Politics Beyond the State : Actors and Policies in Complex Institutional Settings [1 ed.] 9789054874362

Citation preview

Politics Beyond the State.book Page 1 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State Actors and Policies in Complex Institutional Settings

Politics Beyond the State.book Page 2 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State.book Page 3 Wednesday, December 5, 2007 8:41 AM

Kris Deschouwer and M. Theo Jans (eds.)

Politics Beyond the State Actors and Policies in Complex Institutional Settings

Brussels University Press

Politics Beyond the State.book Page 4 Wednesday, December 5, 2007 8:41 AM

The Institute for European Studies is a Jean Monnet Pole of Excellence. It promulgates European Studies in general, and studies of Globalisation, European and Comparative Law, Environment and Regional (European) Integration specifically. The IES is an education and research centre, carrying out research on various European issues, and responsible for the Programme on International Legal Cooperation (PILC), an advanced Masters programme leading to an internationally renowned LL.M. Institute for European Studies (IES) Vrije Universiteit Brussel Pleinlaan 2 B-1050 Brussels [email protected] http://www.ies.be

Cover design: Koloriet, Sterrebeek Book design: Style, Hulshout Print: Grafikon, Oostkamp © 2007 VUBPRESS Brussels University Press VUBPRESS is an imprint of ASP nv (Academic and Scientific Publishers nv) Ravensteingalerij 28 B-1000 Brussels Tel. ++ 32 2 289 26 50 Fax ++ 32 2 289 26 59 E-mail [email protected] www.vubpress.be ISBN 978 90 5487 436 2 NUR 754 Legal Deposit D/2007/11.161/026 All rights reserved. No parts of this book may be reproduced or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the publisher.

Politics Beyond the State.book Page 5 Wednesday, December 5, 2007 8:41 AM

Contents 1.

Politics Beyond the State Maarten Theo Jans

7

2.

Electoral Politics in Multi-Level Settings Kris Deschouwer

3.

Governing Coalitions in Multi-level Settings: State-wide Parties and the Sub-national Arena in Spain 45 Irina Stefuriuc

4.

Torn Between Two Levels. Political Parties and Incongruent Coalitions in Belgium 71 Jo Buelens & Kris Deschouwer

5.

The Weakest Link? Interest Representation by Eurogroups in a Multi-level Governance System 91 Joke Wiercx

6.

The Role of Interest Groups in Fostering Citizen Engagement: The Determinants of Outside Lobbying 109 Christine Mahoney

7.

The Third Level in Brussels? Regional Information Offices and the European Policy Community 139 Michel Huysseune & Theo Jans

8.

Sinking the Port Directive. Belgian Trade Unions’ Mobilization against EU Policy 157 Martine Van Assche & Kris Deschouwer

9.

Deliberation Beyond the State. Towards a Deliberative Interpretation of EU Consultation Mechanisms 173 Irina Tanasescu

10. Delegation Beyond the State? Isabelle Bedoyan

25

193

5

Politics Beyond the State.book Page 6 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

11. Involving Private Actors in Policy-making: Reflections on Gender Mainstreaming and Governance 217 Petra Meier 12. EU Policies on Violence against Women. Contested Policy Areas and the Potential of a Governance Approach 235 Marjolein Paantjens 13. Multi-level Governance at Work: The Case of the European Integrated Product Policy (IPP) 251 Isabelle Bédoyan, Theo Jans, Irina Tanasescu Bibliography

271

List of Contributors

6

295

Politics Beyond the State.book Page 7 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State* Maarten Theo Jans

States, national institutions and policy processes guide our understanding of politics. Individuals are citizens of states and develop their private, professional and public lives in a national context(s). Political science concepts relate to national phenomena or are directly contrasted to the national state-based context. Processes of globalization and regional integration increasingly push politics beyond the state. Global economic operators and supranational institutions give rise to significant volumes of collective decision making occurring outside or only vaguely related to the national context and its state-based institutions. The centrifugal migration of politics away from state-based institutions does not only occur in an upward fashion towards the international level but also endows decentralized and private actors with policy making powers. The resulting picture is one in which state institutions and policy processes seem to be a less appropriate unit to analyze politics. Both in- and outside the state public choices are made that affect and influence state politics. State-based political analysis appears illequipped to come to grips with political decision-making originating beyond or outside the realm of the state. Adapting the political science toolkit to the challenges of non-state based politics is central to furthering of our understanding of politics. Capturing the changing nature of collective decision-making both its relatively novel supranational form as well as within the state, is very much on the agenda of many research domains within political science. The umbrella concept that seeks to capture the changing texture of politics is “governance”. The scope of the governance concept is very broad and seems to cover many apparently unrelated issues. Governance is used in the context of transnational decision-making or regional integration efforts. Governance is also equated with public-private partnerships in the provision of public goods. Another use of the concept refers to the use of ‘soft law’ and/or voluntary instruments. The concept may also refer to the

*

This book is part of a Concerted Research Action funded by the Research Council of the Vrije Universiteit Brussel.

7

Politics Beyond the State.book Page 8 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

concerted action of different policy levels on a given policy issue (e.g. multi-level governance). Some authors present governance as the emergence of a networked polity, in which flexible and fluid networks of public and private stakeholders blur formalized institutional linkages and provide new patterns of collective decision making. The appeal of the concept and its broad use and coverage may also be its main weakness, in that the concept is stretched to point that it covers any phenomenon that falls outside our traditional understanding of state-based politics. As Pierre and Peters point out, the term governance is not new and its appearance can be dated back to the fourteenth century French term ‘gouvernance’ (Pierre and Peters, 2000). The term is clearly dated, the same may apply to the content and the assumptions linked to the term. The novelty of politics beyond the confines of the state must be qualified. The role and functions of states have gone through changes and there never was a period in which political activity was entirely confined to the state. The study of international relations, while recognizing the role of states, identified the importance of international regimes and non-state actors. Even the most Realist reading of international relations will acknowledge groups and organizations other than national states as players in the international arena. It is also fairly obvious that international policies will directly impact on states and domestic politics otherwise there would be little incentive for states to engage in international affairs. The effects of international politics on the domestic state and the presence of non-state and/or sub-national actors on the international forum are by no means new or unprecedented. The interwoven and complex web of national-, supranational and sub-national interactions that is presented as an indication of governance or as politics beyond the state seems to fit in comfortably with existing theories and approaches to international relations. In the domestic context the institutional and procedural complexity may also increase as a result of policy interaction between multiple policy levels. Increased entanglement and complexity at domestic policy levels may be considered a new fact that warrants the use of the governance concept. However, the phenomenon of complex and multi-level interaction is well known and studied within federations. Varying cooperation and coordination between different tiers of government (federal, regional, local) are common practice in many federations. Here too the message conveyed by the governance approach seems to point at what already exists in many states with multiple tiers of government.

8

Politics Beyond the State.book Page 9 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

The active involvement of private actors in the pursuit and execution of collective choices is generally considered to be the defining feature of governance1. Whereas government refers to elected officeholders who within public institution carry out public policies, governance relates to the achievement of collective goals through public/private cooperation. The public authority shares some of its policy-making power with private actors who perform varying tasks ranging from the provision of technical expertise to the effective implementation of public policies. Although private involvement in public policymaking is deemed to be a core feature of governance it is by no means excluded in state governments. Lobbying is a standard form of interest representation in a number of western democracies. The lobbying effort may entail more than merely influencing public decisions but may turn lobbyists into effective decision-makers. Continental European countries with strong neo-corporatist traditions have systematically provided private interests with a role in policy-making. The system of collective bargaining provided preferential access and power to employers and trade unions. Consociational democracies took the inclusion of private interests in the public realm to a new high by transferring broad decision-making powers in the field of education, cultural and social policy to confessional organizations structured in the so-called pillar system. Public policy has never been an entirely public activity from which the private actors and interest were excluded. The above examples indicate that the characteristics of governance may not be entirely novel. Several features of governance were part and parcel of traditional state-based politics. Governance as the complex pursuit of collective goals through multiple tiers of government both within and beyond the state and with a heavy reliance on private resources and interests can also be found in the traditional state context. Although governance may not be the alleged the “a changed condition of ordered rule” as Rhodes2 claims it to be, it would be equally unwarranted to dismiss the concept altogether. There is no evidence to assume a radical transition from state-based government to governance. States and public authorities continue to levy taxes, to impose regulations, to conduct international relations and to generally structure societal interaction. States have not lost

1.

2.

Peters and Pierre highlight the public/private dimension as a central characteristic of governance in Peters, G. B., Pierre, J. (2004) Multi-level governance and democracy: a Faustian bargain? in Bache, I, Flinders, M., (eds.) Multi-level governance. Oxford Scholarship Online Monographs, pp. 75-91. Rhodes R. cited in Stoker, G. (1998) Governance as theory: five propositions. International Social Science Journal, vol. 50, no. 155, pp. 17-28.

9

Politics Beyond the State.book Page 10 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

their relevance and continue to be at the centre of collective decisionmaking both in domestic as well as in international contexts. Rather than a radical break it seems more fruitful to assume a gradual but systematic redefinition of the function, roles and operation modes of states. Under the influence of global markets and economic operators, global communication and global problems, states redefine their roles in function of their existing capabilities and weaknesses. It is recognised that domestic problems (unemployment, pollution, energy security, etc.) are influenced by factors beyond the control of the domestic state government. Regional integration efforts can be interpreted as attempts of domestic states to regain control over policy problems that could not be adequately tackled at the state level (Moravcsik, 1994). Regional integration entails ceding and constraining state power at the supranational level but supranational decisions guided by state interests may also amount to states regaining control over their domestic problems through the supranational policy level. The apparent loss of direct political power at state level is compensated by the enhanced capacity at the supranational level to deal with state problems. The migration of state authority to other supra- and sub-national policy levels and towards private actors can be understood following the same reasoning. In view of increasing constrains (externalities, resources, scope of problem) to address policy problems, states cede power in an attempt to use the resources and capacities of other policy levels and actors to achieve domestic goals. The state continues to be the main frame of reference but the state increasingly resorts to governance to establish and implement collective decisions. Whether governance engenders a weakening or a strengthening of the state remains to be analysed. In any event, the effects of states resorting to governance are unlikely to be uniform across policy domains and policy actors. Governance may (re-)establish the pre-eminence of states in some domains, it may prove futile in other areas. The suggested approach to governance, is one in which the state does not become obsolete but rather changes its functioning and operation mode. The state uses less traditional ‘government’ but experiments with ‘governance’ to achieve outcomes that no longer seem attainable through classic state methods. Governance refers to redefining and remodelling the state. Such gradual but fundamental reforms of the state have, it should not be forgotten, occurred in the past. The shift from a minimal 19th century state to the 20th century welfare state has been momentous. A new shift seems to be in the making with the adoption and diffusion of governance across states. Governance will build on what is already available and

10

Politics Beyond the State.book Page 11 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

present in states. Hence the constitutive elements of governance are not entirely novel but drawn from existing practices within and between states. Public authorities will to varying degrees resort to governance, either by conscious choice or because of contextual requirements. Government and governance will thus coexist within the state, the balance may tilt towards ‘governance’ but public authorities in the state will continue to play a role in the decision making, albeit a different one.

The state, public authorities and government Governance was presented as a change in the functioning of the state. Before we dwell on the nature and scope of that change it is necessary to clarify what we understand to be the functioning and nature of the state. To understand where governance will lead us, we need to develop an understanding of where we are coming from. European statehood is associated with the following characteristics: a) A territorial organisation of authority: states are defined by borders that delimit the territory governed by the state. Within the state additional territorial spheres of authority can be discerned. All states have some form of territorial decentralisation: the regions, provinces, counties, municipalities, … are all based on territorial delimitations. b) The territorial organisation of authority shows a remarkable symmetry between the territorial units in terms of their power and authority. Decentralised local governments tend to be indicative of that symmetry. The powers and structures of Flanders’ smallest local government (68 inhabitants) are essentially similar to those of Flanders’ largest municipality Antwerp (400.000 inhabitants). This symmetry can typically be found in decentralised territorial units. In federations the symmetry between the constituent units may be less pronounced. Federal regions can have different institutions and may even have different prerogatives (e.g. Spain). The institutional asymmetry found in certain federations is often an attempt to accommodate the diversity of population groups on the state territory. Federations with different tiers of asymmetric governments will develop the complex and entangled policy dynamics that are also associated with governance. c) The territorial organisation of authority entails defining borders and a capacity to control and maintain those borders. Though less important for the decentralised units, the capacity to control peoples, capital and markets through border control is vital to craft, implement and enforce collective decisions in the state.

11

Politics Beyond the State.book Page 12 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

d) The state and its territorial components are formalised institutions whose composition, roles and powers are grounded on a constitution or a basic legal norm. The different institutions at the different levels perform tasks within the confines set out in the basic norm (or in legislation derived from the basic norm). e) The pre-defined roles of the public institutions and the ensuing division of labour between institutions and between policy levels lead to a system of state-based checks-and-balances. Monopolisation of power or the capturing of state power by sectional interests is constrained through a constitutionally guaranteed distribution of functions and powers. f) The linkages and relations between institutions and authority levels are clearly specified and often formalised in legal texts. Informal interaction patterns exist and will be important for the functioning of government but they accompany rather than replace formal interaction procedures. The predominance of formalised linkages between institutions creates a certain uniformity, repetitiveness and predictability to the complex functioning of the state. Predictable, repetitive and known interaction patterns tend to be controllable and to inspire trust and confidence. The organizational structure of the state resembles a rather linear or chain structure. Segregated subsystems perform pre-defined functions in sequential order based on direct information and a thorough understanding of the processes involved. g) The predefined roles and division of labour in states entail that public institutions at all levels operate on the basis of mandates (popular or constitutional) that define the radius of their autonomy, where they participate and when they are subordinate. Hence, public authorities do not jump levels and do not engage in activities without a legal framework enabling them to do so. The resulting inflexibility and sizeable administrative burden gives rise to the image of a static bureaucratic state. h) The relations between institutions and policy levels are characterized by autonomy and hierarchy. Public authorities perform a number of functions autonomously without oversight or control by other authorities. In other areas that do not fall under the autonomous powers of an authority, hierarchy applies as the predominant interaction mode. The central state government stands out as the main hierarchical actor but decentralised of federal units can exercise similar hierarchic controls over subordinate decision making levels. The hierarchy/ autonomy combination prevalent in state government is evidenced by the role of local or regional governments in a state. The decentralised territorial governments are in charge of regulating specific local matters 12

Politics Beyond the State.book Page 13 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

and derive considerable autonomy from their local prerogatives. The local autonomy is constrained by a central government determining the limits and modalities of the local autonomy. The central government monitors and possibly overrules local decisions, this hierarchical relationship is more prominent when local authorities act to execute and implement central government policies. i) Territorial public authorities are the main policy makers in the state. Private actors may be consulted or contracted but collective decisions are taken by elected office holders. Public authorities at all levels of the state make use of particular instruments to have their decisions applied and enforced. Taxation and command-and-control regulations stand out as the favoured policy instruments in the state. Public authorities formulate prescriptive norms, they execute and implement the measure, they monitor the execution and compliance, in case of non-compliance sanctions are imposed. Taxation as well as command-and-control requires significant know how and administrative capacity. Adequate rule formulation imposes important information requirements and a thorough understanding of cause-and-effect of the envisaged policy measure. Implementation, monitoring and the imposition of sanctions by public authorities leads to large bureaucracies that turn the state into a major employer as well as an administration with considerable inhouse expertise. Taxation and command-and-control regulation presuppose many of the prior features of states, namely, border control, an unambiguous institutional division of labour and hierarchical interinstitutional relations. j) States were and still are very important economic actors. In the mixed European economies, publicly owned enterprises dominated key economic sectors. The utility, energy, telecom and rail transport sectors were largely owned and run by public authorities in most post-war western democracies. The emergence of the welfare state decommodified many welfare provisions (health, insurances, education) and turned the state into the main service provider. In addition to the state as a direct provider of goods and services, public authorities intervened directly on the market through monetary and economic policies. Policies aimed at maintaining functioning markets (anti-trust measures), market access and product standards provided public authorities further leverage on economic operators within the state. k) Its capacity to levy taxes, to regulate the economy and to hierarchically impose policies, enabled state to set in place considerable redistributive policies, which focussed on welfare entitlements as well as on setting up major infrastructures.

13

Politics Beyond the State.book Page 14 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

l) The penetrating and comprehensive impact of state authority is supported by three main sources of legitimacy, namely, identity, performance and democratic legitimacy (Beetham and Lord, 2001). Performance legitimacy refers to the states’ problem-solving capacity, the extent to which public authorities manage to formulate paretooptimal policy outcomes. Congruence between identities and authority can render public decision more readily acceptable. Democratic legitimacy is based on elections, majoritarian decision making and accountability. Elected office holders are held accountable for their policies and overall performance by territorial constituencies. Given the institutional division of labour, the pre-defined tasks, responsibilities and electoral promises, territorial constituencies are able to asses the performance of elected office holders and to attribute blame or rewards accordingly. The government performance is evaluated at regular intervals by electoral constituencies and enables alternation in government office. Political elites can be confirmed or dismissed at the polls. The capacity to select and dismiss elected officials in public authority based on their performance in office requires clarity on what governments are allowed, able and willing to do. The institutional organisation of the state with its formalised roles, prerogatives and responsibilities provides that clarity. m)The state, its institutions and borders set the framework for a national public sphere. Collective choices are presented, debated and evaluated by media, political parties, interest groups, experts and citizens. The public sphere is the crucial interface between the government and the citizen. Through public debates and interaction, government choices and problems are relayed to the citizen. The public sphere conveys information but also allows public authorities to build or assess the support for their policies. State and institutional borders tend to define the limits of the public sphere. Broad cross-national political debates rarely take place and tend to be confined to several, separate national debates with little cross-national exchange. Even institutional frontiers within the state can segment the public sphere. For instance, in the Belgian federal state public debates occur and are structured on the two main regions (Flanders and Wallonia) with little nation-wide communicative interaction. n) Political parties are essential actors in state politics. Political parties operate on all fronts of the state. They aggregate interests, articulate goals, set policy agendas, formulate policies, co-ordinate and steer different levels of government. Political parties play a vital role in the public sphere by structuring debates and information. Political parties socialize and mobilize electorates. Due to their multiple functions 14

Politics Beyond the State.book Page 15 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

political parties are the drivers and organisers of politics in the state. Political parties have developed in the context of the state, their internal organisation, functions and strategies have been entirely tailored on the territorial organisation of state authority. Given the symbiotic relationship between the state and political parties, changes in the functioning of the state are likely to constitute major challenges to political parties seeking to consolidate their central role in the state. Our understanding of state government was discussed in the previous paragraph. States encompasses more than what was discussed but we chose to highlight the features that are most likely to be affected by the increased presence of governance. The post war democratic state is a territorial organisation of authority based on a constitution characterised by a formalised division of labour, by formalised inter-institutional links which stress institutional uniformity and symmetry. It has a capacity to hierarchically control and maintain borders, to impose taxes and binding norms formulated by elected office holders using problem-solving and democratic justifications. These features have enabled states to intervene in markets as regulators or as producers of goods and services and to formulate major re-distributive policies leading to large bureaucracies and the welfare state. The encompassing, penetrating and binding nature of state activity generates a national public sphere in which political parties are a crucial interface between the state and the populace.

The context of governance The spread and intensification of governance in the state is part of the broader context of different but interrelated processes. First, at the risk of stating the obvious, there is the process of globalisation. The implosion of the soviet regime and the subsequent globalisation of market economy is a major underpinning variable. The rapid development of global communication technology, consider for instance the exponential growth of the internet (1996: 16 million users, 2005: 1 billion users), global media, the acknowledgment of global threats (environment, energy, terrorism, crime, etc.) and increased mobility (tourists, labour, capital, etc.) have permeated state borders and reduced the state’s ability to regulate these global forces in the national state context. Second, regional integration and the development of supranational authorities have lead to a gradual but important transfer of state powers to supranational bodies. The European Union stands out as the cardinal case of regional integration. The European Union imposes binding decisions on 15

Politics Beyond the State.book Page 16 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

states in important policy domains. Due to the broadened application of qualified majority rule the states have effectively lost autonomy and the control over certain Europeanised policies. More than the actual transfer of prerogatives it is the market-oriented nature of European integration that has deeply affected European states. The establishment of a common market, of internal market regulations (VAT, health and safety standards, anti-trust measures) and of common currency governed by a supranational central bank, has curtailed the states autonomous capacity to regulate markets. Moreover, the EU’s sustained effort to abolish state-owned monopolies, preferential treatment of domestic producers and market segment sheltered from competition, has lead to the gradual privatisation of European mixed economies (telecommunications, airtransport, energy, utilities) (Scharpf, 1999). The establishment of a common market in which capital and labour can ‘vote with their feet’ in combination with increased global mobility and communication has further constraining consequences on the states’ ability to tax and regulate economic operators. Even in areas where the EU has no authority and where the state remains fully competent to regulate and act, national states find it increasingly difficult to intervene in and correct economic processes without generating competitive disadvantages for domestic producers or without discouraging potential investors. Whether by deliberate design or by incremental and unintended changes, the emergence of supranational policy-making has severely reduced the state’s role as an economic operator and regulator. Third, transfer of state prerogatives does not only occur towards supranational authorities but sub-national authorities have also been on the receiving end of central state power. Shifting decision-making power to regional, local or other public institutions (for instance, agencies) does not necessarily entail a loss of state authority since the transfer occurs between public authorities. Rodrigues-Posé (2003) discerns a global trend towards decentralisation since the early nineties. Vincent Wright links the decentralisation trend to the growing difficulty of central governments to impose taxes in a global market economy. Central governments are caught between domestic constituencies with high and thus costly welfare expectations (cheap education, cheap medical and health service, high pensions and welfare benefits) and increasing difficulty to raise tax revenue because of popular discontent or capital flight. In attempt to get out of the conundrum central governments seek to transfer the spending burden to lower tiers of government. The dispersal and decentralisation of spending and taxes reduces the visibility of higher taxes and/or and probability of broad (state-wide) contestation of policies curtailing expensive services. Irrespective of the reasons for decentralisation, the result of the process is

16

Politics Beyond the State.book Page 17 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

that many regional or local authorities become stakeholders and effective policy-makers in policy areas previously regulated from the centre (Wright, 1998). The fourth process that sets the context for governance refers to the privatisation of the public sector. As was discussed above, European integration contributed greatly to the liberalization process, but the process is not limited to European states. The neo-liberal concerns with ‘big government’, government failures and the general ambition ‘to roll back frontiers of the state’ to reap the full benefits of unhampered, functioning markets translated in an overall downsizing of state bureaucracies and the introduction of New Public Management. The new managerial approach in the public sector introduced market mechanisms in the civil service (performance management, fixed term mandates, contracting expertise and know how) and stimulated outsourcing of public functions to private actors (Christensen and Laegreid, 2003). The liberalisation wave privatised in-house expertise and endowed private actors with important policy making functions (policy formulation, execution). The increased involvement of private actors in policy-making generates public dependency on private resources and capabilities. Public authorities seek to develop their capacity to steer and monitor private service providers and move away from direct policy execution and implementation. The combined effects of globalization, supranational integration, decentralisation and privatisation stimulate extensive fragmentation of state authority. Prerogatives, administrative capacities, financial and other resources, information and expertise are dispersed over different policy levels and diverse public and private organisations. The increased fragmentation forces state governments to gather and mobilize institutions and private actors at different levels to deploy their resources and capabilities for the purpose of collective decision-making. Not only central but all levels of government are confronted with increased interdependencies or the awareness that own resources and policy tools do not suffice to deliver the desired policy outcomes. Globalisation and the permeability of state borders have strengthened the perception that many domestic problems are in fact cross-national and global challenges which render state action futile or undesirable. Hence, state governments may choose not to address the issue as a policy problem requiring political action, or more likely, the state government points at the supranational level as the appropriate level of undertake action. Responsibility shifting and refocusing state policies to domains with

17

Politics Beyond the State.book Page 18 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

limited interdependencies or external effects are common occurrences in the governance context.

The state and governance Territorial organisation of authority was identified as a defining feature of the state. Although the territorial definition of the state remains relevant additional and potentially competing territorial definitions emerge. The empowerment of regional, local and supranational institutions increases the significance of different territorial delineations of power. Certain crossborder interregional forms of cooperation (eg. Eurostar region, eurregion, etc.) have emerged. Although these interregional cooperation efforts largely originated from EU regional development programs (eg. Interreg), some cross-border collective action has taken root and lead to sustained and endogenous interregional cooperation (eg. the Belgian-Dutch transnational university Limburg). The European Union is the main source of territorial reconfiguration. First, there is the impressive increase in scale of the Union. 27 member states with a total of 493 million citizens entails an major increase in diversity to be addressed by the polity. Formulating effective, adequate and uniform policies for a diverse and large society generates important cognitive and logistic difficulties. Moreover, the ongoing enlargement process which is likely to bring the EU up to some 30 members states, leads to a repeated shifting of geographical boundaries and hampers the creation of cohesive European identity. Second, the EU’s sectoral policy approach has stimulated the creation of EU-wide interest groups in which sector-specific rather than national interests dominate. Tailored on EU policy programs, regions and local communities have invested in direct representation to the European Union, bypassing the national channels, in the form of issuespecific interest groups (e.g. maritime regions, legislative regions, automotive regions, etc.). Collective decision-making through governance does not reduce the relevance of territorial authorities. It does point at different territorial configurations emerging within and beyond the territorial definitions of the state. The cooperation between territorial entities is not mediated or controlled by the state but occurs as regions, local or supranational authorities seem to discern benefits in by-passing the state. Government was associated with a constitutional division of labour between public institutions with pre-defined roles, formalised links and hierarchic control. In a context where state authority is increasingly 18

Politics Beyond the State.book Page 19 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

fragmented and dispersed, public authorities need to mobilise resources, capabilities and support from other authorities and private actors to carry out policies. The increased interdependence through fragmentation of authority is experienced by all levels of government. Supra-, sub- and national governments are all confronted with limitations of their powers and their dependence on third-parties for effective goal achievement. The European Union is a clear example of this interdependence and its consequences. The European Union is a potent and effective regulator, which over the years produced an impressive “acquis communautaire” of some 90.000 pages of legislation directly applicable to the member states. Although European institutions have a capacity to formulate binding rules, their capacity to implement and give effect to these rules is limited. The EU depends on the member states to have its policies executed and implemented. The European Commission is a comparatively small administration (about 24000 officials) without the authority, means or know how to organize the execution of its policies. The revenue of European Union is determined by the member states and the EU has no authority to generate revenue through taxation. Moreover, even for the development and formulation of binding norms the European institutions are heavily depend on the information and expertise provided by member states and private actors. Hence, the success of many supranational policies is heavily dependent on the willingness of national, regional and local authorities to actively contribute to and to adequately implement EU policies. The institutional interdependence is directly related to the exercise of power. Interdependence tends to balance out power relations and elevates formally subordinate actors to more co-ordinate levels. The point is highlighted in Fritz Scharpf’s classic article on German federalism (Scharpf, 1988), where he discusses how the constitutionally subordinate German Länder used their implementation and stalling capacity to force the federal government to negotiate on issues which were formally exclusive federal prerogatives. The direct consequence of fragmented authority and its resulting institutional interdependence is that public authorities at all levels engage in more co-ordinate and negotiated relationships. Formal hierarchy persist but becomes less relevant when levels of government and authorities depend on and can sanction each other. In an analysis of the European regional development policy, Hooghe and Marks (2001) point out two features which further clarify the nature of governance, namely, the variation and diversity of actors involved in the execution of (regional development) policies. The EUs emphasis on ‘partnerships’ and co-financing as conditions for receiving European

19

Politics Beyond the State.book Page 20 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

structural funds gave rise to very different and diverse domestic coalitions and networks. In some member state central governments firmly controlled all features of the development projects (e.g. Greece, France), in other states idiosyncratic cooperation occurred involving regional, local and city governments, development agencies, local business community, social economy and European officials. These collaborative networks varied in composition in each development area and the participants jointly crafted the regional and local development plans and projects. The study highlights the emergence of ad hoc, context or policy specific networks with varying public and private actors drawn from different levels and institutions carrying out public policies. The variation and the public/private composition of these collaborative networks blurs the conventional image of the state with a fixed institutional division of labour and pre-defined roles and functions. The governance approach emphasizes a more entrepreneurial and pro-active attitude of public authorities devising ad hoc and flexible cooperation formats with other public institutions and private actors. Similar coalitions of public/private actors drawn from different levels can be found in other European policy areas (social and employment policy, environmental policy, research and development policy) but it remains unclear whether these variable and multi-level dynamics also occur in the domestic context without EU instigation. Taxation and command-and-control regulation were discussed as essential tools for the operation of government. The ‘governance’ approach can also be associated with specific policy instruments. Public authorities have a broad arsenal of policy instruments at their disposal to alter behaviour and achieve collective goals. Van den Heuvel (2001) distinguishes instruments based on coercion (law, regulation, levies, taxes), transaction-based instruments (contracts, commitments, subsidies) and persuasion-based instruments (information, propaganda, reviews). Public authorities will resort to a mix of policy instruments to implement policies. However, the legitimate use of coercive instruments (law and levies) is strongly associated with ‘government’. Classic ‘command-andcontrol’ regulation imposed in a top down fashion by a central government corresponds with the standard operating procedure associated with the ‘government’ concept. This form of classic regulation entails the formulation of a rule, the monitoring of its implementation and a possible sanction in case of non-compliance. The governance approach refers to policy instruments, which, in one way or another, differ from coercive instruments or prescriptive and

20

Politics Beyond the State.book Page 21 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

enforceable legislation. The ‘soft’ instruments associated with governance can (a) be voluntary or non-binding, (b) have no clear sanctions in case of non-compliance, (c) have no specified procedure, or (d) have no clear goals3. The soft policy instrument will lack at least one and often a combination of the features of classic regulation. Instead, soft policy instruments provide room for private actor involvement in different phases of policy formulation and implementation. The prescriptive and coercive nature of the instrument is replaced by increased stakeholder involvement. Compliance is sought through direct private actor involvement in the policy formulation, rather than through imposition and sanctioning. Examples of these ‘soft’ and ‘governance’ instruments can be found at national and supranational policy levels. Moreover, use of the ‘soft’ instruments seems increasingly popular. Policy making through covenants, quality labels, management- and audit systems, open method of coordination (EU economic and social policies), partnerships (EU regional policy), tripartite contracts (Vara Aribas and Bourdin, 2006), are examples of policy techniques that deviate from classic regulatory methods and explicitly depend on extensive private actor involvement. The explanations offered for the rising importance of governance instruments is to some extend biased by the policy domain and level where the policy instruments are applied. Without seeking to be comprehensive the following reasons for the increased use of ‘soft’ policy instruments can be found in the literature: • The instruments involve little or no administrative costs for public authorities (there is less need for extensive prescription, monitoring and control). • The instruments provide opportunities to systematically tap into the information and know how of private actors (producers, consumers, non-governmental organizations) for public purposes. • The instruments are used to avoid the disadvantages (time, resources) and the risks (deadlock) of cumbersome and veto-sensitive legislative processes (for instance, the co-decision procedure). • The instruments provide an alternative source of government legitimacy because stakeholders and target groups are directly involved, through formats ranging from consultation to self-regulation, in the formulation and execution of the policy.

3.

See Jordan, Wurzel and Zito (2005) for this and other typologies on voluntary instruments.

21

Politics Beyond the State.book Page 22 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

• The flexibility of the instruments allows for context-specific policy measures adapted to the policy situation at hand. • The voluntary and non-coercive nature of the policy instruments and the involvement of those affected by the policy, reduces the probability of contestation and protest. Formulating and implementing public policy in a context of increased diversity, scale and interdependence presents public authorities with formidable cognitive and information requirements. Public authorities at national, and especially at supranational levels, often do not have the means to overcome the challenges of uncertainty and incomplete information. Moreover, constraining legislation may prove to be a competitive disadvantage to domestic producers operating in a common European market and a globalized economy. Public authorities will seek to circumvent the transaction costs linked to classic regulatory instruments and resort to ‘soft’ policy instruments that put the governance approach into practice. The soft policy instruments introduce private actors and target groups as policymakers. The stakeholders provide information and legitimacy both are much needed assets in multi-level political systems. Soft instruments can constitute a solution to the problem of incomplete information, they also reduce the probably of damaging conflict. Public authorities can avoid protest and discontent through the use of voluntary and less constraining policy tools. Despite their advantages, voluntary and soft instruments raise many concerns and can turn out to be a double-edged sword. First, voluntary instruments heavily depend on the participation of private actors, which can be expected to pursue private rather than public goals and interests. Public authorities have limited means to ensure that sectional, and resourceful private interests dominate the process at the detriment of weaker and public interests. Second, voluntary instruments amount to a delegation of public authority to private actors. The delegation requires that public authorities, as ‘principals’ manage to monitor and steer the private ‘agent’. However, when agents monopolize information and expertise and when there are few or no viable alternatives than the services offered by the mandated agent, principal-agent relationships tend to be reversed. Instead of the principal steering the agent it is the agent who takes controls and guides the decisions of the principal. Third, supranational experiences with voluntary and non-binding instruments have been rather disappointing. The instruments work when there is congruence between public goals and private interests. Such parallelism between private and public interest can be achieved by supporting measures which stimulate private actors to adopt voluntary measures. 22

Politics Beyond the State.book Page 23 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

However, these supportive measures are often the threat of binding legislative action or the adoption of levies and taxes, which brings us back to the traditional instruments of government. Accountability and the alternation of elected office holders combined with a capacity to solve social problems are the cornerstones of state legitimacy. Globalization presents social and economic challenges which are beyond the control of the state and force public authorities to reform and downsize some of their key functions (e.g. welfare provisions) and seek policy solutions through supranational action. In this way, performance legitimacy of national states seems to be significantly reduced. European states struggle to formulate answers to industrial delocalization, high unemployment levels, immigration, environmental pollution, international terrorism, while acknowledging that these phenomena can not be fully tackled at the national level but require supranational measures. Governance also generates tensions with regard to democratic legitimacy. As policies originate at different levels and result from complex interactions between a multitude of public authorities and private actors, constituencies find it increasingly difficult to evaluate the performance of governments and to hold policy-makers accountable. Moreover, supranational policy levels are poorly, if at all (e.g. European Commission), connected to electoral constituencies. The accountability problem is magnified as elected officeholders shift policy making functions to private actors. The image of electoral accountability of policymakers and alternation in office in function of electoral policy evaluations seems to loose its relevance in the dynamics and processes of governance. Governance thus leads to the development of alternative sources of legitimacy. Deliberative democracy or the systematic and continuous consultation of stakeholders on all aspects of public policies is presented as a possible alternative to compensate for the loss of traditional democratic control. The following chapters in this volume will focus on the abovementioned features and aspects of the shift from government to governance. The impact on national politics of multi-tiered structures is discussed in the first chapters by Deschouwer, Stefuriuc and Buelens. These chapters focus on the impacts of different levels of government on electoral results and on coalition formation in national settings. The chapters by Wiercx, Mahoney, Huysseune and Van Assche focus on the representation of domestic interests at the European level. Wiercx’s chapter discusses the strengths and weaknesses of Eurogroups and their capacity to aggregate local

23

Politics Beyond the State.book Page 24 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

interests. Mahoney’s chapter discusses outside lobbying and the extent to which interest groups manage to mobilize local and domestic constituencies. The chapter by Huysseune and Jans discusses the role and functions of regional representations in Brussels. The chapter by Van Assche and Deschouwer presents the complexity of national and supranational interest representation in the case of the Port Services Directive. The remaining chapters discuss different features of governance in policy making processes. Tanasescu’s chapter on deliberative democracy discusses the EU’s increased recourse to deliberation in order to compensate for its lack of direct democratic legitimacy. Bedoyan’s chapter discusses the delegation of policymaking functions to private actors in the case of the EU’s new approach legislation and highlights some its disadvantages in the case of the Packaging Waste directive. Meier’s chapter on gender mainstreaming presents the heavy reliance of public authorities on gender experts and its consequences. Paantjes chapter on violence against women shows some of the agenda-setting and –framing opportunities that arise in diffuse and multi-tiered policy contexts. The joint chapter by Jans, Tanasescu and Bedoyan on the EU’s integrated product policy points at some of the pitfalls of governance approaches, namely, the lack of tangible outputs or results.

24

Politics Beyond the State.book Page 25 Wednesday, December 5, 2007 8:41 AM

Electoral Politics in Multi-Level Settings Kris Deschouwer

Regions and electoral politics: three approaches Electoral politics has developed and has become institutionalized within the context of the national state. Expansion of voting rights and the development of mass political parties is indeed one of the crucial developments of modern politics that have been shaped by the state context in which it happened. Voters are voters of the state in which they live and political parties mobilize these voters to win office (or to influence the policies) in that national state. Even the parties that can be identified as belonging to broader party families – present in more than one country – are in the very first place national party organizations active in their own national context. Caramani (1996; 2004) has convincingly shown the extent to which this national state context has influenced the development of electoral politics. The ‘nationalization’ of electoral politics is a process in which both voters and parties have increasingly been using the national state as their primary reference point. The evidence shows an increasing territorial homogeneity of voter’s responses to the party offer. Several indicators can be used to illustrate that point, and they all show a similar trend. The first is voter turnout, where clearly the variation between the constituencies of one single country gradually decline. The second is the party offer. Here the evidence shows that an increasing number of parties opt for a national strategy, i.e. they present candidates or lists in an increasing number of constituencies, with all the major parties indeed spreading their presence over the totality of the national territory. And the third is the response of the voters to the party offer. The results of the political parties become increasingly homogeneous, i.e. their scores become less different from one constituency to the other. Nationalization was a process that occurred quite early. Although there are interesting variations between countries and between party families, the nationalization is a rapid evolution that seems to be almost completed after the First World War. During the 20th century the process of nationalization slows down, but still moves in the same direction. The First World War as the end of rapid nationalization is an interesting finding. It coincides with 25

Politics Beyond the State.book Page 26 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

the date given by Lipset and Rokkan (1967) to identify the ‘freezing’ of the European party systems. World War One is for them the end of the mobilization of the voters. Parties and party systems – all within the context of national states – have by then been developed and institutionalized. After 1918 the parties have obviously changed, adapting to their changing societal environment, but the basic structure of the competition has remained very much the same (Mair, 1997). It is this basic structure – the mechanics of the party system – that is part of the visible and stable national party political context in which voters and parties make their strategic choices. If this national institutional environment is indeed so important to understand the way in which politics function, the gradually changing institutional architecture of politics must have very visible effects on political actors and strategies. If we are today witnessing a move towards a more ‘regional Europe’, towards a (re)invention of the regions as meaningful levels of decision-making, towards an increasing importance of the European Union, then the effects on (among others) party politics must be evident. The general mood in the literature of the past two decades is undoubtedly one in which the regional fact is being explicitly acknowledged, especially in countries where regions are fairly new or where they have recently received more substantial competencies (e.g. Harvie, 1994; Keating, 1998; Loughlin, 2001). For party politics or more broadly for electoral politics this leads logically to three possible types of analysis. The first one is the analysis of the regional level itself: regional elections, regional parties, and regional party units. The new level is analyzed in its own logic, with its own electoral system, its own parties and party system, its own pattern of competition (e.g. De Winter & Türsan, 1999; Hough & Jeffery, 2005). This approach does not differ much from the analysis of national party systems, although the presence of (sometimes strong) regional and regionalist parties can give a special dynamic to regional party politics. Regionalist parties claiming more autonomy do indeed link the regional party system to the national one. The second type of analysis looks much more explicitly at the regional level interacting with the national level ands vice versa. It acknowledges the fact that both levels can only be fully understood by (also) looking at the way in which they influence each other. It is often assumed that this mutual influence is hierarchical, that there is one level than can be considered to

26

Politics Beyond the State.book Page 27 Wednesday, December 5, 2007 8:41 AM

Electoral Politics in Multi-Level Settings

be the ‘core’ level, the one giving meaning to what is happening at the other level. And usually the core level is then the level of the national state. This approach has entered the political science literature after the first direct election of the European Parliament in 1979. These elections were labeled ‘second order national elections’ (Reif & Schmitt, 1980; Reif, 1984), explaining how indeed these elections were contested by national parties and given meaning at the level of national politics. This notion of unequal interaction is however older than the first elections of the European Parliament. It goes back indeed to the analysis of regional elections in Germany (Dinkel, 1977). The concept of second order elections has now recently been brought back to this type of comparison, looking at the way in which and the degree in which local or regional elections have a logic and dynamic that finds its origin at the national level (e.g. Heath et.al, 1999; Jeffery & Hough, 2003; Pallarès & Keating, 2003). In this chapter we want to explore a third approach for looking at the consequences of regionalization of electoral politics. It links directly to the notion of nationalization, the evolution towards homogeneous party actions and homogeneous voter responses. One could expect that the long process of nationalization is being reversed in countries where the regional level has become more important. Indeed, if the regional level is important, it means that there is always more than one relevant level in the institutional environment of electoral politics. And rather than looking at the way in which the regional electoral politics are different from or influenced by the national politics, one can try to see to what extent the national electoral politics are being affected by the presence of the regional level. Our research question is therefore: can we see an effect of the regionalization of political systems on the national electoral outcomes. In other words: is there evidence of a slowing down of the nationalization processes or even of a regionalization of national electoral politics? As said above, that should especially be the case in countries where the regionalization is very important, where regions are very visible and strong political actors. For that reason we will explore the regionalization of national electoral politics in two countries: Belgium and Spain. In these countries we do expect to see evidence of regionalization. Belgium has been transformed from a unitary state into a complex federal-type system, taking into account the ethno-linguistic differences between the Dutch speakers of the north and the French speakers of the south of the country (and of the awkward location of the frenchified capital city Brussels in the northern region). Moreover, there are no national political parties, but only parties competing in their own language group. Even for national elections, the

27

Politics Beyond the State.book Page 28 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

electoral arena is split into a northern one and a southern one. That must lead to a strong regionalization of national electoral politics. Spain is the second case, and also one in which the regional fact is highly important and relevant. The granting of substantial political powers to the Autonomous Communities as part of the process of democratization after the Franco regime has from the outset created a political system in which the political parties have to be aware of regional differences. The regional party systems in Spain display a high degree of variation, mainly due to the presence of a fairly large number of regional parties. The importance of the regional level for electoral politics has also been clearly illustrated by the high degree of dissimilarity between regional and national elections (Jeffery & Hough, 2003; Pallarès & Keating, 2003). Actually, regional elections in Spain appear to be first order elections, having a meaning of their own (and possibly affecting national Spanish politics). For both countries we will ask two related questions. First we try to see whether and to what extent the national electoral outcomes are reflecting the regional differences. Since we look at very decentralized countries, we do assume that we will find these regional differences. The second question is then: do these differences increase? If the regional fact and the regional level are becoming more important in the course of the last few decades, we should see an increasing regionalization of the national electoral politics in Belgium and Spain. For both countries we will use similar measurements, although we also need to take into account the peculiar characteristics of both countries. One of the crucial differences is the number of regions. In Belgium – although there are three language communities (Dutch, French and German) and three regions (Flanders, Wallonia and Brussels) – the division of the country is basically one between north and south. This bipolar logic means that the national level is the aggregation of the two parts of the country, and that any comparison of a region with the national level is simply a comparison of one region with the other. In Spain each region can indeed be compared with the rest of the country. The other major difference is the absence in Belgium of national political parties. That is a very special feature. As we will discuss below, it could possibly stop our research at the very beginning: since there is no national party level, there can only be regionalized party politics. We will for Belgium use the party family as our unit of analysis, in order to assess if and to what extent there are still national patterns than can be discerned.

28

Politics Beyond the State.book Page 29 Wednesday, December 5, 2007 8:41 AM

Electoral Politics in Multi-Level Settings

The Belgian case Belgium seems to be an obvious case of strong regionalization of electoral politics. Indeed, between 1968 and 1978 the three major Belgian parties – Christian democrats, Liberals and Socialists – fell apart into two separate unilingual parties, only competing with the parties of their own language group (except for the central Brussels constituency). Subsequently, all new parties were created in one or the other language group, even if some of these new parties – the Greens in particular – are present on both sides (but as two different party organizations). The coverage of the territory by the Belgian parties is therefore quite low, since there is not one single (significant) party present in all the constituencies. As a consequence, there are two party systems. Voters of each language group are confronted (via unilingual mass media) with patters of competition that are specific to that one language group. In the north the party system is very fragmented, and the once dominating Christiandemocratic party is now competing with three other parties for the number one position in Flanders: the Liberals, the Socialists and the right-wing extremist Vlaams Belang. In the south the party system comes closer to a two-party mechanism, where a strong Socialist party is being challenged mainly by the Liberals. This split party system should be a strong factor working against the nationalization, since the voters receive their major cues from what happens – in terms of party politics – in their own region only. Furthermore, the political institutions of Belgium have changed radically. The unitary system was transformed gradually into a federal-type system, with autonomous language communities and autonomous regions receiving a large catalogue of competencies. Major policy domains like education, culture and arts, public transport, public works, environment and economy (including the international relations for all these matters) have been devolved. That is another institutional element suggesting that the major reference for the voters is increasingly what happens in their region only and what is delivered by the parties of that region only. At first sight it might then seem quite pointless to take Belgium as a case where regionalization of electoral politics can be tested. The regionalization goes without saying. Belgium’s self-image is also one in which the discourse of ‘increasing difference’ prevails. The general belief is that the north and the south of the country are constantly growing further

29

Politics Beyond the State.book Page 30 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

apart, wanting different policies, displaying different attitudes and political cultures. It leads to a constant questioning of the future of the system. Yet the Belgian political system (still) exists, and its existence does mean that national factors remain present in the way in which it functions. The most important federal institution is undoubtedly the federal government. One of its important characteristics (so far at least) is its symmetrical composition. That means that the federal cabinet (for which there is a constitutional obligation to be composed of an equal number of Ministers from each language community) not only produces a clear division between governing and opposition parties, but also between governing and not governing party families. The competition for government thus keeps a similar character in both communities. Even if there is a specific dynamic in both party systems, the parties of the same family go to the polls on both sides of the language border either as parties of the incumbent government or as parties of the opposition. If this is a meaningful element, it should have a homogenizing effect on the electoral fate of party families all over the Belgian territory. In this analysis of the Belgian case, we will therefore take the party families as the unit of analysis. It is also an interesting (and easier) way to present a truly longitudinal analysis, since the party families can be seen and measured irrespective of the presence of a unitary party. One of the basic ingredients of the difference between north and south in Belgium is the difference in the party landscape. That raises indeed the question whether parties of the same family display different evolutions in north and south. We will compare the electoral results since 1946 in the two major regions of the country: Flanders and Wallonia. That is not the only way to divide the country. Indeed, the regional division does not fully coincide with the division in language communities. For Flanders the electoral results of the Flemish region are almost the same as those in the Dutch speaking (Flemish) community. The only difference is the votes for Dutch speaking parties in Brussels. Since more than 97% of the votes for Dutch speaking parties come from the Flemish region and not from Brussels, this difference between region and community is very marginal. On the Francophone side things are a bit more complicated. Indeed, Wallonia is not the same as the French speaking community. Between 15% and 20% of the votes for Francophone parties come from Brussels. We will however look at the Walloon region alone. Opting for the community would mean that we have to be able to count the francophone votes in Brussels, and this is only possible after the split of the Belgian parties. Before the split the unitary Belgian parties received votes in Brussels, but it is impossible to know from

30

Politics Beyond the State.book Page 31 Wednesday, December 5, 2007 8:41 AM

Electoral Politics in Multi-Level Settings

which language group they came. We also leave out the Brussels region. The problem here is that it is not possible – until 1995 – to aggregate election results to the level of what is today the Brussels region. The electoral cantons in Brussels – the lowest level at which votes are counted – did include local communities of what is today the region of Flanders. That also means that the results for Flanders are not absolutely perfect: some of the votes in Flanders (a very small proportion however) are cast until 1995 in cantons that include parts of Brussels.

