of institution to lobby (i.e. whether the national executive, national parliament and/or the execu- tive EU institutions– the European Commission and European Council, as well as the European Parliament); and (2) which lobbying strategies they adopt (i.e. which methods and techniques) to achieve their objective. Based on our review of the literature and analysis of the data from the INTEREURO Multi-Level Governance Module (MLG) (www.intereuro.eu; Beyers et al., 2014), we tested the following four determinants which we expected to shape the strategies of interest organisations: (1) the type of interest organisation (whether economic or non-economic); (2) the policy field (the empirical data on the 20 most salient legislative proposals from the period 2008 – 2010 covers three policy fields – (i) environment and energy, (ii) human rights and (iii) finances); (3) the country variations in interest organisations’ input into EU policymaking (data was gathered for five coun- tries: Germany, the Netherlands, Slovenia, Sweden and the UK); and (4) the EU funding of interest organisations and their membership of EU umbrella organisations. In the following sections we will present a brief overview of the litera- ture in the field before turning to our detailed research questions, the data we used, our methods of analysing it, and our main empirical findings. The multilevel lobbying strategies of national interest organisations With the intensive development of the EU political system as a regional, post-national political system, a new structure of multi-level governance (MLG) has also emerged (Marks, 1996). Interest organisations have gained TEORIJA IN PRAKSA let. 53, 3/2016 608 M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA a more complex institutional opportunity structure (Princen and Keremans, 2008), where they are faced with a variety of venues for promoting their interests. Indeed, some interest organisations which can afford to be active at the various levels of the EU political system are able to choose the best venue to influence EU policymaking (Mazey and Richardson, 1993; Grant, 2000; Beyers et al., 2008; Eising, 2008). EU lobbying may not only involve lobbying EU-level institutions directly, but also via European umbrella organisations (Beyers, 2002; Greenwood, 2003). Until now, Western scholars in particular have been interested in the fac- tors influencing interest organisations’ activities at the various institutional levels and their possible combinations of activities at the various levels (i.e. lobbying both at the national level as well as at the EU level). The search for such factors has identified the following at the macro and mezzo levels: insti- tutional factors (Falkner, 2000; Eising, 2006; Beyers and Kerremans, 2012); the distribution of powers in terms of horizontal and vertical fragmentation (Klüver, 2010); the embeddedness of interest organisations in their domestic (national) milieu (Beyers, 2002; Klüver, 2010; Beyers and Kerremans, 2012); policy issues and/or policy sectors (Beyers, 2002; Wallace, 2005; Beyers and Kerremans, 2012); and the different modes of interest representation (Schmidt, 1999; Beyers, 2002; Klüver, 2010). At the micro level, the follow- ing variations in IO characteristics are taken into account: an IO’s resources (including EU and national funding, and personnel); and the type of IO (whether an economic or non-economic IO) (Greenwood, 2003; Beyers and Kerremans, 2007; Eising, 2007; Klüver, 2010). In spite of the prolifera- tion of research into the factors influencing interest organisations’ activities, the various findings have remained fragmented and inconclusive. There remains a lack of insight into the adaptation of national IOs from post-socialist EU member states to the multi-level political system. Never- theless, some contributions so far have highlighted the EU’s top-down impacts on interest group formation as well as the EU’s top-down impacts on national interest group activity within the framework of the national political system (e.g. Cellarius and Staddon, 2002; Hicks, 2004; Forest, 2006; Gąsior-Niemiec, 2010; Kutter and Trappmann, 2010; Stanojević, 2012; Fink- Hafner et al., 2015). Other research has emphasised the ineffectiveness of taking the European route (i.e. lobbying the EU level directly) even in the case of economic interest organisations such as trade unions (Kusznir and Pleines, 2008; Landgraff and Pleines, 2015). Such research tends to focus on the weak capacity of interest groups from Central and East European EU member states to participate in EU governance (Obradovic and Pleines, 2007). Some authors also highlight other factors, such as Central and East European EU member states’ satisfaction with EU policies and a subsequent lack of perceived need to act at the EU level (Pleines, 2010). Nevertheless, M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 609 the research highlights the crucial fact that limited resources (such as organi- sational resources) and a lack of negotiating capacity hinder the ability of national actors from both the older and newer EU member states to influ- ence policy at the EU level (see Perez-Solorzano Borragan, 2003; Eising, 2004: 217). Although the primarily focus of our research is not national IOs from post-socialist EU member states, our analysis nevertheless provides some new insights into this subject due to our comparison of IOs from one par- ticular post-socialist country (Slovenia) with IOs from countries which have not experienced socialism. Furthermore, our research is unique in that it simultaneously includes national interest groups from both western and eastern EU member states as well as national interest groups from both old and new EU member states and small and large EU member states which enables us to test the variations in national interest organisations’ engage- ment at both the national and EU level. Consequently, our study contributes to the project of identifying the specific conditions under which the activi- ties of national interest groups at the national level in various member states spill over to the EU level. Research design Our comparative research is unique in that it studies both the national and EU-level activities of national interest organisations from five countries (three older and two newer member states, which also simultaneously rep- resent one eastern and four western member states) within the framework of the EU directive-making. In order to identify the differences among the national interest organisations in their multilevel endeavours, we opted for an exploratory orientation. We searched for answers to the following three research questions: Q1: How do interest organisations choose their strategies in terms of the input venues? Here, we differentiate between the national venue, where IOs lobby national institutions to convey their interests up to the EU level, and the European venue, where IOs lobby EU institutions directly. Q2: How do interest organisations choose their strategies in terms of lob- bying methods and techniques? Q3: Which factors influence interest organisations’ choices of strategies? We answer this question from the following hypothetical determinants that indicate interest representation biases in EU policymaking – based on the literature and as permitted by the available data: – the type of interest organisation (whether economic or non-economic); – the policy field (the empirical data covers three policy fields: environ- ment and energy, human rights, and finances and economy); M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 610 – the country variations (data on interest organisations’ input into EU pol- icymaking was gathered for five countries: Germany, the Netherlands, Slovenia, Sweden and the UK); – the EU funding of interest organisations; – and their membership of EU umbrella organisations. In order to answer these research questions we analysed the data col- lected within the framework of the INTEREURO Multi-Level Government Module and we focused our analysis on the involvement of national-level actors in EU policymaking processes. This includes interviews conducted with national officials and non-state stakeholders in Germany, the Nether- lands, Sweden, the United Kingdom and Slovenia based on the 20 most sali- ent legislative proposals during the period 2008 to 2010 (for more infor- mation on the data collection and sampling see Beyers et al., 2014). The current analysis, however, is based on the data gathered from interviews with national interest organisations. The interest organisations interviewed are not representative of the full population of interest organisations in the countries selected; rather, they represent those interest organisations actually involved in EU policymaking on selected EU-level policy proposals. In other words, all of these IOs were involved in forming national positions on the selected EU-level policy pro- posals. While some of them have pursued an exclusively national lobbying strategy or a combined European and national lobbying strategy, none of them have pursued a strategy of lobbying European political institutions to the exclusion of national institutions. The research design is composed of the following stages: 1. In the first stage, we construct six composite indexes to measure lobby- ing intensity at various venues and two indexes to measure the intensity with which two types of lobbying strategies were used. 2. In the second stage, we use linear regression analysis to reveal the fac- tors determining lobbying intensity at the various venues (i.e. the levels of government). 