Party families in Flanders and Wallonia In figure 1 the electoral results for six political families since 1946 are displayed. They show how Christian democrats have always been stronger in Flanders, and the Socialists in Wallonia. It is also quite interesting to see that the long term trends per family in both regions are fairly similar. The most striking exception to that rule is the right wing extremists, who have indeed a different story in north and south. But in general it is mainly the strength of the party families that differs per region. That is a first indicator of the fact that – apparently – the electorates in north and south seem to respond in a similar way to what happens at the level of national politics.

31

Politics Beyond the State.book Page 32 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Figure 1 The evolution of the score for the party families in Belgium between 1946 and 2003

If the level of support for each party family is different, that must add up to a different overall result in Flanders and Wallonia. The question is then: is this difference increasing? Are the two regions growing further apart? That can be tested in a very straightforward way. We have for each election since 1946 calculated the index of dissimilarity (the volatility index), summing up the differences per party family (and dividing by 2). The result of this

32

Politics Beyond the State.book Page 33 Wednesday, December 5, 2007 8:41 AM

Electoral Politics in Multi-Level Settings

can be seen in figure 2. The highest difference was in 1946, and the linear trend goes down, be it not very steeply. The two regions do indeed produce very different electoral outcomes. That has always been the case and still is valid today. Figure 2 The difference between Flanders and Wallonia (results per party family)

The stability or slight decrease of the difference is however a rather counterintuitive finding, not supporting the generally accepted idea that the regions are becoming increasingly different. Of course – this caveat remains valid throughout this Belgian story – we are looking at party families and not at parties. Behind this more or less stable difference between north and south is the story of parties falling apart. Yet while they did indeed fell apart or – for the new ones – developed independently from each other in both regions, their electoral fates follow similar curves. Table 1 The difference between electoral results in Flanders and Wallonia per party family

1946 1949 1950 1954 1958 1961

CD 29.14 20.02 26.57 21.50 21.47 19.35

SOC 8.97 14.06 18.60 18.97 18.42 16.80

LIB 1.58 1.56 2.03 0.92 0.62 0.36

COM 15.99 9.00 5.33 5.17 4.35 5.29

REG

ECO

XR

3.24 0.00 3.89 3.40 5.96

33

Politics Beyond the State.book Page 34 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

CD 19.25 18.11 17.03 17.03 18.19 16.74 12.52 12.06 8.26 4.39 4.91 5.58 5.84

1965 1968 1971 1974 1977 1978 1981 1985 1987 1991 1995 1999 2003

SOC 10.60 8.51 15.71 15.71 16.67 15.66 15.51 15.59 19.58 19.61 13.54 14.02 12.51

LIB 8.80 10.50 1.88 1.88 4.59 0.52 0.47 6.50 3.54 0.58 2.31 1.40 3.24

COM 8.64 1.46 0.44 0.44 4.12 3.91 2.94 2.01 1.21 0.00 0.00 0.00 0.00

REG 9.97 6.66 1.91 1.91 7.58 2.67 10.69 12.65 12.22 8.26 7.38 8.17 4.84

ECO

XR

0.61 1.06 1.98 0.01 0.94 5.61 3.15 6.84 3.39

2.04 1.77 2.20 3.04 8.10 6.01 10.46 12.36

Table 1 shows how the index of dissimilarity is composed per election year. The figures are the difference between the results of each family in Flanders and Wallonia. And it becomes immediately very clear that the substance of the more or less stable difference has been changing quite dramatically. In 1946 there is a difference of almost 30% between the Christian democrats of the north (polling 56% of the votes) and the south. Another major difference is the left vote, with a 15% difference for the Communists and another 9% for the Socialists. This huge gap between the Christian democratic results has gradually been disappearing, which is due to a very large extent to the spectacular decline of Christian democracy in Flanders. To the remaining difference of 15% for the left vote, there was in the late 1980s first a large difference between the votes for the regionalists in Flanders and Wallonia. While the regionalists gradually disappeared in the south, they remained a visible political force in the north until the late 1980s. In 2007 the difference is due to the higher score of the extreme right in Flanders and still the difference in the vote for the Socialists. The two regions have thus always been very different and are still very different. The reason for the difference has changed over time and has become more varied, more fragmented.

Territorial homogeneity of the electoral results Electoral results are homogeneous if the territorial differences for the parties are limited, i.e. if the outcomes per party do not vary extensively over the territory. There are several ways in which this territorial

34

Politics Beyond the State.book Page 35 Wednesday, December 5, 2007 8:41 AM

Electoral Politics in Multi-Level Settings

homogeneity can be measured, and most measures have some (serious) biases. We have opted here for a very simple and intuitive one: a standardized coefficient of variation (CV). It computes the standard deviation of the results per party and per electoral constituency and divides it (for the purpose of standardization) by the mean of the party. As units of analysis we take the results per party and per electoral canton. The cantons are the lowest level at which the electoral results are counted. Results per canton are further aggregated at the level of the electoral constituencies for the distribution of seats. In order to get a figure per election year, we compute a weighted (size of the party) average CV per party. Figure 3 Homogeneity of electoral results per canton in Belgium, Flanders and Wallonia (1945-2003)

Figure 3 shows the summary of the results for Belgium as a whole and for the two major regions. The overall trend is the same: electoral results since 1946 have gradually become more homogeneous. The results for the Christian democrats have always been very homogeneous in both regions, and the decline of this party family has not altered this. Liberals and Socialists were more heterogeneous in the early post-war years and have become much more homogeneous. The regionalist parties have always had scores that varied considerably between the cantons. Only at their highest point in the 1970s they reached a territorial homogeneity similar to that of the Liberals and the Socialists. The extreme right parties both started with

35

Politics Beyond the State.book Page 36 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

very heterogeneous results, but these have also become now quite homogeneous (slightly more in Flanders than in Wallonia). This general picture fits well into the story of increasing nationalization of electoral politics. Indeed, the territorial homogeneity of the electoral results is one of the major indicators of it. If regionalization instead of nationalization would be the story to tell, we would expect the results per region to be much more homogeneous than those for the country as a whole, and we would expect the homogenizing trend to be stronger at the regional than at the national level. Our results for Belgium do not confirm the expectations of regionalization. The level of heterogeneity is indeed a bit higher in Belgium than in each of the two regions, but these differences are very small. Of course we need to keep in mind that when we look at Belgium we look at party families and not at parties. If parties would be the unit of analysis, the heterogeneity of Belgium would be obvious.

Electoral swings Another important indicator of the nationalization of electoral politics is the territorial homogeneity of electoral swings. We are looking then for similar electoral changes in all parts of the territory. If a party loses, it should lose everywhere. If it loses heavily, it should lose heavily everywhere. That is indeed an indicator of the degree in which the electorates respond to national events1. For Belgium the analysis of the electoral changes between elections and per party family are very relevant. Indeed, the very fact that the national government is the crucial link between the two party systems makes the electoral swings an important aspect of the legitimacy of government formation. Although cabinets only need to secure a majority of the seats in the national parliament, there is a psychological element that can not be discarded. When a party (or party family) ‘lost’ the election, i.e. secured less votes than the previous time, it is much more difficult to claim the right to (continue to) govern. And since party families have so far always governed together in symmetrical national governments, a common trend makes it at least easier to legitimize the choices made for cabinet formation.

1.

One can argue however that if responses of the electorate are different in different parts of the territory, that does not mean the voters are not responding to national clues (Caramani, 1996).

36

Politics Beyond the State.book Page 37 Wednesday, December 5, 2007 8:41 AM

Electoral Politics in Multi-Level Settings

In the Belgian public debate it is – as we said above – generally accepted that the electoral results in both regions are increasingly different, sending different messages to the political decision-makers. We have already seen that there is a large difference in the results indeed, but that these are not increasing. How about the electoral swings then? The direction of the swings in north and south are indeed interesting to look at. For the three large party families there are 18 elections for which this can be compared. Of these 54 cases, 17 show a swing that is different in north and south. Eight of these are for the Liberals, 6 for the Socialists and 3 for the Christian Democrats. Thirteen of them are between 1968 and 1987. There were only two differing swings before 1968 (Christian democrats and Socialists in 1949), and two after 1987 (Liberals in 1991 and Socialists in 1995). The conclusion needs to be once more that the differences between north and south are not a recent affair. Actually the major differences can be seen during the ‘hottest’ period of ethno-linguistic conflicts. The most recent elections show a much more homogenous picture with respect to electoral swings. Figure 4 Homogeneity of electoral swings per region in Belgium, Flanders and Wallonia

We have also looked at the electoral swings at the level of the cantons. The question is then whether these swings are homogeneous across the two regions. We followed the same logic as for the homogeneity of the results. 37

Politics Beyond the State.book Page 38 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

We computed the coefficient of variation for the swings per canton (standard deviation divided by average) and then a weighted average (size of the parties) per election year. The summary results are shown in figure 4. They tell us that three periods can be distinguished: a period of increasing heterogeneity between 1946 and 1968, a period of increasing homogeneity of the swings per canton between 1968 and 1987, and then again a period of increasing heterogeneity. The values are higher in 2003 than in 1987 for all the parties, except for the right wing extremist Front National and Vlaams Belang. The most striking result for our purpose here is however the similarity between the two regions. The major movements of the homogenization of electoral swings are exactly the same in Flanders and Wallonia. Differences or increasing differences between the regions, or an increasing regional specificity of electoral politics in north and south can therefore not explain the results. There must be something Belgian going on, something national.

The Spanish case In Spain the most obvious indicator of the presence of an important regional level with its own dynamics, is the variety of the party systems between the regions and the differences between regional and national elections. The crucial ingredients of these differences are the regional parties. Of all the Spanish parties, only a few cover the whole territory. Most of the parties – but these are small parties seen from the national level – limit their activities to one or two regions. The institutional setup of Spain also allows for more regional variation. The ‘vertical’ logic, in which the Autonomous Communities each receive their own autonomous statute from the centre, and the differences made in the early days between regions on the fast track and regions on the slower track towards autonomy all allow for quite some variation between the regions. We do not need sophisticated measurements to know that national electoral results in Spain differ between regions, exactly because of the importance of these regional parties. In some regions like Catalunya or the Basque Country they are the major players in the game. For Spain our second question will therefore be more important. If there is indeed a high level of regionalization of national electoral politics, can we see and say that this is increasing? Is there an ever stronger impact of the regional fact on the national elections? 38

Politics Beyond the State.book Page 39 Wednesday, December 5, 2007 8:41 AM

Electoral Politics in Multi-Level Settings

Territorial homogeneity of the results For Spain we will follow the same logic as for Belgium, although we will need – given the different kind of regionalization – measurements and techniques that are slightly different. The first question is the most general: are the electoral outcomes of the parties homogeneously spread over the territory, and more in particular over the Autonomous Communities? To check for this, we obviously need the national election results broken down per Autonomous Community (and not per province). The data used are those published by Caramani (2000) and – for 2000 and 2004 – by the Spanish Ministerio del Interior (http://argos.mir.es/MIR/jsp/resultados). We use again the simple standardized CV to compute the degree of heterogeneity of each party, and then compute a weighted average per election year. In this average, the statewide parties will carry the highest weight, since the regional parties generally limit their activities to only one of the regions. The level of heterogeneity of the system will however still be influenced by the very skewed results of the smaller parties. We will focus on the two major statewide parties: the PSOE and the PP (and the UCD till 1982). Figure 5 Regional heterogeneity in Spain. Evolution per party and weighted average for the country

39

Politics Beyond the State.book Page 40 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Figure 5 shows us that the overall trend is certainly not one of increasing regionalization. The linear trend is slightly going down. The same goes for the trend for the PSOE, where only the elections of 2000 – elections lost to the PP – show a territorial variation that is as high as in 1979. It is especially the PP that has become more evenly spread over the Autonomous Communities. It has remained however – except again for 2000 – more heterogeneous than the PSOE. The conclusion can thus be that Spain displays quite some regional variation in its electoral results, but that this variation is not increasing over time. To the contrary, for the statewide parties the variation tends to decrease. To evaluate the degree in which regions indeed produce electoral results that are different and specific for the regions, we need to use a bottom-up measure. That should tell us to what extent the result in one Autonomous Community differs from the nationwide results. Since these nationwide results also comprise the results in the region to be analyzed, we propose to compare the regional results in each Autonomous Community with the results in the country minus those in the Autonomous Community for which we measure the specificity. For comparing the regional results with the results in the rest of the country, we have also opted for a very simple and straightforward measure. It comes close to the way in which we compared the Belgian regions. For Belgium we computed an index of dissimilarity – the volatility index – by comparing one region to the other. Thus: for Spain we compute for each region and for each election the index of dissimilarity between the regional results and the results in the rest of the country, minus that region. Table 2 The regional specificity of national elections in Spain (volatility index comparing region with rest of the country)

Andalusia Aragon Asturias Baleares Baskenland Canary Islands Cantabria Castilla-La Mancha

40

1977 10.79 16.00 11.86 19.47 40.03 27.10

1979 17.76 15.63 12.82 17.55 50.53 35.25

1982 17.35 13.38 14.45 20.72 54.09 23.91

1986 20.09 13.35 11.59 15.59 54.54 20.61

1989 25.23 13.82 13.23 20.45 59.72 22.65

1993 18.73 19.29 10.77 18.55 49.07 25.74

1996 17.66 16.13 11.18 15.16 46.51 25.24

2000 18.33 15.02 10.54 16.37 38.37 30.02

2004 20.26 13.83 12.17 17.10 45.18 24.33

Avg 18.47 15.16 12.07 17.88 48.67 26.09

15.57 19.43 17.95 12.28 15.47 14.21 13.21 13.25 14.18 15.06 13.91 15.11 13.56 13.93 17.20 15.59 14.08 15.92 14.26 14.84

Politics Beyond the State.book Page 41 Wednesday, December 5, 2007 8:41 AM

Electoral Politics in Multi-Level Settings

Castilla y Leon Catalunya Extramadura Galicia La Rioja Madrid Murcia Navarra Valencia Ceuta Melilla

1977 1979 1982 1986 1989 1993 1996 2000 2004 Avg 22.29 21.39 20.91 19.80 21.14 15.31 15.80 13.79 13.28 18.19 38.11 17.57 28.59 18.58 12.29 12.75 37.87 11.89 16.08 24.80

30.27 18.52 31.04 21.31 12.58 13.29 35.56 11.53 25.34 25.34

27.30 11.14 27.54 19.81 17.19 15.05 24.18 34.83 13.63 16.84

33.33 12.99 23.44 14.37 14.31 13.34 19.56 10.50 13.77 22.54

36.22 16.39 23.46 16.04 20.93 13.48 24.52 11.57 21.88 30.23

37.14 14.00 19.48 14.17 16.59 12.99 14.63 12.04 20.11 20.76

35.93 12.90 23.26 13.57 19.65 11.73 18.47 11.66 22.08 18.28

35.60 14.47 29.59 11.36 14.34 15.55 16.50 12.11 30.76 27.57

39.79 13.71 21.28 13.72 13.61 20.89 17.97 15.69 21.35 16.70

34.86 14.63 25.30 15.88 15.72 14.34 23.25 14.65 20.56 22.56

The results of this are in table 2, containing of course no real surprise. We see indeed that some of the Autonomous Communities have electoral results that are very specific, while others have results that are more or less the same as in the rest of the country. The most specific regions are the Basque Country, Catalunya, The Canary Islands, Galicia and Navarra (and also the two small entities Ceuta and Melilla). The question here is however whether these specificities increase. Indeed, if there is an increasing regionalization of electoral politics in Spain, we should see these regions producing more and more typically regional results. In figure 6 the trend for a number of crucial regions is shown. And the conclusion is: there is no trend towards a larger regional specificity. Only for Catalunya there is a slight increase, and a stronger one in Andalucia. There is also a sharp decrease of the regional specificity in Castilla y Leon.

41

Politics Beyond the State.book Page 42 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Figure 6 Regional specificity of some Autonomous Communities (1977-2004) – Dissimilarity-index comparing results in region with rest of the country

We have finally also looked at the electoral swings in Spain. We want to check whether the parties move in the same direction between elections in all the Autonomous Communities. We computed again this index of change: votes for party at t1 / votes for party at t-1. We already saw that both PSOE and (increasingly) PP have scores that look very homogeneous across the Autonomous Communities. And the same goes for the electoral swings. Both parties indeed tend to lose or win in a fairly homogeneous way. Figure 7 shows the evolution for PSOE and PP. In the early days of the new Spanish democracy the two statewide parties displayed swings that did vary between the regions, however not spectacularly. After a few 42

Politics Beyond the State.book Page 43 Wednesday, December 5, 2007 8:41 AM

Electoral Politics in Multi-Level Settings

elections the homogeneity of the swings stabilized at a low level, but increased again in the elections of 2000 and 2004. The PSOE was more homogeneous in the beginning, while the PP – that stabilized and institutionalized later than the PSOE – remained at a higher level of heterogeneity. In 2000 and 2004 however the heterogeneity of the swings was higher for PSOE. But this all happens at levels than can be labelled low. Figure 7

If we look at the direction of the swings, the picture is one of fairly large homogeneity. The last three elections can illustrate nicely what is going on. In 1996 the PSOE lost slightly (index 0.97) while the PP won (index 1.13). The PP won everywhere, except in Isles Baleares (0.97), while the PSOE lost in most regions, but not in Aragon, Asturias (both 1.01), Catalunya (1.13) and Valencia (1.00). In 2000 there is a bit more homogeneity. In 2000 the PP won again (index 1.13) while the PSOE lost more heavily (0.85). The PP only lost votes in Aragon (0.97) and in the two small units Ceuta (0.88) and Melilla (0.97). The PSOE lost everywhere, except in Madrid (1.04). And 2004 is – as far as the direction of the swings is concerned – very homogeneous. The PSOE won clearly (1.33) and only lost in Extramadura (0.95). The PP lost everywhere, except in Extramadura (1.08), Ceuta (1.24) and Melilla (1.10). The general picture is one of parties moving in the same direction in a great majority of the regions. The deviant results of Ceuta and Melilla are of course not very important.

Conclusion This chapter has presented data on only two countries. It is a first attempt to tease out the notion of regionalization and to see how it can adequately 43

Politics Beyond the State.book Page 44 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

be measured. We did choose Spain and Belgium however because we thought these are countries where evidence of regionalization would be available. The results are not very convincing for those who believe that the presence of (strong and important) regions has clear and increasing effects on the national electoral politics. For Belgium we must of course acknowledge the very absence of national parties, which is a very strong indicator of a purely regional logic. Yet looking at the party families has revealed a number of slightly surprising results. In the first place we have seen that the party landscape in north and south is very different, but that this difference is not becoming more important, but rather of a different kind, more complex. We have also seen that the electoral swings in north and south have been diverging during a specific period: between the late sixties and the early eighties. Before and after this period the general logic is one of parties of the same families moving together up and down in north and south. We have also found a pattern in the homogeneity of the swings: first more homogenous and then again more heterogeneous. This pattern is exactly the same for both regions. For Spain we have found the expected presence of a fairly strong regional variety in the national results. But this variety is not increasing. It has – like in Belgium – always been there and remains rather stable. We have also seen that for the two major statewide parties – PSOE and PP – the electoral results tend to become more homogeneously spread over the Automous Communities. The homogeneity is higher for the PSOE than for the PP, but for both the trend is one towards increasing national homogeneity. Electoral swings have been for both parties high at the very first elections, and gradually going down. The 2000 and the 2004 elections show more heterogeneous electoral swings. Yet the direction of the swing is basically the same in all the Autonomous Communities. And that means we have a regionalized country with regional specificities present in national elections, but with a trend moving rather in the direction of increasing nationalization.

44

Politics Beyond the State.book Page 45 Wednesday, December 5, 2007 8:41 AM

Governing Coalitions in Multi-level Settings: State-wide Parties and the Sub-national Arena in Spain Irina Stefuriuc

The study of governing coalitions is by now one of the classics in political science. There is a solid theoretical body for studying various aspects of coalition politics and national governments have been the subject of extensive empirical research. The appeal of studying local and regional coalitions, if only for the simple fact that they can provide the researcher with an increased number of cases for testing existing theories and at the same time with the possibility to control for a series of systemic factors, has been early pointed out in the coalition literature (Dodd 1976, Mellors and Breary 1986, Laver 1989). However, so far only a handful of studies have proceeded to research sub-national coalitions.1 Nevertheless, the need for researching sub-national coalitions is becoming obvious for empirical reasons too. In the last decades of the 20th century, several Western European countries engaged in processes of political decentralization, devolving governing and legislative jurisdictions to subnational elected assemblies and regional governments. The territorialization of electoral competition in decentralized states gives rise to a series of cross-level differences in what regards the electoral performance, the organization and the strategies of political parties. Elections often result in asymmetrical governing majorities (Hopkin 2003; Jeffery and Hough 2003). This requires parties to adapt to a dual logic, as the governing and the opposition experiences might overlap in time across levels, and so might the governing-alone and the governing-in-coalition experiences. This chapter builds on the existing body of literature to elaborate a theoretical framework for studying coalitions in multi-level settings and tests several of its predictions using data from the Spanish regions, focusing

1.

See for example Reniu 2005; Bäck 2004; Mershon and Hamann 2000; Downs 1998; Colomer and Martínez 1995.

45

Politics Beyond the State.book Page 46 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

on the patterns of governing that Spanish state-wide parties engage in at the sub-national level. Generally, comparative coalition literature excludes countries like Spain and Great Britain for the simple reason that national governments always consist of a single party, regardless of this party’s majority status (Müller and Strøm 2000). At the regional governing tier however, while single party governments are still the most frequent occurrence, coalitions have formed in Spain in about a third of all cases since 1982. Just as well, 39% of all regional governments held minority status (see Table 1). Table 1 Government size and majority status: 1982-2005

One party Government Two parties size More than two parties Total

Majority status Minority Majority 39 59 9 25 8 4 57 88

Total 97 34 11 145

* See section 2 for an account of all data sources.

This chapter explores some basic aspects of regional coalition formation in Spain, comparing the governing strategies that the two main Spanish statewide parties, the Partido Socialista Obrero Español (PSOE) and the Partido Popular (PP), employ at the regional level. Besides the fact that the Spanish national government has always been formed by either the PSOE or the PP, these two state-wide parties2 have a strong governing presence at the subnational level. As reported in Table 2, only 21 sub-national cabinets, that is 13 percent of all cabinets, did not include any state-wide party. The two main state-wide parties governed in 85% if all the single-party governments that formed, and also in 85% of all the coalitions that were formed at the sub-national level in Spain.

2.

For a definition of state-wide parties see section 4.

46

Politics Beyond the State.book Page 47 Wednesday, December 5, 2007 8:41 AM

Governing Coalitions in Multi-level Settings

Table 2 Presence of state-wide parties in government

No SWP in government At least one SWP in government Total

Single party government Coalition 7 14 (14.89%) (14.29%) 40 84 (85.11%) (85.71%) 47 98

Total 21 (14.4%) 124 (85.5%) 145

Note: entries in parentheses percentages of the total number of cases represented in the column category.

The chapter is organized as follows: Section 1 sketches a theoretical framework and briefly reviews the relevant literature on the topic. Section 2 presents methodological issues pertaining to concepts, data and measurement. After outlining the specific hypotheses I set out to test on Spanish data, Section 3 discusses the empirical findings. The final section concludes and proposes an agenda for further research.

Multi-level systems and governing coalitions: fundamental theoretical assumptions The theories of government formation and general coalition behavior have been developed to apply to national governments. All of them should, insofar as they are based on specific assumptions regarding the goals of the main actors involved in government formation and the life of governments, be testable against data on regional governments (Laver 1989; Mellors and Breary 1986). However, in order to be sensibly applicable to multi-level systems, any such theory must fulfill several basic conditions in what its fundamental assumptions are concerned. These requirements will be presented in what follows. This attempt to sketch a “theory of theory” does not consist in adjusting existing general coalition theories to fit particular cases. Rather on the contrary, the framework outlined in this section is selecting from the existing body of coalition theory those propositions which could be a priori applicable to both single- and to multi-level systems.

47

Politics Beyond the State.book Page 48 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Multiple arenas, multiple party goals The first fundamental assumption of any theory dealing with coalitionmaking in multi-level system is that political actors might simultaneously pursue multiple different goals at different levels (Downs 1998). These goals will depend on the context of competition, and on the individual stakes involved in bargaining outcomes. Take, for instance, the case of a moderate regionalist party with relatively strong electoral support in its region. If the degree of regional governing autonomy is substantial, or if the party perceives participation in regional government as a means to enhance autonomy for the region, then this party’s motivation to enter government in what it perceives to be the most efficient governing formula will be of highest intensity. In coalition parlance, the party’s dominant goal in regional politics will be office. Moving to the next level, due to its strong and territorially concentrated support, this party might find itself in a pivotal position in the national legislature. It might therefore be offered portfolios in the national government. However, the party might find participation in national government too costly, as it will have to engage in shared government responsibility and thus it might have to dilute its regionalist appeal and risk to loose voters at the next elections. Or the organizational costs of participating at two levels of government might be to high for the size of its organization. This party would therefore find it more beneficial to take advantage of its pivotal position and condition its parliamentary support to a minority national government on the implementation of a few crucial policies related to territorial autonomy. Its refusal to participate in national government would thus be explained by motivations pertaining to policy and votes. Thus, the party would pursue simultaneously and with equal intensity both office, at one level, and policy and votes, at the other level. The example above is also useful to illustrate three other fundamental prerequisites of a coalition theory applicable to decentralized systems. One is that such a theory should assume that bargaining (formation) games are iterative (Franklin and Mackie 1983). The second is that these games are nested (Tsebelis 1990), or interconnected between levels (Downs 1998). Finally, the third is that political actors are future-oriented. Let us take each one separately.

Formation games are iterative For quite a long time coalition research treated each government formation case as a snap-shot scenario, meaning that the only actor attributes 48

Politics Beyond the State.book Page 49 Wednesday, December 5, 2007 8:41 AM

Governing Coalitions in Multi-level Settings

considered relevant were the ones they had at that particular moment in time (i.e. usually ideological outlook and parliamentary weight). However, previous coalition experience is part of the minimal information baggage with which actors enter the bargaining game: “the formation of a governing coalition should be viewed as part of a historical sequence of events in which past experience plays an important role” (Franklin and Mackie 1983: 276). The model proposed by Franklin and Mackie is based on four key variables: size and ideology (the classical ones) as well as familiarity, that is, the information from past experiences, and inertia, which would, by virtue of the “immediate past experience” predict the reformation of the same coalition.3 Their model is among the first ones to explain why many real-world coalitions are neither small-sized nor ideologically compact, attributes which summarize the predictions of the classical size and policy schools of coalition theory.4

Coalition formation games are nested Tsebelis (1990) introduced the concept of “nested games” to refer to games played by the same actors in different arenas. Basically nested games are a network of games that are interconnected. Coalition formation in multilevel systems need to be modeled in this terms. Parties need to coordinate their choices in the two arenas (Downs 1998) – be it because they wish to present a coherent ideological image by coalescing with the same range of parties at both levels, or on the contrary, because they wish to maximize gains in one arena rather than the other.

Actors are future-oriented If actors look in the past to assess the viability of specific coalition formulae (see Assumption 1.2), they should also be expected to assess the possible long-term future consequences of the various coalition formulae they are considering. Thus, certain coalitions will not form, even if momentarily

3. 4.

See also Warwick 1994. The minimal-winning coalition is at the core of the size (or office) school (Von Neumann and Morgenstern 1952; Riker 1962). The policy school predicts that connected (i.e. ideologically cohesive) coalitions are the ones most likely to form (de Swaan 1973; Axelrod 1970).

49

Politics Beyond the State.book Page 50 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

attractive, because their perceived future costs, usually in terms of votes and policy, are simply too high (Mershon 2002; Strøm 1990).5

The policy space is bi-dimensional A fifth pre-requisite for a good explanatory theory of government formation multi-level systems is that the policy space is modeled as having at least two distinct dimensions. The territorialization of political competition implies a (re)awakening of the center-periphery cleavage (Lipset and Rokkan 1967). Classical formal coalition theories tended to model the ideological space in which parties compete and bargain as onedimensional.6 One of the most celebrated predictions of this model construction is the median-legislator theory, according to which the party containing the median legislator should always be included in the governing coalition, as its policy position is unbeatable by any parliamentary majority. However, it is doubtful that even in centralized systems the space of political competition can be usefully reduced to a single dimension. An early bi-dimensional account is that of Luebbert (1986), who posited that parties prefer forming coalitions with partners whose policy preferences are not proximal, but orthogonal to their own. Schofield (1993), Schofield and Sened (2005), Laver and Schofield (1990) and Laver and Shepsle (1996) developed what have remained until now the most elaborated models of government formation in bi-dimensional spaces.7 The thrust of their arguments is that where it exists, the party occupying the core of the policy space8 will always be included in the governing coalition.9

5.

6. 7.

8.

9.

Colomer and Martínez (1995) base their argument on coalition formation in multiparliamentary systems on these two behavioral assumptions: parties engage in iterative coalition bargaining games, in which past and perceived future experience is an important determinant of strategic choices. This is mainly because multi-dimensional models are extremely difficult to formalize. Besides rendering his formal model more realistic by building in two policy dimensions, Schofield’s merits lie also in the fact that his model can incorporate future-orientations of actors and is linked to a general model of party competition. The core is basically the intersection of the two medians in two-dimensional spaces. Unlike the simple median, the core is rarely occupied by an existing party (see Schofield and Sened 2005; Warwick 1994). This assumption of the two-dimensional character of the policy space is inherently built in nearly all accounts of sub-national coalition politics. See for example Reniu 2005; Mershon and Hamman 2000; Colomer and Martínez 1995.

50

Politics Beyond the State.book Page 51 Wednesday, December 5, 2007 8:41 AM

Governing Coalitions in Multi-level Settings

Parties are not unitary actors A sixth demand for our theory is that it relaxes the assumption that parties are unitary actors. The unitary-actor assumption lays at the core of most theoretical models of coalition formation and it is only very recently that scholarship has attempted to relax it for better explaining real-world coalition behavior. The strongest argument in favor of the unitary-actor assumption was made by Laver and Shepsle, who argued on the basis of empirical evidence from Western Europe that "parties both enter and leave cabinet coalitions as unified blocks" (Laver and Shepsle 1996: 25). But internal party politics, and most notably factionalism, have also been found to matter (Laver and Gianetti 2004; Mulé 2001; McGillivray 1997; Pridham 1986). And if the unitary actor assumption is questionable in centralized systems, in countries with de facto decentralization, political parties are subject to substantial centrifugal forces. In some cases, the territorial arrangement of the state is mirrored by the organization of political parties. In others, their response to the centrifugal tendencies induced by territorial decentralization is a tightening of organizational centralization. These are two different strategies that target the same problem: how to contain or manage territorial factionalism. The first strategy proposes accommodation by quasi-federalizing the party organization and devolving powers to the regional/federate party units. The second one proposes containment and attempts to keep regional branches under the strict control of the center (Hopkin and van Biezen 2005; Deschouwer 2003; Detterbeck and Renzch 2003; Hopkin 2003). Of course, both organizational strategies are attempted with more complex aims than simply counteracting factionalism. However, they are also indicating that territorial division is a problem that is viewed seriously by political parties. Differences in party organization should be expected to matter for how parties design coalition strategies too.10

Institutions are constraining actors’ strategic choices Finally, any good model should take into account the effects of institutions. Strategic action always takes place in certain parametric conditions, and it has already become commonplace that certain electoral institutions as well as institutions shaping the executive-legislative linkages matter for 10. Two excellent accounts of how internal party politics (i.e. the non-unitary character of political actors) plays a role in the formation of sub-national politics are those of Bäck (2004), who looks at Swedish local coalitions and Downs (1998) who compares sub-national coalitions in Belgium, Germany and France.

51

Politics Beyond the State.book Page 52 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

government formation and survival (Martin and Stevenson 2001; Mershon 2000; Strøm, Budge and Laver 1994; Warwick 1994; Bergman 1993; Strøm 1990). In multi-level systems institutions and their effects can vary across levels. National and sub-national elections might be regulated by the same proportionality rules for example, but variations in district size across levels can result in substantial deviations from proportional outcomes and thus in very different parliamentary party constellations and viable coalition alternatives across levels.11 Likewise, differences in parliamentary rules, government investiture procedures and confidence vote regulations are important for government formation in both unitary and multi-level settings. To summarize, this section sketched the main assumptions that theory of coalition formation applicable to decentralized contexts should rely on: 1. Parties may simultaneously follow different goals (votes, office, policy) in different arenas (state-level vs. regional); 2. Coalition formation games are iterative; 3. Coalition formation games are inter-connected between levels; 4. Actors are not myopic (long-term future consequences are assessed); 5. The legislative policy space is at least two-dimensional; 6. Parties may be behave like non-unitary actors; 7. Institutions are constraining actors' strategic behavior. The next section presents a several empirical hypotheses derived from these assumptions. Before getting there, in what follows section 2 describes and clarifies methodological problems pertaining to concepts, operationalization, data and measurement.

Concepts, data and measures The first issue that must be clarified before proceeding to the analysis of Spanish data regards the units of analysis. In this chapter, the units of analysis are individual regional cabinets. In Spanish regional politics, cabinets are the so-called consejos. They are led by a presidente who must pass a vote of investiture in the parliament. The rules that decide when a new cabinet begins are those usually employed in comparative research of national governments. Thus, a new cabinet is counted every time (1) a new prime-minister (i.e. presidente) is invested; (2) following elections; (3) a change is recorded in the party composition of the cabinet (Müller and 11. For applications to sub-national coalitions see in particular Mershon and Hamman (2000) and Downs (1998).

52

Politics Beyond the State.book Page 53 Wednesday, December 5, 2007 8:41 AM

Governing Coalitions in Multi-level Settings

Strøm 2000) and (4) a change has occurred in the majority status of the government and this latter has passed from a minority to a majority government or the other way around (Reniu 2005). This analysis is focused on the governing strategies of state-wide parties. A state-wide party is defined here as a political party which is contesting both regional and national elections in all or nearly all regions of the country, largely under the same electoral banner; if regional or national elections are fought in certain regions by a regional organizational division of a party which competes under a different banner, but this organizational division is not competing against any other organizational division of the same party and it does not form a separate parliamentary group in those parliamentary contexts in which it co-exists with the national division, it will be counted as the same “state-wide party.” This last specification is particularly important for the Spanish case. Researchers still argue over the “correct” classification of cases like that of the Partit dels Socialistes de Catalunya (PSC). The PSC is formally part of the federal organization of the Partido Obrero Socialista Español (PSOE). However, it enjoys a special status within the party federation, benefiting from extended autonomy in what regards electoral strategies and candidate selection (Colomé 2004). National elections are fought in Catalonia under the PSC-PSOE banner, while autonomous elections are fought solely by the PSC. For certain analytical purposes thereof (i.e. party organization, campaign strategy, etc.), it is perfectly adequate to consider PSC as a separate party unit. In reference to coalition-related aspects at the two levels of government, it makes more sense though to take the PSC and the PSOE as the same single party organization. This is justified by the fact that in effective parliamentary politics, the two party levels are practically indistinguishable, forming the same party group in the national parliament. One must also add to the argument the fact that Spain is a country where parliamentary indiscipline is quasi-inexistent: once in the same party group, regional MPs follow the party line in their legislative voting behavior (Fernández Riveira 2003; Tomás Mallén 2002; Sanchez de Dios 1999). By contrast, a “non-state-wide party” (NSWP) is a party that contest either regional or national elections, or both, in a limited territory of the country (one or several, but never all regions) and which retains a separate parliamentary group organization in sub-national parliaments. If two or more non-state-wide organizations run elections on a common list and following elections form one single parliamentary group, then, for the same justifications as above, they are considered a single NSWP.

53

Politics Beyond the State.book Page 54 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Note that these definitions apply two criteria of classification, one pertaining to electoral politics (votes) and another one to parliamentary representation (seats). Just as well, they take into account the issues of territorial pervasiveness (number of regions) and territorial presence, or spread (type of elections contested) (Deschouwer, 2006). Regarding dimensionality, I follow the method suggested by Mershon and Hamman (2000). They rely on the assumption that the left-right dimension is automatically present in all regional legislatures, and compute a threshold of two-dimensionality that indicates the existence of a second ideological division, defined by the opposition between regional autonomy and state centralization (Heller 2002). The threshold is computed and employed as follows: if non-state-wide parties (regionalist parties thereof) hold above 3% of the total number of parliamentary seats, the system is considered two-dimensional. If regionalist parties fail to reach this threshold of representation, that legislature is considered one-dimensional. The congruence of government composition is a variable that measures whether or not regional cabinets included the same state-wide party that was governing at the national-level at the moment of their formation. The concept of cabinet, or coalition, congruence was originally introduced by Roberts (1989). In Robert’s version, the variable was measuring whether strict correspondence of cabinet membership at the federal and at the subfederal level occurred. The original concept was modified here because, as explained in the introduction, Spanish parties never form coalitions at the state-level. Finally, in what regards the congruence of electoral results across levels, an index of dissimilarity was computed for each regional elections, comparing its results at the regional level of aggregation with those of the previous most recent national elections (at the same level of aggregation). The reference point are thus national elections, namely those of 1979 (for the few cases in which autonomous elections were held already in 1980 – Catalonia – or 1982 – the Basque Country12), 1982 (compared with the first autonomous elections of 1983 held in most ordinary status regions), 1986, 1989, 1993, 1996, 2000, and for a few cases, 200313.

12. For the 1982 regional elections in Galicia a different computation applies. See Appendix for details. 13. See Apendix for a detailed discussion of the general rule and the exceptions in calculating the index.

54

Politics Beyond the State.book Page 55 Wednesday, December 5, 2007 8:41 AM

Governing Coalitions in Multi-level Settings

The exact formula used for computing the index of dissimilarity is:

The index was computed for each regional-national election pair and its values assigned to every government that was formed following the regional election the index was computed for and before the next regional election held. (For example, the 89-91 index in Aragon was assigned to all the governments that were formed in Aragon in the interval between the 1991 and the 1993 regional elections). Ceteris paribus, the index is a straightforward measure of “how divergent electoral outcomes at different levels are” (Jeffery and Hough 2003: 209). Calculated this way, it basically computes the percentage of voters that would have to vote differently in regional elections so that the results of these latter should be identical with the results of the previous most recent national election in that region. However, one should bear in mind that factors operating at the national level as well as the time lapse between the two different-level elections are likely to be responsible for at least part of the value the index of dissimilarity takes (Pallarés and Keating 2003). That is to say that indices of volatility computed for same-level elections (national and regional) must be also looked at. This section concludes with giving an account of the sources that were used to build the dataset. Data on cabinet composition and parliamentary weight was taken from Reniu (2004).14 Electoral data was taken from official statistics posted on www.eleweb.net and http://argos.mir.es/MIR/ jsp/resultados/index.htm. Data regarding parliamentary institutions (size of parliaments, regulation of party groups, investiture and no-confidence rules, etc.) as well as data on parliamentary party groups was retrieved from the official websites of regional parliaments. Data on electoral laws (district magnitudes, thresholds of representation and number of electoral districts in the region) were retrieved from the website of the Spanish Ministry of the Interior, http://www.mir.es/sites/mir/medium/pciudad/elecciones/

14. Several differences occur with respect to the original data posted by Josep ReniuVillamala on http://www.ub.edu/grepa/ and they are due to calculation corrections. These corrections have been cross-checked with Josep Reniu (e-mail communications, June-July 2005).

55

Politics Beyond the State.book Page 56 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

normativa/comunidadesautonomas.html. All other variables employed were computed by the author departing from these sources.

Hypotheses, findings and discussion Starting from the assumptions that have been elaborated in the first section of the paper, several specific propositions can be derived and tested: 1. Parties pursue different goals in different arenas We assumed that political actors pursue different goals in different arenas. This leads to the expectation that many of the governments formed at the subnational level will be incongruent both in what concerns their composition and in what concerns their majority status. 2. Coalition formation is an iterative game If bargaining is an iterative game, and parties rely on familiarity and inertia when forming coalitions, particular patterns of coalition composition are likely to occur more frequently than others. In other words certain coalition formulae are likely to be formed repeatedly. 3. Coalition formation is inter-connected between levels If bargaining is an iterative game and the information baggage of parties extends to experience acquired in both the regional and the national arena, pay-offs for support at one level are likely to be observable at other level. 4. The policy space is bi-dimensional If the policy space has two or more relevant dimensions, we can expect that the party occupying the core of the policy space will always be included in the coalition formula (Mershon and Hamann 2000) or form the government alone even in minority situations. These two hypotheses will, nevertheless, not be tested here as the current existing data about party ideological positioning in Spain is unsatisfactory to this purpose.15

15. In order to test these two hypotheses, it is necessary to have data about the ideological positioning of political parties at different election times. Ideally, this data should be generated either from party manifestoes, or from expert surveys (or from both), for each level of party competition separately. It will not necessarily always be the case, but the territorialization of party competition also means that parties emphasize different policy positions at different levels. This specification is important. In lack of better resources, existing research relies on data generated at one level, i.e. the national one, for studying coalitions at the other level, i.e. the regional one.

56

Politics Beyond the State.book Page 57 Wednesday, December 5, 2007 8:41 AM

Governing Coalitions in Multi-level Settings

5. Parties are non-unitary actors In federal and quasi-federal arrangements, regional party branches enjoy various degrees of autonomy. We can expect that the degree of autonomy of regional branches as decision-makers in coalition politics will be reflected in the coalition choice. Thus, in those regions where autonomy is high some variation from the general pattern of coalition formation will occur. However, this hypothesis will not be tested in this chapter as sufficient systematic data about branch organizational autonomy in the Spanish regions is not yet available. 6. Institutions are constraining actors’ strategic behavior Finally, the institutions that are likely to impact on government attributes fall in three categories: (i) institutions specific to multi-level systems (autonomy status, electoral timing, etc.); (ii) electoral institutions (district size, threshold of representation); (iii) legislative-executive relations (vote of investiture, vote of no confidence, and their respective decision points). I hypothesize that: Incongruence is more likely to be present in those regions whose institutions are “highly regionalized”. By “highly regionalized” institutions I mean special autonomy status, own electoral calendar, and so on. This expectation follows from the fact that special regions also foster particular party system configurations. The higher the electoral threshold and the smaller the district size, the more likely it is for single-party majorities to form (Mershon and Hamann 2000). This is one of the classical expectations of the effects of electoral system characteristics on government formation. It is based on the logic that high electoral thresholds produce a favorable bias towards large parties, which are thus more likely to secure majority representation in the parliament on their own. A small district size acts in the same direction. Minority governments are less likely to form in legislative settings characterized by investiture requirements (especially when the decision rule for the vote is majority rather than simple plurality) and by the requirement of constructive no-confidence for government termination16 (Mershon and Hamann 2000; Budge, Strøm and Laver 1994; Bergman 1993). 7. Other variables: electoral results and party system attributes The formation of government coalitions is primarily conditioned by electoral results and party system characteristics. The higher the level of 16. This hypothesis will not be tested here, as this institutional characteristics do not vary across regions in Spain. Cases from more countries should be included in order to provide a meaningful test.

57

Politics Beyond the State.book Page 58 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

regional electoral volatility and cross-level electoral dissimilarity, the more likely it is for incongruent governments to emerge. This is because in highly volatile regional settings which are furthermore also characterized by strong cross-level dissimilarity in voting patterns, different parliamentary party constellations are expected to emerge. In what follows we will take each of these hypotheses in turn, presenting and discussing empirical evidence.

Party goals and congruence across levels To explore government congruence in party composition across levels, I created a dichotomous variable that takes the value of “1” if the regional government contains the same state-wide party as the national government of the country at the time of regional government formation, regardless of the formula in which this party is present (alone or in various coalitions). As expected, the number of incongruent governments is rather high (see Table 3), with more than fifty percent of all governments formed being incongruent (i.e. governments not containing the party governing at the state level at the moment of their formation). Table 3 Government composition congruence across levels Congruence Incongruent Congruent Total

No. of governments 74 71 145

% 51.0 49.0 100

Of course, one could suspect that this proportion is artificially inflated by the government of the “fast-track” regions in Spain. Table 4 shows that even when excluding from the analysis the cases from Andalusia, Catalonia, Galicia, the Basque Country17, as well as those from the Canary Islands18, the proportion of incongruent governments still reaches above 40%.

17. The four special status regions have incongruent governments in a proportion of 70%. 18. The Canary Islands have a highly regionalized party system (Pallarés and Keating 2003).

58

Politics Beyond the State.book Page 59 Wednesday, December 5, 2007 8:41 AM

Governing Coalitions in Multi-level Settings

Table 4 Government composition congruence across levels (selected cases only)*

Congruence

incongruent congruent N

No. of governments 40 55 95

% 42.1 57.9 100

* excluding Catalonia, Galicia, the Basque Country, Andalusia, and the Canary Islands.

But how much of this can be explained by party strategic behavior? Patterns of congruence are varying quite substantially when we look at the two major state-wide parties separately. As shown in Table 5, the PSOE was part of many more regional governments that were congruent with the state-level one than was the PP. Table 5 Government congruence by party

Congruence Incongruence N

PSOE in regional government 76.1% 23.9% 63

PP in regional government 40.6% 59.3% 59

It also clearly appears that having PSOE included in the regional government is a strong determinant of congruence. Table 6 below shows the results of two logistic regression models. Both include structural variables that could theoretically affect governmental composition congruence across levels: the dissimilarity index, the regional volatility index as well as the number of dimensions of competition in the regional parliament. Besides these, the first model includes a dichotomous variable that takes the value of “1” when PSOE is in government at the regional level and “0” otherwise. The second model includes the same variable for PP presence in regional government.