3. In the third stage, we use bivariate analyses (comparison of means) to demonstrate variations between countries and policy fields regarding the ‘venue strategies’ and the types of lobbying strategies. Our units of analysis are the interest organisations interviewed6. Using 6 We conducted 101 interviews with those interest organisations that were the most active on selected policy proposals in their particular country. In Germany 19 interviews were conducted, in Netherlands 14, in Sweden 11, in United Kingdom 17 and in Slovenia 40 (since 5 state institutions were excluded, 35 Slovenian IOs were ultimately included in our analysis). We acknowledge that we cannot use our research data from these interviews to draw generalisations about the whole population of interest organisations. M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 611 this data, we developed two types of indexes to measure the intensity and the extent of lobbying strategies: a. We formed six ‘venue strategy’ indexes for the following venues: the nati- onal executive, the national parliament, the European Parliament, the European Commission, the European Council and permanent represen- tations, and civil society. The indexes are based on data that records how frequently and to whom interest organisations provide information for the purpose of influencing the legislative outcome. To obtain this data, we collected answers to the following question(s): ‘The following que- stions concern the different political institutions to which your organi- sation may have provided information with the purpose of influencing the legislative outcome of the directive proposal. I will show you a list of different political institutions and I would like to know how regularly you have provided these institutions with information. Was this never, rarely, sometimes, frequently or very frequently?’ Answers were recorded on a scale of 1 to 5, however, for the purposes of our analysis the scale was recoded to values from 0 (never) to 4 (very frequently). The index value for each venue was calculated as an average value of all institutions (tar- gets of lobbying) included in the index. b. Based on the results of the principal component analysis and the hie- rarchical cluster analysis, we also formed two indexes for ‘methods and techniques strategy’ for the following two outsider lobbying strategies: (i) media and publishing – ‘information politics’ (Beyers, 2004); and (ii) the mobilisation of members and supporters – ‘protest politics’ (Beyers, 2004). Here we did not focus on insider lobbying strategies (access to policymakers) since this is covered in our first type of indexes (previous paragraph). (For a more detailed explanation, see the further presenta- tion of indexes below). To explore the differences between interest organisations regarding the use of these strategies and methods we applied comparison of means and a multivariate regression analysis with the following possible factors: the type of interest organisation; the EU funding of the interest organisation; their membership of an EU umbrella organisation; the policy field in which they operate; and their country of origin (see research question Q3). M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 612 The key findings The Factors Determining the Intensity of Lobbying at the Various Venues Based on the regression analysis7 (Table 1), the key determinant of multi- level lobbying is the interest organisation’s country of origin, i.e. whether an older or newer EU member state. Indeed, interest organisations from older EU member states (Germany, the Netherlands and the UK) are systemati- cally more active in all venues we assessed than IOs from newer member states. The other factors we might have expected to have been systemati- cally relevant, such as the finances which IOs receive from EU programmes and their membership of European umbrella organisations, proved not to be so. Rather, they were relevant only for certain particular venues. For instance, the characteristics of an IO (whether economic or non-economic IO) appear to be relevant only in relation to the IO’s interaction with civil society and with the European Parliament (see Table 1). Based on this analysis the following factors are important: – The factor of old/new membership is the single factor that is significant for IO activity at all levels. – The factor of old/new membership is the sole significant factor of activ- ity at the EU level. – The policy field is only significant when IO activity relates to national par- liaments: for matters concerning rights, IOs’ targeting of national parlia- ments is higher when comparing to those active within other two policy fields (finance and economy, environment and energy); while IO within the financial and economic policy field target national parliaments less than those within the other two fields (rights; environment and energy). – Membership of an umbrella organisation is significant in the case of providing information to civil society (significant at p < 0.05) and to the European Commission (significant at p < 0.10). More precisely, in the case of civil society, members provide more information than non-mem- bers; in the case of the Commission, members provide less information than non-members. – The type of interest organisation (namely, whether economic or non- economic) is significant when providing information to civil society (significant at p < 0.05) and to the European Parliament (significant at p < 0.10). More precisely, in both cases (civil society and the European Parliament) economic interest organisations provide less information than non-economic interest organizations. 