59

Politics Beyond the State.book Page 60 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Table 6 Determinants of congruence

PSOE in regional government PP in regional government Dissimilarity index Regional volatility index Dimensionality Region status Constant N Nagelkerke R-square

Dependent variable = Congruence Model 1 = PSOE Model 2 = PP B (SE) Odds ratio B (SE) Odds ratio 1.550*** 4.709 (0.435) -0.352 0.703 (0.424) -0.015 0.985 -0.021 0.979 (0.462) (0.019) -0.045* 0.956 -0.044* 0.957 (0.030) (0.028) -0.556 0.574 -0.665 0.515 (0.549) (0.522) -0.705 0.494 -0.668 0.513 (0.517) (0.507) 1.358 3.888 2.438 11.453 (0.856) (0.812) 122 122 0.315 0.210

Significance levels: ***0.01, ** 0.05, * 0.1 Notes: B is the unstandardized regression coefficient. Standard errors shown in parentheses. For variable codings see Appendix.

The results are rather interesting. While having a PP government does not have any significant effect on congruence, a PSOE government increases the odds of congruence significantly: the odds of having a congruent situation are 4.7 times higher if the regional government includes the PSOE than if it does not include it. Electoral arithmetic also plays its due role, but unexpectedly, dissimilarity – which is also a measure of regionalization of elections – is insignificant in both models. Instead, in systems with twodimensions and high regional volatility, the chances of congruent governments are reduced, as one would intuitively expect. It also appears that the PSOE is more eager to participate in governments at both levels simultaneously, even if this means forming minority governments: 33.3% of all the regional governments including PSOE had minority status, as opposed to the 25.5% in the case of PP. The picture looks similar if we consider only the single-party governments formed by each party: 25% of the PSOE regional governments had minority status, 60

Politics Beyond the State.book Page 61 Wednesday, December 5, 2007 8:41 AM

Governing Coalitions in Multi-level Settings

while only 18% of the PP ones did so. Of course, these conclusions are only tentative, as no real strategy component was introduced at this level of research. Further qualitative analysis must check whether this is indeed a matter of party strategy and provide answers as to why these differences occur.

Coalition formation as an iterative game The first hypothesis related to the second theoretical assumption states that “if bargaining is an iterative game, and parties rely on familiarity and inertia, a particular pattern of coalitions is likely to occur more frequently than others.”. To be more specific, one can expect that the most frequent coalition pattern that state-wide parties would seek to form at the regional level would be a formula including themselves and one or several nonstate-wide parties.19 There are good reasons to expect this in the Spanish case. The main such reason is that except for the Izquierda Unida (IU) and, for a short time span in the early '80s, the various successors of the Unión de Centro Democrático (UCD), the only two parties that are effectively competing in regional level elections are the two big opponents: the PSOE and the PP. UCD's successors practically disappeared from the regional parliamentary arena very early, therefore a general pattern of coalition formation including them and covering the whole period up to 2005 would not make sense. For basic ideological reasons, the IU could only coalesce with the PSOE, but due to a long-term personal animosity between the two parties' respective leaders in the '80s, such a coalition formula is very unlikely for most of the time span covered by this analysis (Reniu 2001). And indeed, of the 49 coalition governments that were formed at the regional level in Spain between 1980-2005, 77.5% were formed by one or more state-wide parties and one or more non-state-wide parties. There were two instances in which two SWPs coalesced formally without including a third or fourth NSWP: the current Asturian PSOE-IU coalition and the 1989-1991 PP-CDS coalition in Castilla y León. The remaining coalitions were formed by NSWPs only. But is there any difference at the level of individual parties in what regards the “favorite” formula of governing? Table 7 below suggests that both parties apply the same logic in this respect.

19. Data on government supporting parties in cases of single-party minority will also need to be included in further analysis for a more solid test of this hypothesis.

61

Politics Beyond the State.book Page 62 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Table 7 Coalitions vs. single-party governments: preference by party*

Single-party Coalition (all) Coalition with NSWPs Total

PSOE in regional government 69.8% (44) 30.1% (19) 29.5% (18) 63

PP in regional government 67.8% (40) 32.2% (19) 27.1% (16) 59

* absolute numbers shown in parentheses.

Also, two logistic regression model were run having as dependent a dummy expressing whether the government was a coalition between one or more state-wide parties and one or more non-statewide parties, and including as explanatory variables PSOEin and PPin respectively besides a number of control variables (see Table 8). The only explanatory variable that appeared statistically significant was the PSOEin, which appears to increase the odds of encountering this particular coalition pattern. Curiously so, except for dimensionality which runs in the expected positive direction, none of the control variables held significant coefficients, and therefore the observed pattern does not appear to be due to structural conditions such as region status, dissimilarity or regional volatility.20

20. Very similar conclusions can be reached when running the regression with “single party government” as a dependent variable (results not shown).

62

Politics Beyond the State.book Page 63 Wednesday, December 5, 2007 8:41 AM

Governing Coalitions in Multi-level Settings

Table 8 Party strategies as determinants of coalition formula

PSOE in regional government PP in regional government Dimensionality (continuous) Dissimilarity index Regional volatility index Region status Constant N Nagelkerke R-square

Dependent variable = SWP+NSWP coalition Model 1 = PSOE Model 2 = PP B Odds ratio B Odds ratio 1.324* 3.759 (0.488) 0.368 1.444 (0.464) 0.041** 1.042 0.034** 1.035 (0.014) (0.014) 0.006 1.007 0.002 1.002 (0.017) (0.017) 0.028 1.022 0.019 1.019 (0.026) (0.025) -0.718 0.398 -0.458 0.632 (0.651) (0.651) -2.728*** 0.084 -2.019*** 0.133 (0.632) (0.556) 123 123 0.206 0.151

Significance levels: ***0.01, ** 0.05, * 0.1 Notes: B is the unstandardized regression coefficient. Standard errors shown in parentheses. For variable codings see Appendix.

One can therefore reach two conclusions. One is that, as intuitively expected, those legislative contexts where the autonomy vs. center dimension is active, a coalition of state-wide and non-state-wide parties is more likely to form. The second is that a such a coalition is substantially more likely to occur having the PSOE, rather than the PP as state-wide partner.

Coalition formation is inter-connected between levels The hypothesis related to the assumption that government formation is a game that is inter-connected between levels states that pay-offs for support at one level must be observable at other levels. Obviously, pay-offs can be of various kinds, more or less observable and operationalizable by the researcher. Moreover, complete data on parliamentary support for minority governments should be analyzed before any final conclusion can 63

Politics Beyond the State.book Page 64 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

be reached. A very basic way to provide some evidence regarding this hypothesis is to simply see if the number of NSWPs in regional government increases substantially when at the state level there is a minority government. In the case of Spain, this is justified by the fact that state-level single-party minority governments have always relied on the parliamentary support of NSWPs (Reniu 2005: 2). Table 9 NSWP parties in government * majority status national government

No NSWP in regional government At least one NSWP in regional government Total

majority status national government minority majority government government at state level at state level 35 54 22 33 57

87

Total

89 55 144

As Table 9 shows, in situations of state-level minority, 38.5% of the regional governments formed included at least one NSWP. This is only negligibly higher than the 38% of the same kind of governments formed in majority situations. Somewhat more interesting is to break down this picture at the level of the two state-wide parties that we are interested in in this paper, the PSOE and the PP. Contrary to any intuitive expectations, during the Aznar I government, 48% of all the regional governments that formed included a NSWP or more, whereas during the last González and the current Zapatero minority governments, only 34% did so. Nevertheless, this does not mean that PP replicated its national support policy at the regional level more than the PSOE, because actually PP itself formed only 3 coalitions with non-statewide parties while governing Spain in a minority status, one in Cantabria and two in the Canary Islands. At the same time however, there were five minority governments formed solely by NSWPs during Aznar I, one in Catalonia and the other four in the Basque Country. Just as well, there are only two instances in which the PP governed in minority at the regional level during Aznar I, but they must be treated with caution, as they are two consecutive cabinets in Navarra, where the previously regionalist Unión del Pueblo Navarro (UPN) had fused with the PP in 1993, and ever since elections were run by the PP under the UPN banner.

64

Politics Beyond the State.book Page 65 Wednesday, December 5, 2007 8:41 AM

Governing Coalitions in Multi-level Settings

What about the González IV and the current Zapatero minority governments? During these two periods, PSOE also formed only three regional coalitions along non-state-wide partners. It formed rather many minority governments: two in Andalucia, two in Aragon, one in the Asturias and another one in Extremadura. Except for the case of Aragon, where the PSOE could have allied with two NSWPs to ensure formal majority in 1995 (the PAR and the Chunta Aragonesa), there was no viable coalition formula between PSOE and non-state-wide partners. Finally, similar to the Aznar I period, Catalonia and the Basque Country, and in addition the Canary Islands this time, were governed by non-state-wide minority governments.21 There is thus some mixed evidence in support of the hypotheses that payoffs for support at one level are observable at other levels, and they are more of a qualitative than of a quantitative nature. We don't necessarily find more NSWPs in government in national minority situations What we do find though is specific NSWP minority governments formed by precisely those parties that support the PP and PSOE at the state-level when they are in minority situations: the Basque Nationalists in the Basque country (PNV-EAJ), the Convergencia i Unió (CiU) in Catalonia, and the Coalición Canaria in the Canary Islands. Of course, in this first exploration I opted for the bluntest way to operationalize the pay-offs for support, namely support for participation in regional government. Further research must provide a more refined measure of pay-offs, including substantive policy concessions to non-state-wide demands at the regional level.

Institutions This part of the empirical analysis addresses the role of institutions. A first hypothesis states that incongruence is more likely to be present in those regions with highly “regionalized institutions” (hypothesis 3.4.a). Table 10 below shows the results of a logistic regression in which congruence was entered as the dependent variable and “own electoral timing” as a proxy for 21. However, the case in the Canary Islands is more complicated. While at the beginning an informal coalition between the PSOE and the CC was struck, shortly after the 1993 elections the PP made a counter-offer that the CC accepted and actually the CC minority government formed in 1995 in the Canary Islands functioned with the parliamentary support of the PP, and not the PSOE. This situation lasted only for one year, for the agreement between the CC and the PP provided that if the PP would win the following general elections in 1996, the agreement would be formalized and the PP would join the government, which is what actually happened (García-Rojas 2003).

65

Politics Beyond the State.book Page 66 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

regionalized institutions. Dimensionality was entered as control variable and, due to the nature of the proxy taken as the independent, the analysis was limited to those cases of governments that were formed following elections. Table 10 Institutions as determinants of congruence across levels

Own electoral timing* Dimensionality Constant N Nagelkerke R-square

Dependent variable: Congruence B Odds ratio -1.270** 0.281 (0.410) -1.022** 0.360 (0.496) 2.288** 9.860 (0.694) 145 0.195

Significance levels: ***0.01, ** 0.05, * 0.1 Notes: B is the unstandardized regression coefficient. Standard errors are shown in parentheses. For variable codings see Appendix. * Andalusia is coded as 0 on this variable, although it has it's own electoral timing separate from the other 13 ordinary-status regions. However, Andalusia has almost without exception held elections on the exact same day with national elections, and its dissimilarity index scores are, with the exception of the 1996-1999 score, below the average of the 17 ACs. It would be misleading thus to characterise it as a highly regionalised autonomous community

Although the regression coefficient is quite low (0.195), the proxy behaves as expected: as we move from regions with their own electoral calendar to ordinary regions, the odds of congruence are decreasing. Just as well, as we move from legislative situations where the left-right dimension is predominant to situations in which the second dimension, autonomy vs. centralism, is also present, we are more likely to encounter incongruent governments. What about the classical electoral variables that are expected to affect size and majority status of governments? Unfortunately, the effects of variables such as the requirements of investiture and constructive no confidence (Strøm, Budge and Laver 1994) cannot be tested on this data, as Spanish regions display no variation in this respect. Tables 11 shows the effects of district size and electoral threshold on the majority status and size of regional cabinets, while controlling for the number of dimensions (hypothesis 3.4.b). 66

Politics Beyond the State.book Page 67 Wednesday, December 5, 2007 8:41 AM

Governing Coalitions in Multi-level Settings

Table 11 Institutions and dimensionality as determinants of government size and status

Threshold/district District magnitude Dimensionality Constant N Nagelkerke R-squared

Dependent variable: single party majority B Odds ratio -0.088 0.916 (0.069) -0.003 0.997 (0.009) -2.149*** 0.117 (0.417) 3.553*** 34.909 (0.775) 145 0.319

Significance levels: ***0.01, ** 0.05, * 0.1 Notes: B is the unstandardized regression coefficient and S.E. its standard error. For variable codings see Appendix.

Contrary to the expectations, the electoral threshold and the district magnitude have absolutely no significant effect on whether the government that formed when we control for the number of dimensions at work. Instead, this variable has a strong negative impact on the dependent, the odds of encountering a single party majority government in a twodimensional legislature being substantially lower than in a onedimensional legislature.

Party system attributes I hypothesized that the higher the level of regional electoral volatility and cross-level electoral dissimilarity, the more likely it is for incongruent governments to emerge. The effects of party system characteristics on the congruence of government composition are also running as expected. As indicated in Table 12, both volatility and dissimilarity have statistically significant effects, be it only at the 0.01 probability level. The fast-track status of the region, entered as a control variable as we have shown previously that it affects congruence significantly, maintains its significant negative effect on congruence.

67

Politics Beyond the State.book Page 68 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Table 12 Dissimilarity and volatility as determinants of congruence across levels

Dissimilarity index Regional volatility index Region status Constant N Nagelkerke R-square

Dependent variable = Congruence B Odds ratio -0.032* 0.969 (0.017) -0.044* 0.957 (0.027) -0.766* 0.465 (0.087) 1.462 4.316 (0.473) 122 0.177

Significance levels: ***0.01, ** 0.05, * 0.1 Notes: B is the unstandardized regression coefficient. Standard errors shown in parentheses. For variable codings see Appendix.

Thus, as we move from an ordinary autonomous community to a fast-track status one, and as the regional-level electoral volatility and inter-level electoral dissimilarity increase, the odds for encountering congruent governments decrease rather substantially. However, one should treat these conclusions cautiously, as we have seen in section 3.1 of this chapter that when we also introduce in the regression model dimensionality and one variable which reflects whether the PSOE was or not in the regional government, both the dissimilarity index and the region status variable loose all statistical significance (see Table 6 above). This only means that, rather than being institutionally determined, congruence across levels is the result of party strategy, which is, of course, conditioned to a certain extent by electoral results.

Conclusions This research revises existing theories developed for the study of national coalition governments to make them applicable to government formation in decentralized countries. It further tests several empirical implications of this theoretical framework, breaking down the picture at the level of the two main Spanish state-wide parties to depict differences in their coalitionbuilding strategy.

68

Politics Beyond the State.book Page 69 Wednesday, December 5, 2007 8:41 AM

Governing Coalitions in Multi-level Settings

Although the chapter considers the determinants of a handful of classical dependent variables (government size, majority status and government composition), its main focus is analyzing the factors that have effects on the congruence of government composition across levels. This aspect is peculiar to decentralized political systems: territorialized political competition and ticket-splitting often give rise to asymmetrical parliamentary weights across levels of parliamentary representation. As the contribution on Belgian political parties (Buelens & Deschouwer in this volume)) also shows, incongruent governing positions are problematic for party organizations, often causing intra-organizational strain and disequilibrium. Supporting evidence was found for most of the theory's implications. Thus, parties do seem to follow different goals in different arenas, they do appear to rely on familiarity and inertia when forming regional coalitions, and payoffs for parliamentary support at the national level do seem to be visible at the sub-national one. Just as well, whether the regional legislature is characterized by one or by two dimensions of competition appears to play an important role in what regards congruence, government size and government status. Congruence itself appears to be substantially influenced first of all by party strategy. Electoral dynamics, as measured by regional electoral volatility and cross-level electoral dissimilarity, mediate the effect of strategy. The strongest finding of this research is that there appear to be important differences in party strategy – with the Socialists being eager to form congruent governments even at the expense of governing with minority support only. On the contrary, the Popular Party in Spain has been more cautious in forming minority governments at the sub-national level. This can probably be explained by the comparative difficulty of the PP to obtain informal parliamentary support from NSWPs. This is in line with another finding in the research, which is that a coalition formula including a statewide and one or more non-state-wide partners is strongly and positively determined by whether the state-wide partner is the PSOE or not. However, more research needs to be done in order to reach definitive conclusions. First and foremost, for a more refined analysis, additional data needs to be collected pertaining to the ideological positioning of parties in each parliamentary setting at each election point. Just as well, parliamentary voting data on government supporting parties as well as coalition agreements must be accessed in order to provide a more substantial test for some of the hypotheses. Finally, further research will need to supplement the results from this quantitative analysis by a

69

Politics Beyond the State.book Page 70 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

qualitative study of parties' coalition strategies, paying attention to organizational characteristics and the particular contexts in which specific governing strategies were developed.

70

Politics Beyond the State.book Page 71 Wednesday, December 5, 2007 8:41 AM

Torn Between Two Levels. Political Parties and Incongruent Coalitions in Belgium Jo Buelens & Kris Deschouwer

Incongruent coalitions In a large number of liberal democracies coalition government is a very normal and widespread feature. A proportional electoral system and some degree of party fragmentation generally limit the chances of one party winning the elections and being able to govern alone (De Winter & Dumont, 2002). Becoming part of a coalition or staying outside is therefore one of the crucial strategic choices many political parties have to make. Entering a coalition allows for the control of office and for the direct influence on policy making, but it is a limited and shared control of office and policy and therefore also contains the risk of being blamed more than the others for things going wrong (Müller & Strøm, 1999). Opting for the half way house between government and opposition by selectively supporting a minority government is – at least for smaller parties – a way to make the choice between governing or not governing less hard (Strøm, 1990). Federal and regionalized states have a larger number of governments. That does not mean however that all these governments are of the same type. The party landscape and the voting behaviour can differ sharply between regions and between regions and the central state, allowing for quite some variation (Watts, 1999). While Spain has generally a one party majority or minority government, a number of Autonomous Communities are governed by coalitions (see Stefuriuc in this book). In Switzerland, with always a grand federal coalition, the cantons are fairly small and generally more homogenous, allowing for one party government at that level (Linder, 1994). In regionalized or federal systems governments at the lower level can therefore have a fairly different party composition, eventually because of a clear divergence in the strategies followed by the federal parties and by their regional branches. The Scottish Labour Party formed a coalition with the Liberal Democrats in Scotland, with the latter remaining an opposition 71

Politics Beyond the State.book Page 72 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

party in Westminster. The German SPD has always refused to even think of forming a federal coalition with the PDS, but has formed SPD-PDS coalitions in several former East German Länder. The Austrian SPÖ fiercely rejects the idea of forming a coalition with the FPÖ (now BZÖ) at the federal level, but did enter a SPÖ-FPÖ coalition in Carinthia, with the FPÖ party leader Jörg Haider as Landeshauptman (Prime Minister). In all these cases the regional and the federal government (or coalition) are not congruent. That means that some parties find themselves in a double and possibly ambiguous position. They govern at one level and are in the opposition at another level. They have a partnership relation with some parties in a coalition at one level, while they oppose the policies of (some of) these same parties at the other level. These tensions – both between and inside parties – or potential tensions are the focus of this chapter. Incongruent coalitions do indeed raise a number of interesting and so far almost completely unanswered questions. Why and when do parties opt for this double role? Is it a deliberate choice or rather the result of institutional pressure? And what are the consequences of the choice? We will explore these questions by looking at the Belgian case. Belgium is a fairly new federation, where therefore incongruent coalitions are a fairly recent phenomenon. It is still possible to trace back the choices made and to investigate the reasons and consequences of incongruence. On the other hand the Belgian federation is quite peculiar though, which limits the possibilities to generalize from its analysis. The bipolar nature of the federation and especially the absence of federal parties create a very specific institutional environment, which makes it a special but highly relevant case. This analysis will be explorative, but will offer a number of interesting insights in the way in which complex and multi-layered institutions affect the strategic options of political actors.

The Belgian federation Belgium is a very young federation. Actually the unitary Belgian state was transformed gradually into a decentralized and finally fully-fledged federal state. The last stage was reached in 1995. That was the date of the very first direct election of regional parliaments. It was not the beginning of regional governments though. Regional governments have existed as separate entities since 1981. Earlier a number of ministers of the national government had explicit regional competencies. After 1981 regional governments were formed, while regional parliaments were not elected

72

Politics Beyond the State.book Page 73 Wednesday, December 5, 2007 8:41 AM

Torn Between Two Levels. Political Parties and Incongruent Coalitions in Belgium

directly. The regional parliaments were composed of the members of the national parliament elected in one of the regions. Furthermore these early regional governments were not really coalition governments. During a transition period that lasted till 1995 the regional governments were proportional, i.e. each party having a minimal number of votes had the right to enter the regional coalition. Since this was not the case for the federal coalition, there were obviously situations of incongruence. These situations of incongruence will not be dealt with it this chapter. We focus on the recent phase of the fully fledged federation after 1995. This period after 1995 can be divided in two stages. The first runs from 1995 till 2003. It covers actually the elections of 1995 and 1999. These were both elections for the federal parliament and for the regional parliaments (1999 also European Parliament) organized on the same day. In 2003 federal elections were organized separately for the very first time. It means that for the very first time only a federal coalition had to be formed, while the regional governments remained in power. In 2004 regional elections were held while the federal coalition was and remained in power. This disconnection of the two elections from 2003 on has created a totally new institutional setting, to which the political parties are trying tot adapt. Before going deeper into the dynamics of the post 1995 coalition formation, we need to discuss a few more elements of the institutional characteristics of the Belgian federation (for more details: Deschouwer, 2005). The first is the double nature of the federation. Belgium has indeed two types of substates: regions and communities. It is both a federation of three language communities (Dutch (or Flemish), French and German) and of three region (Flanders, Brussels and Wallonia). These two types of substates overlap to a great extent, but not completely. The reason for having a double federation is the geographic location of Brussels and its linguistic status. Brussels is geographically and thus historically located north of the border between Dutch speakers and French speakers, but its role of capital city of a country that has for a long time been ruled in French only and in which French was clearly a higher status language has transformed it into a very francophone city. Its official status though is bilingual. Actually the Flemish community can offer its services (like education) to the inhabitants of the Flemish region and of the Brussels region. The French community can do the same in Wallonia and in Brussels, except for a small area where the German community offers its services. Inhabitants of Brussels thus have the choice between the two large communities.

73

Politics Beyond the State.book Page 74 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

The relation between region and community is however not organized in the same way in north and south, making the Belgian federation also asymmetrical. In Flanders, one single institution – Flanders – has been created, with one government and one parliament, dealing with both the regional matters for the Flemish region and the community matters for the Flemish region and for Brussels. Actually in Flanders the difference between region and community is not that important. The difference between both is the group of Dutch speakers of Brussels, being roughly 3% of the Dutch speaking population in Belgium. On the Francophone side the regions and communities remain separated. That means that there is a Walloon region (with government and parliament) and a French community (with government and parliament). On the Francophone side Brussels is much more important and different (also geographically) from the Walloon region. Almost 20% of the French speakers in Belgium (the French community) live in Brussels. This all means that there are six governments in the Belgian federation: – The federal government. The Constitution includes an obligation to have an equal number of Dutch speaking and French speaking ministers (7 of each). The Prime Minister (generally a Dutch speaker) is not counted. – The Flemish government. – The government of the Walloon region – The government of the French community. The parliament of the French community is composed of all the Walloon regional MPs and of 19 francophone members of the Brussels regional parliament – The Brussels regional government. It is composed – like the federal government – of an equal number of French speakers and Dutch speakers. The regional Prime Minister (always a French speaker) is not counted. – The German community government. Its parliament is elected directly. The distribution of competencies in the Belgian federation gives a high number of tasks to the federated entities. Major policy domains like education, culture, environment, public transport, public works, economy, health, housing and area planning have been devolved to regions and communities. They also have the right to conduct their own international relations in these policy areas. Their financial and fiscal powers are however limited. The federal state receives and redistributes most of the taxes (Deschouwer, 2005: Deschouwer & Verdonck, 2003). Although the distribution of competencies has produced fairly watertight departments – with obviously a number of overlapping competencies,

74

.

Politics Beyond the State.book Page 75 Wednesday, December 5, 2007 8:41 AM

Torn Between Two Levels. Political Parties and Incongruent Coalitions in Belgium

mainly in the area of labour and economy – the difference between regional policy, community policy and federal policy is not very visible. Obviously there is a deliberate blurring of the difference between region and community of the Flemish side, due to the fact that the Flemish elite has always preferred the community logic (based on linguistic identity) over the regional logic. But the difference between the regional level and the level of the federated entities is equally vague and unclear in the minds of the citizens (Cambré, Billiet, et.al., 1996). One of the most important reasons for this is the way in which the political parties are organized, and thus the way in which the political elite – federal and regional – communicates with the citizens.

The Belgian parties The most striking and – for us – also crucial characteristic of the Belgian parties is the absence of federal parties. All the political parties are unilingual, which also means that they only compete in one of the two large language communities. Flemish parties are present at the polls in Flanders and in Brussels. Francophone parties are present in Wallonia and in Brussels. Until the 1970s there were still national political parties. Yet in 1968 the Christian democrats fell apart to form two new unilingual parties. In 1972 the liberals followed suit and in 1978 the socialist party split. New parties created afterwards, have all been rooted in one of the two unilingual party systems, even if they belong – like the greens or the right wing extremist parties – to the same party family1. 1.

Most Belgian political parties have recently adopted new names. In this chapter we will – for the sake of clarity – always use these new names, even when we refer to earlier periods when the parties used older names. The Belgian parties that will be present in our story are: CD&V: Flemish Christian democrats (formerly CVP) CDh: Francophone Christian Democrats (formerly PSC) PS: Francophone socialists Spa: Flemish socialists (formerly SP) VLD: Flemisch liberals (formerly PVV) MR: Francophone liberals (electoral alliance of PRL and Brussels Francophone party FDF) Ecolo: Francophone greens Groen: Flemish greens (formerly Agalev) Vlaams Belang: Flemish right wing extremist party (formerly Vlaams Blok) Volksunie: Flemish regionalist party (was dissolved in 2001) N-VA: Conservative Flemish regionalists (official successor party of Volksunie) Spirit: Left-liberal Flemish regionalists (part of former Volksunie)

75

Politics Beyond the State.book Page 76 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

It is exactly this absence of federal parties and this split party system that accounts to a large extent for the absence of a clear distinction between the levels. Indeed, whether the election is federal or regional, the same parties with the same party leaders go to their part of the electorate. When in 1995 and in 1999 both elections – federal and regional – were organized on the same day, the confusion between the levels was total (Versmessen, 1995; Deschouwer, 1999). Parties also used the elected members as one single personnel pool in which they could go and pick the right members for the governments in which they participated. Of all the regional and community Prime Ministers in 1999, not one had been elected at the regional level. They had all run for a seat in either the federal House of Representatives or the Senate. Each party in Belgium is a ‘community’ party, competing in only one language community. For each party there is only one party organization, contrary to (many) parties in other federal states having a federal party organization and regional party organizations. Since the Belgian parties are community parties, and therefore covering two regions – both party groups compete in Brussels – there is at least some difference to be made between the community level of the parties and their regional level. For the Flemish parties the Brussels regional branch is very small. As said above, only 3% of the votes for Flemish parties are cast in Brussels. The Brussels branch of the Flemish parties operates furthermore in a peculiar environment: one in which coalitions have to be formed with the Francophone parties. This lack of weight of the regional branch and its very special institutional environment should make it fairly independent from the central (i.e. community level) party. On the Francophone side the situation in Brussels is different. Although the parties are indeed organized at the community level, the Brussels regional branch is important. All Francophone parties do therefore explicitly acknowledge the presence of a Brussels branch. That is an indicator of a certain autonomy, but on the other hand this Brussels branch is too important (in electoral terms) to just let it go its own way. Going its own way is also made a bit difficult by the fact that the overlap between Walloon region and French community is quite important, and that two different strategies in two parts of the French community can be fairly difficult to manage. It is however one possible way to create incongruence, which does not need to be – in the Belgian double federation – an incongruence between federal and federated level only. Incongruence is also possible (on the Francophone side) between Walloon region and French community.

76

Politics Beyond the State.book Page 77 Wednesday, December 5, 2007 8:41 AM

Torn Between Two Levels. Political Parties and Incongruent Coalitions in Belgium

Even if the Brussels branch has some autonomy in the parties, the leadership remains clearly centralized. The party leader is the party president of the central (community) party. The leaders of the regional branch have some visibility for the Francophone parties in Brussels, but they are hierarchically lower than the party president. Party presidents have always played a very important role in Belgian politics (Fiers & Krouwel, 2004), and party leadership has always been very centralized. Coalition negotiations are lead by the party presidents. Ministers are selected by the party presidents.

Incongruent coalitions In 1999 Belgium witnessed a fairly dramatic change of the governing coalitions. After the federal and regional elections, the Christian democrats appeared to have lost their leading position in Flanders and therefore also in Belgium. On the Francophone side the liberals and the socialists had already agreed that they would form a coalition together. In Wallonia and in the French Community that would be (mathematically) no problem, but the federal level would be more difficult. The Flemish liberals, now the largest party of Flanders and of Belgium quickly realized that this was their chance to remove the Christian democrats – in power since 1958 – from all the coalitions. Mathematically that was however not that easy. In order to form a government without the Christian democrats in the Flemish region and community, liberals and socialists needed the greens and the Volksunie. They formed a four party coalition. At the federal level liberals and socialists had not enough seats in the Parliament, and needed one of the green parties. Since the greens had agreed to stay together in or out of government, both were added to the federal coalition, which was then oversized. The Francophone greens were added to the Walloon and French Community Government, not because they were needed for a majority, but simply to keep the life of all the parties easier by having congruent coalitions. By including the greens at the federal level and in the Francophone coalitions, all the federal coalitions and the regional and community coalitions were to a very large degree congruent. Apparently oversizing seemed a better solution than incongruence. Yet incongruence there was. First there was the Volksunie. It was part of the Flemish government, but not of the federal government. In the Brussels region the plans were to form the same liberal-socialist-green government. The agreement was indeed negotiated, but the members of the Brussels green party section rejected it, which meant that Ecolo was in all 77

Politics Beyond the State.book Page 78 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

governments but the Brussels one. Here it is the regional branch of the party deciding to stay out. At the Flemish side in Brussels there was incongruence too, since that was the only place where the Christian democrats remained in power. They were mathematically needed to form a coalition without the right wing extremist Vlaams Belang. In Flanders the Christian Democrats still had the largest group in the regional parliament, but since they had no chance to be part of the federal government, they decided not to govern in Flanders. They refused to even try to form an incongruent Flemish coalition. In 2003 the first federal elections were organized while the regional and community coalitions had still one year to go. That created quite some confusion, since the community parties engaged in an electoral campaign with all the available personnel, including the regional and community ministers. Socialists and liberals did very well, while the greens lost heavily. The Flemish greens even lost all their federal MPs. The new federal coalition was formed again by liberals and socialist, but without the greens. The greens had however to go on at the regional level. In Flanders their leaving the coalition would have ended it. The green ministers were replaced, but the party decided to remain in the government until the regional elections of 2004. Ecolo had the choice whether it would remain in the Walloon and French Community government. These were indeed oversized and could survive without the greens. Ecolo did however decide to stay, and opted thus freely for the incongruent position until 2004. The regional elections of 2004 – the first separate regional elections – brought new dramatic changes. Again all party elites, irrespective of their position, engaged in the regional elections (including the federal Prime Minister). Especially the Flemish results were surprising. Not only did the radical right wing Vlaams Belang poll an impressive 24%, the Christian democrats – allied with a fraction of the by then disappeared Volksunie – came out number one and claimed the leadership of the government. Given the size of Vlaams Belang, a three party coalition of Christian democrats, liberals and socialists had to be formed. The Flemish Christian democrats then opted for incongruence, since they were not in the federal government, while their two partners were. At the Francophone side the socialist party won the elections in Brussels, Wallonia and the French community, allowing the PS to choose its partners. While it had continued the federal coalition with the liberals in

78

Politics Beyond the State.book Page 79 Wednesday, December 5, 2007 8:41 AM

Torn Between Two Levels. Political Parties and Incongruent Coalitions in Belgium

2003, the PS opted in 2004 to form regional coalitions with the Christian democrats, leaving the Francophone liberals in an incongruent position. Table 1 gives an overview of all these coalitions since 1999. It allows us to identify all the cases of incongruence. They are the following: – Volksunie 1999-2002 (Flanders – not federal) – CD&V 1999-2004 (only Brussels) – Ecolo 1999-2003 (Federal, Wallonia, French community – not Brussels) – Groen 2003-2004 (Flanders – not federal) – Ecolo 2003-2004 (Wallonia, French Community – not federal) – MR 2004-… (Federal – not Wallonia, not Brussels, not French community) – CD&V 2004-… (Flanders and Brussels – not federal) – CDh 2004 – (Wallonia, French Community, Brussels – not federal) – Ecolo 2004 (Brussels, not Federal, not Wallonia not French community) Table 1 Composition of the coalitions since 1999 (party of Prime Minister in bold)

Federal Flanders Wallonia French Community Brussels

1999 Regional and federal elections VLD – MR – PS – SPa – Ecolo – Groen VLD – SPa – Groen – VU PS – MR – Ecolo PS – MR – Ecolo

2003 Federal elections

2004 Regional elections

VLD – MR – PS – Spa VLD – SPa – Groen – VU PS – MR – Ecolo PS – MR – Ecolo

VLD – MR – PS – SPa CD&V – NVA – Spirit – SPa – VLD PS – CDh PS – CDh

MR – PS – SPa – PS – MR – SPa – PS – CDh – Ecolo – CD&V – VLD – VU CD&V – VLD – VU CD&V – VLD – SPa – Groen – Groen

Although the disconnection of regional and federal elections is still a rather recent phenomenon, incongruent coalitions are very much part of the game. At first sight this is rather surprising. Given the absence of nationwide Belgian parties, incongruence should be a fairly difficult situation to deal with. Indeed, the coalition partner at any level is not the national or regional section of the political party, but always – except for Brussels – the political party itself. Having a coalition partner at one level

79

Politics Beyond the State.book Page 80 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

that is an opposition party at another level therefore seems to be an awkward situation for both governing and opposition parties. Only those being in the same position at both levels find themselves in a strategically more easy and manageable situation. Governing in the region with a party you oppose at the federal level means that you agree on for instance educational policies but not on labour market policy. The parties and especially the party presidents have to constantly shift roles in the public debate. And when the next election comes, they need to explain why they keep on governing with a party they want to kick out of government at the level for which the electoral campaign has been launched. And yet incongruence is not a marginal or exceptional phenomenon. Parties do opt for incongruent positions. And that raises interesting research questions. One should assume that political parties are free to choose for an incongruent coalition. Incongruence always means being in power on at least one level, and being in power is not a default situation but the result of a deliberate choice. Since entering or accepting to remain in a incongruent coalition is a deliberate choice, we should try to see and understand why that choice is being made. Which are the advantages offered by incongruence that outweigh the disadvantages that we have described above? We have so far only very few theoretical guidelines to engage in this type of research. The analysis of parties in multi-level settings is still a fairly new research domain (Hough & Jeffery, 2006; Hopkin, 2003; Thorlakson, 2001, Deschouwer, 2006), and coalition formation as a specific party activity has so far not being analyzed in a systematic way. What we want to do in this chapter is explore a few interesting cases of incongruence in Belgium to discover the mechanics and the motivations of it. Our results can then be used for a further, more systematic and also comparative analysis of the phenomenon of incongruence. We will discuss five cases for four political parties. The first case – CD&V in 1999 – is actually not a case of incongruence but one in which incongruence was avoided. The second case is the CD&V in 2004. We will also look at the Volksunie in 1999 and at the two green parties in 2003. This limited number of cases does offer some interesting variations. First we have a large party and smaller parties. We have – as we said – one case where incongruence was avoided and a number of cases where it was accepted. We have two cases – CD&V in 2004 and VU in 1999 – of parties opting very deliberately to enter a coalition that will be incongruent, while we have also two cases (the greens) where a change in the coalition at the

80

Politics Beyond the State.book Page 81 Wednesday, December 5, 2007 8:41 AM

Torn Between Two Levels. Political Parties and Incongruent Coalitions in Belgium

federal level confronts them with incongruence and with the question whether they should remain in the regional coalitions. Each of these cases raises a number of questions. The first is obviously the choice. Is incongruence really (perceived as) a free choice, or do parties experience some degree of pressure coming form the institutional environment in which they have to function? A second set of questions refers to the motivation, the reason for entering or staying in an incongruent situation. Why does CD&V avoid incongruence in 1999 and not in 2004? What sort of arguments is being used? Is staying on board a matter of office (keeping the positions and the money that comes with being in government), a matter of policy (finishing the job) or a matter of electoral strategy (keep the visibility of being in power)? In the following paragraphs we will describe the coalition formations and give some first answers to these questions. Our analysis is based on four types of sources: the yearly chronicle of Belgian political life in the political science journal Res Publica, journalistic accounts of what happened during the period that we discuss (Bouveroux, 1996; 2003; Van Peteghem, 2001), books written by politicians (Durant, 2003; De Bast, 2002) and interviews with politicians who were involved in the formation and life of the coalitions.

CD&V in 1999 and 2004 In 1999 regional and federal elections were organized on the same day (with also European elections). Since 1987 the core parties of all coalitions at all levels had been the Christian democrats (CD&V leading the governments of Belgium and of Flanders) and the socialists (PS leading the governments of Wallonia, French Community and Brussels region). Since 1987 both liberal parties had been in opposition at all levels, although they did slightly improve their score at almost every election. In 1999 again the liberals went to the polls with the clear strategy to break this ‘natural’ coalition of the strongest parties of the north (Christian democrats) and the south (socialists). The francophone liberals and socialists had actually formally agreed to form a so-called ‘purple’ coalition (red and blue) wherever it would be possible. Given the strength of both parties in the Francophone electorate and the relative weakness of the francophone Christian democrats that was a coalition agreement that was easy to realize and that would weigh on the coalition formation at the other levels and even in the Flemish part of the country. This pre-electoral francophone agreement for a purple coalition thus becomes part of the institutional 81

Politics Beyond the State.book Page 82 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

environment in which the federal and the Flemish coalition would have to be formed. The results of the elections of 1999 were quite mixed. In Wallonia the PS remained easily the number one. But a very good result for the MR in Brussels (where it became the number one) brought it very close to the PS if all francophone votes are taken together. In Flanders the VLD managed to take the number one position by very closely beating the Christian democrats. The VLD also became the largest party of the country. Yet for the election of the Flemish parliament the victory went – also extremely closely – to the CD&V that remained the largest group in the Flemish parliament. That actually gave them in principle the right to make the first move. Yet the Flemish Christian democrats never moved. Their electoral defeat was not the first one, but this was the first time they lost their leading position in Belgium. The purple alternative that was believed to be a significant change and opportunity for political renewal – the Christian democrats had been in power since 1958 – immediately became the obvious coalition to be formed. The possibility for the CD&V to form and lead another coalition in Flanders was therefore merely a theoretical one, exactly because of the multi-level game being played. It could eventually form a coalition with the VLD, but that party was going to be the leading party of the federal coalition marking a symbolic end to the domination of the Christian democrats. That would have been an extremely uncomfortable situation for both parties, and especially also a situation that could hardly be explained to the voters. CD&V thus simply opted for the opposition at all levels. Yet it did remain in the Brussels regional coalition, since the agreement to exclude the right wing extremist Vlaams Belang from power made the presence of the CD&V indispensable at that level. In Brussels the party ‘chose’ for incongruence, but the choice was a consequence of the political context in which one party was and remained a pariah in the system. Being in the opposition for the more important governments of Flanders and Belgium was however clearly seen as the best solution. It was believed to be a clear and unambiguous situation from which the party could recover and get rid of the role of the party that could be blamed for anything going wrong. In this clear and unambiguous situation the CD&V could prepare for moving back into power at the next elections. The next elections – federal elections in 2003 – were however a disappointment. liberals and socialists did very well, and the Christian democrats went further down. The next goal was therefore the regional 82

Politics Beyond the State.book Page 83 Wednesday, December 5, 2007 8:41 AM

Torn Between Two Levels. Political Parties and Incongruent Coalitions in Belgium

elections of 2004. The four party coalition at the Flemish level had by then been weakened. The Volksunie had fallen apart in two new parties and Groen had lost all its federal seats after the 2003 election in which the Ministers of the Flemish government were the leading figures of the campaign. Without these two parties the purple coalition would not be able to survive in Flanders, unless the liberals and socialists did even better than in 2003. A very wise move of the CD&V was the formation of an electoral cartel – i.e. presenting common lists – with N-VA, one of the two parties (the most radical one) resulting from the split of the VU. A newly introduced threshold of 5% per province (Hooghe, Maddens & Noppe, 2006) would have made it quite hard for the N-VA to gain representation on its own. While the opinion polls were showing that this so-called ‘Flemish cartel’ was going to win the regional elections, the question of the possibility and consequences of incongruence made its way to political agenda. Indeed, the question was raised whether CD&V would be able to govern in Flanders without ‘breaking into’ the federal government. Clearly the general mood was that congruence would be easier than incongruence. In the end CD&V with the N-VA did win the regional elections and became again the largest group in the regional parliament. Again with a very strong Vlaams Belang (24%) being excluded from power, there was no winning coalition possible without the CD&V and no other choice but the one to accept the responsibility to lead the Flemish government. With the greens announcing that they would opt for the opposition, the Flemish coalition would have to bring together the Christian democrats, the VLD (still leading the federal government) and the SPa (also in alliance with one of the two post-VU parties). The CD&V would thus have to keep on opposing its Flemish regional partners at the federal level. The CD&V accepted the incongruent position. The only way to get out of it would be a change in the federal coalition. That was not the easiest solution. Psychologically it would have been quite difficult for the VLD (and for Prime Minister Verhofstadt in particular) to take the Christian democrats on board of a government that had in the very first place been formed to remove the Christian democrats from power. Adding the CD&V would also have meant that the francophone Christian democrats of CDh should be allowed to enter the government. One of the unwritten but important rules for the formation of a federal government has been the requirement of symmetry, i.e. having both parties of the same family always together in or out. The difficult relation between the Flemish parties

83

Politics Beyond the State.book Page 84 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

would then have required a reshuffle on the Francophone side, which was again not an easy solution at all. And therefore the CD&V ‘chose’ for the incongruent position. Avoiding it was far too difficult, and keeping the congruence by not governing in Flanders was no option. All the CD&V could do then was trying to live with the incongruence and making the best of it. Torn between two levels the party had to invent a wise strategy to survive on both levels. That meant in the first place a lower profile at the federal level. Fierce opposition at the federal level would lead to constant conflict with VLD and SPa, the partners needed for a Flemish regional majority. Second part of the strategy was then obviously to focus as much as possible on the former party president who had become the Flemish Prime Minister. The party should be seen in the very first place as the party governing (well) in Flanders. That choice is also very much inspired by the incongruent position and by the knowledge that the next federal elections of 2007 are an important deadline. Getting back into power at the federal level is thus a path going via visibility of the party at the Flemish level. That position also allows the party to be on a rather safe side whenever negotiations between the two levels are failing. The parties being part of both governments, and not the party being in only one of them can actually make sure that the relations between the two governments are smooth. This blame avoidance did work a few times. When for instance the federal government failed to find an agreement on an electoral reform asked by Flemish government, the liberals and socialists had to take the responsibility for the failure. It created quite some tensions in the Flemish government, yet nobody had any other option but to stay where they were. This story of the CD&V dealing with incongruence reveals a number of interesting elements. It is clear that incongruence requires a learning process. While in 1999 the situation was almost inconceivable, it was a real option in 2004. It did require however some creative thinking about the way in which the inter-party relations could be organized. The learning process is not over yet. Only when parties in incongruent positions learn what the electoral consequences of their double role can be, they will be a bit more secure about the choices they make. For this to become clear the Belgian federation needs a few more electoral cycles and a few more incongruent coalitions. The CD&V is a large and traditionally governing party. It always has a number of options available. Yet the institutional context does reduce these options quite dramatically. Exclusion of a pariah party and the obligation to have a linguistically balanced government both at the federal level and

84

Politics Beyond the State.book Page 85 Wednesday, December 5, 2007 8:41 AM

Torn Between Two Levels. Political Parties and Incongruent Coalitions in Belgium

in Brussels are elements that limit the options available. Incongruence then appears as an option that is not a first choice but the best choice within the realistically available options.

Volksunie in 1999 The Volksunie is a smaller and not a traditionally governing party. The VU has however participated in a number of federal and regional coalitions. When it became involved in the coalition formations in 1999, it was not a first time experience. The position of the party was not easy though. As said before the agreement between francophone socialists and liberals and the general mood that it was time for a government without Christian Democrats lead to the formation of coalitions with liberals and socialists at all levels. On the francophone side the combination of liberals and socialists was always winning. On the Flemish side the situation was a bit more complex. For the formation of a Flemish government with VLD and SPa two more parties were needed: Groen and Volksunie. That put the VU in a quite powerful position. Without its consent a Flemish government was impossible and the radical renewal by the purple coalitions would fail. The VU was however also quite eager to see this political renewal. Accepting to join the Flemish government was not too difficult. Yet exactly the fact that the party was not needed at the federal level to form a winning coalition led to some hesitation. There were three options: staying out of both governments, entering only at the Flemish level or entering at both levels. In the latter case the VU could use its power position in Flanders to request governmental power at the Belgian level. Further devolution of competencies to Flanders – one of the main goals of the VU – had to be achieved at the federal level, since only the federal parliament can change the constitution. But on the other hand the presence in the Flemish government could be a way to put pressure on and keep and eye on those parties that would have to implement the constitutional changes at the federal level, without the VU itself being involved. liberals and socialists were on the other hand not too eager to take the VU on board at the federal level, because that would make the formation of the federal coalition a longer and more painful process. At the federal level the parties wanted to show that the purple renewal could be put into place rapidly and therefore also in an atmosphere of ethno-linguistic peace. The VU was quite divided on these different options. It finally opted for the incongruent solution: governing in Flanders and not in Belgium. Although the choice was only to a limited extent a fully free choice, the party did see 85

Politics Beyond the State.book Page 86 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

the advantages of the incongruence. At the federal level the party was not needed and therefore its power would be extremely limited. At the Flemish level it was needed, and that allowed the party to actively participate in the policy-making at the level that had after all been created as a result of the pressure of the VU since the 1960s. After having realized the Flemish autonomy it felt the duty to be present when that Flemish autonomy was put at work. Being present only in Flanders also offered the possibility – like for the CD&V after 2004 – to clearly mark the difference between the Flemish and the federal government. And if the constitutional reforms at the federal level would fail, the VU would not be the party to blame. There is clearly a potential for a sort of blackmail strategy for parties in an incongruent position. The rest of the story is however not very glorious. When a new party president had to be elected in 2000, the faction defending opposition at all levels defeated the outgoing president who had opted for the ‘blackmail’ and incongruent strategy. The party did remain in the Flemish government, but the gap between its cooperative Flemish strategy and its opposition strategy at the Belgian level (then new president was a national MP) appeared difficult and finally impossible to manage. When an agreement was reached for a constitutional reform, with the Flemish government highly involved, the party president and the party group in the federal parliament decided to vote against it. The split between the ‘governing’ faction and the ‘opposition’ faction, coinciding to a large extent (but not completely) with the split between the party at the Flemish level and the party at the Belgian level finally led to a split of the party. Two new parties were created: N-VA and Spirit, with Spirit remaining in the Flemish government (be it now with one and not with two ministers). The story of the VU tells us that incongruence seems to be much more difficult to manage for a smaller party than for a larger or leading party. While the CD&V has been after 2004 a bit torn between two levels, the VU was literally squeezed between two levels. Yet we should not too easily generalize from this single case. Another major difference between the VU in 1999 and the CD&V in 2004 is the absence of a clear and coherent strategy for the VU. The party was divided, and the choice for governing and thus for incongruence was a narrow one. The party president stood clearly on the side of the choice to govern and was soon removed by a new president defending the option of full opposition. The hesitation between governing and not governing had actually always been a matter of fierce internal debates in the VU, and the choice for doing both at the same time has certainly speeded up the termination of the party.

86

Politics Beyond the State.book Page 87 Wednesday, December 5, 2007 8:41 AM

Torn Between Two Levels. Political Parties and Incongruent Coalitions in Belgium

The greens in 2003 Both green parties were after the federal elections of 2003 suddenly confronted with the problem of incongruence. Their very poor score, with the Flemish Groen even losing all its federal MPs, put an end to the federal coalition of liberals, socialists and greens. liberals and socialists went on without the greens. At the regional level however, the greens were still part of the governing coalition. In Flanders Groen was furthermore mathematically still needed for a winning coalition. In the French Community and in the Walloon region Ecolo was not mathematically needed. It had actually never been needed, since its presence in the regional governments on the Francophone side was only a consequence of the necessity to have at least one green party for the federal majority in 1999 and of the unwritten rule (and deal between the two green parties) that they would either both govern at the federal level or both be in opposition. The situation for Groen was dramatic, with not one single federal MP. Its first option would have been to also leave the Flemish government. There were good reasons to do so. The extremely bad federal score had been realized in a campaign in which the Flemish green ministers had taken the lead (and were effective candidates). The Flemish voter had disapproved the Flemish green ministers. The party did hesitate, but very quickly decided to stay and thus to go for the incongruent position. Here again we can see how advantages and disadvantages are being compared, with in the end the incongruent solution being preferred. There were several elements tipping the balance towards staying in the Flemish coalition. The two green ministers were however replaced and the party leader resigned. In the first place there was the – officially most important – reason that the job was not finished. The greens had entered with enthusiasm both the federal and the regional new and purple coalitions and felt that they had been able to make the difference in some policy decisions (Buelens & Deschouwer, 2002). The more strategic reasons for staying on board were more negative, and show the fear for even more damage. One of the reasons for the bad performance at the federal level was the idea that in the end the greens were not very reliable partners. Leaving the Flemish government and thus taking away its needed parliamentary majority would certainly not help to counter that image. The party preferred to show loyalty and respect for the governmental agreement, rather than slamming the door. Another rather negative motivation was the fear to disappear completely from the political debate. At the federal level there were no MPs any more, and in 2004 new regional elections were scheduled. Opting for the 87

Politics Beyond the State.book Page 88 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

opposition would have meant going to the next elections after a year with very few opportunities to show and explain what a green policy would be. The party had removed its very experienced Flemish ministers and needed enough media attention to remain in the picture. Governing was therefore the chosen option. And finally the loss of the parliamentary group and of the presence in the Belgian federal government was a serious financial blow. In the Belgian tradition of personal collaborators of the ministers, a party in government always has an interesting number of jobs available for experts close to or member of the party. The MPs and the parliamentary groups also have a number of collaborators paid by the parliament. And direct subsidies to parties are paid only to parties having at least one elected member in both houses of the federal parliament. Groen had to fire many of these collaborators. Leaving the Flemish government would have reduced even further the resources and personnel available to the party. One can say to a certain extent that the choice to remain for one year in this incongruent position did pay. It was a very difficult year, with lots of internal debates and lack of clear central leadership. But in the end Groen went to the polls alone in 2004 and managed to pass the 5% threshold. It was represented again in the Flemish parliament, but announced immediately that it would not go on governing. Incongruence had been a conscious choice, but the party went out of it as soon as it was feasible. For the Francophone Ecolo the situation was slightly different. It could leave the regional governments without hurting them. Yet it also decided to stay. The reasons and arguments are similar to those of Groen. First there was the policy-related argument that the job was not finished, and that the opportunities to control policy outputs should not be given away because of the federal defeat. And second there was also the resourcerelated argument. Ecolo still had a parliamentary group in the federal parliament and kept some federal party subsidies, but both were at a lower level. Leaving the Walloon and French Community governments would have hurt even more. And finally there was the visibility argument. By remaining in power the party felt that it would be in a better position to prepare the regional elections of 2004. Maybe the double role of government and opposition was a bit easier to play for Ecolo, because it had since 1999 already opted for a slightly ambiguous attitude. The party had accepted to govern but tried to present itself still as an alternative party, as a party not governing in the traditional way. That was one of the ways in which the party could keep some internal peace between those very much in favour of governing and those fearing 88

Politics Beyond the State.book Page 89 Wednesday, December 5, 2007 8:41 AM

Torn Between Two Levels. Political Parties and Incongruent Coalitions in Belgium

that governing would kill the real spirit of the party and the real influence of the grass roots. While Groen had a strong party leader consciously coordinating the activities in the two governments, Ecolo had collective and shifting leaderships, with an ongoing debate about the decision to participate in the federal and regional coalitions. It is important to remind that the Brussels members had rejected the entry of the party in the Brussels regional government.

Conclusions The starting point of this explorative analysis was that incongruent coalitions are an uneasy and not very comfortable situation for political parties. We assumed that especially for Belgian parties incongruence would be the source of difficult tensions and would therefore be avoided. The absence of federal parties indeed means that incongruence obliges parties to collaborate with each other at one level and oppose the same parties at the other level. But the inventory of coalitions formed since the transition of Belgium from a unitary to a federal state shows that incongruence is not a rare or marginal phenomenon. We do see parties deliberately choose for entering incongruent coalitions or for staying in coalitions that have become incongruent. Apparently incongruence is not only the source of potential tensions, but also of potential advantages. The cases discussed in this chapter have certainly revealed a number of advantages that make parties choose for or accept incongruence. In the first place there is the advantage of government as such. Governing offers a party some degree of policy influence, visibility to the voters and access to both financial and personnel resources. Being in power at one level is then better than having to play the opposition role at all levels. In the second place we have seen that coalition formation in a federal state is indeed a game that is being played at more than one level, where incongruence can offer interesting strategic opportunities. A party being in coalitions at both levels also carries the burden of the smooth coordination and cooperation between the two governments. Being only in power at the regional level allows a party to shift the blame to the others. Being only in power at one level actually gives a partner in a coalition an extra instrument to put pressure on the other parties. Being in power at only one level is also a way to combine governing and opposition, and that can be an interesting situation for newer parties (in our case the greens and 89

Politics Beyond the State.book Page 90 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

the Volksunie) for whom the decision to govern is not easily accepted by the rank and file. It is therefore not possible to reach a simple conclusion about advantages or disadvantages of incongruent coalitions. Actually, incongruence alone does not explain the choices of the parties. The size of a party (and eventually its need to survive as an organization), its position in the party system (traditionally governing or occasionally governing), and some characteristics of the party system (like the presence of a pariah party) are variables that certainly have to be taken on board of a more elaborate and comparative theory of coalition formation in multi-level systems and of incongruent coalitions in particular.

90

Politics Beyond the State.book Page 91 Wednesday, December 5, 2007 8:41 AM

The Weakest Link? Interest Representation by Eurogroups in a Multi-level Governance System Joke Wiercx

Introduction The spread and intensification of governance in and beyond the state has not only had influence on the organisation and function of policy making actors, but also on the organisation of interest actors. As policies originate at different levels, it is important for interest organisations to be present ‘where the action is’. Interest organisations nowadays are not only active at the local/national level but also increasingly at the European level. Different forms of interest organisation in the European Union can be discerned. Besides the numerous consultants and direct firm lobbyers who are active at the European level, there are the traditional national organisations and since the beginning of the European integration process interest groups have also been organising in a transnational way. These transnational forms of organisations are generally called ‘eurogroups’. Ideally, these eurogroups are composed of members of all EU member states, but also more regionally based groups do exist1. Members can be organisations themselves, individuals or firms, although the first form is the most frequent. The multi-level governance context of the European Union is an important factor for the analysis of eurogroups. A considerable amount of research focuses on the question why and how eurogroups mobilise. Although different push- and pull-factors in the European integration process can be identified here, it has often been argued that the multi-level governance context has to be taken explicitly into consideration. The multi-level governance context of the European Union is not only important to explain the mobilisation of eurogroups, but also to understand their way of working. The European institutions themselves have often considered eurogroups as a means to overcome the democratic 1.

“A minimum criterion used by the Commission for the designation of an interest group as ‘EU level’ includes the involvement of members from three or more EU or candidate countries.” (Greenwood & Halpin, 2005:3)

91

Politics Beyond the State.book Page 92 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

deficit, since they are considered as key actors to increase consultation and participation of civil society. In the literature this role for eurogroups has been discussed at length. Not only arguments on the difficult mobilisation of eurogroups, but also arguments about the legitimacy of these groups have made several authors to conclude that eurogroups are only a weak form of interest representation and that other forms, such as national organisations or single firm representation are the better options to work at the European level. In this chapter, we will summarize and discuss the arguments on the strengths and weaknesses of eurogroups. Strength and weakness is a combination of several aspects. Ideally, an interest organisation has members who are satisfied with the internal decision-making procedures and the general programme of the organisation. On the other hand, the ideal interest organisation is also able to pursue its objectives, which means that it represents its members’ interests effectively and efficiently. Another important element would be the access of the organisation to political decision-making procedures and actors. To be strong, an interest organisation should ideally combine all these factors, although in practice this is very difficult. Consequently, in this chapter it is considered important for an interest organisation to have a good balance between the relation with its members and the interest representation towards the political actors. The literature on eurogroups is rather elaborate and very diverse. Most research on eurogroups is case study-based and consequently rather fragmented, or is concentrating on one type of eurogroup, mostly business interests or public interests. This makes a general discussion of eurogroups a bit difficult. Still, several arguments on the strengths and weaknesses of eurogroups can be found in the literature on different types of eurogroups. When there are arguments only referring to one single type of interest, we will say so. The chapter is divided in three parts. First, after drawing a general picture of the amount and variety of eurogroups present in the European Union, we will briefly discuss the different demand- and supply-side arguments for the mobilisation of eurogroups. In the first part we will also discuss the characteristics of the multi-level governance system and its consequences for eurogroups. The second part will discuss more in detail the ideas of the European Commission on European governance as formulated in its white paper (CEC, 2001e), and more precisely the role it sees for eurogroups in this debate. The arguments on the legitimacy of eurogroups will also be discussed in the second part. In the third part, after discussing the 92

Politics Beyond the State.book Page 93 Wednesday, December 5, 2007 8:41 AM

The Weakest Link?

legitimacy of eurogroups, we will deal with the central question of this paper: are eurogroups indeed a weak form of interest representation? The rise of eurogroups: push- and pull-factors and the particularities of the multi-level governance system

Push- and pull-factors in the mobilisation of eurogroups Eurogroups have been around since the very beginning of the European integration process. And although it is difficult to give exact numbers (different directories mention different figures), it is clear that the numbers have risen since these early days (Greenwood, 2003). That growth did not follow a linear trend, but jumped at every expansion of the competencies of the EU. Milestones in the evolution of eurogroups are thus the ECSC (1951) and EEC (1957) treaties, the Single European Act (1986) and the Treaty on the European Union (1992) (Göhring, 2003). Approximately 1500 eurogroups can be found today, representing all sorts of interests in the European institutions. Some authors argue that the mobilisation of eurogroups has now reached its upper limit. The sometimes highly specialized topics indicate that further mobilisation of interests is quite unlikely (Greenwood, 2003; Mazey & Richardson, 2001). If one looks at the landscape of eurogroups, it is clear that some types of interest are more present than others: 66% of all eurogroups defend business interests (Greenwood, 2003). This includes not only comprehensive cross-sectoral and sectoral eurogroups, but also a large number of highly specialised European business associations (Göhring, 2003). Least organised is the public sector, including territorial interests, with only 1% of all interest groups (Greenwood, 2003). Table 1 Classification of eurogroups according to types of interest represented (Greenwood, 2003: 19) Type Business Professions Trade union Public interest Public sector Total Total number

% 66 11 3 20 1 100 1450

93

Politics Beyond the State.book Page 94 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

The mobilisation of the different types of interests did not occur at the same time. Especially the business interest associations organised themselves at an early stage of the integration process. By 1962 already 270 European level business groups had been founded and in the build-up to the Single European Act and the Treaty on the European Union a new boost in the number of business groups was seen. Public interest groups, although some of these did mobilise already in the early days of the integration process, only started mobilising at a larger scale after the establishment of the Treaty on the European Union (Smets and Winand, 2000). Even though the European treaties seem to have had an influence on the timing and the type of mobilisation, it is still quite surprising that interest groups do mobilise at all at the European level. There are indeed quite some barriers making transnational mobilisation fairly difficult. Competencies of the European Union are also still limited in some policy areas, like for example social policy. Yet this did not prevent social interests mobilising at the European level (Cullen, 2004). There are also numerous practical barriers to cross. Not only the differences in the various national organisational forms and strengths, but also the differences in the relation between organisations and national governments can be obstacles for a good and smooth European cooperation (Helfferich & Kolb, 2001). We should therefore in the first place explain why this mobilisation of eurogroups did occur. In the literature on European interest representation, supply side arguments (push-factors) and demand side arguments (pull factors) are given to explain the eurogroup mobilisations. We will discuss briefly both sets of arguments. Supply-side arguments (push-factors) The explanation of the emergence of eurogroups can rely on two theoretical approaches for the study of social movements: rational choice and neo-institutionalism. Rational choice explanations for collective behaviour refer to the selective incentives as motivations for individual actors to participate in an organisation (March & Olson, 1989; Jordan & Maloney, 1996). Consequently, when looking at eurogroups, this theory suggests that individuals or national organisations will mobilise collectively at the European level on the basis of a cost-benefit analysis. The rational choice theory builds on the assumption that actors have exogenous preferences that exist independently of their environment. This has been contested by neo-institutionalist theories. Rational neoinstitutionalists on the one hand still recognise the exogenous preferences 94

Politics Beyond the State.book Page 95 Wednesday, December 5, 2007 8:41 AM

The Weakest Link?

of actors but state that these preferences are bounded by an institutional context, which means that only a maximization of their preferences within this context is possible (Bursens, 1998: 235). On the other hand, historical neo-institutionalists claim that “the preferences of the actors choosing to embark upon collective action at the EU level are considered to be endogenous, or developed within the institutional framework of the European Union” (Cram, 1998: 64) Whether the rational choice theory or the neo-institutionalist theory is used, the EU’s relevance for a group can always be considered to be an important push-factor for European level mobilisation of interests. This EU relevance does not necessarily have to be proven. Cram (1998) argues that “[t]he symbolic dimension of the EU may also constitute an incentive for groups to organise at European level (Göhring, 2003: 185). Moreover, mobilisation of eurogroups can also be triggered by the mobilisation of other eurogroups. In this case interest actors rather apply a risk avoidance strategy than a rational reasoning (Mazey & Richardson, 2001). Two other supply-side arguments, besides the rational choice and the neoinstitutionalist theories, are formulated in the literature on transnational interest mobilisation. They refer to the globalisation thesis and the group characteristics thesis. They both focus on the question why groups want to cooperate at an international (and in this case European) level. The group characteristics-thesis states “that group resources and group ideology, rather than common globalization forces, influence which groups actually participate in transnational networks” (Rohrsneider and Dalton, 2002: 516). This implies that there will be differences in the levels of activity in transnational cooperation not only between countries, but also within countries and within interest sectors. One interpretation of the globalisation thesis builds on the idea that interest groups will mobilise at an international level because national governments are not capable any more to deal with certain transnational problems (Rucht, 1999). This is certainly the case in the European Union, as member states did cede a number of competencies to the European institutions. Another interpretation of the globalisation thesis, mainly formulated by Keck and Sikkink (1998), says that international governance systems give opportunities to interest groups to circumvent national governments. This so-called boomerang strategy is however contested by authors like Rohrsneider and Dalton (2002). On the basis of a survey of environmental organisations, they conclude that “the groups that are most active in dealing with national governments through conventional

95

Politics Beyond the State.book Page 96 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

channels are also more likely to engage in each form of international activity” (p. 528). Demand-side arguments (pull-factors) Next to the push-factors playing a role in the mobilisation of interest groups at the European level, different pull-factors are also mentioned in the literature on eurogroups. Several researchers have indicated that the role of the European institutions, and especially of the European Commission, has been important in the mobilisation of eurogroups. The European institutions have their own agenda, and therefore they have their own needs vis-à-vis interest groups. The European Commission has been from the beginning actively stimulating interest groups to establish European level organisations. Three main reasons for this mobilisation are mentioned by several authors: the need for information, the need for legitimization and the Commission’s own demand for expanding competences (Aspinwall and Greenwood, 1998; Cram, 1998; Smets & Winand, 2000; Mazey & Richardson, 2001; Greenwood, 2003). First of all, eurogroups are a practical instrument for the Commission to receive direct and clear information on certain topics, which is important for the development of policies. Because of the technical nature of legislation and of the limited staff available to the Commission the European institutions are often dependent on the information given by interest groups (Smets & Winand, 2000). Second, eurogroups can also be used as an instrument to respond to the criticism of the democratic deficit. The European Parliament and the European Commission hope that interest groups, as potential links between decision-makers and citizens, can legitimize the actions of European institutions. They are conscious though of the fact that not all contacts with interest groups are necessarily legitimizing. “Being very close to an interest group is not considered to be acceptable when the external partner is from industry” (Kohler-Koch, 1997: 9). We will come back to this point further in this chapter. Thirdly, the pressure of eurogroups can be used as a means to increase European competencies. By organising platforms and subsidising some groups (mostly public policy interest groups) the European Commission did not only get involved in the mobilisation of eurogroups, but had also some influence on the form in which these groups developed (Cram, 1998). For example, different social eurogroups did organise themselves in an umbrella-platform under impulse of the European Commission (Geyer, 2002). 96

Politics Beyond the State.book Page 97 Wednesday, December 5, 2007 8:41 AM

The Weakest Link?

Characteristics of the multi-level governance system Several authors have focused their attention on the question whether European interest representation is evolving towards a corporatist or a pluralist system (see among others: Streeck, 1992; Mazey and Richardson, 1993). Especially the implementation of the ‘social dialogue’ did give the impression that a European corporatism was emerging (Greenwood, 1998), but empirical evidence for both interpretations is found. Michalowitz points out that “the empirical evidence suggests that neither of the two concepts is completely compatible with the European structures. A basic condition, essential to both pluralism and neo-corporatism, is the existence of a single point of reference: the ‘state’” (Michalowitz, 2002: 42). The European Union, with its multi-level governance structure, can hardly be called a ‘state’. Grande (1996) was one of the first to recognise the influence of the particularities of the multi-level governance system on European interest representation. He argues that “as a consequence of the new framework of multi-level decision-making, the logic of influence between public and private actors may be changing significantly” (Grande, 1996:331). An important consequence of the multi-level governance framework is the increase of access points for interest representation (Grande, 1996; Michalowitz, 2004; Kohler-Koch, 1997). There is not one central stateactor to which interest groups can express their grievances and try to convince of their opinion. Instead, interest groups have to take into account that several and different state-actors all have their own part in the decision-making process. Although this can open up opportunities for them, it is not always in favour of the interest groups because this also increases the complexity of interest representation. There is not only the complexity of the different levels of the multi-level governance framework (local, national and European) but also the complexity of the decision-making procedures themselves at the European level. The multiple arenas of policy making (European Commission, European Parliament and European Council of Ministers) often lead to unpredictable outcomes. While the Commission is known for its technocratic basis, the Parliament is lead by its democratic basis and the Council is known for its trade-offs in decision making. In each of these arenas different sorts of interests can have a different sort of impact. Business interests will more easily influence the European Commission with their more technically founded arguments, while citizen groups will have a larger impact in the European Parliament as they can emphasise their representative basis. (Greenwood, 2002; Kohler-Koch, 1997) 97

Politics Beyond the State.book Page 98 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Taking into account the multiple levels and multiple arenas, it is important for groups with the same interest to cooperate intensively at the different levels. Such cooperation requires a very good coordination of the actions undertaken by the interest groups at the local or national levels. Nevertheless, national and local interest groups all have a different background and work within different political contexts. This makes cooperation sometimes difficult, as opinions on certain topics can differ between these interest groups (Kohler-Koch and Quittkat, 1999; Helfferich and Kolb, 2001; Falkner, 2001). These difficulties in coordination, found in several case studies, have then led some authors to formulate this argument of ‘weak eurogroups’ (McLaughlin et al., 1993). This issue will be discussed further in this chapter.

Legitimacy of eurogroups Tackling the ‘democratic deficit’ by using eurogroups? During the 1990s, several events indicated a growing discontent with the European integration process. Consecutive difficulties with the ratification of the treaties of Maastricht, Amsterdam, and Nice and the degrading enthusiasm of the European citizens as measured by the consecutive Eurobarometers did uncover, among other factors, a democratic deficit in the European Union. In the literature, two main reasons for the democratic deficit in the European Union are being discussed: the mode of political representation and the nature of policy outputs. The mode of representation is problematic in the sense that only one of the four main decision bodies of the European Union – the European Parliament – is elected directly. Moreover, the turnout at the EP elections is low and campaigns focus hardly on European issues. Not all institutions are thus directly accountable to the voters and those which are, are not taken seriously (Majone, 1998). As for the nature of policy outputs, there is a problem with the fact that “EU directives and regulations promote wider and deeper markets, while providing only a truncated range of compensating and counterbalancing policies of regulatory protection or social welfare guarantees” (Moravcsik, 2002: 605). Conscious of these critics, the European Commission has tried to tackle the democratic deficit by putting emphasis on a more intensive involvement of the ‘civil society’. In 2001 the European Commission

98

Politics Beyond the State.book Page 99 Wednesday, December 5, 2007 8:41 AM

The Weakest Link?

published the ‘White Paper on European Governance’ (CEC 2001e). An important goal of the paper was ‘creating better involvement and more openness’. Among others, the white paper states that a larger involvement of civil society is needed as well as a more effective and transparent consultation. (CEC 2001e) The notion of ‘civil society’ has indeed become popular in European Union politics, but the definition is not univocal. “[D]epending on the approach chosen, civil society may be defined as a network of associations either located between the state and the private sphere, or between the state, the economy and the private sphere” (Göhring, 2002: 120). In the literature and even within the European Union both definitions are used. For example, in the DG Trade’s civil dialogue, the working definition of ‘civil society’ includes all public and private interest organisations. This makes the scope of who is included in the consultation of interests very broad. The definition as it is formulated by the Economic and Social Committee (ESC) and also used by the European Commission has a more limited scope, although it is still a broad definition. Here civil society organisations are defined as “…organisational structures whose members serve the public interest through discussion and function as mediators between the public authorities and the citizen” (Economic and Social Committee 2000:107). In this definition the concepts ‘civil society’ and ‘intermediary organisations’ are interchangeable (Göhring, 2002). In the White Paper on European Governance (CEC, 2001e) the Commission stressed two action points to increase the involvement of civil society: the adoption of a code of conduct that sets minimum standards for consultation, and the development of more extensive partnership arrangements in certain sectors. With a code of conduct the Commission wants to prevent a one-sided consultation of interests. This code of conduct, although brought back to a limited version of what was proposed in the white paper, has been established and is applied as from 2003 (CEC 2002a). The extensive partnerships should provide additional consultations with organisations and should guarantee transparent internal structures, openness and representativeness and prove the organization’s capacity to relay information or lead debates in the member states (CEC 2001e). Considering the rather negative reactions to the proposal of the extensive partnerships, this has been no further point of discussion (CEC, 2002b). Nevertheless, within some Directorate Generals one can see restricted consultation practices that could be described as de facto ‘extensive partnerships’. One example is the continuous dialogue between DG

99

Politics Beyond the State.book Page 100 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Development and the Confederation of European Non-Governmental Organisations (CONCORD). Greenwood and Halpin (2005: 16) argue that, although these de facto extensive partnerships pre-date the White Paper on Governance, “none have been rejected or weakened as a result of the Commission officially rejecting partnership arrangements following the Parliament’s expression of disapproval of them”. When it defined the modalities of consultation, the European Commission did take the particularities of eurogroups into consideration. For example, a consultation period of at least 8 weeks has been established. This must enable eurogroups to consult properly their members in the different member states (CEC, 2002a). Moreover, the European Commission says explicitly that the consultation of eurogroups is preferred above the consultation of national interest groups or individuals: “The Commission would like to underline the importance it attaches to input from representative European organisations” (CEC 2002a: 11). In its current ‘guiding principles’ to govern its relations with interest groups, the Secretariat-General of the Commission also expresses this preference for eurogroups (CEC 2006). We already noted earlier that the European Commission is a demanding party for the mobilisation of eurogroups. One of its main arguments is that the consultation of eurogroups contributes to the democratic value of decisions. The mobilisation of eurogroups and their privileged position in the consultation process is in this way seen as a possible solution to the ‘democratic deficit’ of the European Union.

Problems of legitimacy in the eurogroups The privileged position of eurogroups in the consultation procedures is however considered problematic in the literature on European interest representation. In the document on minimum standards of consultation, the European Commission states that not only the Commission itself has to be more transparent, but that openness and accountability also has to be assured by the organisations when they are seeking to contribute to EU policy development. Organisations have to be open on which interests they represent, how inclusive that representation is and how accurately they reflect those interests (CEC 2001e). In other words, the European Commission expects that interest organisations are representative and legitimate. Researchers have indicated several problems with the legitimacy and representativeness of eurogroups.

100

Politics Beyond the State.book Page 101 Wednesday, December 5, 2007 8:41 AM

The Weakest Link?

First, it is clear that a number of eurogroups are almost directly created by the European Commission.Examples of this ‘guided’ mobilisation of interest groups are found in all sorts of interests, whether these are professional, business, labour, public or territorial interests. The professional interest organisation CEPLIS (European Council of the liberal Professions) was the idea of a Commission official and was founded in 1974 (Greenwood, 2003: 128). The European Round Table of Industrialists (ERT), a business interest group with strong influence especially in preparation of the Single European Act (1986), mobilised on the initiative of the European Commissioner Etienne Davignon (Greenwood, 2003: 80). And also more recently the Platform of European Social NGOs, which was set up in 1995, started as an initiative of the European Commission proposed in the Green Paper of 1993 (Platform of European Social NGOs, 2003). This ‘guided’ mobilisation implies that these organisations did not emerge naturally out of the national civil societies. As such, one could question the legitimacy of these organisations. Do they really represent the constituency they claim to represent? The legitimacy of an organisation can be guaranteed in two ways. Firstly, formal legitimacy exists in the cases where the president and/or the board is elected democratically (Dastoli, 2002). The election of a board or a president gives it the powers to take decisions and to act in the name of the members of the organisation. There are no figures available on the number of eurogroups having an elected president and/or board, but Dastoli (2002) notices that most European public interest groups do not have one. Nevertheless, if the European Commission stresses the importance of the legitimacy of the eurogroups, it does not refer to this formal form of legitimacy as this does not as such guarantee a larger involvement of civil society. The second option to guarantee legitimacy is making sure that the group represents its constituency. This implies that within an organisation there are decision-making mechanisms that guarantee the participation of all members. Göhring (2002) claims that this type of representativeness is not desirable because it would impede these groups to be flexible enough to pick up and react to trends in civil society. She argues that exactly the flexibility to react to changes in society is necessary for an organisation to be representative. However, a distinction must be made between the representativeness of the grassroots support, the members of the organisation, and the larger target group the organisation tries to represent. The institutionalisation of decision making procedures is necessary for the representativeness of the members of the organisation, but will indeed not 101

Politics Beyond the State.book Page 102 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

take into consideration the representativeness of the larger target group of the organisation.2 Dastoli (2002) and Göhring (2002) do make these claims only for public interest groups, but there are indications that the same claims can be made for other interests. The institutionalisation and professionalisation of eurogroups, mostly under impulse of the European Commission, tempers the flexibility of these groups. Approximately two thirds of the eurogroups take the structural form of (con)federations of national (federal) organisations. This form obliges the organisation to develop formal decision making structures, which makes it slow in formulating policy positions (Greenwood et.al., 1999). This is a problem intrinsic to all federated organisations. But the fact that in eurogroups the member organisations have different national backgrounds and consequently normally work in different political contexts can complicate these decision making procedures. Furthermore the national members of the eurogroups are often also federations themselves, which means that first they have to consult their different member organisations before they can formulate their opinions on issues. One could however question this need for fully representative European interest organisations. On the one hand some interest actors, whether these are national interest organisations, firms or individual actors, try to by-pass the lack of flexibility in eurogroups. One finds an increase in direct eurolobbying by firms and in the establishment of ad hoc coalitions between different groups of interest actors (Mazey & Richardson, 2001). In this case eurogroups are non-crucial actors in the interest representation process. On the other hand, all actors do not always want to get involved in decision making processes of the eurogroup. Research of Warleigh (2001) unveiled that the internal governance of European public interest organisations (which he refers to as NGOs) has an elitist character, since the members are not actively involved in shaping policies, campaigns and strategies, but “most NGO supporters do not actually want to undertake such a role” (Warleigh, 2001: 365). In these cases there is no problem of representativeness as members do agree without giving an explicit consent.

2.

In this sense, the argument made by Greenwood and Halpin (2005) that “groups that advocate for constituencies, that are without the capacity to exercise accountability over representatives, authorise representatives or be present themselves” could only pursue a solidarity based legitimacy and not a representative based legitimacy, does not hold as we are only looking at the representativeness of the members of eurolevel organisations.

102

Politics Beyond the State.book Page 103 Wednesday, December 5, 2007 8:41 AM

The Weakest Link?

Not only federated eurogroups have problems with decision making structures. Greenwood (2003) points to the fact that some direct membership euro-organisations only have pseudo-representativeness. Especially in direct membership business organisations one can find internal procedures where contributions are dependent on the size of the company and which favour the largest contributors. Bouwen (2003) speaks in this context also of ‘fig leave’ associations. Besides the lack of formal legitimacy, the lack of flexibility and the difficulties with internal decision making rules, several authors mention the absence of a true European civil society as a problem for eurogroups (Marks & McAdam, 1999; Göhring, 2002; Michalowitz, 2004; Koopmans, 2004). In its stimulation of eurogroups, the European Commission expresses its belief in more debate and dialogue as a solution for more democracy in the European Union, but Koopmans (2004) shows in his empirical study of claim-making in newspapers of different European countries that the debates on European integration as they appear in the media are very elitist. One could expect that fully Europeanised policy domains are the first to be discussed in a European public domain. Yet it are the least contested policy domains such as monetary policy that are the most Europeanised. As such, a real European civil society does not seem to exist. The belief of the European Commission in more debate and dialogue as a solution to obtain more democracy is indeed a rather elitist approach (Göhring, 2002).

Eurogroups as the weakest link? We have identified five problematic points concerning the legitimacy of eurogroups: the problem of a ‘guided’ mobilisation, the lack of formal legitimacy, the lack of flexibility, the difficulties in internal decision making rules and the absence of a true European civil society. Besides these critics related to the legitimacy vis-à-vis the European institutions in general and the European Commission in particular, other problems with eurogroups as interest actors have been mentioned in the literature. One of these is the fact that European interest groups are ill-resourced to perform their task properly3. Compared to most national interest organisations, eurogroups indeed operate on a much more restricted resource base. Yet this statement must be qualified. Eurogroups do often 3.

See among others McLaughlin et al. (1993) on business groups and Geyer (2001) on social interest groups

103

Politics Beyond the State.book Page 104 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

also have a more limited role than national organisations. They specialise more in political representation, while national organisations also offer all kinds of services to their members. Especially European business and professional organisations have this more restricted task of pure political representation. They also face another limitation. Although the general claim is that public interest groups have fewer resources than business and professional interest groups, a survey of Greenwood et al. (1999) has shown that European business and professional interest organisations are not so wellof as public interest groups, not only in purely financial terms but also in the number of staff employed. Yet some public interest groups are dependent on funding by the European Commission which can then reduce their room for manoeuvring (Geyer, 2001; Union of International Associations, 2002). A second problem for eurogroups is the multi-level governance system that complicates the influencing of policy decisions. We already mentioned the fact that the multiple levels and multiple arenas in the European Union make it rather difficult, if not impossible, to dominate the policy debate and to assure the outcome of the policy decision. Exactly because of this complexity of the multi-level governance system and of the structural problems eurogroups face in their internal decision making procedures, quite a large group of non-collective actors have been created. Using inhouse lobbyists or consultancies, individual or national interest actors can easily by-pass their eurogroup if their positions on certain policy proposals differ (Greenwood, 2003). All these arguments against eurogroups together give the impression that eurogroups are indeed weak and not very appropriate interest actors. But this is a fairly one-sided image of eurogroups. The broader context of the multi-level setting of interest representation in the European Union has to be brought into the picture to fully evaluate the role of eurogroups. Comparing the different forms of interest representation (individual action, collective action, third party) without looking at how they interact with each other in a multi-level governance system will always lead to the conclusion that European interest representation is not effective. National interest actors can choose between grosso modo three possible influence routes. First there is an influence route via the domestic channel where the interest actor tries to convince the public authorities to defend its position. A second route is a direct supranational route via euro-level public actors, where the interest group tries to by-pass the domestic public actors. Thirdly, there is the indirect supranational or transnational route, 104

Politics Beyond the State.book Page 105 Wednesday, December 5, 2007 8:41 AM

The Weakest Link?

the route going via the eurogroups. These three routes do not have to been seen as exclusive routes, but can be combined by the interest actors (Beyers, 2002). Bouwen (2003) studied different organisational forms of representation within business interest and he argues that each form has its appropriate role in the process. He distinguishes three sorts of access goods, i.e. sorts of information that is crucial in the EU policy-making process and which interest actors can deliver. A first one is the expert knowledge, the technical expertise and know-how needed to decide on technical policies. Individual actors, like companies, or in lesser extent consultants, are most appropriate for delivering this kind of information. Mazey and Richardson (2001) point to the fact that European officials often directly contact the source of technical expertise on which the eurogroups themselves usually draw, although the general rule is that the eurogroups are consulted. Information about the domestic situation is the second sort of access good Bouwen distinguishes. In European legislation the national differences have to be taken into account, and therefore information on national particularities is needed. National associations are the most appropriate actors to deliver this kind of information. The third sort of access good is the information about the European situation. Not only information on national particularities, but also information on which impact a policy could have at the European level as a whole is important. Eurogroups could be considered as the most appropriate actors to deliver this kind of information because they can aggregate the information from their members. As a consequence they can enhance their potential to contribute to the implementation of EU legislation (Bouwen, 2003). Bouwen did concentrate his research on business interest, but this functional differentiation of interest representation between different forms of representation can also be found for other interests. Michalowitz (2004) interviewed mainly Brussels-based lobbyists with different backgrounds (different sectors, nationalities, forms of interest representation) on patterns of interest representation. On the basis of these interviews, she confirms the existence of the functional differentiation between the different forms of interest representation also for other interests than business. According to her interviews, interest representation in Brussels is very goal-oriented. She states that “based on these responses we can suppose that private actors use in-house lobbyists, consultants, eurogroups and –to a lesser extent- regional offices simultaneously and in combination in order to arrive at an optimal level of satisfying the institutional demand” (p.164). Mazey and Richardson do agree with this as they argue that “The more complex the multi-venue 105

Politics Beyond the State.book Page 106 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Euro-policy game becomes, the greater the need for flexibility and manoeuvrability by interest groups if they are to create policy win situations. Concentrating resources on one type of (federal or confederal) organ of representation is a very risky strategy and likely to fail” (p. 227). The extent to which this functional differentiation takes place, is probably dependent on the policy sector. Falkner (2001) argues that public-private interactions at the national level will show some cross-sectoral differences since the European Community is a particularly strongly sectoralised system. Similarly, one could expect a growing sectoral difference in European interest representation in general. Consequently, dependent on the degree of Europeanisation of the policy sector, the different forms of interest representation will be more or less developed (Kohler-Koch, 1997). In his research on the European adaptation of domestic interest associations, Beyers (2002) did not find exactly this sectoral difference, but he concluded that “many actors are indeed trying to diversify their strategies, but also that many are not successful in this” (p. 607). He found that especially those interests which have already “privileged access at the domestic level have the opportunity to invest in euro-level networking” (p. 608). But, whether differences can be found in policy sectors or in the embeddedness in domestic interest representation, it does mean that there is a clear functional differentiation of interest representation in the European institutions.

Conclusion Can we conclude that eurogroups are indeed a weak form of representation in the European system? We have shown in this chapter that eurogroups are a relatively new form of interest representation. Their emergence and development can be explained both by the need for specific forms of representations in a political system like the EU and by some explicit attempts by the European Commission to create channels for consultation. Eurogroups also have – as federations of organizations – a very specific form and they have to operate in a very specific environment: multiple layers and multiple and varying political actors. That means indeed that the life of eurogroups is not very easy. They have to be at the same time legitimate and representative organizations for their members and effective defenders of the interests of their members. That might make them one of the weaker links. Yet in a system like the EU, 106

Politics Beyond the State.book Page 107 Wednesday, December 5, 2007 8:41 AM

The Weakest Link?

where a wide variety of inputs and expertise is needed, they can play a significant role. That role is certainly not that of the strongest link between interests and European decision making, but maybe strong links are not something to be expected in the complex and multi-layered setting of the European Union?

107

Politics Beyond the State.book Page 108 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State.book Page 109 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement: The Determinants of Outside Lobbying Christine Mahoney

Introduction1 Interest groups play a critical role in modern democracies by acting as ‘mobilisers’. By monitoring policy developments and mobilizing the public when policymakers act contrary to their wishes, advocates2 promote citizen engagement. Arnold refers to actors acting in this regard as ‘instigators.’ He argues the availability of an instigator to help reveal citizens’ stakes in an outcome affects the probability that citizens will notice if a policy is being proposed that is important to them (1990). Through ‘outside lobbying’ tactics such as mobilizing letter writing campaigns, media work, public advertising campaigns in print, radio and TV media outlets, organizing grassroots meetings, demonstrations and other outreach programs, organized interests foster citizen engagement. Thus citizens need not follow the policy developments of every proposal, but if advocates are monitoring the halls of government and notifying their constituents; citizens can become engaged on those issues that are most important to them. In short, organized interests often serve as instigators through the use of ‘outside lobbying’ – playing Socrates’ proverbial gadfly to the sleeping mule that is the public.

1.

2.

Thank you to Frank Baumgartner, Donna Bahry, Jeffrey Berry, and the participants in the CONNEX meeting Bled, Slovenia, May 2005 for comments on an earlier draft of this paper. Thank you also to the US State Department for the Fulbright Fellowship that supported the data collection and to the Pennsylvania State University for the Fellowship that supports the analysis. Much debate surrounds the proper term to use when studying lobbying. Interest groups, organized interests, civil society organizations, and lobbying groups, all connote some type of ‘group’ leaving out the other important players in all lobbying communities such as individual firms, institutions, other governmental units and lobbying, law and PR firms. The term ‘advocate’ – any entity attempting to influence the policymaking process – successfully captures all these actor types.

109

Politics Beyond the State.book Page 110 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

The degree to which civil society organizations perform their role as ‘mobilisers’ varies. I argue that three levels of factors must be considered in an explanation of that variation: institutional, issue and interest characteristics. The institutional structure of a political system can facilitate or constrain the use of outside tactics. The nature of the issue – for example whether it is highly salient or arcane – also affects the ability of advocates to engage in outside lobbying tactics. Lastly, advocate characteristics – i.e. whether it is a mass membership organization or a corporation – also play a role in the decision to ‘go public.’ Previous literature on EU lobbying has suggested advocates are generally not employing tactics that lead to mobilisation beyond Brussels (Greenwood 1997, 2000; Pedler & Van Schendelen 1994; Van Schendelen 1993; Pedler 2002; Bouwen 2002; Rucht 2001; Watson & Shackleton 2003). But little empirical research has been done to systematically measure this general observation, nor to assess the determinants of the phenomena (for an exception see Beyers 2004). Literature on the lobbying community in the United States has empirically documented the use of outside tactics as well as some of its determinants (Scholzman & Tierney 1986; Baumgartner & Leech 1997; Kollman 1998; Goldstein 1999; Gerber 1999), but without a comparative framework we are unable to discern the role of institutional structure in the outside lobbying process. Thus, the primary research question is: What determines public mobilisation by interest groups? This, in turn, leads to many other more specific inquiries: Do some types of groups mobilise the public more than others? Do lobbyists on certain issues exhibit a higher degree of outside tactic usage? How do advocates in the US and the EU systems compare – do we see more or less citizen mobilisation by interest groups? In order to answer these questions and systematically assess the role of institutional structures, issue specific factors and interest group characteristics on the decision of lobbyists to mobilise the masses; I present data from a larger project on the lobbying activities of 149 advocates in the United States and the European Union active on a random sample of 47 issues. After briefly touching on the literature discussing the role of social capital in modern democratic societies, I review the literature on EU and US lobbying. I highlight the expectations regarding the influence of 1. institutions, 2. issues and 3. interest group factors on the decision to outside lobby and in doing so engage the public in the policymaking process. Regarding the first I discuss the necessity of a comparative research design and the specific benefits of the US-EU comparison. Regarding Issue and Interest Group characteristics, drawing on previous

110

Politics Beyond the State.book Page 111 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

literature I build the case for considering both of these factors in addition to the institutional structures to work toward a complete model explaining advocacy behaviour. Outside lobbying is not always an optimal strategy or even feasible, issue and interest characteristics are fundamentally important in determining if a given advocate uses outside lobbying tactics on a given policy debate. The second half of the paper describes the research design and presents the data and empirical findings.

Institutions, issues and interests: factors influencing the decision to mobilise the people There is little scholarly debate as to whether ‘social capital’ is normatively a good thing. Rather the questions surround: is it on the decline? Or the rise? Where does it fare better? And worse? What factors promote it? (Putnam 1993; Berry et al 1993; Mintrom 2001) Considering a significant amount of research on the positive effects of social capital, this paper assumes social capital – or civic engagement – to be a positive force in a healthy democracy. In addition, I proceed from the argument, as laid out in the introduction, that interest groups foster civic engagement through outside lobbying tactics – aimed at mobilizing the public to influence the policymaking process. If we assume that social capital is desirable and that organized interests can foster it through their outside lobbying campaigns, the question then becomes; what influences the lobbyists’ decisions to go outside? I argue that three levels of variables are critical in understanding the decision to use outside lobbying tactics: institutional structures, issue specific characteristics and interest group characteristics. I discuss each of these levels in turn.

Institutions First, it is important to recognize why a comparative research design is critical for researching the effect of institutional design on outside lobbying. The EU is a unique system in that there is a great deal of institutional variation within the system across the three primary political institutions. This range of institutional design makes the EU an opportune venue to research the effect of democratic institutional design on advocacy. In the larger project from which this paper stems, I exploit the variation in democratic institutional design (the European Parliament – popularly elected; the European Commission – not elected but responsible to the EP;

111

Politics Beyond the State.book Page 112 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

the European Council – not elected and not responsible to an elected institution) to analysze its effects on a range of other advocacy activities. However, when it comes to outside lobbying, by its very definition – being outside the halls of government – it is nearly impossible to determine the target of outside lobbying and thus impossible to parse out the effects of the specific EU institutional differences. (One could hypothetically consider the timing of outside lobbying campaigns, in relation to the stage of the legislative process, i.e. if an ad campaign coincided with a vote in Parliament, then the target could be deduced, however the Council and the EP work in parallel, making the task extremely difficult). To assess the effect of instutional system characteristics on outside lobbying, it becomes critical to do so by comparing across systems. A number of recent scholars have come to a similar conclusion after assessing the state of the literature. In Woll’s review of the literature she concludes “In all of these studies, there is a consensus that EU lobbying cannot be understood without looking at the institutions and policy context in which groups are trying to act….Increasingly, scholars have thus turned their attention to cases beyond Europe to understand more about EU lobbying” (6, 2005). It is not difficult to make the case that to study the effect of institutions; you need variation in institutions, which is lacking in a single-system-study. But the question then becomes: which systems to compare and the answer for many is the EU with the US. Michalowitz makes a strong case for a EU-US comparison in researching influence, arguing that “Firstly, the question as to why EU lobbying may be less influential than it appears from literature is difficult to answer from looking at the EU alone…Since US literature provides data already, it appears most useful to begin with a comparison of EU and US lobbying before going into more detailed studies of EU interest group influence” (2005: 2). In addition: “The availability of US literature on the topic is the second reason to decide for a comparative analysis of the EU and the US” (ibid). And of course she notes there is the simple fact that observers – academic and practitioner alike – are increasingly making the comparison and noting similarities. This can be seen in countless magazine, news articles and reports (Euractive, “EU and US approaches to lobbying” 170205; Legal Week, “Playing the Lobbying Game in the EU and US” as well as a number dealing with regulating lobbying and drawing EU-US parallels: CEO “Curbing Corporate Lobbying Power” 251004; Malone 2004). In short, lobbying in the EU and the US appear similar. Woll similarly echoes that “a comparison between the two largest lobbying industries, Washington, D.C. and Brussels” is justified in that “lobbyists have to interact with a fixed set of institutions that are comparable in terms 112

Politics Beyond the State.book Page 113 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

of the roles they play during the policy process” (2005: 3). In addition to the strong similarities in the group systems, Kreppel (2005) analyzes the legislative systems in detail concluding: “This brief comparison between the two legislatures demonstrates that, despite the common assumption that the two are fundamentally different, there are some important similarities that are generally ignored in analyses of the EP that use the legislatures of the EU member states as the basis for comparison. Moreover, the similarities that exist between the EP and the House of Representatives are some of the most familiar aspects of the American legislature that are generally thought to set it apart from its democratic cousins in Western Europe, suggesting that comparisons between the EP and the parliaments of the EU member states may simply be misguided” (7, 2005).