7 We used linear regression analysis. M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 613 By contrast, the financing of IOs from EU programmes appears to have no significant effect on activity at any level. Table 1: THE RESULTS OF REGRESSION ANALYSIS – THE FACTORS OF LOBBYING INTENSITY AT THE VARIOUS VENUES (STANDARDISED REGRESSION COEFFICIENTS) Lobbying indexes: Provision of information to the… Factors: … n at io n al ex ec u ti ve … n at io n al p ar lia m en t … E u ro p ea n P ar lia m en t … E u ro p ea n C o m m is si o n … E U C o u n ci l an d /o r p er m . re p re se n ta ti o n s … ci vi l s o ci et y old EU member ***0.374 ***0.482 ***0.498 ***0.493 ***0.511 ***0.438 Policy field – rights 0.056 **0.220 0.070 -0.118 0.018 -0.045 – finance, economy -0.168 **-0.198 0.002 -0.093 0.015 -0.123 – environment, energy ref ref ref ref ref ref financing from EU programmes 0.172 -0.089 -0.141 -0.023 0.043 0.081 member of an umbrella organisation 0.065 -0.054 0.013 *-0.188 0.041 **0.220 organisation type: economic 0.009 -0.039 *-0.193 0.071 -0.012 **-0.214 N = 93 93 93 93 93 93 Adjusted R2 = 0.115 0.322 0.275 0.193 0.210 0.244 F = 3.001 8.280 6.805 4.660 5.084 5.960 *** p < 0.01 ** p < 0.05 * p < 0.10 Note: Shaded regression coefficients are statistically significant at the level p < 0.05 Source: own analysis based on data from INTEREURO Multi-Level Governance Module. Venue Strategies and the Differences between Countries From the national point of view, only two main interest organisation strategies could be identified: a national strategy and a combined national and European strategy. The majority of interest organisations reported hav- ing contact with national institutions. Meanwhile a sizable portion of inter- est organisations also report having contact with EU institutions and civil society. However, we can observe a significant difference between the IOs from older and newer EU member states regarding the proportion of IOs M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 614 which lobby at both levels. All interest organisations from the United King- dom reported lobbying at both levels, as well as 92.9 % from the Nether- lands and 89.5 % from Germany. In stark contrast, only 45.5 % of IOs from Sweden and 31.4 % of IOs from Slovenia reported the same. In general, the European Parliament appears to be the most lobbied institution at the EU level (Table 2). This comes as some surprise given that the literature suggests that it is the European Commission which dominates EU policymaking (Greenwood, 2003). On the other hand, the European Par- liament is a more approachable institution than the Commission due to the fact that it has national contact persons – MEPs (Kohler Koch and Quittkat, 2016). Here we should note that the policy proposals included in our analy- sis were those which were adopted with the participation of the European Parliament. Interest organisations differ less when deciding whether to adopt a national or European lobbying strategy than they do when selecting which institutions to target. In particular, they differ in deciding whether to give information to (1) the national executive and/or the national parlia- ment and/or to (2) the European Parliament, the European Commission, the Council and national representations in Brussels. Table 2: ‘VENUE STRATEGY’ INDExES – AVERAGE VALUES FOR COUNTRIES (AVERAGE VALUES ON A SCALE 0–4) Provision of informa- tion to… National executive institu- tions National parlia- ment & parties European Parlia- ment and EU par- ties European Commis- sion EU Council and/or perm. repr. Civil society United Kingdom 1.167 0.902 1.136 0.673 1.368 1.779 Germany 0.912 1.211 1.413 0.322 0.632 0.500 The Nether- lands 1.333 0.786 0.654 0.111 0.589 1.643 Sweden 0.546 0.470 0.343 0.061 0.409 0.591 Slovenia 0.767 0.143 0.046 0.038 0.136 0.579 Total 0.924 0.620 0.632 0.220 0.550 0.932 F 3.634 8.859 11.355 12.260 15.618 15.128 p 0.009 0.000 0.000 0.000 0.000 0.000 Eta2 0.138 0.280 0.333 0.350 0.407 0.399 Source: own analysis based on data from INTEREURO Multi-Level Governance Module. As shown in Table 2 and Figure 1, IOs from the UK stand out with an overall outstanding provision of information to all national and EU-level institutions as well as to civil society. IOs from Germany are notable for their above-average lobbying of their national parliament and the European M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 615 Parliament. Meanwhile, IOs from the Netherlands are above average in their lobbying of civil society and in provision of information to the national executive. Slovenian IOs mostly tend to provide information to their national government and to civil society, but even these two strategies are underdeveloped compared to the IOs from the other countries studied. All in all, the differences in activity of IOs from the different countries can be characterised as giants compared to dwarfs. Figure 1: ‘ VENUE STRATEGY’ INDExES – AVERAGE VALUES FOR COUNTRIES (AVERAGE VALUES ON A SCALE 0–4) Source: own analysis based on data from INTEREURO Multi-Level Governance