Menon, in a paper calling for caution among scholars that may be tempted to simply transplant American theories and methods to studying politics to the EU, concedes that “The tendency to compare the EU and US is easily understandable. After all, they share several traits, including: potentially comparable institutional landscapes; apparently parallel developmental trajectories and strikingly similar institutional dynamics stemming from comparable federal structures” (2005: 3). In sum, the US and the EU share numerous similarities, in their institutions, norms, and lobbying communities. Moreover, the wealth of scholarly treatment of lobbying in the two systems provides a strong foundation for pursuing comparative research. This is not to say, however, that there are not differences – there are; their existence is the very advantage of the comparative design. Now that we have determined the need for a comparative framework for considering the question of the effect of institutions on outside lobbying, and built the case for an EU-US comparison specifically, I now turn to discussing the specific institutional factors that should have a bearing on whether advocates seek to mobilise the public as part of their lobbying strategy. In recent literature on EU lobbying and the fairly young literature considering lobbying in a transatlantic perspective specifically, two key institutional factors emerge as important: the method of selecting officials and the presence of a broad reaching media system. I discuss each of these in turn, how scholars have suggested they should influence the use of outside lobbying, and finally, my hypotheses related to each. First, there is the democratic accountability of policymakers – or the selection process. While there is variation on this variable within the EU, with the EP being more democratically accountable then the Commission and Council, the EU institutions are largely not democratically accountable to the people – as is often lamented on the large literature on the democratic deficit. And even the democratic accountability of the EP, 113

Politics Beyond the State.book Page 114 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

which is directly elected, is considerably reduced due to the use of party lists by the member states during EP elections (Kreppel 1999). In addition Princen and Kerremans note that the EP “is largely shielded from direct popular control because EP elections are usually decided on domestic themes and popular interest in the EP’s work remains low between elections” (2005, 8). In the US on the other hand, members of both the House and Senate are highly aware of their electoral vulnerability and are thus highly accountable to the people in direct elections. The democratic accountability of an institution will influence which type of tactics a lobbyist pursues on an issue. Some lobbying tactics are designed to evoke the will of the people and tap the fear of policymakers accountable to those citizens. Other strategies are designed to convey technical information about policy proposals. Lobbyists are more likely to employ the former strategy if they are active in an arena were the officials are driven by the re-election motive. That is, I hypotesize that outside lobbying tactics such as grassroots letter writing campaigns, advertising, press releases, press conferences, or protest activity should be more prevalent when lobbying in a venue that is highly democratically accountable. Second, many observers of EU politics have noted the lack of a pan-EU media system, or what some have termed a ‘European Public Space.’ Due to the 19 official languages of the EU, and the lack of any wide-spread EUnewspaper, or television channel, there is no simple way to directly relay information about the EU to the people. It has to be funnelled through the national outlets which vary in their interest and their spin of EU affairs. Rucht in his study of the lobbying and protest strategies of environmental groups notes “Virtually all communication via mass media remains within the boundaries of national languages and discourse. Journalists located in Brussels report only for the national papers of their respective home countries” (2001: 139). Princen & Kerremans suggest that “the EU lacks an integrated ‘European Public Space.’ As a result, there are no or very little European news media and European-wide debates, especially when compared with the media and public debates in the EU’s member states…this puts those groups at a disadvantage that rely on direct political actions to mobilise public opinion and the media” (2005: 8). Similarly Saurugger notes that “What seems to be missing is a European demos with a shared identity, a common deliberative forum and an open system of communication” (2005: 6). Michalowitz sees this lack of a pan-European media space as a result of a lack of a European public: “Outside lobbying is also the reason for a strong significance of the media in US lobbying. Media are needed to create a public for the interest represented…EU lobbyists

114

Politics Beyond the State.book Page 115 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

rarely concentrate on media lobbying, most likely because no European public exists” (2005: 10). Considering previous literature and the logistics of lobbying I hypothesis that certain outside lobbying tactics such as issue advertizements (in print, radio or TV), press releases, or press conferences should be more prevalent when lobbying in an institutional setting with a broad reaching media system. However, not all outside lobbying tactics should be suppressed by the lack of a media machine, specifically mobilizing letter writing campaigns of organizational members and the mass public should still be possible. Both of these tactics can be coordinated at the EU level and the mobilisation of the people carried out by the national associations. Neither requires complex media campaigns and either would foster citizen engagement. In sum, two factors of the institutional setting are likely to influence advocate use of outside lobbying: the democratic accountability of officials and the presence of a broad media system. Knowing the institutional system within which an advocate is lobbying, however, is not enough; in addition we must consider the characteristics of the issue at hand as well as the type of advocate and their resource set. I discuss these factors below.

Issues The second source of variation that influences the use of outside lobbying is at the issue-level. Issue-specific characteristics shape the way advocates carry out their advocacy strategy. In any society the number and range of social problems which the government might attempt to solve – thus forming a “political issue” – is enormous. In the early years of the European Community, the supranational institutions had control over only a limited range of policy issues, mainly those dealing with the common market. However, as the vast literature on European integration demonstrates (see Sandholtz & Sweet 1998 for an overview), over the past fifty years the EU’s competencies or political control has expanded significantly. Today the EU has some level of control over a wide range of policy areas, Sbragia notes “Agriculture, environmental protection, international trade, the movement of goods, services, labour, and capital across boarders, justice and home affairs and foreign and security policy are all affected by the decisions taken in Brussels” (2003: 113). The Commission has established over 25 Directorates Generals to deal with the growth in the EU’s competencies and each of those DGs houses countless issues.

115

Politics Beyond the State.book Page 116 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

A parallel trend is evident in the American system, well documented by Baumgartner and Jones (1993, 2002, 2005). They show that over the postwar period, not only has the volume of issues exploded dramatically but so too has the range of issues on the American agenda. Every issue has characteristics that influence both if interest groups mobilise and their advocacy activity if they decide to become engaged in a political debate. The fundamental power of issues to influence advocacy however has not been fully recognized due to the tendency of previous research to focus on single issues in the case study framework. Looking at advocay on only a single issue holds issue-related determinants of advocacy behaviors constant. Moreover, there is an extreme danger in extrapolating the conclusions that authors reach on case studies to the phenomena of lobbying in gernal due to the strong tendency for scholars to select cases that are ‘interesting’ ‘important’ or ‘visible.’ The very cases that are often selected by scholars for case studies are issues that have extreme values on the important issue variables that influence lobbying. That is, studies on GMOs, abortion rights, or terrorism are interesting reads becuase they are not the norm – but it is the norms and the patterns that we must seek to understand as social scientists. To do so requires that we look at the ‘little’ issues and the not-so-interesting issues, as well as the visible ones. To study the effect of issue characteristics on advocacy in a way that is generalizable to the system at large requires a research design with variation in issue characteristics and that the issues be a random sample. As mentioned in the introduction the research presented here is based on a random sample of 47 issues – 21 in the US and 26 in the EU. Drawing a random sample of issues is difficult in that the universe of issues is undefined: there are theoretically an infinite number of issues that could rise to the political agenda. Using lists of introduced bills or hearings are faulty routes because those issues have already received a considerable level of political attention and they are institution-specific. To capture issues being pushed through all institutions and their sub-units, as well as the issues that never fully make their way onto the active political agenda, I rely on a random sample of advocates to identify the random sample of issues in my study. While students of the US system might have a good grasp of the process from watching Capitol Hill’s “How a bill becomes a law” in the EU there are at least three distinct policy-making processes: The Community Method – dominated by all three supranational institutions; the Coordination Method – a highly cooperative method including the Commission but being more dominated by member state governments in

116

Politics Beyond the State.book Page 117 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

the Council; and the Intergovernmental Method – which is largely member state run though the Council (Stubb, Wallace & Peterson 2003). Moreover, across these broad categories there are “over thirty specific and different legal instruments and procedures at the disposal of the Union” (ibid). Again to compare to the United States, instead of just acts, the institutions of the EU issue: regulations, directives, recommendations, general decisions, individual decisions, opinions, resolutions, declarations, communications and action programmes (Borchardt 1999). These various instruments vary in the degree to which member state governments are bound to implement them, but each can be powerful in that less-binding measures can be the first in-roads into supranational control over a new policy area. For example, in the 1970s the Commission promoted a coordination method on environmental issues which evolved into a binding EU policy in 1987 (Stubb, Wallace & Peterson 2003). Today the EU controls over 90 per cent of the environmental policy in the member states (Greenwood 1997). Thus, advocating for a communication from the EP to the Commission to consider introducing a policy proposal, may not in itself be a significant policy change but may be the catalyst for substantial policy change in the future, thus all issues now matter how “big” or “small,” no matter what EU instruments are involved, were allowed to fall into my random sample of issues. Following the initial issue identifier interview, in order to collect accurate contextual data on an issue, I carried out additional interviews, triangulation and collection of additional, publicly available information – that is I conducted small scale case studies on a large number of issues. In this way I was able to code my 46 issues for a number of critical variables that likely influence advocacy, and the decision to outside lobby specifically. First, is the scope or size of the proposed policy – the magnitude of the impact on individuals (Baumgartner and Leech 2001; Browne 1990). The larger the issue, and the bigger its impact on citizens, the more likely an organization might attempt to mobilise the masses. Second is the salience of the issue to the mass public (Kollman 1998). Clearly, the more salient an issue is to the people – the more they are engaged on the topic and following its developments – the more likely an advocate will go ‘outside.’ Third, the presence of a focusing event is an important factor influencing advocacy (Kingdon 1984). If a crisis has occurred related to the policy, this event can crystallize the need for policy action and create a concrete example with which to mobilise people. The presense of a focusing event should increase the likelihood an organization will engage in outside lobbying. Fourth, the level of conflict also is a determinant of character of 117

Politics Beyond the State.book Page 118 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

lobbying (Salisbury et al. 1987). The hypothesized impact of this variable is that the more conflict the more outside lobbying. A high conflict issue may compel one side of the debate to outside lobby – those that want to expand the scope of conflict (Schattschneider 1960). And the media always like controversy, it gets ratings. While the opposition may not like the coverage, once the scope of conflict is expanded, they will likely also have to go public to combat the other side’s outside communications. Finally, the history of the issue is important. New issues have more room for manipulation of the dimensions of debate and alliance and opposition patterns, compared to older re-occurring issues (McKissick 1995). A new issues, that is a ‘hot’ topic may be used by advocates to spark the interest of the people, thus making outside lobbying a good tactical option. In sum, it is critical to consider the set of issue-specific variables of scope, salience, focusing events, conflict and reoccurrence when investigating the determinants of the decision by advocates to lobby outside.

Interests The third level of analysis provides the final set of independent variables that influence the use of outside lobbying. Organizational characteristics also have an impact on the decisions of advocacy organizations (Scholzman & Tierney 1986; Dür 2005). The range of interests active in the EU has mushroomed along with the expansion of the polity’s competencies. The wide range of group types has been well documented (Greenwood 1997; Balme & Chabanet 2002; Mahoney 2004). Groups vary not only by group type but by a number of factors and each of these have implications for advocacy activity. First, the type of organization, and thus its goals or purposes, determine the whether the entitiy can engage in outside lobbying (Clark and Wilson 1961; Beyers 2002). For example a sectoral asociation of chemical manufacturers might be much less happy with their Washington or Brussels representative organizing a blitz media campaign than an association of green activists. Second, the level of financial resources is a major determinant of outside lobbying. It plays a critical role in the type and number of outside lobbying tactics an organization can engage in (McCarthy & Zald 1978; Schlozman & Tierney 1986). Staff size is a good indicator of the level of financial resources an advocate controls, establishing a capital office (DC or Brussels), and staffing it is an expensive undertaking. Staff has also been shown to highly correlate with other, more difficult to gather indicators of financial resources like annual budget or money spent on lobbying.

118

Politics Beyond the State.book Page 119 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

Third, the size of an organization’s membership is also an important determinant of their tactic options (Bacheller 1977; Kollman 1998). If an actor has a broad membership it is much easier and more likely effective than if an organization is comprised of only a few hundred individuals. Fourth, it is also important to consider the structure of an organization. Federated structures or organizations with local chapters should be better able to reach out to their members, by working through their field offices. In sum, a range of interest-group-specific variables including monetary and membership resources as well as group type and organizational structure can be expected to affect the decisions to go ‘outside’. Institutional, issue-specific and interest-group factors are all critical pieces of the advocacy puzzle and thus each will be measured and analyzed so that a complete image of influences on outside lobbying can be constructed. For clarity, the independent variables for each level are listed below.

Institutional Level • The degree of democratic accountability (low-high) • The presence of a broad reaching media community

Issue Level • • • • •

Scope (size of the population affected) Salience (level of media coverage) Focusing event (presence of) Conflict (level of) History of issue

Interest-Group Level • • • •

Financial resources (staff) Group Type (i.e. Citizen, professional, trade, union, corporation, etc.) Organizational structure Membership size

119

Politics Beyond the State.book Page 120 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Research design and data As mentioned, the data presented in this paper stem from a larger project aimed at analyzing the determinants of each of the stages of the entire advocacy process. I carried out the American interviews during 2002 under the Baumgartner, Berry, Hojnacki, Leech, and Kimball project3. The European field work was carried out during 2004-5. The American actors were randomly drawn from a database created from the Lobbying Reports for 1996 that were filed with the Secretary of the Senate compiled by Baumgartner & Leech (see Baumgartner and Leech 1999, 2000). In the European arena, I developed a sampling frame from the 2004 Registry of the European Parliament, the 2004 Commission registry of civil society organizations (CONECCS) and the 2004 European Public Affairs Directory. The design consisted of randomly drawing advocates from the respective universes of lobbyists. These randomly selected actors were then asked to identify the issue they were most recently working on, forming the random sample of cases. Then snowball sampling was employed to gather information on the other major actors on that case. Other major actors could be any type of actor, if an official was named as being an active advocate on the issue they were allowed to fall into the sample, thus the sample includes a wide range of actor types in both polities including citizen groups, trade and professional associations, business groups, lobbying, PR and law firms, institutions, individual corporations, and policymakers, though to a greater extent in the US sample. This data collection process led to a random sample of 21 issues in the US and 26 issues in the EU4. Interviewees are asked about the background of the issue they were working on, the other major players on the issue, the tactics there were engaging in, coalition participation, the arguments they were employing, the opposition they were facing, the allies they found to support their cause, among others. The coding of their tactical strategy forms the basis of the data for this paper. Each interview transcript is coded (0 – not

3. 4.

Advocacy & Pubic Policymaking Project – NSF grants # SES-0111224 and SBR9905195 A list of the sample of issues can be found at: http://www.personal.psu.edu/users/c/x/ cxm548/

120

Politics Beyond the State.book Page 121 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

mentioned, or 1 – mentioned) as to whether the advocate reported using a range of outside lobbying tactics including: 1. Grassroots mobilisation of the organization’s membership (contacting members, encouraging them to write, email, call, attend town meetings or otherwise communicate with their elected representative); 2. Grassroots mobilisation of the mass public; 3. Media usage (issuing press releases, organizing press conferences, talking with the press, doing interviews); 4. Issue Advertisements; 5. Public education campaigns/ PR (coordinating large-scale public relations strategy including multiple routes to communicate to the public); 6. Protest, demonstration or rally; and 7. Op-ed or editorial (writing op-eds, encouraging elite organization members or other experts to write op-eds). These are the very types of strategies that organizations can use to influence policy making but that simultaneously promote civic engagement and cultivate social capital.

Additional data collection from publicly available sources In addition to data collected through the in-person interviews I also collected information on the individual organizations and issues. In the US I gathered information on group type, founding date, membership size, membership type, staff size, and annual budget from Associations Unlimited – an online directory of Washington organizations. For groups in the EU, I gathered information including group type, founding date, type of membership, and membership size from their websites. As additional measures of resources I coded if a group appeared in the EP registry and if so how many representatives of their office held passes to Parliament; I also coded where the group appeared in the CONECCS database and/or the European Public Affairs Directory. On issues, in the US, I monitored Roll Call and the Washington Post and conducted research on the websites of the House of Representatives, Senate, administration, any relevant agencies and the Library of Congress’ legislative tracking system. In the EU, I monitored the European Voice and Euractive weekly as well as researching the issues on the Commission archives and EP Legislative Observatory OEIL as well as Pre-LEX. This issue research was the basis for the coding of the issue variables of Scope (coded: 0-impacts small sector, 1-impacts large economic sector, 2-impacts multiple sectors and 3-has pan-EU or pan-US impact ); Conflict (coded: 0-

121

Politics Beyond the State.book Page 122 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

only one perspective or view point on issue, 1-multiple viewpoints but not directly opposed, and 2-directly opposing viewpoints); Presence of a focusing event (0-absence, 1-presence); and the history or the issue (coded: 0-for reoccurring or ongoing issue, and 1-new issue). The hypothesized directions are: the larger the Scope the higher the probability of outside lobbying; the higher the conflict the higher the probability of outside lobbying; the presence of a focusing event the higher the probability of outside lobbying and the newer the issue the higher the probability of outside lobbying. A measure of salience was collected for each American issue through Lexus-Nexus searches of the “major papers” – a search of the 50 largest papers in circulation – for the time period of the research. For a salience measure of the EU issues, I conducted Lexus-Nexus searches of the Financial Times and archive searches of the European Voice. These are selected as indicators of salience; it is simply not possible to monitor media coverage of 26 issues in all 19 of the EU’s official languages in the major papers of all 25 of the member states. Therefore I have chosen the FT and the European Voice as measures of media attention to the issues, both are widely read in Brussels and are considered an unbiased presentation of EU news; “With few exceptions (such as the Financial Times or the BrusselsBased European Voice) the media invariably view EU developments through a national prism” (Watson & Shackleton 2003). The expectation is that the more salient the issue to the public, as indicated by news coverage, the more likely it is that advocates will engage in outside lobbying on the topic.

Empirical findings Institutions As expected there is a marked difference in the use of outside lobbying in the US and the EU. Every outside lobbying tactic is used more by lobbyists in the US than the EU. Quite notable is the difference between grassroots mobilisation of organizational membership, at 25% in the US and 1% in the EU; and the difference between media usage, at 40% in the US to 23% in the EU.

122

Politics Beyond the State.book Page 123 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

Table 1 Outside Tactics by Institutional System US Grassroots – Organizational Grassroots – Mass Public Media Issue Ads Public Education/ PR Protest/ Rally Op-ed/ Editorial Total

EU

Number 16 3 26 4 5 1 9

Percent 24% 4% 40% 6% 7% 1% 14%

Number 1 1 19 2 3 1 2

Percent 1% 1% 23% 2% 4% 1% 2%

65

100%

82

100%

To look at the results a slightly different way, I calculated a summary variable – an additive index of the total number of outside lobbying tactics used by each advocate. Only 51% of advocates in the US reported not engaging in any outside lobbying activity, whereas in the EU outside lobbying was not employed at all by 76% of lobbyists. In addition, in only 6 times out of 82 did EU advocates pursue more than one outside lobbying tactic, while combined strategies were much more prevalent in the US at 23%. While the quantitative evidence is striking, the numbers alone cannot convey the difference in intensity of outside lobbying between the US and the EU. An advocate in the US and the EU may both be coded as using a media strategy, but for the American advocate this may be a concerted media effort, whereas in the EU it is a simple press release. In many instances media usage in the US is a full fledged campaign, as one lobbyist active on the debate over anti-terrorism regulations explained: We were constantly commenting on these issues in the media, speaking personally I probably gave a dozen newspaper interviews on this particular subject and half a dozen to a dozen radio interviews, I can’t remember if I did any TV on this particular issue and our executive director is the one that most often is the spokesperson for the organization, I’m sure you could say that he appeared on nationally televised programs a number of times. Also we’ve had editorial published on the topic. We relied pretty heavily to advance our agenda through the media not to the exclusion of trying to do some things with Congress.

123

Politics Beyond the State.book Page 124 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

A similarly aggressive media campaign was launched on the cloning prohibition case, as a professional association lobbyist described: We’ve had a very aggressive media strategy both as part of the coalition and as individual organizations. I could give you about three or four names to do a Google search on and you’d see there are about three or four of us that have become spokespeople for it. Like me becoming the spokesperson on behalf of ASRM or for the coalition and so just speaking personally I’ve done talk radio all over the country, I’ve done CNN a couple of times and we have a very strong relationship with a half dozen or so sort of leading national reporters that are covering the issue…. And you know we did an op ed campaign, and we did paid advertising, we did a lot of talk show appearances everybody from me to Christopher Reeve, and anybody in between, anybody that could get anybody on. And again the best people to do it are patients, or a really, really good scientist if you have somebody that is both a Nobel prize winner and can handle the media then that’s very effective.

Media is often an important part of an advocacy strategy in the US and this is facilitated by the presence of a vigorous media community in the capital. In the EU, the media-intensive tactics are more difficult to employ. A humanitarian organization active on the trafficking in children case explained that the Brussels press corps is often only covering the narrow topic of what Dossiers are up for a vote: “We haven’t done any press, Brussels media is interested in what’s on the legislative agenda so there is not much interest in this, but that is just a fact of what Brussels news covers; not about if the issue is interesting; but they just follow the legislative agenda and there hasn’t been much legislative action on this.” A trade association lobbyist was more openly skeptical of the use of pursuing a media strategy in the EU: Journalists come to us for quotes, so they can have the other side, to fact check, but I don’t go to them, what’s the point? Who are you going to reach with that? It is good for internal lobbying but not external, real lobbying, it shows your members you are doing something – they see oh look our organization is in the news, they are doing something. But that’s all it’s good for. No officials are going to read that and be convinced of that. I get these publications calling me all the time “we’ll give you a full page ad for the price of a half page!” – what am I going to do with that – it is a waste of money. Like this publication The Parliamentarian – they say “oh this is going to go to all MEPs, and the Council and the Commission” and I’m like it is going to go straight in the bin. They say I’m being cynical, but I can ensure you maximum 10, probably more like 5 MEPs will look at that, the rest it goes straight to the bin. They say it’s good for profile raising – that is a waste of money, that money would be better spent by organizing a cocktail or visiting more MEPS.

While the presence or absence of an extensive media system is important, electoral accountability also plays a role. The electoral connection was quite apparent in a number of discussions with American lobbyists. One

124

Politics Beyond the State.book Page 125 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

citizen group advocate active on the TANF reauthorization explained the importance of the constituency link: I always look at myself as the middleman. Part of my job is to go to the Members of Congress and the staff and tell them what people with disabilities need. The other side of what I do is to go to my constituency and say here's what you need to be telling them. They need to hear from you. It's important that I go and talk to them. There are times that I talk to them, but until they hear from you, people who are actually out there, living there, breathing this, eating and sleeping it every day of your life, they need to hear from you to better understand and to really bring home that impact of what's happening at the state level…

A professional association active on the math and science education debate echoed the point that mobilizing the constituents matters: One of the most valuable lobbying tools we have is our legislative action network and other associations have similar networks, we have 7000 of our members, not very many compared to our 160,000 total, but 7000 members are essentially dedicated to looking at action alerts when we send them out. Five or six times a year we will send out an action alert to 7000 of our members saying Congress is about to vote on this math and science education bill, here’s the background, here’s our position, here’s the key players, please go to our website, type in your zip code, read more about the issue, and if you support our position then send an email to your Member of Congress. And we give them a draft letter that we hope they will modify, some of them do, some of them don’t it’s a lot easier just to send what we give them. So the goal there is quantity, I’ve worked on the Hill long enough to know that if one month you’re getting two letters on math and science education and the next month you’re getting 12 letters well that’s something you’re going to check out and hopefully it will get to the Member of Congress…What we do know is that all those Members of Congress do listen to constituents…So that’s grassroots, getting your wide membership involved in the process, getting them to weight in.

A trade association described a similar on-line system to promote citizen engagement, as did a number of the other lobbyists interviewed: We’ve used our website, on it you can write your Member – we call it “Making Waves” – it is through Capitol Advantage a lot of associations use them, but if you go there you can click on the Recreational Marine Employment Act and here we explain what the issue is, and you can even tell a friend, and you can compose a letter and the letter is written for you and you can edit it and then you can either mail it or email it to your Member of Congress.

A professional association made the point about the importance of member mobilization on a more abstract level: Even if it’s just to compete with the other groups, we tell our members, we know the business groups are out there lobbying, we know the education groups and the environmental community are out there, that’s why [our community] needs to express it’s voice, if for no other reason then to be sure that our information gets fac-

125

Politics Beyond the State.book Page 126 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

tored in. You can’t be above it all. Senator Simpson used to say “Take part or be taken apart” and that is a little strong but that’s what happens. Unless you are getting your information in there, you’re going to be short changed somehow.

The level of constituent mobilization found among US advocates is markedly absent in the EU arena. Thus, we see support of both parallel hypotheses – media presence and electoral accountability – predicting less outside lobbying in the EU. However, the two tactics of grassroots mobilisation of organizational members and the mass public allow us to parse out the effects to a degree. That is, the two grassroots mobilisation tactics do not require a broad based media system. Both tactics could be accomplished by EU organizations encouraging either directly or through their national associations for citizens to write, email, fax or call policymakers. The fact that both of these tactics are used to a much smaller degree in the EU suggests the first institutional characteristic – the democratic accountability of the institutions – is also at play, and that the lack of media tools are not the only explanation for the limited use of outside lobbying.

Issues In order to see how the use of outside tactics varies by issue characteristics, I tabulated each of the seven dichotomous variables of outside lobbying tactics by: Scope, Conflict, the presence of a focusing event, the recurring / new nature of the issue and Salience. Tables 2 A-C report a subset of the results for the US system. The table shows, for example, that 25 percent of all groups used Grassroots tactics associated with their own members. Part A shows that this percentage was only 13 percent for groups involved in small sector issues but rose to 30 percent for groups involved in systemwide issues. The last column shows the Ns on which each of the individual sets of percentages is calculated.

126

Politics Beyond the State.book Page 127 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

Tables 2 A-C. Outside Tactics by Issue Characteristics in the US A. Scope Grassroots Members Public Media Ads PR Protest Op-ed Small sector 13% 4% 30% 4% 4% 0% 13% Large Sector 27% 5% 46% 14% 14% 0% 18% Multiple Sectors 40% 10% 20% 0% 10% 10% 10% System-wide 30% 0% 70% 0% 0% 0% 10% Total – N Total – percent

16 25%

3 5%

26 40%

4 6%

5 8%

1 2%

9 14%

N 23 22 10 10 65

B. Conflict

None Multiple P Opposing Total – N Total – percent

Members 19% 33% 27%

Public 5% 0% 5%

16 25%

3 5%

Grassroots Media Ads PR Protest Op-ed 33% 0% 0% 0% 5% 33% 33% 0% 0% 0% 44% 7% 12% 2% 20% 26 40%

4 6%

5 8%

1 2%

9 14%

N 21 3 41 65

C. Salience INSIDE – WP 0 stories 1-5 stories 6-20 stories 21 or more Total – N Total – percent

Grassroots Members Public Media Ads PR Protest Op-ed 9% 5% 18% 0% 0% 0% 0% 36% 5% 36% 9% 9% 0% 14% 25% 0% 38% 0% 0% 13% 13% 31% 8% 85% 15% 23% 0% 39% 16 25%

3 5%

26 40%

4 6%

5 8%

1 2%

9 14%

127

N 22 22 8 13 65

Politics Beyond the State.book Page 128 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

OUTSIDE – NYT 0 stories 1-5 stories 6-50 stories 51 or more

27% 14% 36% 20%

7% 0% 9% 0%

23% 36% 55% 80%

7% 7% 0% 0% 9% 18% 10% 10%

0% 0% 9% 0%

7% 7% 27% 30%

30 14 11 10

Total – N Total – percent

16 25%

3 5%

26 40%

4 6%

1 2%

9 14%

65

5 8%

There is a clear upward trend for Scope of the issue with regard to grassroots mobilisation of the membership, grassroots mobilisation of the mass public and media usage. As the issue increases in size each of these outside lobbying tactics is used by a larger percentage of advocates. There are no discernable patterns in Advertising, Public Relations, Protests, or Op-ed writing. For the issue characteristics of Conflict, as issues exhibit higher conflict, the outside lobbying tactics of grassroots mobilisation of the masses, media use, protest, PR and op-ed writing all increase in usage. Only Advertising and member mobilisation do not exhibit any clear patterns. It is understandable that we see more outside lobbying as conflict increases. As one side of the debate gets coverage, the other side responds in kind, as one citizen group explained: I don’t want to say most importantly, but very important was also getting this information out to the media. Because this issue had not been debated in a really comprehensive way, in over ten years…and there was a need to really rebut the industry’s arguments and help people in the media and policymakers and the public kind of see an alternative point of view. There is less evidence of a clear pattern for the presence of a focusing event; only two tactics trend in the hypothesized direction: Media and Op-ed writing (shifting from 33% to 55% and 11% to 20% respectively as we move from an issue with no event to an issue with a related focusing event). All the others tactics exhibit no pattern, or as in the case of Mobilisation of the grassroots, suggests the opposite direction: this type of outside lobbying happens on issues that have no focusing event. Similarly for the variable of issue history – whether the issue is a new topic or an older reoccurring issue – the evidence does not support the hypothesized direction. Save for the slightly higher usage of Media on new issues, the data would suggest the opposite is in fact the case: that the older the issue the more likely outside lobbying – with every other outside lobbying tactic registering higher usage percentages when the issue is older.

128

Politics Beyond the State.book Page 129 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

For the issue characteristic of Salience, I distinguish between ‘inside’ and ‘outside’ salience. Some issues are highly discussed among politicians and among political observers and in the news outlets followed by those that are engaged in day-to-day politics but that are not of interest to the population at large. While there is a high correlation between the two (Pearson’s R of.91 in the US and.47 in the EU), they are not exactly the same. Thus, to parse out the effects of the interest of political elites and the interest of the public at large on outside lobbying, I coded issue coverage by an ‘inside’ news outlet (The Washington Post) and an ‘outside’ news outlet (The New York Times). Regarding inside salience, the hypothesized direction is exhibited for every outside lobbying tactic: the more salient the issue to the policymaking community, the more advocates engage in outside lobbying tactics. Similarly for outside salience: the more salient the issue is to the mass public, the more advocates are engaging in outside lobbying techniques. It is extremely difficult to parse out the causation here but the fact that letter writing campaigns among the public, grassroots mobilisation of the membership and protest activities are all also trending upward, in addition to the specifically media-oriented tactics, suggests highly salient issues correlate with outside lobbying by advocates. Now we turn to the same analysis but for issues in the EU system. As in the US arena, the use of outside lobbying tactics by advocates is related to the five issue characteristics of interest. In some cases, due to the much lower prevalence of outside lobbying in the EU, patters are difficult to discern. Tables 3 A-C. Outside Tactics by Issue Characteristics in the EU A. Scope Grassroots Members Public Media Ads PR Protest Op-ed Small sector 0% 0% 13% 4% 0% 0% 0% Large Sector 4% 4% 39% 4% 9% 0% 9% Multiple Sectors 0% 0% 21% 0% 5% 0% 0% System-wide 0% 0% 19% 0% 0% 6% 0% Total – N Total – percent

1 1%

1 1%

19 23%

2 2%

3 4%

1 1%

2 2%

129

N 24 23 19 16 82

Politics Beyond the State.book Page 130 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

B. Conflict

None Multiple P Opposing Total – N Total – percent

Grassroots Members Public Media Ads PR Protest Op-ed 0% 0% 19% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 2% 2% 27% 3% 5% 2% 3% 1 1%

1 1%

19 23%

2 2%

3 3%

1 1%

2 2%

N 16 7 59 82

C. Salience INSIDE – EV 0 stories 1-5 stories 6 or more

Grassroots Members Public Media Ads PR Protest Op-ed 0% 0% 8% 0% 0% 0% 0% 2% 2% 27% 5% 5% 0% 5% 0% 0% 35% 0% 6% 6% 0%

N 24 41 17

Total – N Total – percent

1 1%

1 1%

19 23%

2 2%

3 4%

1 1%

2 2%

82

OUTSIDE – FT 0 stories 1-2 stories 3 or more

0% 5% 0%

0% 5% 0%

9% 35% 30%

3% 5% 0%

0% 5% 7%

0% 0% 3%

0% 5% 3%

32 20 30

Total – N Total – percent

1 1%

1 1%

19 23%

2 2%

3 4%

1 1%

2 2%

82

In the EU case we see essentially no pattern between issue Scope and the use of any of the seven outside tactics. For Conflict there is generally an upward trend from zero for six of the seven tactics, with the exception of Media, which trends upward but with no cases registering in the middle category of ‘multiple but not conflicting view points.’ For the variable of the occurrence of a focusing event, the majority of the cells are very low usage, however, the two tactics that have a more significant level of use, confirm the hypothesized direction: As in the US, Media and Op-ed tactics are employed more on issues that have had some type of attention grabbing event related to them (shifting from 21% to 75% for Media and 1% to 25% for Op-ed). Regarding the variable of issue history, it similarly trends in the 130

Politics Beyond the State.book Page 131 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

expected direction for Media and Op-ed but in the opposite direction for all other five tactics. Finally, to look at the role of Salience on outside lobbying usage in the EU, I coded media coverage in the European Voice for an indicator of inside salience and The Financial Times London Edition for a measure of outside salience. For salience to the Brussels policy making community all seven outside lobbying tactics trend upward the more salient the issue. For example the average use of media was 23%, for advocates working on issues with low salience only 8.33% of them engaged in the tactic, on those active on more salient issues 26.83% percent used this outside lobbying technique and those lobbying on the most highly salient issues 35.29% were issuing press releases, trying to talk with reporters and doing interviews. This same pattern is exhibited for all seven tactics in the case of outside salience. When the public isn’t concerned with an issue it is difficult for advocates to engage the media, as one trade association lobbying against the packaging environment indicator on the Packaging and Packaging Waste case noted: “We don’t use the media much, our targets are the institutions, so maybe sometimes the European Voice, but the media isn’t interested in packaging waste, and if you went to your average Joe on the street, and tried to talk with him about packaging and packaging waste! – good luck!” It should be noted, another issue characteristic that is often mentioned in the ability to use an outside lobbying strategy is ‘complexity.’ Issues related to arcane or difficult to understand problems are more difficult to refine into a sound bite that the media can pick up. As one US advocate described: “We didn’t do any media, it was too hard of an issue and too complicated of an issue to actually break down into sound bites. I think if we had done anything like that it probably would have made the situation worse because you would have gotten a story printed in one of the major papers that wouldn’t give all the necessary details.” However, the perceived complexity of an issue is very much tied up with framing and issue definition and an advocate’s position on an issue. As another noted, on a different issue “…the problem is that a lot of the other side, the opposition, have these handy little sound bites that are often completely wrong but it’s much easier when you say ‘this thing lets terrorists in the country’ and everybody goes ‘oh well that’s horrible’ and then when you say ‘well actually that’s not true and these people have this visa’ and you have to explain the whole process, it’s a much harder sound bite to sell.” While one version of their position may be “harder to sell” a different conceptualization might be easier, for example something like “this country was built by immigrants” and so visa regulations should not be too burdensome. Thus, the 131

Politics Beyond the State.book Page 132 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

complexity of an issue is very much determined by how lobbyists can frame the issue, rendering it quite impossible to objectively code whether an issue is ‘complex.’ In conclusion, we see support for the first two hypotheses that lobbyists active on larger issues and highly conflictual issues are more likely to employ outside lobbying tactics – working to engage the citizens in the policymaking process. The support for the hypotheses regarding the second two issue characteristics is more mixed suggesting knowing the issue’s history and whether a focusing event occurred may not be quite enough to predict lobbying strategies. Finally, there is evidence supporting the hypothesis that there will be more outside lobbying on more salient issues, both inside and outside the policymaking communities.

Interests The final level of independent variables that should have an effect on the use of outside lobbying is organizational characteristics. As readers of the literature would expect, in the US citizen groups are much more inclined to employ outside lobbying tactics: they use them to a greater extent than the average on every tactic except Advertisements (and protests, but only one advocate reported employing a protest strategy out of all 65 respondents). They are the only type of actor mobilizing the mass public (at 22%) other than trade associations, of which only 8% reported doing so. Professional associations are also large users of outside strategies, exhibiting higher-than-average use of five of the seven outside lobbying strategies. This category of actors is especially inclined to use grassroots mobilisation of the membership with 57% of professional associations reporting that they used this tactic on the issue on which they were interviewed. Also interesting to note is what types of actors are dominating certain tactics, the use of issue advertisements is used by only 6.15% of actors on average, but it is the professional associations and the corporations that are the only types of actors engaging in this activity; 43% of professional organizations are advertising related to their policy fights, and 14% of corporations. Advertising requires a significant amount of resources, to design, produce and air. One advocate described the blitz advertising campaign being financed by industry in the CAFÉ debate: A big part of their ad campaign, before the vote in Congress this past year was, they had ads running in rural America saying: Congress is going to take away your pickup truck and you’re going to be hauling hay in the back of a little Pinto, kind of thing…And in suburban America, they were running ads on the radio and newspapers and other places saying: Did you know the government’s going to take away

132

Politics Beyond the State.book Page 133 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

your SUV? …It’s kind of, if you’re familiar with the Harry and Louise ads, it was a similar kind of format: Husband and wife talking at the kitchen table and the wife says, “Honey have you seen this? The Senate is getting ready to vote to take away my SUV. I need my SUV to take the kids to school and go to the market, and da da da da” I mean, it is absolutely absurd, it doesn’t even pass the laugh test. So it was all, motherhood and apple pie and pick up trucks, and you know don’t take our fucking pick up trucks.

All the think tanks reported pursuing media strategies and a third also reported engaging in public relations campaigns and writing editorials. Multinational corporations were not engaging in any form of outside lobbying. One company lobbyist described how they try to stay out of the media: “We generally stay behind the scenes, we are not one to go grab a microphone or a spot light very often, we will call people, rather than go see them, we are a company that draws a lot of criticism because we are so big, …So we basically keep to ourselves, sort of lurk in the shadows and try not to put our fingerprints on anything.” Regarding the interest group characteristic of staff5, the data does not clearly support the hypothesized relationship – larger staff sizes seem to be related to a higher use of outside lobbying tactics only in the case of media where we see 43% of small offices reporting media tactics, 50% of mediumsized offices and 57% of large offices. For both types of grassroots mobilisation and protest, it appears you need to cross some threshold of staff resources to engage in these outside tactics but that the largest offices are not ‘going public’ as much as the mid-range offices. This finding might not be as surprising as it seems at first blush, for it is the well heeled lobbying firms, corporations, and business associations that have the well staffed offices – the very types of advocates that are unlikely to engage in large public mobilisation campaigns. Thus it appears, when you don’t have the money and inside the beltway staff to engage in incessant direct lobbying, you rely on the masses and mass membership to communicate your message to lawmakers. However, you need enough staff to put into motion the outside strategy. Organizational structure seems to have a clear influence in the decision to go outside or not, advocates that have sub-units at the regional, state or local level use outside lobbying strategies at a higher level than groups with no local groups, and this goes for every outside lobbying tactic except for op-ed writing, which is often done by expert members or Washington staff. This finding follows logic, it is easier to mobilise the members, citizens in 5.

Table 4B presents an N of 46 rather than 65 because Congressional advocates are excluded.

133

Politics Beyond the State.book Page 134 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

the communities, coverage in local papers, public education campaigns and protests if you have established offices already active in the field. This was exemplified by one of the environmental organizations I interviewed; as their lobbyist explained: We are the largest grassroots environmental organization in the country we have about 750,000 members and the real strength of our organization is the ability to mobilize people across the country, so we often help in the release of other organization’s reports because they don’t have that field staff, they don’t have the regional offices, and the chapter office in every state and volunteers that can put on a press conference in just about any city you would want. So we do a lot of co-releases, because that’s really one of our niches is being able to get media attention, being able to have that national coverage and be able to get the word out.

To assess the relationship between membership size and outside lobbying, I distinguished between individual members and organizational members – be they organizations, corporations or institutions6. The larger the individual membership the higher the percentage of outside lobbying usage – and that goes for ever single lobbying tactic. A similar pattern also holds for corporate membership – the more members the more outside lobbying – but only for the tactics of grassroots mobilisation of organizational members, grassroots mobilisation of the public and media use. The other four tactics are not used by organizations with corporate or institutional members, regardless of how large that membership is. Turning to the EU and investigating the relationship between group type and outside lobbying, a pattern similar to the US emerges: domination by the citizen groups, it is they who use the majority of outside lobbying tactics more than the average of all groups. In addition, they are the only category of actors mobilizing the mass public. Second to citizen groups though are trade associations, using four of the seven outside tactics at a higher-than-average rate. As in the US it is the industry associations that are the only category of actors using advertising. A number of actor categories reported using no outside tactics: professional associations, business groups, lobbying firms, and the governmental actors. While the governmental actors are, of course unlikely to mobilise letter writing campaigns or protest events, media, advertising, public relations and op-ed tactics are all open to them. When it comes to the relationship between staff and outside lobbying the opposite of the hypothesized relationship is exhibited, with more outside lobbying being done by the smaller and mid-range offices. Again, as in the 6.

Twenty actors are excluded from this analysis since they are governmental actors and the concept of membership is not applicable to them.

134

Politics Beyond the State.book Page 135 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

US, this makes sense, since it is the well-heeled offices that can engage in aggressive inside lobbying strategies, they do not need to resort to outside mobilisation. Again, as expected, organizational structure has a clear relationship to outside lobbying. Organizations without sub-units in the member states use only one type of outside lobbying – media – and they do so at a level lower than the average. It is the organizations that have a federated structure – composed of member associations – that exhibit higher-thanaverage outside lobbying on every type of tactic. These types of umbrella organizations can spread the message and call for action through their member associations down to the individual members. The one organization that reported mobilizing the masses and membership described their strategy on the live animal transport debate: So we try to coordinate with our member associations, and get the media done at the same time, so we did that on this issue. Citizens write to their MPs and to their MEPs. MEPs say that animal welfare is the largest number of letters they get, that is, the most letters they get about one topic is on animal welfare. People care about this, people care about animals and they write.

The EU differs from the US in that there are few mass membership organizations. Only one organization that fell into my sample had direct individual members, of which there were only 91. This organization was not pursuing any outside lobbying tactics. Thus, for the EU arena, the membership analysis is limited to organizations with institutional members, which could be associations, corporations, or institutions7. In the EU arena, differing from the US findings, it is the mid-range category of advocates – those that have members but not a large number of members that are employing outside lobbying tactics at a higher rate. In sum, the findings fit with previous suggestions by political observers, it is the citizen groups, and those with a week insider presence that resort to outside lobbying. In addition, organizational composition also plays a role. Those groups that have local offices in the field are better able to engage in outside lobbying and thus draw the public into the policy debates in the capitals. In the US organizations with more members, whether individual or institutional are using outside lobbying at a higher rate, while in the EU, it is the groups with medium sized institutional membership.

7.

Five actors are excluded from this analysis since they are governmental actors and the concept of membership is not applicable to them.

135

Politics Beyond the State.book Page 136 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Multivariate Analysis To parse out the effects of issue and interest characteristics in the two systems I conducted a multivariate analysis. It is necessary to keep the models fairly simple due to the small N in both systems. Thus, the models presented are OLS regression models estimating the level of outside lobbying. The dependent variable here is an additive index of outside lobbying: the number of outside lobbying tactics the advocate used on the issue, it thus has a minimum of zero to a maximum of seven. The independent variables include a dichotomous variable for advocate type that indicates whether or not the advocate was part of the business community (professional, trade or business association, individual corporation, multinational corporation, or PR, law, or consultancy firm all coded one; other types of organizations coded zero); the size of membership (none, low, high; and in the case of the US no membership or membership); staff size; and the four issue characteristics of Scope; Conflict, Salience and whether a Focusing Event occurred8. The unit of analysis is the advocate. Table 4 OLS Regression of Outside Lobbying Use – US and EU

Type Membership Staff Scope Conflict Event Inside Outside

Coef. -0.52 0.58 0.00 0.09 0.12 -0.60 0.06 -0.03

US Std. Err. 0.46 0.26 0.01 0.29 0.28 0.59 0.03 0.03

z -1.12 2.23 0.08 0.3 0.44 -1.01 2.08 -1.22

Constant

0.71

0.59

1.21

N=46 R-squ=.32

8.

*

**

Coef. -0.32 0.43 -0.01 -0.17 0.17 0.53 0.05 0.01

EU Std. Err. 0.18 0.18 0.01 0.10 0.11 0.37 0.03 0.02

z -1.76 2.41 -1.05 -1.76 1.58 1.46 1.95 0.51

0.04

0.22

0.21

* * *

**

0.837

N=82 R-squ=.21

Due to multicollinearity the variable of whether the issue was new or reoccurring had to be dropped from the analysis.

136

Politics Beyond the State.book Page 137 Wednesday, December 5, 2007 8:41 AM

The Role of Interest Groups in Fostering Citizen Engagement

As the results in table 4 show, membership is significant in both models, suggesting having members, and having more members than other organizations leads to higher use of outside lobbying strategies. This is inline with the bivariate analysis presented earlier, but here we have controlled for issue factors. Also significant at the.1 level in both models is the salience of the issue in the policymaking world – the inside salience measure. The more salient the issue is to policymakers and observers the more advocates engage in outside lobbying strategies, controlling for actorspecific characteristics. Then there are two additional factors that only register as significant in the EU arena. First, the scope of an issue is also significant, but the sign on the coefficient is negative, suggesting that the for a unit increase in the scope of an issue (so say, from moving from an issue that only touches on one industry to an issue that effects multiple sectors) we should see a decrease in the cumulative number of outside lobbying tactics by.17 (recall from the bivariate analysis no clear pattern was evident). We also see a decrease in outside lobbying with the other factor that is significant uniquely in the EU: whether the advocate represents business interests. If an advocate is part of the business community in Brussels, they decrease their cumulative outside lobbying activity by.32, controlling for the nature of the issue

Conclusion Thus we see evidence that the political context as well as individual interest group characteristics play a role in determining if a lobbyist pursues an outside lobbying campaign. Knowing any single factor is not enough to predict if outside lobbying tactics are the way to go. It is the melding of factors from all three levels that influences the ultimate tactical strategy. The finding showed that civil society groups are most likely to use outside lobbying, and members of the business community are less so inclined. Small and medium staff sizes also suggest an outside approach. In the US a large membership base further suggests more outside lobbying, while in the EU it is the mid-range groups that are pursing outside tactics rather than the largest membership organizations. And in both polities we saw that an established on-the-ground network of local and regional offices facilitates going public. At the issue level, evidence from both polities supported the hypotheses that more outside lobbying is exhibited on issues that are high in conflict and salience. In addition, in the US issues large in scope also lead to more outside lobbying. A clear pattern was lacking when it came to the history 137

Politics Beyond the State.book Page 138 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

of the issue and if a focusing event was linked to the issue by a policy entrepreneur. It is clear from comparing the EU and the US findings that EU advocates are not mobilizing European citizens as much as they might be. The institutional structure plays a large role in explaining the lack of outside lobbying. And the requirements of the various outside lobbying tactics – some requiring a media system, and some not – have allowed us to confirm both institutional factors are likely at play: the presence of a media system and the selection of officials. Even if the constellation of organizational and issue characteristics are those which would suggest high use of outside lobbying, in the EU such an advocate is still unlikely to mobilise the masses, to foster citizen engagement. Two institutional reforms could bring about a change in that tendency, which would open the door for more citizen involvement via outside lobbying. First, the lack of a pan-EU media machine could change in time, with the rise of the blogosphere and other on-line papers, it may be more and more possible to bypass the slow-moving traditional forms of media and spread a pan-EU message though the internet and mass emails. The Commissioner for Communication, Margot Wallstrom, called for just this mid-June following the constitutional crisis and budgetary crisis of 2005: “innovative steps must be taken to create cross-boarder ‘public space’ for debate at the European level. This might include translation facilities, venues for exchanging articles and exchange programs for journalists” (21 June 2005, International Herald Tribune). If this is the case we may begin to see more outside lobbying in the EU, but orchestrated through virtual news outlets rather than old-school print editorials. However, as the data showed, even the two outside lobbying tactics that do not require a pan-EU media system – mobilisation of members and the mass public to produce grassroots communications – are also less prevalent in the EU. Suggesting the democratic responsiveness of policymakers is also very much at play. Thus, changes in the democratic accountability of the primary political institutions seems to be required – to impose a reelection concern on policymakers making them more responsive to communications from the citizenry.

138

Politics Beyond the State.book Page 139 Wednesday, December 5, 2007 8:41 AM

The Third Level in Brussels? Regional Information Offices and the European Policy Community Michel Huysseune & Theo Jans

Introduction This chapter will discuss a very particular group of supranational actors in Brussels, the so-called RIOs (regional information offices). Representations of local and regional authorities at the European Union have been present in Brussels since the mid-1980s. These representations are difficult to classify, as they represent public authorities, while they have no official status in Brussels. They represent regional and local governments but often operate as lobbyists and interest groups. As an expression of the new opportunities that the process of European integration (presumably) offers to sub-national authorities, they have attracted broad scholarly attention (see e.g., Badiello 1998, 2000; 2002; Gallup 2003; Heichlinger 1999; Jeffery 1996a, 1996b, Larsson 2000; McAteer & Mitchell 1996; Nielsen & Salk 1998; Smets 1998; Smith 1998a, 1998b; Vos 1999). Since these offices have no official status in Brussels, their activities may be compared to those of other interest groups, attracted to Brussels by the opportunity structure provided by EU institutions (see for example Sandholz & Stone Sweet 1998; Mazey & Richardson 2001; Greenwood, 2003). From an interest representation perspective, their practices resemble those of other lobby groups at the EU policy level (e.g., Panebianco 2000, pp. 57-65, Van Schendelen, 2003). The literature on interest groups highlights a number of effects of their presence in Brussels that we will verify for the RIOs. These include a process of Europeanisation, understood as the construction, diffusion and institutionalisation of formal and informal rules, procedures, policy paradigms, styles, and shared beliefs and norms (Radaelli 2003, p. 30). Literature on RIOs does suggest that cooperation within the European policy space has led to a convergence of behavioural patterns transcend the differences in national backgrounds (John, 2001). We will further observe in which measure the spill-over effect – that when one interest group

139

Politics Beyond the State.book Page 140 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

arrives in Brussels, similar groups will be attracted – also occurs among local and regional authorities. The RIOs position as representations of sub-national interests may also be interpreted within the framework of multi-level governance (MLG) theory (see e.g., Benz & Eberlein 1999; Bullman 1996; Hooghe & Marks 2001; Marks et al. 2002¸ Nielsen & Salk 1998). This theoretical perspective presumes that European integration has given rise to the formation of a third sub-national policy level, whose institutions — in the first place the regions — are affirming themselves as actors embedded in the European policy process. Although acknowledging that the sub-national level remains secondary to the other (supranational and national) levels, the MLG approach does assume that regions exercise a direct influence on the EU decision-making processes and in particular on the implementation of EU policies (especially concerning regional development). In addition the MLG perspective refers to the concept of “governance”, i.e., the realisation of public goals through forms of public/private cooperation. Regional representations may to varying degrees cooperate with private actors (businesses, chambers of commerce, etc.) to achieve their goals in the Brussels’ context. Clearly regional authorities (especially from countries where they have a strong, institutionally embedded position, e.g. Germany, Belgium) have been interested in the creation of a third tier in European governance. The creation of the Committee of the Regions (CoR) in 1994 heralded the arrival of regions as relevant players in EU policy process. Although RIOs, as representations of those authorities, can not be considered as principal actors in decision-making processes, we can nevertheless investigate whether they are in any way involved in a so-called third level of governance, or whether they are trying to contribute to the formation of such a third level. In addtion, we will look at the measure in which they take part in forms of public/private cooperation to achieve those goals. In what follows we will analyze the role and the functions of the RIOs in Brussels, on the basis of a research carried out in 2004 by the Vakgroep Politieke Wetenschappen and the Institute for European Studies of the Vrije Universiteit Brussels for the Brussels-Europe Liaison Office. We will first develop a typology of these representations and discuss their material characteristics. Thereafter, we will analyze how these representations describe the goals of their activities and we will also point out how these goals evolved during the two decades that such offices have been present in Brussels. Thirdly, we will look at how these offices intervene in the policy process and identify their interlocutors within the institutions of the

140

Politics Beyond the State.book Page 141 Wednesday, December 5, 2007 8:41 AM

The Third Level in Brussels?

EU. In the conclusion, we will outline why RIOs constitute a particular and separate category of policy actors in the EU policy community.

A typology of representations According to our research, RIOs are essentially representations of public authorities (see table 1). No RIO represented only private (local or regional) actors and only a relatively small percentage (13.8%) of RIOs are public-private partnerships. Among the represented entities, regions are the dominant group. For cities, the opening of an independent representation seems to be an exceptional choice, only made by the largest cities (4.9% of our respondents). Cities are not necessarily absent in the Brussels environment: their representation may either be integrated within an umbrella regional representation, or they may be represented through consortia of local authorities or by specific interest groups like Eurocities (7.3% of our respondents). A similar observation can be made concerning (local) authorities below the regional level: they also tend to be represented in Brussels as a constituent part of a regional representation, and only rarely set up independent representations.1 Regions (to be understood as the first level under the national level) are indeed the dominant presence. For regions to set up a representation in Brussels is indeed becoming the norm, rather than the exception. This is most evident in the case of the regions from the 15 “old” member states. With the exception of the regions from traditionally highly centralized states (Ireland, Portugal and Greece), practically all regions of these countries are represented in Brussels. We may observe that, as is generally the case for interest groups, a spill-over effect has occurred; the presence of some regional authorities in Brussels has incited most other regions to follow suit and set up their own representation in Brussels.

1.

These include representations from French departments (e.g. Moselle), and Spanish provinces (Barcelona, Alméria, Huelvas). The offices of English counties, among the first to open representations in Brussels, are nowadays integrated within representations of broader regions, within which they nevertheless preserve a great degree of independence.

141

Politics Beyond the State.book Page 142 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Table 1 Types of representations 2

Type of representation Public regional authorities Public local authorities Public/private partnerships Other public representations2

Frequency Percentage 69 56.1 17 12.2 17 13.8 22 17.9

A similar “spill-over” effect seems to be at work for the regions of the new member states. Polish regions in particular have been very eager to open representations in Brussels, but regions from Czechia, Slovakia, and Hungary were also well represented at the time of our field research. The other new member states are catching up although in an initial phase local and/or regional authorities often tend to be represented through a collective body (e.g. the Association of Estonian Cities, the Latvian Association of Local and Regional Governments, the National Union of County Councils of Roumania). Other regions from the new member states, from the candidate member states and from the Wider Europe are exploring the possibilities of opening representations or have already opened an office (cfr. three RIOs from Croatia). We may assume that the characteristics and the activities of the RIOs in Brussels, besides the differentiation between representations of local and regional authorities, are also reflective of the type of authority that is being represented. As a general rule, regional authorities from centralised states are slower to open up offices in Brussels than countries with stronger regional authorities. It can be assumed that the latter have more to win (and to loose) from the actions of the EU, especially if they have legislative authority that may be affected or curtailed by EU legislation. Therefore, in what follows we will systematically differentiate regions with legislative powers from the other representations.3 Because of their very different historical background, but also because of their shorter experience with the EU, representations from new member states were distinguished from the other RIOs. A third element that may have an impact on the functioning of RIOs in Brussels are the economic conditions of the region they represent.

2. 3.

This category includes consortia of regional and local authorities and international networks. The category always involves public authorities. We also included regional representations in which the administration of a legislative region is a partner. This includes public/private partnerships, but not the representations of local authorities located within that region.

142

Politics Beyond the State.book Page 143 Wednesday, December 5, 2007 8:41 AM

The Third Level in Brussels?

For this purpose, we differentiate representations from regions entitled to structural funds (Objective 1) from other regions.4 Table 2 Characteristics of representations by type Type of representation

Staff members

Regional representations (n=76) Local representations (n=17) Regions with legislative competencies (n=41) Representations from new Member States (n=14) Objective I regions (n=49) Average for all representations

5.8

Office space (m2) 281

4.5

232

35

387

109

6.9

8.0

397

75

519

1201

10.7

1.8

71

34

127

177

1.9

4.7

163

44

288

334

6.1

5.3

258

56

382

635

7.9

Seminar Budget (in Visitor Age (in visitors thousands g) home years) region 68 364 819 8.5

In general, it may be observed that RIOs are relatively modest organizations (table 2). They employ a limited number of people (their staff averages 5.3 persons) and have rather limited annual budgets.5 The presence of a small number of large representations moreover influences these averages: more than 60% of these offices count in fact less than 5 employees and 15% are one-person enterprises. Budgets likewise range from 42,000 to 1,981,700 Euro. The differentiation between offices is in fact relatively important. Especially the representations of regions with legislative competencies distinguish themselves by their higher budgets, greater number of employees and larger office spaces. Offices of new member states are on the other hand smaller than average. Other differentiations are less outspoken: regional offices are somewhat larger than average, representations of local authorities and of regions eligible for structural funds (Objective I) are somewhat smaller. Differences between countries 4.

5.

We considered each representation from old or new Member States active in regions entitled to Objective 1 funds, even if only a part of the territory is able to call on these funds, or where these funds are phased-out. The profile of these latter regions does not significantly differ from those that are able to fully call on these funds The results on the budgets of RIOs have to be read with caution, since only 56 of 123 respondents answered this question.

143

Politics Beyond the State.book Page 144 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

are also often important. The German offices are the largest with an average of 9.3 staff members.6

What has brought RIOs to Brussels? The process of opening up regional information offices originated in the 1980s, and mainly concerned regional and local authorities from a limited number of member states. The reasons for opening representations were in an initial phase clearly determined by national factors. In the case of the United Kingdom, the offices were opened by cities or counties, disposing of limited means and few competencies. Coming from a country with a strongly entrenched lobbying culture, they came to Brussels to find resources to finance projects, often in partnership with private institutions (Jeffery 1996a). The process in Germany was driven by the federal regions, the Länder. The process of Europeanization eroded their constitutionally assigned powers, which motivated them to acquire the right to be involved in the European policy-making process (Börzel 2002, pp. 53-87). This process essentially took place in Germany itself, but already in 1957 the Länder organised an informal collective body at the European level, to inform them about relevant developments in Europe. The status of this Beobachter der Länder bei der EU was officialised in 1988 (Börzel 2002, p. 61). During those years, regions also opened individual representations in Brussels, which were recognised by the federal government in 1992 (Börzel, 2002, p. 77). The presence of the German regions was also motivated by their expectation that the European Union would provide them a more important institutionalised position at the supranational level. The creation of the Spanish representations followed a similar logic. Spanish regional authorities discovered that their country’s accession to the EEC (1986) could erode their recently acquired autonomy. As was the case in Germany, this process led to tensions between the regions and the central government. The regions decided to establish a direct line with Brussels via informal representations (organised as private structures).

6.

The very high standard deviation (8.7) also shows the strong variation in staff for German representations. A number of West German representations in particular have a very high number of employees (among our respondents: Baden-Württemberg 35, North Rhine-Westphalia: 26). Smaller regions, regions from the former East Germany and local authorities have much more modest representations.

144

Politics Beyond the State.book Page 145 Wednesday, December 5, 2007 8:41 AM

The Third Level in Brussels?

Their intention was to work directly with the European institutions, bypassing the Spanish central state. Treaty changes, the deepening of European integration and its increasing impact on the sub-national level also made the establishment of a Brussels’ office a more attractive and/or necessary option (Badiello 2000). An important event in this process was the adoption of the Single European Act in July 1987. The Single Act expanded the prerogatives of the EU to the area of regional development, and thus stimulated the development of a European regional policy. It also provided the EU with authority over policy domains, which in some Member States belonged to the authority of local and regional governments (e.g., environment, social policy, R&D and industry). The resulting cohesion reforms (especially the Delors II report 1988) also incited a number of regional authorities (Catalonia, Basque Country, Brittany, Wales) to open offices in Brussels to exercise influence at the source regarding the distribution of these cohesion funds (Schmitter 1996, p. 138; Panebianco 2000, p. 61). A number of events in 1993 further triggered the opening of more regional representations in Brussels. The development of the internal European market increasingly impacted on regional and local authorities. The Maastricht Treaty introduced the principle of subsidiarity, confirming the importance of the regional dimension in the European policy process. More concretely, the establishment of a Cohesion Fund for disadvantaged regions, and the creation of the Committee of the Regions (1994) seemed to give regional authorities a more prominent position within the Union. The Maastricht Treaty also allowed regional ministers to represent their country in the council of ministers, an option that has been used by three federal states: Germany, Austria and Belgium. Especially the changes introduced in 1993 were an important catalyst for the opening of representations of local and regional authorities in Brussels. According to our research, 11.6% of the respondents opened their Brussels’ branch before 1990, 41.3% before 1995, and 66.9% before 2000, and the period between 1992 and 1994 witnessed the opening of no less than 26 offices. However, the emergence of new RIOs did not cease in subsequent years. Especially in 2002 and 2003 many new offices were established by cities and regions from new Member States. The increasing presence of RIOs in Brussels undoubtedly reflects the window of opportunity that the process of European integration provided to local and regional authorities. A number of quite different reasons may explain the arrival of these representations in Brussels: a search for funding

145

Politics Beyond the State.book Page 146 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

opportunities, the possibilities to lobby for regional interests, but also the growing pressure to be present in Brussels in order to expand or preserve regional powers in the home country. While regional representations may have been drawn to Brussels to make use of a myriad of opportunities, their experience has often obliged them to redefine and even downsize their original ambitions. German regions initially intended their offices to be actors in a Europe of the Regions. Regions were expected to become central and institutionalised players in European decision-making. These ambitions have not materialised. These regions have therefore reduced their ambitions, although they still perceive their representations as the equivalent of an embassy. The German regions continue to lobby pro-actively for the extension of the influence of regional authorities in Brussels. Spanish regions initially hoped to use their offices in a strategy whereby they would maintain and develop direct relations with the EU, bypassing their national government. This strategy failed, and the Spanish regions were obliged to recentralize their actions and to organise a permanent cooperation with their national government when addressing EU institutions. The UK offices that came with less political goals in mind also changed their approach. Their original approach, essentially based on obtaining funding, became more policy-oriented over the years. A funding-driven policy was replaced by a search for more policy-driven funding, whereby policy orientations take precedence over the exploitation of funding opportunities. The representations of local and regional authorities in Brussels have been opened for a broad variety of reasons, often determined by the national context from which RIOs emanate. The Brussels’ experience has frequently forced the RIOs to modify their initial intentions. At the same time, working in a supranational policy context and the contacts and cooperation with other offices has generated a collective socialisation process based on the mutual learning and exchange of information. The outcome of this process is clearly expressed in the type and the broad range of activities that offices are nowadays involved in.

146

Politics Beyond the State.book Page 147 Wednesday, December 5, 2007 8:41 AM

The Third Level in Brussels?

Table 3 Importance attached to activities according to type of representation Item Information about EU legislation Information about funding opportunities Ties with other representations Liaison between region and EU Explaining the region’s position Raising awareness of region in Brussels Increasing knowledge about the EU Responding to requests for assistance from people of the region Influence decision making in EU Gain influence for region Promotion of region towards private actors

mean Regions (n=123) (n=76) 4.5 4.6

Local Legislative New Objective (n=17) (n=41) (n=14) I (n=49) 4.5 4.9 4.5 4.7

4.5

4.6

4.6

4.6

4.9

4.4

4.2

4.1

4.3

4.0

4.5

4.1

4.5

4.5

4.5

4.6

4.4

4.3

4.3

4.4

4.6

4.6

4.2

4.3

3.9

4.0

3.9

4.2

4.4

4.0

3.6

3.8

3.1

3.9

4.0

3.4

4.1

4.2

3.8

4.4

4.1

3.9

4.3

4.3

4.7

4.6

3.9

4.3

3.9

4.0

3.8

4.3

4.0

3.9

2.9

3.0

2.4

3.3

4.1

3.1

Table 3 shows (on a scale which evaluates the operational importance of each item from 1 – minimum = unimportant – to 5 – maximum = very important) the average importance representations attach to a set of activities.7 The most striking result is the high average score of almost all the items, which reveals that, generally speaking, the offices are now attributing importance to all of these activities (in contrast to their initial orientation towards more specific and limited goals). With the exception 7.

The list of items reproduces those proposed in a study carried out in 1999 (Marks et al. 2002). To these items, we added an eleventh item regarding regional marketing, not included in that study. The ten items withheld in both studies generally obtain similar scores (a difference less than 0.5 point), with one exception: in our study the item “gaining more influence for your region more generally” scores significantly higher (3.9 instead of 3.2). The most plausible explanation for this increase is that the increasing experience of RIOs in Brussels increases their confidence and thus heightens their ambitions.

147

Politics Beyond the State.book Page 148 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

of the last item, the promotion of the region towards private actors, the scores of all items are above average (3), and are thus considered to be rather important. Both gathering information, building up ties between the town/region and the EU are considered crucial, but the creation of networks and exercising influence on the EU policy process are deemed almost as important. Only to regional marketing towards private actors do RIOs attribute less importance. Although RIOs find the promotion of their town and region to be an important goal, this process remains mainly oriented towards promoting their region to other public actors (or to other RIOs). Region-marketing to broader audience is perhaps less attractive to them because it requires significant resources (CEEG 2002, pp. 50-51), or because offices represent public authorities rather than private interests, or as a result of the delegation of this task to specific bodies (which in many cases cooperate with the representations) (Caciagli 2003, p. 84). The above results confirm that offices have become multifunctional: their sometimes limited original goals have expanded to include a broad variety of items. This has happened regardless of the size of these offices: the material conditions in which they operate do not correlate at all with the strategy and goals of these offices.8 The importance attributed to the various goals does, however, differ according to the type of representation. The most striking difference concerns offices from regions with legislative competencies. Not surprisingly, and confirming an observation made by Marks et al. (2002), they emphasise more than average “influencing EU policy” as a key priority in their work. They give high scores to influencing the EU policy process, to providing explanations regarding the points of view of the region and to gaining influence for regions in general. These representations, however, score higher than average for all the items (except “ties with other representations”). We may therefore assume that their goals, efforts and activities reflect the fact that legislative regions have more to win and to loose in Brussels and are thus committed to a pronounced role of their representation in Brussels. Concerning their priorities, other types of representations have a less outspoken profile, Representations of new Member States distinguish themselves by the greater importance they attach to information about funds, promotion of the region (especially vis-à-vis private actors), and increasing the public knowledge on the EU. They are less interested in influencing the general EU policy process, but attach just as much importance as other regions to gaining influence for their region in 8.

None of the measured material features of the offices correlated with any of the eleven statements reported in table 3.

148

Politics Beyond the State.book Page 149 Wednesday, December 5, 2007 8:41 AM

The Third Level in Brussels?

Brussels. Local and regional representations do not differ much from the standard profile, and the same holds true for representations of regions eligible for structural funds. Remarkably, the objective I representations do not even attribute a higher importance to “information about funding opportunities”. These representations obviously attribute importance to this topic – as do representations in general – but like other representations they have developed a multifunctional profile, without an exclusive focus on structural funds. The core activities of the offices – the ones to which they attribute the highest importance – can be summarised around four axes: information, networking, liaison between local and regional authorities and the EU, and the influencing of EU policy. The literature on RIOs has revealed that collecting information on EU policy tends to be their most important goal. For subnational authorities, it is of crucial importance to obtain information about EU legislation and the future development of EU policies. For these authorities, having an outpost in Brussels is advantageous, because it allows them to obtain, besides the official documents that are made accessible through official websites, crucial non-official information on the subject. The issues or policy domains that these offices are most interested in reflect the competencies that typically belong to the remit of local and regional authorities: regional policy, social cohesion policy, R&D, agriculture, environment, transport, industry and energy policy (Badiello 2000). Gathering information is the starting point for the intervention of the RIOs, including for their other activities like project and funding opportunities, participation in trans-national networks, and influencing EU policy. At the same time, however, RIOs also play an important role in offering information to the EU. Because of the small scale of their own administration, EU officials themselves seek information and expertise, and RIOs are (as “grass roots” or “civil society” representatives) perceived to be a valid and legitimate (public) source of (regional) information and data. RIOs are relevant partners for European Commission (EC) officials that enable them to develop programmes that meet the actual needs perceived at the grass-roots level. That RIOs nevertheless attribute more importance to gathering than to providing information (Gallup 2003, p. 6) is confirmed by our results: gathering information about legislation (4.5) or fund-raising (4.5) score higher than explaining the position of the region to EU policy makers (4.3), but the difference remains limited. RIOs are also strongly involved in networking and in the development of ties with other representations, an activity they consider to be almost as

149

Politics Beyond the State.book Page 150 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

important as gathering information (score: 4.2). This involvement reflects the value the EC attributes to collective and trans-national representatives of specific interests (Mazey & Richardson 2001). Regions in Europe have been involved in the creation of such networks around specific interests well before establishing representations in Brussels. Through their location in Brussels, however, offices are able to play an important role in sustaining and developing such policy networks. These networks share some features of trans-national lobbies and private interest groups, but are nevertheless more modest in their goals. They are more oriented towards the exchange of information and the dissemination of best practices, rather than towards exercising political pressure. Besides such interest-oriented networks, RIOs (and their supporting regional administrations) also developed specific trans-national partnerships with a limited number of other regions and organizations in Brussels. Such partnerships often express the strategic choice of regional administrations to promote international cooperation. Sometimes, such partnerships lead to the creation of a collective inter-regional representation in Brussels. This is for example the case for the regions of Wielkopolska (Poland), Aquitaine (France), Emilia-Romagna (Italy) and Hessen (Germany): although each representation deploys its own particular activities and disposes of its office space within the building, the common location reflects a strategic choice for close cooperation and exchange of information between these regions. The pooling of several RIOs in one common location occurs increasingly with the arrival of new RIOs in Brussels, as it allows for more swift integration of the newcomers in the new policy environment. Offices further act as a liaison between the region and the EU. The importance attributed in the literature on RIOs (Badiello 2000; CEEG 2002, p. 51; Jeffrey 1996a, pp. 196-197) to this task is confirmed by our research. Liasing between the home region and the EU institutions is a crucial function of RIOs. The staff members of the representations frequently visit their hometown/region. They are in close contact with officials in local and regional authorities, but also with a number of key economic, social and political actors from their territory. Through a broad variety of means (internet site, newsletters …) they communicate information concerning the EU to their home constituency. They are wellpositioned to provide assistance to actors of the home region on EU projects and programmes: they may mobilise local expertise to contribute to EU policy development, they may assist in partner searches, or even give practical assistance concerning contact persons and procedures in the EC. The offices also act as a contact point in Brussels for administrators, 150

Politics Beyond the State.book Page 151 Wednesday, December 5, 2007 8:41 AM

The Third Level in Brussels?

interest groups and even private citizens from the home constituency. They in fact receive a considerable amount of visitors from the home region, on average 635 per year. The RIOs are seen to be more constituency-oriented and “user-friendly” than many other institutions in Brussels. Regional representations are involved in influencing EU policy. In general, it is assumed that national governments still continue to act as gatekeepers between local and regional authorities and the EU, although sub-national actors do play a (subordinate) role in the decision-making process. In this process, the RIOs do not necessarily play a central role: as a general rule, sub-national authorities attempt to influence EU policy at the national level, through the national government. RIOs nevertheless include influencing policy in their priorities, although acquiring political influence scores lower (3.9) than explaining the position of the region to EU policymakers (4.3) and influencing the policy-making process in favour of the region (4.3). These data seem to confirm that offices are more oriented towards “soft politics”: instead of acting as a third tier of European governance they are rather involved in trying to exercise influence by providing information (Badiello 2000). However, a comparison with study results for Marks et al. (2002) dating from 1999 reveals that according to our data offices nowadays attribute much more importance to influencing EU policies.

How do RIOs operate in Brussels? As actors in the EU policy community in Brussels, RIOs are primarily concerned with acquiring a good reputation. Credibility can be achieved by proving to be reliable and relevant partners, able to provide useful and correct information to the EU institutions. Reflecting an egalitarian ideology that the quality of the intervention matters more than the quantitative weight of interest groups, they do not attribute a crucial importance to material means to reach this goal. To be efficient and successful in their intervention, offices emphasise the importance of a welldefined strategy towards the EU, a focus on limited goals and a (lobbying) plan with clear targets. Qualified staff members with a large and practical experience in working with EU institutions are deemed to be the most important asset for the success of a RIO in Brussels. RIOs are clearly oriented towards exercising influence on the EU institutions. They act as other interest groups attempting to influence EU decisions, but as emanations of public authorities they also express the desire to become institutionally recognised actors of the EU policy 151

Politics Beyond the State.book Page 152 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

community. Although this latter goal is not a priority for all offices, many of them and especially the representations of regional authorities with legislative competencies express this desire. For them, the informal position of RIOs is indeed problematic. The DG Regional Policy organises formal meetings with the RIOs (twice every year) but these meetings do not lead to any form of institutionalised policy participation of these offices who, like other interest groups, remain informal and unofficial partners in the EU policy process. The Committee of the Regions (CoR) offers regional entities their own institutional framework within the European Union, but representations are often sceptical as to the efficiency of this consultative body, and sometimes view it as competing with representations. RIOs therefore only attribute a limited importance to the Committee of the Regions. Although it is supposed to be the natural spokesperson of local and regional interests within the union, most RIOs consider the CoR to be an interlocutor of limited importance, which does not warrant a lot of attention when they seek to influence policies. Following previous research on this issue (Badiello, 2000; John, 2001, Gallup, 2003), our questionnaire confirmed the relative importance (3.5) offices attribute to contacts with this CoR. As a matter of fact, not all offices maintain contacts with the CoR, although a majority of them (74.6%) does. Although organisational reasons (mainly because regional governments at home monopolise them) may explain why not all RIOs have contacts with the CoR, this result does reflect the secondary position of the CoR within the EU. The reasons given by the offices for working with the CoR confirm this: only 51.2% of them explicitly mentions the participation in the Committee’s policy making activities, while 26.2% refers to “membership” without further specification, 35.7% to assistance to members of the Committee from their own region. 28.6% mentions participation to PR activities and events like the CoR’s Open Days.9 These data confirm the limited relevance of the CoR and suggests that the role it plays in Brussels is comparable to that of an advisory body and information source rather than an institution sought after for its policy impact. This reveals the weakness, if not the absence of a Third level of governance at the EU. In the absence of this Third Level, RIOs act as other lobbies do at the EU level. They contribute to the policy process (providing information, producing position papers, participating in expert groups, public hearings, consultations, etc.), as other interest groups in Brussels seek to do, aiming 9.

Representations often mention more than one of these items, resulting in the sum of these percentages exceeding 100.

152

Politics Beyond the State.book Page 153 Wednesday, December 5, 2007 8:41 AM

The Third Level in Brussels?

to become credible interlocutors for the EU and EC officials in particular. Although offices, as representations of public authorities are sometimes reluctant to describe their activities as lobbying (which for them may have the connotation of “advocating private interests”), they are nevertheless clearly interested in influencing European Union policies in those fields that are of direct interest to their home regions. Following the EC’s preference for collective and trans-national interlocutors, RIOs increasingly seek to further their interests through trans-national thematic networks. While many of these networks are of mixed composition and are not merely comprised of regions, the Brussels RIOs have played a crucial role in creating a number of thematic networks, such as EPRO (Environment Platform of Regional Offices in Brussels), ERRIN (European Regions Research and Innovation Network), and Cities for Cohesion (a network of major cities confronted with social dualisation, with the objective to achieve the goals of the Lisbon agenda) (Goergen 2004). While these networks clearly offer opportunities to become important partners within the EU policy community, they do not institutionalise the position of RIOs. Moreover, the creation of these networks itself has often been triggered more by the needs of the EU institutions rather than by those of the offices. The EC played a crucial role both in the creation of ERRIN, sponsored by DG Research (Goergen 2004, pp. 161-162), and EPRO (created under the impetus of Commissioner Wallström, to meet the need of DG Environment for local information on environmental issues). For the moment, the creation of these networks is essentially reactive, in response to EU initiatives. Like other interest groups, RIOs act in the first place to become credible partners for the EC. Not surprisingly, the EC is for them, like for other interest groups, the most important interlocutor in the institutional triangle. Contacts with the EC are considered to be of crucial importance by almost all RIOs (an average of 4.8 on a scale from 1 to 5; see also Badiello, 2000, Gallup, 2003, p. 6). They do not only develop such contacts individually, but also through the networks to which they belong. Like other interest groups, however, RIOs also increasingly develop contacts with the European Parliament (EP), whose growing importance is emphasized. Although contacts with the EP also take place through the thematic networks it is worth noting that RIOs in the first place lobby through the national or even more frequently the regional MEPs. While RIOs undoubtedly act as other interest groups, our research also revealed that, as outposts for public (regional) authorities, they also dispose of other channels to exercise influence. A first possibility is related

153

Politics Beyond the State.book Page 154 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

to the option offered by the Maastricht Treaty that allows representatives of regional authorities to participate to the Council of Ministers (see also see Kovziridze 2002). Of the three countries concerned, however, only the Belgian representations in Brussels are directly involved in this activity. The representations of the Belgian regions are integrated into the national Permanent Representation. The embedded regional representatives participate fully, through the Belgian Permanent Representation, in the Council of Ministers and its preparatory meetings (eg. working parties and committee of permanent representatives). In the two other concerned countries, Germany and Austria, the regional intervention towards the Council of Ministers is organised in the home country, and the German RIOs in Brussels only play a secondary role in this process. Regional offices have generally become much more involved in cooperating with their respective national permanent representations. In recent years, representations from several countries have started to develop a national coordination of RIOs, which works in close symbiosis with their respective Permanent Representations, with which more or less systematic forms of exchanges of information, coordination and/or policy planning are developed. Through their Permanent Representation, RIOs have indirect access to the Council of Ministers, and therefore dispose of a channel of influence that is unavailable to private actors. How such forms of coordination are organised varies very much depending on the national context and the specific central-regional relationships prevalent in each of the states. For the reason outlined above (coordination of regions taking place in the home country) the offices from the federal states Germany and Austria have not developed strong forms of coordination in Brussels. The offices from the Netherlands provide an example of how local and regional representations from a single country without a federal tradition have created strong forms of cooperation with each other as well as with their corresponding permanent representation. After relocating in the same building, the House of the Dutch provinces, in 2000, these representations have discovered the practical advantages of cooperation, since it has significantly improved their efficiency and use of resources (pooling of resources, thematic specialisation, visibility, etc.). The operation of the House has shown that the original mistrust as regards sharing of information was unfounded. Cooperation created significant mutual benefits for all provinces (and cities) involved. It strengthened the legitimacy of their lobbying efforts towards the EU institutions and the Dutch permanent representation, because the House of the Dutch Provinces conveyed the views of all the Dutch Provinces (rather than those of individual provinces). In recent years, similar forms of national 154

Politics Beyond the State.book Page 155 Wednesday, December 5, 2007 8:41 AM

The Third Level in Brussels?

cooperation have also been developed in other countries, although not necessarily in such a systematic way as in the Netherlands.

Conclusion An analysis of the functioning of RIOs in Brussels reveals a strong convergence between RIOs regarding their goals, priorities, strategies, institutional contacts and perceptions. The RIOs focus on similar activities and they identify the same priorities. They have clearly been socialised within the EU policy community, the cooperation and experience-sharing between the RIOs has been instrumental in this regard. The networking and cooperation efforts constitute a central activity of the RIOs, which has also had the effect of harmonizing their activities and approach to the Brussels’ policy environment. A spill-over effect has attracted most regional authorities from the “old” and several from the “new” member states to open offices in Brussels. As informal members of the European policy community, they are involved in the process of providing and gathering information. As public actors representing local interests they are credible and legitimate interlocutors for EU officials and institutions. Within the EU policy community, RIOs nevertheless occupy a particular position. They represent public authorities, but failed to obtain official status and certainly do not appear to be an expression of the institutionalised position of sub-national authorities in a developing multilevel polity. Nevertheless, their role as outpost of regional authorities gives them possibilities to intervene in the EU policy community (through their national permanent representation which enables them to intervene indirectly in the Council of Ministers) in ways that are not available to private interest groups. A specific dimension of the RIO activities concerns their role in forging close links between the home town or region and the EU policy community. A core activity of RIOs concerns their function as a gobetween between the EU and their home constituency. On the one hand RIOs disseminate information about EU policies and programmes, explaining how the EU works and which opportunities it offers. On the other hand RIOs offer a relatively easy access to EU institutions, possibilities for contact, partnerships, common projects and funding opportunities. RIOs perform a fairly central role in ‘driving European

155

Politics Beyond the State.book Page 156 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

policies and opportunities home’, and in offering access to the European supranational policy community.

156

Politics Beyond the State.book Page 157 Wednesday, December 5, 2007 8:41 AM

Sinking the Port Directive. Belgian Trade Unions’ Mobilization against EU Policy Martine Van Assche & Kris Deschouwer

Representation and mobilization in a multi-layered context Political representation and mobilization are concepts and especially practices that were developed in a very specific institutional environment. That environment is the national state. Pressure groups have become crucial actors in the modern states, linking – together with the political parties – the authorizing demos on the one hand to the authorized political elite on the other. The challenges parties and pressure groups are faced with today can certainly be attributed to the growing interdependence and globalization affecting the role and position of the national states. Central functions of the state, like its internal organization, its forms of intervention and also its modes of representation are being thoroughly transformed (Bartolini, 2002; 2006; Luther & Müller-Rommel, 2002). The European integration process is one of the crucial expressions of this changing environment. The institutions of the European Union are the results of an answer to these challenges in one particular part of the globalizing world. The European integration process indeed affects several aspects of the national states, among which the forms of participation and representation are not the least important. This raises concerns about the legitimacy of the integration process and fuels the debate about the democratic quality of the European Union. Some authors believe that the EU is in fact as democratic as it could or should be. Others say that a democratic polity requires real competition for political leadership and a true public debate about the policy agenda, exactly like in national states (Follesdal & Hix, 2005; Lord & Magnette, 2004). This link between democracy and public debate is also voiced in the European Commission’s White Paper of European Governance: “Democracy is synonymous with dialogue, debate and belonging to a political community in which the individual’s rights are carefully protected by law” (p. 7). And further: “Democracies have to create the conditions for

157

Politics Beyond the State.book Page 158 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

an active exercise of citizenship … only in this way, it is argued, can politics be brought closer to people’s real concerns” (p. 29). The aspect of democratic governance and legitimacy that is central is this debate is the linkage of citizens to decision-making, the transmission of demands from society to political elite. It is exactly the function that has been fulfilled in the national state by political parties and by pressure groups. In a complex and multi-layered political system, where competencies are spread and scattered over different levels of decisionmaking and over a variety of institutions in each of these levels, this transmission of societal demands is obviously less easy. But it is needed for the polity to fulfill this minimal and important condition of democratic legitimacy. The growing impact of EU policies on the daily life of the citizens and on the structuring of national states has at least the potential of being the source of many societal conflicts and debates. It is generally accepted that two lines of debate are present: conflicts about the form of the European polity itself and conflicts about the policies produced by the European polity (Hix & Lord, 1997; Mair, 2005). Balme and Chabanet (2002) distinguish four modes of Europeanization of action from national actors: internationalization, externalization, transnationalization and supranationalization. The first is mobilization that is essentially local or national in terms of actors, the system used and the target aimed at. Farmer protest typically belongs to that type. When local or national actors activate a system of European action in order to bypass national obstacles or to take advantage of complementary resources, we can talk of externalization e.g. co-operation of local governments. The target is then European or national, and the issue at stake is the use of political opportunities offered by the EU. Transnationalization involves the transformation of the actors and the crossing of national boundaries. The target is essentially European and the commonest technique is lobbying or overt protest to influence EU law e.g. multinational enterprises. Supranationalization, finally, refers to a institutionalization of actors at the European level and their formal coopting into the EU decision-making process e.g. the European social dialogue. According to Balme & Chabanet the socio-political configuration now emerging appears fragmented and marked by persisting national styles of interest organization and a mode of functioning that puts compromise first. Imig & Tarrow (2001) also found that the major responses to European policy-making are cases of what they call ‘domestication’.

158

Politics Beyond the State.book Page 159 Wednesday, December 5, 2007 8:41 AM

Sinking the Port Directive. Belgian Trade Unions’ Mobilization against EU Policy

Domestic groups target national or sub-national agents in response to their claims to the EU. In this chapter we will look at the debates that followed the European Commission’s initiative to regulate investments in ports and the competition between ports. That is an interesting case because it did result in quite some mobilization of the societal actors who were in the end able to fully stop the Port Directive. It is also a case that involves ‘traditional’ actors. The Port Directive was affecting the working conditions in the ports and was therefore a matter to be picked up by the trade unions in the first place. And although trade unions do have international and transnational organizations, the core of their organization is national, is linked to the national state in which they have been created. Their organization is also – for many of them – closely linked to traditional political parties whose origin and central point of reference is also the national state. We have therefore a case in which indeed Europe ‘hits home’ (Börzel & Risse, 2000). We have a case in which concerns about a European policy proposal were communicated to the variety of actors involved in the European decision making. We have a case in which traditional national actors were trying to respond to a European initiative. We will focus on Belgium and on the Belgian trade unions. Belgium and its trade unions have indeed been very active. The strong protection of the dock workers in the existing social law (see below) made them very alert for the possible effects of the port directive. Belgium is also an interesting case because it is never fully ‘national’. Belgium’s major harbour – Antwerp – is located in the region of Flanders, and the region itself has some important powers in social and economic matters. Trade unions are ‘unitary’ organizations, with however also regional branches, and political parties are fully split in Flemish and Francophone parties. In the European Parliament – but not in the Belgian Parliament – the elected members of the parties of the same ideological family belong to the same group. That creates quite an interesting institutional context and dynamic in which we will try to trace back what happened, who became involved and who liaised with whom. In the following paragraphs we will first briefly present the case itself. Next we describe the way in which the trade unions reacted. In the concluding section we try to see what lessons can be learned from this case.

159

Politics Beyond the State.book Page 160 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

The directive ‘Port services: market access and financing of maritime ports’ The EC prepared a Green Paper on ‘Sea Ports and Maritime Infrastructure’ in 1997. The main aim of this Paper was to enhance transparency in (public) investments in ports and fair competition between ports. In this early phase the impact of political parties – national or European – is quasi non-existing. The impact of other national actors depends on their (participation in) lobby activities, own expertise, and/or membership in groups invited to participate in the consultation procedure. Employment in ports was actually only treated in an indirect way. It was not the central concern of the proposal. Yet the Commission did advise the International Transport Workers Federation (see below) and all its members to urge their governments to ratify the ILO conventions on work in ports, and that could have raised some questions. This convention indeed allows establishing registers for all categories of dock workers and says that registered dock workers should have priority of engagement. Belgium never ratified the convention because its own national laws provide for a higher protection of workers than the convention did (Baete, 2004). In September 1998 the EP issued a resolution asking for more transparency in port investments. After that, the Green Paper fell into oblivion for a while. Under the co-decision rule, the Council and the EP are the second and third crucial actors in the process (with also an opinion given by the Committee of the Regions and by ECOSOC). On February 13 2001 the Commission submitted a proposal for a EP and Council decision on ‘Market Access to Port Services’ to the EP (COM (2001) -2001/0047 (COD)). The proposal was sent to the responsible Committee on Regional Policy, Transport and Tourism (RETT) on February 28 2001. In this phase we would expect national actors to try to approach their national commissioner, even if he/she is not the responsible commissioner (which was indeed not the case for Belgium). It seems even more obvious that the ‘national’ MEPs will be contacted, and in particular those who are member of the responsible committee or those having a particular mandate in it (rapporteur, shadow rapporteur). Moreover, we would expect the MEPs, and even also their national party, to pay special attention to issues with major implications for their own electorate, and this as early as possible in the policy making process. Because of the small number of Belgian MEPs the interest groups might also try to contact other members. That will depend on the stage the procedure, the voting procedure (thus the number of people that need to be convinced) and the expected

160

Politics Beyond the State.book Page 161 Wednesday, December 5, 2007 8:41 AM

Sinking the Port Directive. Belgian Trade Unions’ Mobilization against EU Policy

outcomes of the deliberations in the institution. Input of expertise remains important here, when amendments need to be prepared. At the level of the Council, it seems obvious that groups will address the responsible minister and his or her national party. This could happen directly, in parliament, via organizations related to the party, etc. Another way of entrance could be the permanent representation. But also here, we expect that interest groups realize that it will not be enough only to lobby the national representatives. Additionally, it might be useful to involve all top politicians at the national level as these are very sensitive to their appeals for mere electoral reasons. The national arena is indeed where the media are. Media can become an important weapon when a European policy has major implications on the national level. Very sensible issues might be picked up both by governing and by opposition parties. The national electoral cycle is of crucial importance here. Yet it is also important to realize that many European issues – including this port directive – are quite technical and not easy to cover by the media or to be used for electoral mobilization. Another important national cycle directly linked to the European level is the presidency of the EU. The chairing country can influence the timing of the procedure, but the chairing country can also be vulnerable. It needs to watch its image vis à vis the other EU member states. In the following section we will have a close look at the way in which the national actors in Belgium have tried to influence the decision-making procedure for the port directive. We will describe and discuss the way in which trade unions and political parties – some of which have very close links to the trade unions – have tried to be both in Europe and in Belgium. It was indeed quite striking to see that this issue did result in a fairly high level of politicization and mobilization. One of the reasons for the salience of the port directive is certainly the very special position that dock workers enjoy in Belgium. One of the most contested elements of the proposed directive was the definition of ‘self handling’. In Belgium, dock workers are protected by the so-called ‘Law Major’ (1972), named after the former president of the socialist trade union Louis Major. This law states that only formally recognized dock workers can work in Belgian ports. The EC proposal would also allow (foreign) workers to carry out loading in European ports. Belgium has some 8000 dock workers. This is not an extremely large number, but these workers are very well organized and the economic damage they can bring to the country is very high. An interruption in the 161

Politics Beyond the State.book Page 162 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

transport supply chain has major consequences on almost all areas of economy. This dock workers statute received major attention during the campaigns against the directive. But the earliest warning on the consequences of the proposal came from the harbor pilots. They were engaged in negotiations with the Flemish regional government (Belgium’s major harbour – Antwerp – is situated in the region of Flanders) and were afraid that they would be included in the package. The statute of the pilots was certainly as important as that of the dock workers, but it was afterwards far less publicized. Belgium has fairly strong trade unions. They reflect the old pillarized structure of the political system, with a socialist (ABVV/FGTB), a Christian-democratic (ACV/CSC) and a liberal union (ACLVB/CGSLB). With 60% of the labour force affiliated to one of these unions, the trade unions are an important force. Their power also results from their position in the neo-corporatist logic of Belgian labour relations. The three large unions are recognized as ‘representative’ organizations, which gives them privileged access to the policy making. Several institutions for permanent negotiation and information exchange bring together trade unions, employers and sometimes also the government. Two of these institutions deal with the port issues. One is the federal ‘Nationaal Paritair Comité Havenbedrijf’ (NPCH, National Port Committee), the other is the Flemish regional ‘Vlaamse Havencommissie’ (Flemish Port Committee). The Belgian social partners soon realized that they had to mobilize the available means to avoid the potential negative impact of the port directive, and this did happen in the national and Flemish port commissions. On February 2001 the NPCH gave the federal Minister of Employment and Labor the advice to accept the request of the Belgian members of the International Transport Federation to ratify the ILO conventions. The NPCH hoped that this convention could be a safety net in case the directive would be accepted. The National Port Committee subsequently remained an important forum for action. It prepared an amendment in June 2001 and issued a declaration to endorse the ‘Major Law’ in March 2003. Coordination of the actions was mainly done in the Flemish Port Committee that issued several critical recommendations on the port directive. From the beginning, the Committee explicitly stated that it did not agree with several points and options in the directive, mainly also the concept of self-handling (Merckx & Neyts, 2005). Its recommendations accompanied all major steps in the procedure. It suggested several amendments on the first proposal, which (among others) made the EP agree on a first text on 14 November 2001. A next recommendation evaluated the text the EC proposed to the Transport Council in February 162

Politics Beyond the State.book Page 163 Wednesday, December 5, 2007 8:41 AM

Sinking the Port Directive. Belgian Trade Unions’ Mobilization against EU Policy

2002. The third initiative prepared the second reading in the EP and this resulted again in the approval of many amendments. In December 2004 the Committee advised the Flemish minister on the second port package. Several recommendations were also made on the working document of the EP transport committee rapporteur. During the first procedure, the Flemish Port Committee asked an external group of experts to make an impact analysis of the directive on the country. The results of this study were discussed in a special working group. During the second half of 2001, Belgium chaired the EU, and that did also influence the way in which the port directive was dealt with. When Belgium holds the presidency of the EU, this is always a major event. Many differences between government and opposition are put aside and all politicians do their best to enhance the image of the country. All potential issues that could disturb a smooth course of the event are avoided. And this is exactly also what happened with the port directive. The federal and Flemish governments were well aware that this directive could seriously disturb the social peace in the ports. All actors involved were invited to assess this risk and all agreed that social problems should be avoided during the presidency. The major strategy of the government was to delay the procedure. There is thus in 2001 a guideline proposal originating from the EU level that could have a major impact on the working conditions in port services. The proposal puts in danger a social group that enjoyed within the national framework a very strong protection. The group is well organized and is part of trade unions that play – given the strong neo-corporatist structure – an active role in policy making and implementation. At the national level they have thus a wide repertoire of possible actions, including strikes that can hurt deeply. Their request to keep the existing protection intact is however generally considered to be legitimate, even for employers and for the government. The issue is also one that can be considered as very European. Removing national peculiarities hindering the creation of a free market belongs to the core of the European Union. It is actually an issue in which both the policy choices and the very ‘raison d’être’ of the EU are involved. For the trade unions it is a nice case to illustrate the negative effects of free market liberalism on employment. It is a case in which the European policies can be framed as part of the classical left-right divide, the typical cleavage of national politics. But the national pressure groups do not have easy access to the European level. It is therefore interesting to see how they have dealt with the issue. 163

Politics Beyond the State.book Page 164 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

How did national trade unions try to play their role, try to mobilize the national rank and file and try to influence the decision-making at the European level? That is what we will discuss in the following paragraphs. We will describe the process of politicization of the port directive issue and the way in which trade unions and parties build links with their members and with the European institutions. We make use of books and documents and of interviews1 with several of the key actors in the process.