Module. Venue Strategies and the Differences between Policy Fields As we have already seen, our regression analysis suggests that the differ- ences between countries are greater than the differences between policy fields (Table 1). The results of our bivariate analysis (a comparison of means and one-way analysis of variance) for the differences between countries (Table 2) and the differences between policy fields (Table 3) further sup- ports the conclusion of our regression analysis. This means that the variabil- ity among interest organisations within a country is lower than the variability within a policy field. Or, to put it another way, the variability among interest organisations from different countries active within the same policy field is M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 616 greater than the variability among interest organisation from the same coun- try which are active in different policy fields. Indeed, interest organisations are more homogeneous within a country than within the same policy field. So, an interest organisation’s country of origin – rather than its policy field – is a more significant determinant of its venue strategy. This is particularly true for lobbying at the EU level (compare Table 2 and Table 3). Table 3: VENUE STRATEGY INDExES – AVERAGE VALUES FOR POLICY FIELDS (AVERAGE VALUES ON A SCALE 0–4) Provision of information to… National executive institu- tions National parlia- ment & parties Europe- an Parlia- ment & EU par- ties Europe- an Com- mission EU Coun- cil and/ or perm. repr. Civil society rights 1.086 1.005 0.886 0.189 0.644 1.114 environment, energy 0.887 0.457 0.435 0.222 0.442 0.888 finance, economy 0.698 0.302 0.688 0.278 0.672 0.688 Total 0.924 0.620 0.632 0.220 0.550 0.932 F 2.017 7.013 2.246 0.267 1.184 1.435 p 0.139 0.001 0.111 0.766 0.311 0.243 Eta2 0.042 0.131 0.046 0.006 0.025 0.030 Source: own analysis based on data from INTEREURO Multi-Level Governance Module. As Figure 2 demonstrates, even if the country of origin is the more sig- nificant factor, an IO’s venue strategy nevertheless also depends on its pol- icy field. This is not surprising when we consider that policy fields do dif- fer with regard to their institutional anatomy and policy modes (Wallace, 2005). IOs active in the field of rights will typically orient their lobbying towards the national executive and civil society but also towards the national par- liament and parties. IOs in the field of finance and the economy typically pay attention to the European Council and permanent representations, the national executive, civil society and the European Parliament and EU parties. However, compared to IOs in other fields they tend to be the least focused on national parliaments. IOs in the fields of environment and energy spread a comparable amount of information equally to the national executive and civil society.8 They seem to pay less attention to the European Council and 8 The interest organisations in our sample were specifically identified based on their engagement in EU policymaking and– according to the literature (Beyers and Kerremans, 2007; Eising, 2008) – are there- fore more likely to lobby at the EU venue. M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 617 permanent representations, the European Commission and the European Parliament/EU parties. Figure 2: ‘VENUE STRATEGY’ INDExES –AVERAGE VALUES FOR POLICY FIELDS (AVERAGE VALUES ON SCALE 0–4) Source: own analysis based on data from INTEREURO Multi-Level Governance Module. Overall, the national executives and civil society appear to be the targets of the most lobbying in terms of the intensity of information provision. The European Parliament and EU parties are the most lobbied actor at the EU level. Lobbying Methods and Techniques and the Differences between Countries We measured the intensity and the scope of the lobbying methods and techniques employed across ten activities.9 Based on analysis, we can speak of there being two dimensions10: 9 Answers to the following question were used: ‘How frequently did [ORGANISATION NAME] under- take the following activities to try to affect or influence legislative outcomes? (never, rarely, sometimes, frequently or very frequently)’ The available answers were arranged on a scale of 1 to 5, however, for the purposes of our analysis the scale was recoded to values from 0 (never) to 4 (very frequently). 