The Trade Unions Getting started The central player in the trade union actions was the European Transport Workers Federation ETF. It is the European branch of the International Transport Workers Federation ITF. The Commission deals directly with ETF, and ETF liaises with ITF. ETF was formed in 1999 and brings together affiliates of the Federation of Transport Workers’ Unions in the EU with the European affiliates of ITF. The relationship between the two organizations is still not very clear. Some European transport unions pursue only actions through ETF, with the ITF as secondary resource. Others see Europe as part of the broader international arena and consider ETF as the regional office of ITF. The transport unions of the Belgian 1.

We have spoken and/or communicated by e-mail with the following persons: Paul Beekman, The Greens/EFA, Adviser on Transport and Tourism, 19 August 2005 Jos Bertrand, Flemish socialist party, staff EP, 20 may 2005. Myriam Chaffart, ACV-Transcom, section water, spokeswoman Port Package campaign, 30 June 2005, August 2005. Jan Dhaene, Flemish socialist party and former member of the Flemish green party (MEP and member of the transport committee), 24 June 2005 Saïd El Khadraoui, Flemish socialist party, MEP and member Transport Committee, 21 June 2005. Marc Loridan, BTB, June 2005 Nelly Maes, Flemish left liberal party and former member Flemish national party, former MEP (1999-2004) and member of the transport committee, present chair of the, 19 August 2005. Kees Marges, Secretary ITF until 01/12/04, 29 June 2005. Bart Staes, Flemish green party, MEP Dirk Sterckx, Flemish liberal party, MEP and member of the transport committee Anne Van Lanker, Flemish socialist party, MEP, 22 August 2005. Amélie Verwissen, Francophone socialist party, staff responsible EU affairs, 16 August 2005. Toon Wassenberg, Flemish socialist party, staff cabinet Minister Van Brempt, 1 September 2005.

164

Politics Beyond the State.book Page 165 Wednesday, December 5, 2007 8:41 AM

Sinking the Port Directive. Belgian Trade Unions’ Mobilization against EU Policy

Socialist and Christian trade union federations ACV/CSC and ABVV/FGTB are members of the ETF: Transcom for the Christian union and BTB/UBOT for the socialist union. The liberal ACLVB/CGSLB has a sector ‘Transport’ that represents all transport sectors in general, but it is not affiliated to the ITF. ACLVB is part of the common trade union front in the port of Antwerp. It has always supported the initiatives but the socialist and Christian unions clearly played a much more prominent role2. In several transport industries, joint committees have been set up (sponsored by the European Commission) between employers and trade union federations. There is no such formalized consultation however for port activities. Both the European Sea Port Association (ESPO, major European ports) and the Federation of Private Port Operators (FEPORT) strongly oppose an organized social dialogue. ETF was consulted for the Green Paper within ECOSOC but the objectives mentioned in the paper were then not recognized as trade union concerns. In July 2000 ETF received a questionnaire from the European Commission on access to the port services market. This questionnaire was discussed and completed during a Dockers’ Conference of ITF (Baete, 2004: 140). For Myriam Chaffart, spokeswoman of the Christian trade union ACV/CSC in the port dossier, the proposal came as a ‘thief in the night’ (interview in June 2005). Her trade union was not aware of the proposal, let alone of its implications for the Belgian dockers. The slow understanding of the implications of the proposal was confirmed by her socialist colleague Bob Baete. He was invited as an expert during the ECOSOC consultation, but it was not clear at all for him what the exact aim of this document was. Later, together with his colleagues of the transport unions, he filled in the questionnaire of the European Commission, still not aware of the potential consequences of the proposals. Both trade unionists admit that at that time they were not at all familiar with the decision making structures and procedures of the EU (Janart, 2004; Baete, 2004). Interviews with MEPs confirm that the trade unions needed the help of the politicians to find their way in the European labyrinth. In 2001, the Dockers’ secretary of ITF Kees Marges warned in the press for a ‘war on the waterfront in Europe’. At the end of 2000 he had received a text of the Commission – not meant for the eyes of the trade unions – that worried him heavily. This press release gave way to a meeting of ETF/ITF with EC officials, once the White Paper was released, and eventually to 2.

ACLVB/CGSLB unfortunately did not respond to our (repeated) request to answer some questions about its involvement in the actions against the port directive.

165

Politics Beyond the State.book Page 166 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

some minor changes to the proposal. The Belgian unions ACV/CSC and ABVV/FGTB were present because they were easy to mobilize on short notice. And so they became closely involved in the EFT actions from the very beginning. In this period the EC still assured that the ‘Law Major’ was not in danger, and therefore the Belgian trade unions did not worry too much yet. The strategy ETF finally agreed upon was the following. National trade unions would lobby MEPs, national ministers and the Commission with core slogans. ETF/ITF delegations would lobby on the EU-level. During all EP meetings there were demonstrations, both in Brussels and Strasbourg. To put pressure on politicians in charge, work stoppages were organized simultaneously in different countries in compliance with each state’s national legislation. These were meant both to educate the dockers and to send a first warning to the EC. There was agreement on one and only one clear objective: to stop the directive. To increase the pressure the European struggle was linked to the worldwide ITF ‘Cargo handling Campaign’, aimed at adding a clause to all collective agreements for seafarers that would prevent them from handling cargo for little money. Thanks to these contacts, the Americans also got involved in the protest. The actions were assessed and re-defined in a co-ordination group. An information network was established to allow the distribution of all information (Kamin, 2005).

Preparation for the first reading (vote EP 14/11/01) The Belgian union sections applied the policy of ETF and indeed lobbied their MEPs and the Belgian minister of transport. Initially, Transcom (Christian union) mentions a meeting with the Christian democratic MEPs, while BTB/UBOT (socialist union) refers to amendments from the Flemish socialist MEPs. Later the contacts were expanded to other MEPs. In September 2001 the first European actions against the directives in the European ports were born. In Flanders, the work was stopped to attend information meetings. The ITF pamphlets that warned against Europeanwide degradation of social protection were however not generally used, probably because at that time the unions still believed the Law Major would protect them. The European component of the problem was not yet fully taken into consideration3 (Baete, 2004: 143).

3.

MARGES, K., e-mail correspondance June 2005.

166

Politics Beyond the State.book Page 167 Wednesday, December 5, 2007 8:41 AM

Sinking the Port Directive. Belgian Trade Unions’ Mobilization against EU Policy

One month later, during the informal European summit in Ghent, the dockers participated in the protest march for a social Europe organized by ETUC, ACV/CSC and ABVV/FGTB. By then the RETT had accepted a report the dockers were not prepared to accept. Their next target was therefore the EP, where this committee report was going to be discussed. The text that was finally accepted in the EP in November 2001 – after being adapted by the rapporteur and heavily amended by the EP – was still unacceptable for the unions. The EC did not agree with the EP’s amendments neither and it issued a new proposal for a directive in February 2002.

Preparation for Council activities 03/02-06/02 and the second reading (vote EP 11/03/2003) A first action was planned in the context of a meeting in Brussels that would be attended by the Belgian minister of Transport and the EC Commissioner responsible for the proposal of directive. Unfortunately, they did not show up but the EC Director of DG Transport assured once again that the wage and labor conditions in the ports were not under threat (Baete, 2004: 145). The Euro-protest march in December 2001 in Brussels, again an organization of ETUC, ABVV/FGTB en ACV/CSC, was (amongst others) meant to convince the Minister of Transport to reject the directive in the Council. The lobbying of MEPs, of the minister of Transport and of the EC went on, both by the trade unions and by the Joint Committee of Ports (which includes the Belgian port employers). Following these protests, the Belgian minister of Transport promised to reject the proposal during the next Council meeting in June 2002 (Baete, 2004: 146-150). Docker unions were also present during the ETUC protest in Barcelona on 14 March 2002. Two weeks before the Council meeting, the Belgian trade unions decided to increase the pressure and they announced a first 24 hours strike in all Flemish ports. The Germans would follow one week later. According to Chaffart, this was an important turning point for the Belgians. From that moment on they took the lead in the EU actions against the directive. Because of the Law Major, they had much more to lose than the dockers in the other countries. And they realized more and more that this law would not protect them ‘against Europe’. Besides that, they felt like they had to do a bit more because of the presence of the EU institutions in the country. The Flemish MEPs had urged the unions to enhance pressure on the Council as well. 167

Politics Beyond the State.book Page 168 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

The Council however accepted the EC-text and a common point of view of the EC and EP was issued in October 2002. This document was sent to the EP for a second reading. In November 2002 the Flemish trade unions participated in a conference of the North Sea Ports where French, German and Dutch unions met with their Belgian colleagues. Other conferences would follow in June and September 2003. The driving force behind these meetings was the Belgian socialist BTB/UBOT. The aim was to form a united front with non-ETF members against the port package4. This approach turned out to be very fruitful. Still in November 2002 the port unions of ETF met again with the EC but it was clear that the EC was not prepared to change its point of view (Baete, 2004: 154). It was only in December 2002, during the International Conference of ETF, where unions, employers, port managers and politicians discussed the directive, that Chaffaert realized that also some European employers organizations had important criticisms against the directive. It was also at that moment that the MEPs asked the trade unions publicly to raise the social pressure. For the MEPs it became increasingly difficult to make their point in the EP if this would not be supported by social pressure. This pressure was indeed provided by the actions of the unions. In January 2003 the unions participated in the first real European port strike, organized in 13 EU member states in the prospect of the next RETTcommittee meeting. During the meeting in February, unions protested in front of the EP building in Brussels. This strike was supported by a ETF protest march in Brussels one month later. The actions were again supported by several MEPs, who had issued many amendments on the text of the directive. The idea was that all amendments of MEPs for the next text of the plenary session of March 2003 would be supported by all other parliamentary groups. Plan A was (amongst others) to redraw the concept of ‘self handling’. If this would not work, plan B aimed to limit this ‘selfhandling’ as much as possible. Yet none of the Belgian amendments were accepted. A next important step was the initiative of the social partners of the port in Antwerp to bring together the port managers of the Flemish 4.

There are two different umbrella organizations competing for dockers affiliates. On one side there is ETF, on the other hand there is the International Dockworkers Council (IDC) to which the majority of the French and the Spanish dockers are affiliated. For the actions against the port package, this problem was solved by the Belgische Transportwerkers Bond (BTB, Belgian Transport Workers Union of the socialist trade union). It convened the so-called North Sea Range Conferences.

168

Politics Beyond the State.book Page 169 Wednesday, December 5, 2007 8:41 AM

Sinking the Port Directive. Belgian Trade Unions’ Mobilization against EU Policy

ports, the port unions, the employers’ organizations, the Belgian political parties and MEPs. Meanwhile, the unions kept the pressure high. Before and during the second plenary session, new strikes and protest marches were organized in Brussels (07/03) and Strasbourg (10/03), both by the Belgian unions and ETF. They managed to get new amendments on the agenda, although the time to introduce them had expired long before. Some important amendments were then accepted in the EP. The EC and the Council though rejected the point of view of the EP and thus a conciliation committee was established.

Preparation for the conciliation process and, eventually, the third reading (vote EP 20/11/03) During the Belgian-Dutch port seminar in February 2003, the Flemish regional Minister of Transport (socialist) assured the trade unions that the Belgian government supported their position. Later on also the regional Flemish minister of Mobility and the federal Minister of Labour Policy publicly declared their support. At the federal level, the social partners issued a declaration in the Joint Committee and declared once more their attachment to the Law Major. In April 2003 the common declaration of the national port dialogue was renewed. The unions also received the assurance of the Flemish liberal MEP and member of RETT, Dirk Sterckx, that social dumping had to be prevented. Given his position in the EP and his voting behavior in the past, this was an important announcement for the unions. Several actions in all European ports supported the position of the EP at this stage. Again, a major European manifestation was organized, this time in Rotterdam. Notwithstanding all this support, the conciliation procedure did not produce a satisfactory agreement for the unions. So, the third plenary session offered a last chance for the unions to communicate their grievances. Meetings were organized with the federal Minister of Transport, the president of the Flemish Christian Democrats, MEPs, the Federal and Flemish Ministers of Mobility, the State Secretary of Labor Organization and others. All agreed that the directive had to be stopped. With the socialist ministers alternatives were discussed for a new strategy if the third reading would result in a failure.

169

Politics Beyond the State.book Page 170 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Also within ETF the strategy was re-evaluated and pressure increased. In Strasbourg all MEPs were invited to an information meeting of the trade unions. The first day of the plenary session in November 2003 all European ports organized actions. Delegations of the trade unions visited all the party groups in the EP to communicate their demands. And the president of the EP received 16000 signatures of European dockers against the directive. In Belgium the homework was already done: most MEPs would vote against the directive. It was now also up to them to convince their colleagues in the EP. After the vote in the EP – 229 MEPs voted against, 209 for and 16 abstained – the trade unions celebrated the withdrawal of the directive and the success of their actions. On October 13 2004 the EC tabled its second proposal for a directive. The chair of the transport committee issued his first working papers and consulted the actors involved to prepare a new report. The second proposal was again not what the unions expected. But the trade unions now kept from the very beginning a close eye on the issue. It was clear that the mobilization machine was now really well-oiled. The second proposal would be again rejected by the EP in January 2006 after again massive protest all over Europe and in Strasbourg on the day of the vote. In January 2006 the Commission finally closed the case and withdrew the proposal.

Conclusions There is quite a contrast between the early days of the first proposal in 1997 and the final mobilization in January 2006. Between the two lies a period during which many actors have learned how to put pressure on the European institutions. Between the two lies also a period that allows us to see how difficult it is for organizations that are in the first place rooted in their national state and national context to combine actions at the state level and at the European level. The transport sector had a representative organization at the European level. It was involved in the consultation process of the Green paper. Yet it did not see what was going on. It is true that the Commission was not very explicit about the possible consequences of the directive, but it is also true that the ETF was apparently not able to fully decode the proposal. The EFT interventions against the port package only really started after the first reading in the European Parliament. The slow start is due to several elements. An important hurdle was the major differences in the legal 170

Politics Beyond the State.book Page 171 Wednesday, December 5, 2007 8:41 AM

Sinking the Port Directive. Belgian Trade Unions’ Mobilization against EU Policy

frameworks the trade unions are functioning in. In Germany, for example, industrial actions against proposed policies are prohibited. The German dockers could thus not participate in the strikes. The Belgian, Spanish, Finnish and French dockers initially condemned this ‘passivity’ of the Germans. It was the role of ETF/ITF to inform its members about these national peculiarities. From the Belgian perspective it is important to point again at the Law Major, that seemed for a while a firm guarantee for the Belgians that nothing serious could happen to them and that Europe would not really ‘hit’ home in this respect. Another problem was communication. Amongst the older trade union leaders, all socialized in their national environment, there were quite a number who did not speak English. That was a problem for internal communication and rapid co-ordination of actions, but also a problem for setting up direct contacts with the office holders in the European institutions. It is clear that the trade unions, both at the national and at the European level, went through a learning process. But once this was over, the ETF not only reacted to the European initiatives but became very pro-active itself. When the second port package was sent into the European decisionmaking process, the protest actions were deadly efficient and the port package was indeed buried. The ETF combined a political approach – lobbying the national and European political actors – with social pressure. And the social pressure was also organized both at the national and at the European level. The MEPs also became aware that the game had to be played and was being played at more than one level. The Belgian MEPs were confronted with a common front of trade unions, employers, port management and politicians in the country. The port management of the port of Antwerp made important statements against the directive from the very beginning. Meetings of all the actors involved were taking place in the so-called ‘Port house’ in Antwerp, the office of its management. Not all actors had the same reasons to denounce the proposals, but all agreed they had to stop it, or at least try to change it. Common declarations of the actors involved were issued in the existing negotiation bodies, both at the national and the Flemish regional level. It was not always easy for the unions to find a balance between the need to use social pressure and the need to co-operate with the port management that (of course) wanted to prevent economic losses. Several politicians played a mediating role between the two.

171

Politics Beyond the State.book Page 172 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

With this Port Directive Europe has really hit home. And the home was reorganized to deal with the European issues. In the end the mode of mobilization and Europeanization clearly reached the level of supranationalism (Balme & Chabanet, 2002). The Port Directive was sunk, and in the short term there will be no fundamental changes in the organization of the work in the European harbours. There will be no policy change. But the actors involved in the sinking of the directive did change. They will not any more be taken by surprise by the European level.

172

Politics Beyond the State.book Page 173 Wednesday, December 5, 2007 8:41 AM

Deliberation Beyond the State. Towards a Deliberative Interpretation of EU Consultation Mechanisms Irina Tanasescu

Introduction One of the main features of governance is the involvement of private actors in public policy, both at the level of policy design and at the implementation stage. The former corresponds to the need of public authorities to tap on the expertise of private actors and to the perceived increased legitimacy that such an inclusive policy-making process entails. The latter, clearly exemplified by voluntary instruments such as ecolabeling schemes or social partner agreements, is a consequence of the assumption that such instruments are less costly, less time consuming and easier to implement than traditional legislation. Policy making at the level of the European Union is a very interesting example of the complex interplay, at different level, between private and public actors. Because of its relatively small size and subsequent lack of in-house expertise, the European Commission (24.000 staff, around the same as the administration of a medium-size European city) not only welcomes but openly invites and encourages input from the different interest groups in the policy design stage. In addition to bringing much needed expertise, such a wide involvement is also a source of democratic legitimacy and a possible answer to the democratic deficit accusations targeted at the EU institutions. In order to understand how the Commission and third parties interact and thus shed more light on private actor involvement in public policy-making, I suggest the use of deliberative democracy, a theoretical approach very much in line with the concept of governance. Deliberative democracy theories have been increasingly developed over the last two decades and have come to supplement the classical liberal theories of democracy, outdated by the developments of the contemporary world. The virtues of deliberation and the search for common good through discussion among equals have been praised both theoretically and in more practical test cases of local governance. It was not long before deliberation draw the attention of EU specialists, who quickly saw the possible link

173

Politics Beyond the State.book Page 174 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

between the European Union and this line of thinking. Thus, a series of authors started to analyze the institutions and processes of the EU through the deliberative lens (Joerges and Everson’s and Joerges and Neyer’s study of comitology for instance), while others created theoretical paradigms based on deliberative principles and assessed EU processes against them (such as Fossum and Menéndez on EU constitutional politics). Despite this rich literature, little, if any, attention has been paid to the possible relevance of deliberative democracy to the relationship between the EU and civil society organizations. Given that the main democratic deficit reproached to the EU stems from the lack of inclusion and participation of citizens in the decision-making process, looking closely at consultation mechanisms through the deliberative lens might get us to a better understanding of the input civil society can bring to the legislative work of the EU. In addition, deliberative democracy could be an inspiration for redressing actual problems and shortcomings of the consultation mechanisms as they stand today. This is a first attempt to bring deliberative democracy and EU-civil society relations on the research agenda. In addition to being a normative principle, deliberative theories can also be a useful approach for describing and understanding real-life policy processes. This is just an initial sketch, a preliminary exercise meant to explore the topic and to set the challenge of investigating it further in the future. After a brief presentation of deliberative democracy and of its use in EU studies so far, I will turn to two concrete examples of EU consultation mechanisms and try to see what deliberative theories can bring to their understanding and amelioration. Social dialogue is the first example. It was picked because it is the most institutionalized mechanism, given that the role of the social partners is inscribed in the Treaty. In addition, trade unions and employers’ organizations involved in the process enjoy powers that no other civil society groups have at the EU level, ranging from mere consultation to the actual drafting of EU directives. Having seen how social dialogue works and in which way deliberative democracy might enlighten our understanding of it, I will turn to the second case study: EU environmental policy-making and, more precisely, the EU Eco-labeling scheme. The choice of this particular instrument is motivated by the growing importance of EU environmental policy, both in terms of sheer volume and in terms of impact range. Moreover, EU environmental policy is the leading policy field in the shift from traditional, hard law, regulatory-based instruments to voluntary, soft law instruments, shift that requires an understanding of how the EU institutions work in this new paradigm. If environmental policy-making is heading this way, then a clear

174

Politics Beyond the State.book Page 175 Wednesday, December 5, 2007 8:41 AM

Deliberation Beyond the State

understanding of the voluntary instruments it relies on is more than needed. Deliberative democracy could, in this case, not only help us theoretically grasp these instruments but also provide ideas for their improvement. Having concluded on the lessons learnt from the two test cases, I will then identify salient issues and propose future research tracks.

Deliberative democracy Stemming from the liberal traditions of the 18th century, deliberative democracy has been increasingly developed over the last 20 years, especially through the contribution of major thinkers like Rawls and Habermas. On the one hand, this development corresponds to a time of declining trust in the political class and in institutions and of subsequent decreasing participation in electoral politics. In this context, deliberative democracy is a way to supplement traditional forms of democracy by advocating for the involvement of citizens in decision-making in ways other than electoral. On the other hand, deliberative theories also developed as a response to the aggregative theory of democracy proposed by rational choice theorists (Rosenberg, 2004). In contrast to the latter, who emphasize the selfishness of the individual in the pursuit of his own interest, deliberative approaches offer a different, more generous view of the individual and of decision-making. Following Rawls, the democratic citizen is considered to be endowed with the cognitive capacity to argue with reasons, to recognize criteria of justification, to understand rules of evidence, to be logical and to reflect on his own presuppositions (Rawls, 1993). Thus, an individual is not only a rational actor who makes choices and pursues his personal interests, he is also an ethical and moral agent who reflects and collaborates. Even if individuals do not have the motivation to put these inherent qualities to good use, theorists argue that the deliberative setting will help them think and interact in a way that is more logical, rational, just, considerate of others, self-critical and oriented to the common good (Rosenberg, 2004). Thus, the assumption is that deliberative democracy is possible because individuals are endowed with the above-mentioned capacities that are not only natural but also enhanced by the deliberative circumstance itself. Following Benhabib, “it is a necessary condition for attaining legitimacy and rationality with regard to collective decision-making processes in a polity, that the institutions of this polity are so arranged that what is considered in the common interest of all results from processes of collective deliberation conducted rationally and fairly among free and 175

Politics Beyond the State.book Page 176 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

equal individuals” (Benhabib, 1996). This is a procedural definition, as it is the process of deliberation itself that ensures the legitimacy of the outcome. Nevertheless, in addition to legitimacy, deliberative practices also generate practical rationality. On the one hand, through taking part in deliberation participants become informed of the others’ positions and of the possible consequences a decision might have on their interests. All are, therefore, more aware of possible conflicts and more prone to compromising. On the other hand, given the nature of the process, individuals must bring good reasons to support their positions. This very process of finding the best reasons brings one to the position of thinking of what would count as a good reason for the others, thus being obliged to think from different perspectives. In addition, the need to publicly state and defend one’s arguments is conditioned by the “civilizing force of hypocrisy” (Elster, 1998), namely the need to provide reasonable and acceptable arguments. To summarize, deliberation is thought to help bring about competent (and, as far as possible, consensual) policy-making through reflection and discussion. In addition, deliberative processes might also produce shared collective meaning and a stronger sense of community (Papadopoulos, 2002). For deliberation to begin, several conditions must be met (Rosenberg, 2003): a) A suspension of action to create the necessary political space for the deliberation to occur. There must be guarantees that no political decision or action will be taken until the deliberation is completed. b) Inclusiveness. All parties potentially affected by the decision or representing all relevant points of view need to be represented. Given the fact that inclusiveness might in some cases lead to scale problems, Parkinson suggests the legitimacy of representation as a possible solution. Thus, he argues that there are legitimate ways of limiting participation and that the legitimacy of representation depends partly on seeing deliberative forums as being embedded in a wider deliberative system in which legitimacy is created in the openness of the linkages between moments, rather than relying on ideal legitimacy conditions for each moment taken separately (Parkinson, 2003). c) Publicity. All those affected but not directly involved need to be informed and have the right to react to the outcomes of the deliberation process. Moreover, publicity is also required in the name of accountability and considered, by some authors, even favorable to deliberative outcomes (Elster, 1995). d) The requirement that the results of the deliberation be binding on those involved in the process. This is needed in order to ensure that deliberative

176

Politics Beyond the State.book Page 177 Wednesday, December 5, 2007 8:41 AM

Deliberation Beyond the State

institutions are not mere talking forums. Quoting Cohen, “the principal virtues of the deliberative conception are allied closely to its conception of biding collective choice” (Cohen, 1998). It is precisely this binding character that enhances the value of deliberation. In addition, the measures agreed upon will encounter lesser resistance as compared to decisions taken in non-deliberative instances. Theorists also seem to agree on a series of standards of conduct that need to be met, such as: a) The autonomy of participants and equality of participation. Participants are and regard each other as free and equal. This means, in Cohen’s perspective, that no religious or moral view provides a defining condition of participation or a test of the acceptability of arguments. In addition, the rules regulating the deliberative procedure do not confer special advantages or disadvantages on individuals. Every participant has and is recognized as having equal status in the deliberation (Cohen, 1998). – “the outcomes of deliberation must be consistent with the associated values of justice as fairness and democracy as governance oriented to the common good and guided by the principles of autonomy and equality” (Rosenberg, 2003) – a “shared appreciation of the truth and right of the reasons for the collective choices being made” (Rosenberg, 2003). The claim is that once these conditions and standards are met the ensuing discussion will be deliberative. “As such, it will consist of a respectful and reciprocal expression, correction, revision and restatement of views. In the process, thinking will become more logical and self-reflection will become deeper and more critical. As a result, personal beliefs, values and preferences will change. At the same time, this will encourage the discovery of a common ground for agreement, one that will yield more just and legitimate recommendations for public policy. This in turn will provide the basis for both a renewal of interest and faith in democratic governance (thus addressing current problems of declining interest and participation in politics) and a means for social reintegration (thus addressing the problem of a socially destructive individualism and a socially disintegrative multiculturalism)” (Rosenberg, 2004). Deliberation is distinguished from other kinds of communication, such as consultation or bargaining for instance, in that “deliberators are amenable to changing their judgements, preferences, and views during the course of their interactions, which involve persuasion rather than coercion, manipulation, or deception” (Dryzek, 2000).

177

Politics Beyond the State.book Page 178 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Deliberative democracy and EU studies The link between deliberative democracy theories and the European Union was not hard to make, taking into account at least two different dimensions. On the one hand, at a general level, the EU’s democratic deficit stems to a large extent from secrecy, bureaucratization and lack of involvement of citizens in the decision-making. The ideas and procedures of deliberative democracy – putting the emphasis on constructive discussion prior to taking a decision – could address the problems of the EU. Following Hoskyns, “deliberative democracy is relevant in this context because it goes beyond these aspects of democracy (i.e. representation) to address more centrally the issue of participation, and to suggest how, if participation is broadened, different perspectives and points of view can be accommodated. In doing so, it both assumes and helps to create a public space for debate and action – features most notably lacking in the EU as at present constituted” (Hoskyns, 2000). This view is also shared by Cohen and Sabel (Cohen and Sabel, 1997) who go beyond deliberative democracy to shape a new concept: “directly-deliberative polyarchy”. They propose in fact a novel form of public governance that “combines the advantages of local learning and self-government with the advantages (and discipline) of wider social learning and heightened political accountability that result when the outcomes if many concurrent experiments are pooled to permit public scrutiny of the effectiveness of strategies and leaders”. Their analysis of this new form and of its relationship with traditional notions of democratic constitutionalism is relevant for the EU legitimacy debates, as the EU might become the ultimate form of directly-deliberative polyarchy. More recently, Stijn Smismans proposed a new concept, “reflexivedeliberative polyarchy”, which brings together the advantages of directlydeliberative polyarchy and of reflexive law into a normative framework to look at governance processes in the EU. Its goal is to encourage both decentralized deliberation among individuals and reflection among subsystems within a polyarchical institutional setup (Smismans, 2005). On the other hand, at a more specific level, deliberative principles have been identified in or tested against a series of institutions, formats or principles of the EU. Joerges and Everson (Joerges and Everson, 2000) for instance see comitology as an instance of “deliberative supranationalism” operating within a novel constitutional framework. They argue that the process of European integration has lead to a series of institutional innovations that question the traditional paradigms of bureaucracy. The committee system is one of these innovations, an “institutional form of regulatory policy” meant to address the novel challenge of regulatory 178

Politics Beyond the State.book Page 179 Wednesday, December 5, 2007 8:41 AM

Deliberation Beyond the State

policy at the EU level. The processes of negotiation and opinion-forming within these committees do not seem to fit within the intergovernmental theories and document the emergence of an incipient European polity. Solutions are found through co-operative approaches and participants go beyond their own state’s interest in a complex interweaving of arguments. Deliberative principles seem thus to be at work in this instance, thus challenging both traditional visions of decision-making in the EU (supranational and inter-governmental) and the pessimistic accounts of comitology as a closed, bureaucratic procedure. John Erik Fossum and Agustin Menéndez (Fossum and Menéndez, 2004) try to shed light on a different European issue by apply deliberativedemocratic principles, namely the constitution-making process (understood as both the IGC-model and the Convention as set up by the Laeken Declaration). Their focus is to see how the basic norms of the EU are deliberated and agreed on and to assess the extent to which the resulting constitutional framework would enhance the democratic character of EU decision-making processes. The deliberative character of the constitutional process is assessed against an ideal model of constitution-making structured around criteria such as inclusion/ participation, the nature of the process, transparency and accountability. They conclude that the Laeken process performs better than the IGC model on most of the criteria and that, if it bears fruit, it would lay the ground for the use of the European constituting power in the future. More recently, Klaus Bachmann argues that the debates in the European Convention revealed a deliberative process at work, despite not fulfilling all the theoretical criteria outlined by Habermas. Nevertheless, the success of the Convention and the broad consensus it reached in many fields, unlike previous Intergovernmental Conferences, can be seen as a consequence of the deliberative character of the debates (Bachmann, 2005). Deliberative principles have also been applied to the study of EU social policy, either to prove that it does not fit deliberative criteria (Wendler, 2003) or to assess the Open Method of Coordination (Smismans, 2005; Armstrong, 2006). When deliberative democratic normative principles are applied to the OMC in the fields of employment and pensions, the results show not only important differences between the two but also the fact that the OMC does not live up to its own democratic ambitions (de la Porte&Nanz, 2003).

179

Politics Beyond the State.book Page 180 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

The deliberative qualities of the EU institutions and principles have also been analyzed in relation to the Council of Ministers1 (Gargarella, 2000), the Conference of Parliaments2 (Blichner, 2000) and the principle of subsidiarity3 (Follesdal, 2000). All the above-mentioned analysis of EU politics and institutions using democratic deliberative lenses can be organized into two main categories. Some authors have focused on inter-state (the Council of Ministers, comitology) or inter-institutional (COSAC) negotiation formats, while others build theoretical models as standards or place their focus at a very general, theoretical level. Little, if any, attention has been paid to the interaction between civil society and the EU institutions from a deliberative perspective. This is an extremely important exercise, not only given the call for more openness and participation, but also in order to assess the democratic quality and the added value of the consultation processes. Given the difficulty to frame theoretically the interaction between the EU and organized civil society groups, deliberative democracy could complement existing models and could help shed more light on the issue.

Towards a deliberative interpretation of interest representation Interest representation has mainly been analyzed using the pluralist and the neocorporatist framework, and, more recently, with the help of policy 1.

2.

3.

Gargarella argues that the Council of Ministers has its philosophical roots in the conservative tradition, characterized by its defending of centralized institutions, low public participation, limited public discussion and many “counter-majoritarian” devices. This ideological rooting needs to be challenged for the institution to be readjusted and made more open and deliberative. Blichner focuses on increasingly numerous contacts between parliaments throughout the EU in different formats, one of which is the COSAC (Conference of European Affairs Committees). He argues that interparliamentary discussions might help meet the democratic challenge of the EU by serving as an intermediary public sphere and by creating the conditions for reaching a discursive agreement. In his view, democratic legitimacy in the EU is dependent on arenas of free discussion, unhindered by national or ideological considerations. COSAC seems to be such an arena, precisely because it is meant to be a “talking” forum, with limited capacity to act. Follesdal argues that the conception of subsidiarity as spelled out in the Amsterdam Treaty conflicts with the principles of democratic deliberation when it comes to institutional arrangements. This conflict is threefold: the accountability of the application of the subsidiarity principle, the unwarranted powers it grants to Member States and the possible hindering of the development of trans-European values and commitments.

180

Politics Beyond the State.book Page 181 Wednesday, December 5, 2007 8:41 AM

Deliberation Beyond the State

network analysis and exchange theory. They have all contributed to enlightening a particular aspect of interest representation, but they have neglected the importance of preference formation within the consultation bodies in which lobby groups are present. More precisely, little, if any, attention has been paid to what happens, for instance, in the interaction between the European Commission and the interests it consults. Therefore, deliberative democracy might prove useful in clarifying this interaction, by helping to understand how groups interact both among themselves, when given the institutional setup to do so (in the EU-level social dialogue for instance), and with the European institutions. The connection between deliberative theories and interest representation was drawn, at a theoretical level, by Jane Mansbridge back in 1992, but has not been followed up or tested with empirical case studies. Her argument is that interest groups function not only to pressure but also to persuade. Traditional research has been picturing interest groups as vehicles for pursuing conflict, power and will in a rent-seeking model in which whatever I get must be taken from you. Nevertheless, Mansbridge argues, politics offers possibilities for joint gain, fostered or smoothened by processes of deliberation. A well-designed system of interest representation would produce as many deliberative gains as possible with as few rentseeking costs. The design of such a system would necessarily imply a regulatory function of the public authority in order to reduce some of the inequalities that otherwise dominate the interest group deliberative system. A laissez-faire market in interest representation would give participants extremely unequal chances of being heard in the process. This need for an active role of the regulator seems to be met in the case of the EU by the position of the European Commission, who spends 1 billion Euros each year to support NGOs (Greenwood, 2005). If this condition is met, and if organized interests are consulted, why would they go for bargaining rather than for deliberation? After all, deliberation was proved to work in other instances of EU policy-making, such as comitology or the Convention on the Future of Europe. A good start for assessing participation and consultation mechanisms through the deliberative lens is to look at two of the most institutionalized forms of participation at the EU level. On the one hand, social dialogue, as the only consultation mechanism inscribed in the Treaty. On the other, voluntary instruments in environmental policy-making, out of which for the purposes of this study I chose to pick the EU Eco-labeling scheme for the multiple venues for stakeholder involvement at different levels.

181

Politics Beyond the State.book Page 182 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Social dialogue Social dialogue practices play an important role in the European Union, not only because they are seen as an integral part of the European social culture but also because, from a purely legal point of view, social partners can intervene directly in the making of EU social legislation. By “social dialogue” I understand, for the purposes of this paper, negotiations and bargaining between trade unions, employers’ organizations and the national government or EU institutions. It can take place under different forms, stretching from inter-sectoral to sectoral dialogue and even to implementation of EU directives. Ever since the Treaty of Rome, workers and employers have had, under one form or another, a say in the EU social policy-making. Their influence increased over the years and evolved from simple consultation to the legislator power they enjoy nowadays. Nevertheless, despite a series of more or less formal meetings during the ‘60s and the ‘70s, the social partners did not manage to arrive to an institutionalized level of European collective bargaining until the mid ‘80s. In 1985 Jacques Delors, then President of the Commission, invited the Union of Industrial and Employers’ Confederations of Europe (UNICE), European Trade Union Confederation (ETUC) and the European Centre of Enterprises with Public Participation and of Enterprises of General Economic Interest (CEEP) to a meeting that would set the basis for a new social dialogue (the Val Duchesse social dialogue process) aimed at involving the social partners in the internal market completion. One of the reasons that had drawn Delors to initiate this process was the consistent use of British veto power in social policy matters (Franssen, 1998:61). In this context, the European social dialogue could have been an alternative path to a social dimension of the Community. The meeting at Val Duchesse is considered to be the true moment of birth of the European social dialogue, given that it brought together the three main social partners and inaugurated a period of continuing dialogue between them. The agreements that the partners reached under this system were mainly non-binding joint opinions, given the reticence of UNICE to engage in bargaining on legally binding agreements. The Single European Act (art. 118B) marks the insertion of “social dialogue” in the Treaty, thus creating a specific treaty basis for the process and making possible the existence of collective agreements at Community level. Nevertheless, no procedures were prescribed and article 118B seemed to be more a political legitimation of the Val Duchesse process than

182

Politics Beyond the State.book Page 183 Wednesday, December 5, 2007 8:41 AM

Deliberation Beyond the State

a clear description of social dialogue mechanisms. The results of the social dialogue after the SEA were a series of joint opinions on general issues such as the economic situation of the community, informing and consulting employees, etc. The social partners decided in 1991 to form an ad hoc group whose task was to study how the social dialogue could develop in the new institutional framework of the future EU Treaty and submit the conclusions to the intergovernmental conference. The outcome was a joint agreement on social policy that was subsequently adopted by all Member States with the exception of the UK and appeared in the social protocol of the Maastricht Treaty. The Amsterdam Treaty will put an end to the UK exception (ruled by a new Labour government) by incorporating the substance of the social protocol in Art. 137-139, equally applicable to all Member States. The agreement between the social partners and their willingness to make consultation a compulsory stage of social policy-making could not have been achieved without a radical change in UNICE’s position. The Treaty revisions seemed to include qualified majority voting in social policy areas. Thus, one veto (that could have been easily obtained by lobbying in favor of employers’ interests) would not be enough anymore. Faced with the threat of legislation, UNICE decided it was better to be part of the decisionmaking process than undergo its effects. A new era began in 2001 with the social partners announcing, in a joint contribution to the Laeken Summit, their commitment to an increasingly independent and autonomous social dialogue. In 2002 the social partners adopted their first joint “multi-annual work programme” for 2003-2005. The role of the social partners in the European social dialogue process is clearly described in Art. 137-139 of the Amsterdam Treaty. Their prerogatives can be summarized in three categories: Implementation of Community directives “A Member State may entrust management and labour, at their joint request, with the implementation of directives adopted pursuant to paragraph 2. In this case, it shall ensure that, no later than the date on which a directive must be transposed in accordance with Article 249, management and labour have introduced the necessary measures by agreement, the Member State concerned being required to take any necessary measure enabling it at any time to be in a position to guarantee the results imposed by that directive.” (Art. 137.3)

183

Politics Beyond the State.book Page 184 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Consultation “ 1. The Commission shall have the task of promoting the consultation of management and labour at Community level and shall take any relevant measure to facilitate their dialogue by ensuring balanced support for the parties. 2. To this end, before submitting proposals in the social policy field, the Commission shall consult management and labour on the possible direction of Community action. 3. If, after such consultation, the Commission considers Community action advisable, it shall consult management and labour on the content of the envisaged proposal. Management and labour shall forward to the Commission an opinion or, where appropriate, a recommendation.” (Art. 138) Self-regulation “1. Should management and labour so desire, the dialogue between them at Community level may lead to contractual relations, including agreements. 2. Agreements concluded at Community level shall be implemented either in accordance with the procedures and practices specific to management and labour and the Member States or, in matters covered by Article 137, at the joint request of the signatory parties, by a Council decision on a proposal from the Commission. The Council shall act by qualified majority, except where the agreement in question contains one or more provisions relating to one of the areas for which unanimity is required pursuant to Article 137(2). In that case, it shall act unanimously.” (Art. 139) We can easily see what a huge impact the social partners can, theoretically, have on the development of the EU social policy. They enjoy a unique position in the institutional system of the EU Treaty, not granted to any other interest groups. Thus, both at the sectoral and at the inter-sectoral levels, the social partners can engage in European-level collective bargaining and see the outcome of their negotiations turned into compulsory EU law. Currently there are 31 sectoral social dialogue committees at the EU level, dealing with specific issues ranging from transports to live performance. They consist of a maximum of 50 representatives of the social partners, comprising an equal number of employers' and workers' representatives and are chaired either by a representative of the social partners or, at their request, by the representative of the Commission. In all cases the latter provides the secretariat for the committees. The work of the sectoral

184

Politics Beyond the State.book Page 185 Wednesday, December 5, 2007 8:41 AM

Deliberation Beyond the State

committees is completed by the cross-sectoral dialogue, in which 4 partners are present: ETUC, UNICE, CEEP and UEAPME4. The concrete results of the social dialogue process do not match its strong legal basis and potential important impact. Out of the 31 sectoral dialogue committees and the cross-sectoral forums, in 13 year after Maastricht, we have 6 binding agreements – 3 cross-sectoral5 and 3 sectoral6 – and a long list of non-binding recommendations and declarations7. The results of the negotiations have been modest, both in quantitative and in qualitative terms. To put it into bargaining language, there seems to be no serious commitment from the employers’ organizations to engage in collective bargaining at the EU level and no real power of the trade unions to force them to do so. Trade unions do not dispose, at the European level, of the same bargaining power they enjoy at the national level given the impossibility of a “European-wide strike” for instance. On their side, employers’ organizations keep on fleeing any kind of binding agreement, which explains the relatively high number of non-binding documents they have signed (as opposed to 6 binding ones). Trying to see the European social dialogue through the deliberative democratic lens, the first thing to notice is the high number of fora available for employers and trade unions to discuss. They are the sole masters of the agenda (they can respond to demands from the Commission or discuss issues of their own choice) and of the rules of procedure. The participants are selected having regard to a series of representativeness criteria devised by the Commission. Thus, in order to be part of the European social dialogue, an organization must: – be cross-industry, or relate to specific sectors or categories and be organized at European level – consist of organizations which are themselves an integral and recognized part of Member States' social partner structures and with the

4. 5. 6. 7.

UEAPME stands for Union européenne de l'artisanat et des petites et moyennes entreprises On parental leave (1995), part-time work (1997) and fixed-term contracts (1999) – all turned into Council directives On work time for seafarers (1998), work time for railway workers (1998) and on work time for civil aviation staff (2000) – all turned into Council directives 179 non-binding texts came out of the sectoral committees between 1997 and 2003 according to the European Social Observatory (2003)

185

Politics Beyond the State.book Page 186 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

capacity to negotiate agreements, and which are representative of all Member States, as far as possible – have adequate structures to ensure the effective participation in the consultation process. Representativeness, while not a problem until recently, is increasingly become one due to the need to include the social partners from the new Member States in the process. Thus, in some cases there are no organized trade unions/employers’ organizations for specific sectors in the new Member States to join the European federations who sit in the sectoral committees. Or there are different organizations competing in the same sector and not being able to decide who is to join the EU federation. The lack of proper coverage of social dialogue in Central and Eastern Europe could be undermining the legitimacy of the European federations and their mutual respect. Going back to the conditions for deliberation mentioned in the second chapter, social dialogue seems to be performing well on inclusiveness (any organization can be consulted by the Commission if it meets the three criteria above) and less well on publicity and binding results. It is even hard to find the texts of the agreements that came out of the sectoral committees (the Commission financed a project meant to compile them all). As for the results, given the non-binding nature of the great majority of documents adopted, the last criterion does not seem to be met. Looking at the standards of conduct compiled by Rosenberg, the position of employers’ organizations is clearly not in line with the “shared appreciation of the truth and right for the collective choices being made”. They might see the necessity of declarations and recommendations, but their constant opposition to binding outcomes is indicative of a willingness to “talk for the sake of talking” rather than “talk in order to achieve results”. This attitude could also be justified by the nature of social dialogue itself. It seems that deliberative procedures have better chances of success when the initial level of polarization is not too high and when the level of politicization is low (Papadopoulos, 2002). Following this argument, EU social dialogue topics might be just too controversial for deliberation. Most theorists, though, do not exclude any kind of topics from deliberation; quite the contrary, they argue that there is always a “common good”, beyond individual interests, that can be achieved. Given these problems, does deliberative democracy shed a new light on social dialogue? I think it does, in the sense that it shows that for deliberation to function properly there is a need for the outcome of the deliberations to be binding to all participants, which might explain the lack 186

Politics Beyond the State.book Page 187 Wednesday, December 5, 2007 8:41 AM

Deliberation Beyond the State

of substantive qualitative results of the EU social dialogue processes. Deliberative theories also show the importance of the framework in which deliberation occurs: representativeness of the participants, constraints over the agenda, etc. The Commission seems to have done a good job in securing the conditions for the social partners to engage in fruitful deliberations. If they have not done so sufficiently it is because of their attitude towards the process. Nevertheless, even if the right attitude were there, the social dialogue process would not be devoid of questioning and legitimacy problems. The European Parliament is completely left out of this process that may ultimately culminate in an EU directive. How legitimate and democratic is it, then, to have EU law done by a handful of trade unions’ and employers’ organizations, leaving out the Parliament, an elected body, and not allowing the Council to make any changes to the text? Defenders of the social dialogue mechanisms argue that the social partners enjoy the legitimacy of their national members combined, which would compensate for the European Parliament being left out of the Art. 139 procedure, but the debate is still open, especially given the representativity issues in the new member states. Despite the outcome of this debate, it is important to mention, in this case, the fact that deliberative democracy cannot be, by itself, the answer to solving the legitimacy problems of the EU. Even if the social dialogue would be a perfect example of deliberative democracy at work, it would need to fit within a bigger framework of “legitimate” inter-institutional relations in order to be unchallenged.