10 We used a principal component analysis and a hierarchical clustering analysis. Based on low com- munalities in the principal component analysis, two variables were excluded: Hiring commercial consult- ants and Publishing research reports and brochures. M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 618 1. Media appearances and contacts, publishing – Organising press conferences or distributing press releases – Participating in debates in the media (e.g. giving interviews, editorials, opinion letters, open forums) – Contacting reporters and journalists to increase media exposure – Publishing statements and position papers (on your own website) 2. The mobilisation of members and supporters – Encouraging members and supporters to lobby policymakers, to initiate letter-writing campaigns or petitions directed at public officials – Using social media to publicise your position – Placing advertisements in newspapers and magazines – Staging protest activities that involve members and supporters (strikes, consumer boycotts, public demonstrations) For further analysis, we developed an index for each of the two dimen- sions.11 Based on our analysis, interest organisations do not appear to adapt their lobbying strategies to the idiosyncrasies of particular institutions (the executive or representatives at the national and EU-level). IOs which are more active at a given level at any institution tend to use more varied meth- ods and techniques. There are no systematic and relevant differences in terms of their targeting national executives, parliaments or European insti- tutions. In this respect, we see the same pattern we previously observed for lobbying strategies, namely: IOs from older EU members more frequently employ various methods and techniques than IOs from newer EU members (see Figure 3 and Table 4). In general, IOs – regardless of their country of origin – employ the media strategy considerably more frequently than the mobilisation strategy. There are, however, certain differences in national patterns resulting from the overall level of lobbying methods and the strategies employed. For instance, although IOs from the Netherlands also employ the media strategy more frequently than the mobilisation strategy, the frequency ratio is notably smaller compared to IOs from the other countries studied. 11 The average value on a scale of 0 to 4 was calculated for all methods included in the index. M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 619 Figure 3: THE USE OF LOBBYING METHODS AND TECHNIQUES – TWO LOBBYING STRATEGY INDExES (MEDIA AND PUBLISHING; THE MOBILISATION OF MEMBERS AND SUPPORTERS) (AVERAGE VALUES ON A SCALE 0–4) Source: own analysis based on data from INTEREURO Multi-Level Governance Module. Table 4: ‘LOBBYING METHOD STRATEGY’ INDExES Media and publishing Mobilisation of members, supporters… United Kingdom 2.044 0.868 Germany 1.658 0.355 The Netherlands 1.500 0.696 Sweden 1.296 0.114 Slovenia 1.200 0.293 Total 1.495 0.445 F 2.297 5.371 p 0.065 0.001 Eta2 0.092 0.191 Source: own analysis based on data from INTEREURO Multi-Level Governance Module. The UK stands out when it comes to mobilisation campaigns combined with an above-average use of media, applying expertise and strategies that use position papers. Germany is notable for its above-average media orien- tation combined with an above-average use of press conferences, the use of M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 620 expertise, contacting journalists and the use of position papers. The Neth- erlands is notable for its above-average social media orientation. Sweden is notable for its above-average use of position papers. Slovenia is idiosyn- cratic in that the activity levels of its interest organisations are below average in all respects. Conclusions Our study addressed the biases of interest representation in the EU polit- ical system from a bottom-up perspective. Our main focus was to identify the differences between certain EU member states in terms of the spill over of national IO activities to the EU level. We can summarise our main find- ings as follows: Firstly, interest representation is not transmitted equally from the national to the EU level. The main difference appears to be between the IOs from the older member states (Germany, the UK and the Netherlands) and IOs from the newer EU member states (Sweden and Slovenia). IOs from the older EU member states are more active at the EU level than IOs from newer EU member states. These findings are in line with previous country case studies (e.g. Lundberg and Sedelius, 2014). Secondly, we can observe national patterns in interest organisation strat- egies in the scope of their targeting the national executive and the national parliament. The Netherlands and the UK are noteworthy for the quantity of information their IOs provide to their national governments; Germany and the UK are noteworthy for the quantity of information their IOs provide to their national parliaments. We can also observe national patterns in interest organisation strategies in terms of the scope of their targeting EU-level insti- tutions for the provision of information: Germany and the UK are notable for the extent to which their IOs target the European Parliament and Euro- pean parties, while the UK is specifically notable for the extent to which its IOs target the European Commission, the European Council and member state representations. Overall, the greatest differences evident are between the IOs from the UK (giants) and those from Slovenia (dwarfs). Thirdly, of the political venues studied, civil society and national execu- tive are the most lobbied, while at the EU level the most lobbied institution is the European Parliament and its political parties (except by Slovenian IOs). This finding is in line with research showing the European Parliament to be a more accessible institution for national IOs to lobby than the Com- mission (Kohler Koch and Quittkat, 2016). The extreme orientation of Slo- venian IOs toward the national executive suggests that further investigation is required. Possible explanatory factors include: the impact of resources on IOs; the lack of socialisation within the EU political system; and the possible M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 621 political and cultural residue of the socialist mindset – that it is the state that takes care of representing all interests. Fourthly, we found that the differences between countries are greater than the differences between policy fields. Interest organisations are more homo- geneous within a country than within the same policy field. The other factors proved to be significant only for certain venues. For instance, membership of a European umbrella organisations is a relevant factor for the provision of information to civil society and the European Commission; and the type of IO (economic or non-economic) is a significant factor in the provision of infor- mation to civil society and the European Parliament. This supports Wallace’s (2005) thesis, that the institutional anatomy of a particular policy and the mode of an IO’s lobbying behaviour are important. We also noted that access to EU programme money has no effect on an IO’s selection of lobbying venue. Fifthly, we observed that IOs utilised two lobbying strategies and tech- niques, namely: (1) media and publishing; and (2) the mobilisation of mem- bers and supporters. Although IOs tend to employ media and publishing strategies more often than they mobilise their members, national idiosyn- crasies do appear to be significant. As already noted – IOs from older EU members more frequently employ various methods and techniques than IOs from newer EU members. While IOs from the Netherlands are char- acterised by a more equal instrumentalisation of both strategies (they are also idiosyncratic in their use of social media), IOs from the UK are notable for the sheer extent to which they instrumentalise both strategies. IOs from Sweden use position papers more often than IOs from other countries, while IOs from Slovenia are below average in their usage of all methods and techniques. The fact that EU-level policy actors receive considerably more information from some member states than from others clearly illustrates a bias in the inputs from national IOs in EU-level policymaking. Our findings are noteworthy for several reasons. They reveal major dif- ferences in the national activity patterns of interest organisations in the pro- cesses of European policymaking. This, combined with the insights into the overall power and influence of certain EU member states at the EU level, translates into major differences in the transmission and representation of national interests in EU policymaking and may lead to considerable biases if these interests are taken into account in the EU policy outputs (such as directives). Identifying these differences in the representation of interests enables us to shed light on the current questions of the EU’s democratic deficit. It also enables us to better understand the difficulties in transposing directives into national laws in member states (national interest organisa- tions blocking the transposition of EU law). This in turn feeds back into the question of how legitimate and efficient EU policies are perceived to be across the entire European Union. M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 622 BIBLIOGRAPHY Beyers, Jan (2002): Gaining and Seeking Access: The European Adaptation of Domestic Interest Associations. European Journal of Political Research 41 (5): 585–612. Beyers, Jan (2004): Voice and Access: Political Practices of European Interest Asso- ciations. European Union Politics 5 (2): 211–240. Beyers, Jan and Bart Kerremans (2007): Critical Resource Dependencies and the Europeanization of Domestic Interest Groups. Journal of European Public Pol- icy 14 (3): 460–481. Beyers, Jan and Bart Kerremans (2012): Domestic Embeddedness and the Dynam- ics of Multilevel Venue Shopping in Four EU Member States. Governance: An International Journal of Policy, Administration, and Institutions 25 (2): 263–290. Beyers, Jan, Laura Chaqués Bonafont, Andreas Dür, Rainer Eising, Danica Fink Hafner, David Lowery, Christine Mahoney, William Maloney and Daniel Naurin (2014): The INTEREURO Project: Logic and Structure. Interest Groups & Advo- cacy 3 (2): 126–140. Beyers, Jan, Rainer Eising and William Maloney (2008): Researching Interest Group Politics in Europe and Elsewhere: Much We Study, Little We Know? West Euro- pean Politics 31 (6): 1103–1128. Cellarius, Barbara A. and Caedmon Staddon (2002): Environmental Nongovern- mental Organizations, Civil Society and Democratization in Bulgaria. East Euro- pean Politics and Societies 16 (1): 182–222. Eising, Rainer (2004): Multilevel Governance and Business Interests in the Euro- pean Union. Governance. An International Journal of Policy, Administration and Institutions 17 (2): 211–245. Eising, Rainer (2006): Interest Groups and Social Movements. In Maarten Vink and Paolo Graziano (eds.), Europeanization: New Research Agendas, 167–181. Bas- ingstoke: Palgrave Macmillan. Eising, Rainer (2007): Institutional Context, Organizational Resources and Strategic Choices. Explaining Interest Group Access in the European Union. European Union Politics 8 (3): 329–362. Eising, Rainer (2008): Interest Groups in EU Policy-making. Living Reviews in European Governance 3 (4). Accessible at http://europeangovernance-livingre- views.org/Articles/lreg-2008-4 (8. 4. 2016). Lundberg, Erik and Thomas Sedelius (2014): National Linkages and Ambiguous EU Approaches among European Civil Society Organisations. Journal of Contem- porary European Research 10 (3): 322–336. Eriksen, Erik O. (2014): The Normativity of the European Union. Basingstoke: Pal- grave Macmillan. Falkner, Gerda (2000): Policy Networks in a Multi-Level System: Convergence towards Diversity. West European Politics 23 (4): 94–120. Fink-Hafner, Danica, Mitja Hafner-Fink, Meta Novak (2015): Europeanisation as a Factor of National Interest Group Political-Cultural Change: The Case of Inter- est Groups in Slovenia. East European Politics & Societies 29 (1): 287–305. Forest, Maxime (2006): Emerging Gender Interest Groups in the New Member M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 623 States: The Case of the Czech Republic. Perspectives on European Politics and Society 7 (2): 170–184. Fossum, John Erik (2016): Democracy and Legitimacy in the EU: Challenges and Options. IAI Working Papers 16/01. Accessible at http://www.iai.it/en/pubbli- cazioni/democracy-and-legitimacy-eu (21. 4. 2016). Gąsior-Niemiec, Anna (2010): Lost in the System? Civil Society and Regional Devel- opment Policy in Poland. Acta Politica 45 (1–2): 90–111. Grant, Wyn (2000): Pressure Groups and British Politics. New York: St. Martin’s Press. Greenwood, Justin (2003): Interest Representation in the European Union. Basing- stoke: Palgrave Macmillan. Hicks, Barbara (2004): Setting Agendas and Shaping Activism: EU Influence on Central and Eastern European Environmental Movements. Environmental Poli- tics 13 (1): 216–233. Klüver, Heike (2010): Europeanization of Lobbying Activities: When National Interest Groups Spill Over to the European Level. Journal of European Integra- tion 32 (2): 175–91. Kohler Koch, Beate and Christine Quittkat (2016): The National Dimension in Multi-level Interest Representation. Presented at WEP special issue Conference in Bochum, Germany, 28–29 April 2016. Kröger, Sandra (ed.) (2014): Political Representation in the European Union: Still Democratic in Times of Crisis? Abingdon: Routledge. Accessible at https:// books.google.si/books?id=hzFEAwAAQBAJ&printsec=frontcover&hl=sl#v=on epage&q&f=false (21. 4. 2016). Kusznir, Julia and Heiko Pleines (eds.) (2008): Trade Unions from Post-Socialist Member States in EU Governance. Stuttgart: Ibidem-Verlag. Kutter, Amelie and Vera Trappmann (2010): Civil Society in Central and Eastern Europe: The Ambivalent Legacy of Accession. Acta Politica 45 (1): 41–69. Landgraff, Christin and Heiko Pleines (2015): Research Questions and Research Context. In Landgraff, Christin and Heiko Pleines (eds.), Interest Representa- tion and Europeanization of Trade Unions from EU Member States of the East- ern Enlargement, 29–44. Stuttgart: Ibidem-Verlag. Marks, Gary (1996): An Actor-centred Approach to Multi-level Governance. Journal of Regional and Federal Studies 6 (2): 20–38. Mazey, Sonia and Jeremy Richardson (1993): Lobbying in the European Commu- nity. Oxford: Oxford University Press. Obradovic, Daniela and Heiko Pleines (eds.) (2007): The Capacity of Central and East European Interest Groups to Participate in EU Governance. Stuttgart: Ibi- dem Publishers. Perez-Solorzano Borragan, Nieves (2003): Coming to terms with European Union Lobbying: The Central and East European Experiences. In Jenny Fairbrass, Alex Warleigh (eds.), Influence and Interests in the European Union: The New Poli- tics of Persuasion and Advocacy, 160–183. London: Routledge. Pleines, Heiko (2010): Is this the Way to Brussels? CEE Civil Society Involvement in EU Governance. Acta Politica 45 (1–2): 229–246. M. HAFNER-FINK, M. NOVAK, D. FINK-HAFNER, R. EISING, D. RASCH, P. ROZBICKA TEORIJA IN PRAKSA let. 53, 3/2016 624 Princen, Sebastiaan and Bart Keremans (2008): Opportunity Structures in the EU Multi-Level System. West European Politics 31 (6): 1129–1146. Schmidt, Vivien A. (1999): National Patterns of Governance under Siege: The Impact of European Integration. In Kohler-Koch, Beate and Eising, Rainer (eds.), The Transformation of Governance in the European Union, 155–172. London: Routledge. Stanojević, Miroslav (2012): The Rise and Decline of Slovenian Corporatism: Local and European Factors. Europe-Asia Studies 64 (5): 857–877. Wallace, Helen (2005): An Institutional Anatomy and Five Policy Modes. In Helen Wallace, William Wallace and Mark A. Pollack (eds.), Policy-Making in the Euro- pean Union, 5th edition, 49–89. Oxford: Oxford University Press.