Environmental policy-making – the EU Eco-labeling scheme The EU Eco-Label is based on Regulation 880/92 (revised in 2000, Regulation 1980/2000) and is part of the broader strategy aimed at promoting sustainable production and consumption. It is a market-based instrument, meant to stimulate both the supply and the demand of greener products. In practice, the process and the division of labor between the EU and national entities can be summarized as follows: the European Union Eco-labeling Board (EUEB) develops ecological criteria for product groups in close collaboration with the Commission. The EUEB is composed of the Competent Bodies (national authorities entitled to award the EU Ecolabel to the products that meet the criteria) and the Consultation Forum (representatives of consumer NGOs, environmental NGOs, trade unions, industry, SMEs and commerce). The criteria developed by the EUEB are submitted to the Regulatory Committee (made up of governmental experts

187

Politics Beyond the State.book Page 188 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

of the Member States) and then endorsed by the Commission. The award of the Eco-Label is made by the Competent Body in each Member State. Public authorities are present, within this framework, only in the Competent Bodies. Nevertheless, the Competent Bodies carry out the most important activities in the Eco-label scheme: leading the ad hoc working group created to evaluate the introduction of new product groups in the Eco-Label scheme, leading the preparatory work to develop or revise criteria, receiving applications and awarding the EU eco-label. In addition, national experts sit on the Regulatory Committee, but their position as representatives of Member States or independent experts is hard to assess. The Commission has the task to adopt criteria (after they have been endorsed by the Regulatory Committee) and to ensure the transparency of the whole process by inviting international observers together with the EUEB. It can also select specific groups of products for the scheme (the EUEB enjoys the same right) and drafts the mandates for developing or revising criteria. From a purely administrative perspective, the Commission finances the lead Competent Body in charge of the preparatory work or the preparation of criteria. The stakeholders have several roles: on the one hand they are members of the Consultative Forum which, together with the Competent Bodies, make up the EUEB. On the other hand, they can be part of the ad hoc working groups formed either for preparatory work or to devise/revise criteria. In addition, they are consulted at national level by the Competent Bodies. The EU Eco-label is, for the moment, the only EU product policy in place and, potentially, a perfectly suited tool for innovation. Nevertheless, there are serious problems with the scheme, arising especially from the degree of stakeholder involvement. On the one hand, as a voluntary scheme, the EU Eco-label never had strong support from economic operators (producers, distributors, services and importers) (Eiderstrom, 1998), fact confirmed by the figures. Criteria have been drafted for some 23 product groups (out of which 5 are under revision) and a further 5 product groups are under development8; 224 licenses for the use of the logo have been awarded throughout the EU so far9. In addition, only a few of the product groups account for most of the products labeled. In 2000, when only 15 product groups were developed, four of them (paints and varnishes, textile products, footwear, tissue paper) accounted for 85% of all products labeled

8. 9.

http://europa.eu.int/comm/environment/ecolabel/product/index_en.htm http://europa.eu.int/comm/environment/ecolabel/whats_eco/greenstore_en.htm

188

Politics Beyond the State.book Page 189 Wednesday, December 5, 2007 8:41 AM

Deliberation Beyond the State

(Taylor Nelson, 2001). Also, at the same time, 75% of the products labeled were concentrated in five countries10. On the other hand, the multiple opportunities for stakeholder involvement in the scheme do not mean that environmental and consumer protection interests are those which prevail. Environmental NGOs have pointed out to the heavy industry lobby to water down criteria11 and to the fact that several industrial sectors systematically boycott the scheme (EEB, 2004). These complaints clearly point to a deliberative problem in the functioning of the scheme. Thus, despite the venues for environmental and consumer NGO involvement in the Eco-labeling scheme, their participation is not on equal footing with industry. On the one hand, their limited resources do not allow for their presence in both the Consultative Forum and the different ad-hoc working groups. Even if they are there, they do not have the resources to commission technical reports or studies on the specific issues under discussion. Not to mention that they are outnumbered by the representatives of industry in every meeting – there is only one representative of environmental NGOs, the EEB, and there should be one representative of consumer NGOs (it used to be BEUC, there is none for the moment). The equality of participation, as one of the standards needed to ensure the deliberative character of discussions, is clearly not met. Nor is the willingness of participants to go beyond their own interest and try to reach a common interest. The author’s own witnessing of a EUEB meeting and the minutes of the meetings prove the rigid positions of both industry and NGO representatives, the first sticking to a clear position, despite the evolution of the discussions, the latter sometimes demanding results that cannot be achieved by industry. The consequence of these non-deliberative discussions is the lack of market support for the scheme, mentioned above. Would the preparatory work for the criteria be more deliberative, then the support of the participants for the outcome would be bigger. Industry would then try to get as many eco-labeled products as possible, while NGOs would promote the label, being convinced of the value of the labeled products. Deliberation seems to be, in this case, the key to market response and not necessarily to increased legitimacy. In fact, being a voluntary instrument, the EU Eco-labeling scheme has been exempted so far of criticism directed at its legitimate/non-legitimate character. Nevertheless, EU environmental policy seems to rely more and more on such voluntary instruments, which means that we should start to question their legitimacy and validity. For instance, in 2003 the Commission released a 10. idem 11. idem

189

Politics Beyond the State.book Page 190 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Communication on Integrated Product Policy (IPP)12, which sets the basis of this new framework policy with a potentially huge environmental and economic impact. Its aim is to minimize the environmental impact of products and services by looking at all the stages of their life-cycle and taking action where it is most effective. The implementation of IPP relies exclusively on voluntary instruments such as the EU Eco-labeling scheme. If European public policy is to be implemented via voluntary tools, it is important to check on their legitimacy and efficiency. The EU Eco-label, from this perspective, is definitely not, at this moment, a reliable tool. Deliberative democracy principles might help transform it into a viable instrument, at least by ensuring equality of participation for all. It might also inspire participants to go beyond private interests in the search of the common good.

Conclusions Taking into account the results of this preliminary exercise, deliberative democracy seems to be a useful tool for looking at EU-civil society relations. It both helps explain the functioning of existing mechanisms and it offers solutions for their improvement. In the case of the social dialogue, it shows the need for the outcome of the deliberations to be binding on all participants in order for the social partners to take it seriously. It also shows the importance of the framework in which deliberation occurs and, in this case, the Commission seems to have understood its role. Nevertheless, deliberative democracy cannot, by itself, explain the functioning of social dialogue. It cannot account for the bargaining dimension for instance, nor can it provide an inspiration for improvement that would solve all the legitimacy problems associated with social dialogue. Thus, as shown above, the fact that the European Parliament is completely left out of a process that can culminate in an EU directive cannot be compensated by the mere participation of the social partners, no matter how deliberative their meetings are. Also, the question of representativity of the trade unions and employers’ federations after enlargement needs to be solved via other means. In the case of the EU Eco-labeling scheme and of EU voluntary-instrumentbased environmental policy in general, deliberative democracy seems to 12. Communication from the Commission to the Council and the European Parliament. Integrated Product Policy. Building on Environmental Life-Cycle Thinking. COM (2003) 302 final.

190

Politics Beyond the State.book Page 191 Wednesday, December 5, 2007 8:41 AM

Deliberation Beyond the State

help more in providing solutions than in explaining current patterns. If at least some of its principles were applied, it could lead both to an increased market support for voluntary instruments and to their enhanced legitimacy. Given that EU environmental policy relies more and more on soft law, it is essential for the implementation instruments to be effective and legitimate. The EU Eco-labeling scheme does not meet these conditions for the moment. To sum up, deliberative democracy is certainly a valuable tool for looking at EU-civil society relations even if, by itself, it cannot provide all the answers. Further research is needed in order to investigate the applicability of deliberative principles to policy areas other than the social and the environmental. My intuitive feeling is that the same conclusions would be reached, at least in general terms. Also, for an in-depth assessment of the deliberative character of specific consultation mechanisms, an analysis of the minutes of the discussions and interviews with the participants would be needed.

191

Politics Beyond the State.book Page 192 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State.book Page 193 Wednesday, December 5, 2007 8:41 AM

Delegation Beyond the State? Isabelle Bedoyan

The EU as a delegation exercise The European Union can be considered as a complex network of delegation systems and as such has given rise to extensive scientific literature. By the term “delegation” we solely refer to the act of delegation whereby a principal selects an agent and entrust him with a task to be performed on his behalf. The primary act of delegation is from sovereign Member States to the EU and its institutions. A secondary type of delegation occurs within the EU between the Treaty institutions: the Council, the European Parliament and the Commission, which occurs while carefully respecting the power balance between the different institutions. Finally, during the executive phase of EU policy making, new delegable tasks are introduced: delegation to the Commission and committees, European Agencies and various public-private cooperation schemes. This category of delegation, where public institutions represent the principals and private actors act as their agents, will be the focus of this chapter. The terminology of principals and agents is derived from the Principal Agent theory: originally developed within the literature of new institutional economics, which focused on transaction costs. Delegation theory developed out of interest in the relationship between shareholders and managers in businesses (Egan, 1998: 487). In the field of political science, the Principal Agent theory has been extensively applied to delegation mechanisms in the context of the US where it was used to study the autonomy of the US bureaucracy vis-à-vis the US Congress and the president (Gilardi, 2001: 5). It was only in the nineties that the approach was transposed to the EU, where it has been used to analyze EU comitology. Based on the American Principal Agent literature, many different reasons for delegation, or allocation of discretion, can be found. Pollack (2002) summarized the finding of these studies as three fundamental facets of the political environment.

193

Politics Beyond the State.book Page 194 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

The first, and most widespread, feature is the need for information and policy relevant expertise, which results from a state of imperfect information1. For example, information and expertise were key reasons behind delegation between the US Congress and its regulatory agencies. If information is indeed a major incentive for delegation, it is to be expected that future political issues, in need of expertise and information, will generate an increase in delegation. Related to the need for expertise and information is the lack of capacity to perform a certain task. It is not primarily the expertise that is sought, but rather a way of fulfilling different tasks under the constraints of limited resources e.g. time (Strøm, 2003: 57). The second point raised by Pollack (2002) is the need for credible commitment and the demand for an independent, credible regulator. At the heart of this argument lies the presumption that delegation to independent agents provides credible commitment when politicians themselves are unable to commit due to electoral or political pressure. An independent agent, free of political pressure, would be able to commit to optimal solutions even though these might be unpopular. Such a mechanism may be observed with the setup of central banks who avoid ad hoc changes of interest rates to boost the economy, increase economic stability, and thus increase credibility of government policy vis-à-vis market players. In addition to these motives, delegation can be a means to ‘shift the blame’. This blame shifting can be used by the principal to hide from unpopular decisions or to ‘dump’ politically risky issues which result in little or no political gain (Epstein and O’Halloran, 1999: 22-23). Besides the motives enumerated above, delegation research points to causes which fall outside the scope of a rational-choice-based Principal Agent theory. Based on the study of independent national banks, McNamara (2002) highlights the importance of socially constructed opinions rather than rational efficiency criteria as a decisive element for delegation to this type of bank. Her conclusions point to the need for additional theoretical explanations to complement the Principal Agent theory. The failure of the Principal Agent theory to explain differences in type and timing of delegation, further supports the need for additional theoretical explanations. To complement the theory, other insights can be used: ‘historical’ and ‘sociological’ institutionalism can help explain the spread of some forms of delegation by identifying non-functional motivations. Sociological institutionalism uses, amongst others, normative 1.

See Majone (2001) for further details.

194

Politics Beyond the State.book Page 195 Wednesday, December 5, 2007 8:41 AM

Delegation Beyond the State?

pressure to explain some acts of delegation and points to the influence of the environment in which delegation takes place (Thatcher and Sweet Stone, 2002: 9-13). The spread of certain types of delegation can also be explained by policy learning and institutional isomorphism. In these cases, existing delegation systems are borrowed from successful experiences at national or international level. Private interest groups actively involved in delegated tasks form communities of professionals (e.g. banking, competition policy), which can support the spread of this type of governance. Isomorphism can also be induced: in the case of European integration, for example, non-majoritarian institutions were generated when the adoption of independent central banks became a basic condition of admission to the European Monetary Union (Thatcher and Sweet Stone, 2002: 12-13). It is clear that the American norm cannot simply be transposed to European cases, however, the basic aspects that have been pinpointed by Pollack and others can be used to shed light on delegation mechanisms in the context of the EU: helping to analyze the motives as well as the risks of the current evolution in delegation.

Problematic features of delegation Delegation towards private bodies, which have varying levels of independence, raises particular problems of control. Unlike public administrations, these private entities cannot be controlled by traditional methods as they fall outside the traditional democratic chain of accountability. The act of delegation to an agent entails the risk of agency shirking or loss, and agency slippage. Agency loss is defined as ‘any form of non-compliance by the agent and results from a conflict of goals’ (McCubbins and Page, 1987: 410). Agency slippage occurs when ‘the structure of delegation itself provides perverse incentives for the agent to behave in ways inimical to the preferences of the principals’ (Pollack, 1997: 108). Agency loss is closely linked to two major characteristics of the relationship between principals and their agents. The first element is the possible discrepancy between the principal’s preferences and the preferences of its agent. This can be shown as a difference of opinion about desired outcome, objectives to be reached, or concerning mode of operation and division of labor (Lupia, 2003: 37). Discrepancies on these 195

Politics Beyond the State.book Page 196 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

topics have a determining influence on the manner and the outcome of delegation. The issue of preferences is closely linked to a second element determining agency loss: lack of knowledge and information. The principal as well as the agent can experience an information deficit, each having a specific impact on the delegation process. In the principal’s position, knowledge concerning the preferences, capabilities and actions of its agent is of the utmost importance. For the agent, knowledge of the task at hand is critical. A principal’s information deficit can cause two serious problems: adverse selection and moral hazard. Adverse selection occurs when a principal is not able to select the right agent due to a lack of knowledge about the agent’s preferences or skills. Moral hazard occurs after the selection of the agent, when the principal is unable to ensure the agent’s honesty and diligence; often linked to a lack of control over the agent’s actions. Any concealment of actions from the principal by the agent ultimately results in the occurrence of hidden actions (Müller et al., 2003: 22-26). Different mechanisms can be used to control the different forms of agency loss. The success of delegation is, therefore, heavily linked to the control instruments that are at the principal’s disposal. These control instruments can intervene either in the run up to the delegation (ex ante) or after delegation has been initiated (ex post). The Principal Agent theory proposes several control mechanisms. The first is oversight. McCubbins and Schwartz (1984) identified two forms of oversight: ‘police patrol’ and ‘fire alarm’. The first is a permanent control and entails serious costs. The ‘fire alarm’ control is an ex post control mechanism, which has the advantage of involving third parties as control agents. Third parties offer extra information and reduce the control cost by triggering political actions on problematic issues. Besides these broad categories of control types more specific control instruments can be found. Administrative control, for example, directly affects the process by laying down the procedures to be followed. This allows the principal to define the ‘rules of the game’ and even the balance between the different players, ultimately influencing the ‘fire alarm’ potential. The important essence of this instrument as a key ex ante control strategy is the contract, between principal and agent: a good contract can sharply define the delegation procedure and objectives, impose ex post controls and can even establish a shared interest between the principal and the agent (grant the agent a fixed percentage of the outcome or regulatory relief).

196

Politics Beyond the State.book Page 197 Wednesday, December 5, 2007 8:41 AM

Delegation Beyond the State?

The strategic appointment of top officials within the agent’s structure is yet another way to influence the agent’s function. Along with the strategic appointment of top officials, the principal might try to control the agent’s organizational structure to secure its loyalty. This control strategy is based on the importance of the selection and the screening of potential agents in order to select those who are the most skilled and loyal. The budget can be another way to control the agent. By controlling the budget the principals can restrict or broaden the agent’s activities and eventually sanction or reward particular behavior. However, this form of control is less efficient when the agents can rely on alternative sources of income. Institutional checks can be introduced by using multiple agents and dividing a particular competence between them, hereby creating additional checks and balances. As a last resort, the principal can depart from the governance approach and choose to use its legislative prerogative. By doing this he can change the nature of the agent or simply overrule its actions. If this option is credible enough, the mere existence of such a threat could serve as an incentive for the agent to comply with the principal’s wishes. Detailed legislation can also serve as an ex ante control over the agent: limiting its discretionary powers. More detailed legislation can also mean more detailed description of the status of the agent. (Gilardi, 2001: 13-14). The type and extent of control that is at the disposal of the principal is partly determined by the objective of the delegation process. When credible commitment is the objective, the possibilities for ex post control are fairly restricted: extended control over the agent would undermine his ability to achieve credible commitment. On the other hand, ex post control is far more relevant when expertise is the main objective: extensive control would not affect the information or expertise that is gained by delegation (Thatcher and Stone Sweet, 2002: 15). Nevertheless, it is important to note that if efficiency and reduction of the government workload is pursued as well, the control cost must be entered into the equation. If the control activities take up too many resources, the point of the entire exercise could be questioned. Finally, the decision to delegate and its outcome is influenced by the balance between three elements: the status quo (SQ) which is the situation before any delegation has been done, the preferred position or outcome of the principal (P) and finally the agent’s preferred position (A). If the gap between the status quo and the principal’s preferred outcome is wide, the

197

Politics Beyond the State.book Page 198 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

principal will be more inclined to compromise with the agent. Indeed, a large gap between the status quo and the principal generates an important loss for the principal, if no delegation in undertaken. In such a situation, the potential loss (gap between P and A) is more likely to be smaller than the loss when no delegation occurs (gap between SQ and P). Therefore delegation becomes more likely and will provide a more positive outcome (Lupia, 2003: 36-40). The control problems raised by delegation discussed above are closely related to a third issue: legitimacy and accountability. If we take into consideration the fact that most delegation tasks are performed to fulfill the objectives of credible commitment, expertise and efficiency, we can conclude that the achievement of these objectives earns the agent some form of ‘output legitimacy’. However, as the agents fall outside the traditional democratic chain, legitimacy might be lost if a deficient control system leads to agency loss and the output no longer attains the objectives. Procedural legitimacy helps strengthen overall legitimacy through increased transparency, openness and accessibility of the agent’s activity compared to those of the principal’s. This procedural legitimacy can be considered a substitute to the principals’ traditional democratic accountability (Thatcher and Stone Sweet, 2002: 18-20). However the transparency, the openness and the accessibility of the agent’s activity should always be carefully monitored by the principal.

Delegation within the EU Students of the European Union gradually borrowed the Principal Agent Theory from US scholars to help describe the delegation mechanism from Member States to the EU as well as within the EU institutions. The first fullfledged attempts to adapt and implement the theory to the European arena were undertaken by Mark Pollack (1997, 2003). His study focused on the EU as a supranational agent to which member states would delegate various tasks. The European Commission (EC), the European Court of Justice (ECJ) and European Parliament (EP) were analyzed from a Principal Agent’s perspective, which described their responsibilities and the different mechanisms to monitor their actions. The theory was effective in clarifying the motives and mechanisms of delegating tasks to the Commission (see below) and the European Court of Justice. The ECJ has been delegated functions related to monitoring and enforcing compliance of EU law and filling in the details of incomplete contracts. Delegation of these functions was primarily motivated by the need for credible 198

Politics Beyond the State.book Page 199 Wednesday, December 5, 2007 8:41 AM

Delegation Beyond the State?

commitments. In accordance with this motive, the ECJ enjoys extensive statutory discretion. In addition, control mechanisms available to the member States are fairly weak and the ECJ’s prerogatives haven proven to be quite resistant. Delegation to the European Parliament, on the contrary, did not fit the rational choice arguments of the Principal Agent theory and seemed to reflect normative desirability and democratic legitimacy motives rather than cost effectiveness. Pollack’s analysis of the European Commission reflects the way he applied the Principal Agent Theory to the EU context. In his analysis, he details the different tasks delegated to the Commission, the motives behind their delegation and the control instruments available to the principal. Pollack labels the Commission’s ‘right of initiative’, which enables it to shape the European agenda, as a delegated task motivated by the need of credible commitment. The credibility of the Member States’ commitment is further strengthened by charging the Commission with the task to monitor compliance with the Treaty and by giving it the means to enforce the Treaty through infringement procedures. The selection of cases and the approach used to handle them is left to the discretion of the Commission. This situation strengthens the independence and credibility of the Commission as guardian of the Treaty, but could lead to agent shirking. Besides this, the Commission was given the task to implement EU policies through the adoption of implementing regulation, the management of EC spending programs and the direct application of EC law (e.g. in competition policy). According to Pollack delegations of these tasks occur to ensure credible commitment and efficiency Pollack further analyses the different control mechanisms developed to monitor activities of the Commission using the Principal Agent theory. First of all, control is exercised through the procedure defining the composition of the Commission. In addition, Member States’ committees perform a ‘police patrol’ oversight known as comitology. The use of comitology is in fact a form of delegation in its own right, which involves non-elected representatives from the Member States, be it scientific experts or representatives from the Member States’ administrations. The control function assigned to comitology is evident by the high importance the actors involved devote to rules and procedures, and their perception that only the most restrictive forms of committees and delegation systems would be suitable (Pollack, 2003: 128-129). The strength of comitology, as a control instrument, is highly dependent on the type of committee chosen in the legislation: advisory, management 199

Politics Beyond the State.book Page 200 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

or regulatory committees. The weakest form is the advisory committee: attributed an advisory function with which the Commission has no obligation to comply. The management committees, in contrast, have the power to oppose any of the Commission’s proposals by Qualified Majority Voting (QMV) after which they are referred to the Council. The regulatory committees are the most powerful: the Commission can only enact its measures after a positive vote (QMV required) by the committee. Another control instrument within the comitology practice is the ‘safe guard procedure’. This procedure allows the Council the possibility to examine, at the explicit request of a Member State, the decisions of the Commission related to temporary safeguard measures. Alternatively, if comitology does not secure a satisfactory implementation, the Council can choose to deal with the implementation directly (Pollack, 2003: 117-118). By doing this the Council forsakes the possible advantages of delegation and steps in to secure the desired outcome. The control function of comitology is clearly reflected in the choice of procedures made during the legislative process. Regulatory procedures are generally preferred by the Council, for maximum control. The Commission tries to avoid the most restrictive procedures, but often anticipates the reaction of its principals and is prepared to make concessions in order to reach its goals (Pollack, 2003: 137-139). The same anticipation technique can be found when implementation measures are proposed under the comitology procedure: the Commission prefers to adapt a draft proposal, in function of the Council’s, and the committee’s preferences, in order to secure a smooth adoption (Ballmann et al., 2002: 571). The European Court of Justice and the Court of First Instance provide judicial review of the decisions by the Commission or other EU institutions. These Courts act as ‘fire alarms’ when EU law has been violated. They also introduce extra participants into the control system by giving public and private actors the opportunity to signal potential contraventions. In addition to the judicial review, financial pressure can be put on the Commission through the revision of its budget. This control instrument is an option, but it is weakened by Commission’s involvement in the setup of the EU annual budget and by the impact this type of instrument has on the overall activity of the agent. Finally, the Commission is monitored by two non-treaty based institutions: the Court of Auditors controls the budget and the Ombudsman controls its administration. The Ombudsman offers private actors an extra opportunity to act as a ‘fire alarm’ (Pollack, 2003: 75-90). 200

Politics Beyond the State.book Page 201 Wednesday, December 5, 2007 8:41 AM

Delegation Beyond the State?

The scope of the delegation is affected by the objectives to be reached (expertise, efficiency, credible commitment) and by to the control mechanisms available to the Principal. Hence, the Commission’s important role as guardian of the Treaty and its extensive discretion are a consequence of the need for credible commitment. The same objective of credible commitment can be found in the extensive agenda setting-prerogative of the Commission in the first pillar of the EU, which demonstrates the Member States’ commitment to a strong supranational policy. Finally, the implementation and regulatory power delegated to the Commission presents features which point both to credible commitment and to the need for efficiency. The areas delegated by the treaties are principally areas in need of credible commitment because of the high cost and diffused benefits they bring about or because they combine regulatory and monitoring features. Delegation mechanism in the secondary legislation, on the contrary, incorporates the need for efficient and rapid implementation, which is a serious argument when one considers the complexity of the traditional legislative procedures within the European Union (Pollack, 2003: 105-106).

Delegation to private actors Delegation within the EU system is not limited to delegation between public actors like the Commission or to EU agencies, which remain under tight scrutiny of the EU institution and Member States. Within the EU, public-private delegation is also used. The European New Approach Directives, for example, is based on the delegation of technical tasks from the European Commission to private European Standardization Bodies (CEN, CENELEC and ETSI)2. The New Approach system of delegation was developed in 1985 to fulfill a double purpose: the completion of the Single Market and the guarantee of a high level of protection for public interest objectives: security, public health and environmental soundness of products. The New Approach was launched after the failure of an ambitious program of regulatory harmonization which was in place since 1958. This policy 2.

CEN: Comité Européen de Normalisation, CENELEC: Comité Européen de Normalisation Electrotechnique, ETSI: European Telecommunications Standards Institute.

201

Politics Beyond the State.book Page 202 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

covered almost three decades, but only managed to tackle a few trade barriers, as it had to cope with both regulatory complexity and conflicting political preferences of the Member States. The EC Treaty itself has several provisions concerning the free movement of goods: custom duties, common external tariff, quantitative restrictions, restrictions on imports and quotas. However, the Treaty also provide an important derogation from these provisions through article 36: permitting restriction based on the protection of public morality, policy or security, the health of human, animal and plant, the protection of national treasures, and the protection of industrial or commercial property. These restrictions are permitted, provided that these restrictions are not arbitrary or discriminating. This derogation was balanced with article 100 introducing the harmonization of laws and regulations. Art. 100 provides the possibility to replace contested national rules by European regulations and thus prevent the undermining of the common market through the use of the derogations in art. 36. Still, the art. 100 itself was held back because of implementation problems, the use of unanimity rule, and because Members States reluctance to harmonization (Egan, 2001: 61-82). Harmonization efforts were made, but in the absence of a legally binding standstill agreement, the Commission had difficulties to keep up with continuous national regulation efforts. The economic crisis of the 1970’s slowed the harmonization process further down. The creation of harmonization directive was burdened by technical complexity: technical experts of the different Member States had to be consulted, preliminary studies were undertaken by working committees within the Commission and the issues at hand often required the coordination of several national departments. As a result the negotiation of draft directives took four years on average before reaching the Council. Moreover, national governments did not consider harmonization a priority issue, so little political impetus was provided. The harmonization had to keep pace with a rapidly evolving technology, which could make regulation obsolete even before its adoption. Finally, the harmonization was not flexible enough to provide for the changing economic production (e.g. the shift to services and the introduction of greater product diversity at lower cost). Increasingly harmonization was perceived as a barrier for technological innovations and a burden for producers dealing with various products and shorter product cycles in a competitive environment (Egan, 2001: 61-82). In the absence of legislative harmonization ECJ rulings filled in the blanks. The Cassis de Dijon case validated the principal of mutual recognition, which meant that harmonization was only necessary in cases where national regulations could not be considered as equivalents. The Cassis case also added the 202

Politics Beyond the State.book Page 203 Wednesday, December 5, 2007 8:41 AM

Delegation Beyond the State?

criteria of causality, proportionality and substitution to evaluate possible exceptions on art. 30. The negative integration the ECJ carried through by invalidating discriminatory national regulations paved the way for subsequent positive integration at the EU level through harmonization. (Egan, 2001: 108). In 1985 the market integration program was given an extra impulse by the publication of a White Paper: Completing the Internal Market. The White paper was build on the case law developed in the preceding years (e.g. mutual recognition principle) and on an assessment of the remaining trade barriers, and was focused on the removal of technical barriers to trade. To deal with the continuous national regulatory efforts and the growing number of standards a Mutual Information Directive was introduced in 1983. This obliged Member States and national standardization bodies to inform the Commission of their activities and allowed the Commission to intervene at an early stage and anticipate trade barriers. To deal with existing technical barriers other techniques were put forward based on the Low Voltage Directive, which used a method of reference to standards provided by standardization bodies in its directive (Egan, 2001: 109-121). This simplified form of technical harmonization provided a significant relief of the regulatory task of the EU and would form the basis of the New Approach to Technical Harmonization launched in 1985. This New Approach to Technical Harmonization combines traditional legislation with voluntarily applied standards. The legislative part of the scheme consists of a Directive, which benefits from the democratic checks and balances provided for in EU lawmaking. The harmonization in the Directive is limited to essential requirements concerning safety, health, consumer protection and environmental protection. The complex and labor-intensive translation of these essential requirements into harmonized European standards is delegated to the three European standardization bodies: CEN, CENELEC and ETSI. A mandate from the European Commission to the appropriate bodies forms the bridge between traditional legislation and the delegated standardization. The mandate is created by the Commission, based on the requirements of the Directive, and is submitted to Committee 98/34 on Standards and Technical Regulations for its opinion. The use of a mandate procedure in combination with comitology, reflects the double delegation exercise present in the New Approach Directives. On the one hand, it depicts the Commission as the principal delegating to private actors, using the mandate as a tool. On the other hand, it represents the Commission as an agent of the Member States controlled through comitology.

203

Politics Beyond the State.book Page 204 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

The standards produced by the CEN, CENELEC and ETSI are published in the Official Journal; as a consequence compliance with the standards provide a presumption of conformity to the requirements of the Directive. From this moment on producers can proof their compliance with the essential requirement through the use of the corresponding standard. We must add that the standards are voluntary instruments which act as a proxy to demonstrate compliance with the directive. Producers are not obliged to use them and can refer directly to the essential requirements in the Directive. Nevertheless, the presumption of conformity attributed to producers using the European standards seems to provide the standards with a de facto monopoly. Products not adhering to the recognized standard are often rejected by distributors because alternative methods of proof are considered too burdensome (Hunter and Molyneux, 2000). Moreover, the presumption of conformity reverses the burden of proof between the producers and Member States: if a Member States denies market access to a product which complies with the standard, it is up to the Member State to prove that the product does not comply with the essential requirements of the Directive. The type of tasks delegated through the New Approach Directives and the complex process of market integration in which these Directives were developed, all point to issues of technical expertise, efficiency and capacity as being the main motives for delegation. The technical harmonization required by the EU market integration could not be achieved through traditional legislative procedures The use of essential requirements in legislation combined with standardization has facilitated a smooth adaptation to rapidly evolving techniques. Delegation to private standardization bodies, in which industrial actors and producers play a significant part, has reduced the risk of contestation of the implemented policy while providing the required technical expertise. Under the New Approach scheme, credible commitment does not seem to be the main motive, delegation rather stems from the need of expertise, efficiency, and capacity concerns. As the fulfillment of expertise, efficiency and capacity goals requires less autonomy from the principal, tight control might be expected. However, despite this, European standardization bodies enjoy an exceptionally independent status and control over their activities is limited. In the New Approach scheme, most of control is performed ex post through the use of a ‘safeguard clause’. This clause is a ‘fire alarm’ type of control mechanism, which can be invoked by Member States or by the European Commission when a standard fails to meet the essential requirements 204

Politics Beyond the State.book Page 205 Wednesday, December 5, 2007 8:41 AM

Delegation Beyond the State?

setout in the Directive. The use of this clause can prevent or revoke publication of the standard in the Official Journal thereby denying presumption of conformity. In practice the safeguard procedure is started by the introduction of formal objection against a particular standard. This formal objection is than transferred to the Committee on Technical Standards and Regulation (Committee 98/34) for its opinion. The Committee 98/34 was created by the ‘Directive 98/34/EEC3 laying down a procedure for the provision of information in the field of technical standards and regulations’. It is composed of Member State’s representatives and is chaired by a Commission representative. Directive 98/34/EEC as well as its committee provides a general framework to deal with technical standards and regulation within a European integrated market, but does not deal with specific topics. Therefore, New Approach Directives developed for specific topics (e.g. Packaging) have to possibility to create their own committee or group of experts. Because of their expertise; most formal objection will be dealt with first by the Directive’s own committee before the opinion of the Committee 98/34 is sought. The DG of the Commission responsible will then write a proposal based on the assessment of the directive’s own committee, the opinions of other Commission services, and the information obtained from Member States, European Standardization bodies and other stakeholders. This proposal shall be transmitted to the Committee 98/34 for an urgent opinion (CEC, 2003b). The final decision, in the safeguard procedure, is taken by the European Commission upon advice of the Committee on Technical Standards and Regulation (Committee 98/34) and the directive’s own committee. The use of committees, composed of Member States’ representatives, as instruments to both defend national positions and control the Commission’s actions points to the Member States as being the primary principals of the entire delegation exercise. However, the Member States’ position as principal is weakened by the advisory character of the comitology committee. Apart from this ‘safe guard clause’, no systematic, ‘police patrolling’ is performed by the Commission on the standards. This situation generates an information deficit concerning the agents’ action leaving the door open for agency loss under the form of moral hazard. Even though the Commission can participate as an observer to the European standardization bodies it does not do so on a regular base. The Commission has repeatedly insisted on active participation by public interest groups, which has improved over the years, but is still deemed 3.

Directive 98/34/EEC is a codification of Directive 83/189/EEC, which already provided for the creation of the Committee.

205

Politics Beyond the State.book Page 206 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

insufficient. This insistence in interest group participation might reflect a tactic to introduce extra ‘fire alarm’ control opportunities: where public interest groups address potential contraventions of essential requirements. McCubbins and Schwartz (1984) described similar tactics concerning Congressional control over US agencies: informal networks of public interest groups acted as ‘fire alarms’, reporting possible infractions of legislative goals to their representatives in Congress. In addition, consultants are hired by the European standardization bodies to ensure that the actors involved in the actual drafting of the standard understand and respect the essential requirements. Overall control over the standardization process can be tightened by well defined mandates and procedural clauses in the guidelines for cooperation between the EU and the European standardization bodies. These mandates often leave room for manoeuvre while the cooperation guidelines do not provide strict procedural codes. Finally, budgetary sanctions can be imposed when contracts between the Commission and the European standardization bodies are not respected (Egan, 2001: 124). However, the standardization bodies are financially independent because of the income generated by membership fees and the sales of European standards. Despite this bleak analysis, New Approach standardization has proved very successful when applied to safety issues: more than 20 directives and 2165 standards have been published (CEC, 2004b: 7).Yet, when the same method is applied to environmental matters, control issues seem to be highlighted. The Packaging and Packaging Waste Directive (94/62/EC) illustrates this concern and points out the difficulties involved.

The Packaging and Packaging Waste Directive This directive is the first New Approach directive explicitly developed to achieve environmental goals. While the actual directive was developed in 1994, and the first mandates to CEN were delivered in 1996, it took until February 2005 for all the mandates to be turned into publishable harmonized standards. Standards can take several years to develop, however extra difficulties seemed to burden the procedure in this particular case. The packaging directive was developed to reduce the quantity of packaging, increase reuse and recycling, and reduce the amount of harmful substances used in packaging. For these reasons a mandate was given to CEN (mandate 200 Rev. 3). The mandate asked for five standards

206

Politics Beyond the State.book Page 207 Wednesday, December 5, 2007 8:41 AM

Delegation Beyond the State?

to be prepared. The first standard concerned packaging requirements specific to manufacturing and composition, and dealt with waste prevention. The second standard had to cover the reuse of packaging. The following three standards respectively dealt with packaging and material recycling, energy recovery, and organic recovery. The mandate was accepted by CEN, who developed six standards based on this mandate, despite the fact that only five standards were directly mandated by the Commission. In addition to the explicitly mandated standards, CEN delivered an “umbrella” standard (EN 13427:2000 Packaging-Requirements for the use of European Standards in the field of packaging) which was designed to serve as a guide to the use of the other standards. In 2000, these standards were adopted by the CEN members and submitted to the European Commission for publication of its references in the Official Journal. Such a publication would grant a “presumption of conformity” to the products applying the standards. At that time, the so-called “safeguard clause“ was put into action. On the basis of article 9(4) of the Packaging directive ( 94/62/EC) Belgium and Denmark filed a formal objection with the Standing Committee on Standards and Technical Regulations (98/34 Committee). The objections contained general remarks which applied to all standards, and more specific remarks related to technical specifications in the individual standards. For example, Belgium raised objections against the formulation and status of the non-mandated umbrella standard, which could affect the essential requirements in the mandated standards by directing their uses. Other objections regarding the lack of participation of environmental and consumer NGOs in standardization activities were also voiced. Participation by public interest groups was especially important, since the need for it was explicitly stated in the mandate. Concerns were raised as to the efficiency of ISO 9000 and EN ISO 14000 series management control systems to guarantee essential requirements and to realize a harmonized internal market. Finally, the lack of technical specifications was addressed, as a requirement appeared to be literally copied from the mandate without further elaboration. In addition, new terms and definitions were used in the standards instead of specific legal terminology.4 These official objections were matched by the criticisms of 4.

Ministerie,van Sociale Zaken, Volksgezondheid en Leefmilieu, Clausule van formele tegenkanting ingediend door België bij het Permanente Comité ingesteld overeenkomstig artikel 5 van Richtlijn 98/34/EG betreffende de CEN-normen voor de invulling van de essentiële eisen van de Verpakkingsrichtlijn 94/62/EG in het kader van Mandaat 200 Rev.3 overeenkomstig artikel 9, § 4 van deze Richtlijn, Brussels.

207

Politics Beyond the State.book Page 208 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

public interest groups defending consumer and environmental issues. The European Association for the Co-ordination of Consumer Representation in Standardization (ANEC) highlighted the lack of consideration for nonindustrial stakeholders in CEN’s Packaging Committee and the predominance of industry on this committee. ANEC also condemned the lack of substantive and verifiable requirements (number of trips for reusable packaging), and the ignorance of the mandates provisions of the CEN standards. In their critique of the individual standards they condemned, for instance, the predominance of marketing and presentation criteria on source reduction for packaging.(ANEC, 2000) Similar criticisms were found in the position paper of the European Environmental Bureau (EEB), which particularly stressed the fact that the management system that had been opted for in the standards would not be able to guarantee essential requirements. The EEB even felt that the publication of these standards as they were, would “encourage CEN and hence industry, to ignore and bypass environmental legislation and write out its own in the future”. According to the instructions of the safeguard clause, the objections of the Member States were considered in the Committee for the Adaptation to Scientific and Technical Progress (created by Directive 94/62/EC on Packaging and Packaging Waste), better known as the Article 21 Committee. Compliance with the essential requirements was discussed during their meetings, but diverging opinions remained. Austria, Belgium and Denmark were clearly against the publication, while France and the UK defended the standardization effort. After consulting the Article 21 Committee, the Commission turned to the Standing Committee on Standards and Technical Regulations (98/34 Committee) with a draft Decision, but no consensus could be reached. After a vote, the Commission deemed it had sufficient support to proceed. It consulted the 98/34 committee again on the 27th of June before publishing its decision. In its decision of 28 June 2001, the Commission published the references of standard EN 13432 concerning recovery by composting and bio-degradation, and EN 13428 concerning prevention by source reduction, although the latter was published noting that it did not fully cover the essential requirements (EEB, 2000).5 The remaining standards were not published, and consequently did not receive the “presumption of conformity” but still kept the status of CEN standards. In

5.

Commission Decision 2001/524/EC of 28 June 2001 relating to the publication of references for standards on packaging and packaging waste – OJ L 190 of 2001-07-12

208

Politics Beyond the State.book Page 209 Wednesday, December 5, 2007 8:41 AM

Delegation Beyond the State?

this way they can be used as an instrument by Member States on a voluntary basis, this was the case in The United Kingdom and France. In coordination with the different committees and the Member States, a new mandate was developed to revise the remaining standards and to incorporate the so-called “umbrella” standards proposed by CEN. Drafts of the revised mandate were distributed among the Member States for comments. In November 2001, the final version of the revised, or second, mandate M317 EN was finalized and reached the relevant CEN technical Committee by March 2002. Although the Commission opted for this second standard, it must be noted that the Commission recognized the need for a fundamental review of the New Approach elements of the Packaging Directive in order to achieve a complete solution (CEC, 2001a). Moreover, during the preceding committee meeting some Member States expressed their skepticism towards a revised mandate and asked for a broad review of the New Approach elements of the directive, while others believed the revised mandate could provide a swift solution. Concerns were also expressed regarding the effective separation of political issues and technical issues to ensure that only technical issues were addressed by CEN (CEC, 2001b). By December 2004, CEN entered a request for publication of the revised standards. No official objection was raised against the revised standards, although an informal objection was raised by Austria and discussed during the Article 21 Committee meeting on February 2, 2005. During this meeting, CEN had the opportunity to present the standards and to answer delegates’ questions. Some Member States were concerned about the capacity of the standards to establish clear boundaries between acceptable and non acceptable packaging. The Commission’s opinion was asked concerning the minimum criteria for rotation of reusable packaging and the presence of hazardous substances. In its response, the chair qualified the management approach as the “best feasible” and explained that it was difficult for the Commission to request fixed values. A new attempt to refine the standards was not expected to make any substantial difference (CEC, 2005a). Finally, in the absence of any formal objections, the references of the revised standards on reuse (EN 13429: 2004), material recycling (EN 13430:2004), energy recovery (EN 13431:2004) and a new version of the partially accepted standard on prevention by source reduction (EN 13428: 2004) were published on 19 February 2005. This publication was completed by the inclusion of the umbrella standard in the second mandate (CEC, 2005b).

209

Politics Beyond the State.book Page 210 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

When we look at the opinions of the public interest groups such as ANEC and ECOS on the matter, we still see a strong rejection of the revised standards. In their joint position paper, they considered that the standards did not satisfy the essential requirements, or the provisions of the second mandate M317, that most of the changes are purely editorial, and that only a few substantial changes were made (ANEC and ECOS, 2005). Regarding the individual standards, the public interest groups criticized the use of management systems instead of clear–cut specifications, the supremacy of marketing criteria above packaging reduction, the inclusion of hybrid systems in the reuse standard and the minimum caloric value that was prescribed. The example of the Packaging and Packaging Waste directive, demonstrates quite clearly the difficulties of using New Approach legislation for environmental purposes. The whole process took more than a decade while multiple objections and a variety of issues were raised by the Member States, public interest groups and even the Commission. In this case, a revision of the New Approach elements of the directive was proposed by the different actors, but time restrictions influenced the choice of a second mandate. Not only did the delays encountered by the procedure ran counter the objective of efficient policy making, it also restricted the option that could be taken by the principal: revision of the directive itself or second rejection of the standards would have caused further delays. With regard to the standardization process itself, concerns were expressed regarding the division of political and technical decisions. When using New Approach legislation, utmost attention must be devoted to the definition of the essential requirements to prevent the migration of political decision to standardization bodies and industry. Clearly defined essential requirements and mandates, as the equivalent of a well tied contract, is the key to a successful delegation. It does not only function as general control instrument, but should also have prevented the shifting of political decision to private agents. The advocacy of the public interest and the control of the public authorities on the matter must be evaluated in the context of “real life”. As we have seen in the packaging case, lack of meaningful involvement has been raised, both by the Member States (cfr. formal objection of Belgium and during article 21 committee meetings) and the public interest groups (position papers ECOS and EEB). The lack of meaningful involvement jeopardizes the role of public interest groups as third party control agents and as a consequence the control potential of the principal. As to the public authorities, we can note that the “safeguard clause” was successfully

210

Politics Beyond the State.book Page 211 Wednesday, December 5, 2007 8:41 AM

Delegation Beyond the State?

applied to object to the first standards. However, the European Commission still opted for the formulation of a new mandate instead of a thorough revision of the directive. Furthermore, the revised standards were accepted even though important doubts still arose as to the fulfillment of the essential requirement. This turn of events points out that even though the principal is aware of the actions of its agents he is not always able to rectify the situation: time pressure, restricted capacity and information deficit can seriously hinder such an effort. Finally, the role of the European Commission seems to be more important than that of the Member States. The Commission has to acknowledge the formal objections and consult the relevant committees cited above, but the output of these committees is non-binding (CEC, 2001c). As a consequence, the final decision is reserved for the European Commission. Regarding the “safeguard procedure”, we must note that its outcome is influenced by the characteristics of the different committees. The Article 21 Committee of the Packaging directive, for example, is composed of different types of delegates: members of the Permanent Representations to the EU and members of national agencies, or national administrations. As a consequence, the knowledge of the issues at hand can vary substantially and delegates can be restricted by the mandates they receive. Even though perfect information is not needed to make delegation work, a significant level of expertise and a follow up of the matter by all members involved in the committee are essential to perform an efficient control. When we consider the public interest viewpoint, we can conclude that control by public authorities is rather limited. This is the case within the European standardization bodies, as we have stated before, but also within EU institutions in the case of the safeguard procedure. In this procedure, the final decision is in the hands of the European Commission, as the opinions of the committee (composed of national delegations) are nonbinding. Moreover, contextual factors can work in favor of the agent, as expost control systems can cause serious delays. Despite the efforts to include public interest groups in the process, their position does not seem strong enough, at the moment, to guarantee the systematic advancement of public interests. They often lack the means to defend these interests sufficiently and as it was the case in the packaging case, feel their views are not adequately taken into consideration within the standardization committees. Finally, the case study described above underlines the importance of a consensus on preferences between principal and agent. If such a consensus

211

Politics Beyond the State.book Page 212 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

is present, control mechanisms and information deficits become less relevant: the agent’s actions are more likely to be in line with their shared objectives. But, if such a consensus cannot be found control costs will rise and the delegation exercise itself can become compromised.

Conclusions The review above presents us with two different types of delegation – from Member States to Commission and from the Commission to private Standardization bodies – as well as two different applications of the Principal Agent theory. In both examples the theory helped describe the motives for delegation, types of tasks delegated and the particular discretion the agents were attributed. Delegation to the Commission is the more comprehensive, as the Commission has been delegated agenda setting, control, executive and implementation tasks. As an agent, the Commission is expected to provide efficient policies, as well as, credible commitment. In spite of the latter objective and with regard to extensive delegation, important control mechanisms have been put in place. Judicial review provides ‘fire alarm’ types of control, while the comitology system makes ‘police patrolling’ possible. These powerful control tactics complemented by financial control provide a strong toolbox for the principals to use in order to prevent agency loss. The second type of delegation steps outside the inner circle of public actors to delegate tasks to private actors. A public-private partnership is set up between the Commission and the European standardization bodies to translate essential requirements into technical provisions. The European standardization bodies are private bodies, which are situated outside the EU institutions, yet they are delegated tasks of major public interest. The outcome of their activities is incorporated in the EU legal framework through publication of standards in the Official Journal, which provides a presumption of conformity. Such standards have a de facto legislative status. In return, the European standardization bodies offer a significant level of expertise and information, and assist in the operation of an efficient system of technical harmonization, which cannot be realized through traditional legislation. Besides the primary delegation between public and private actors, the New Approach procedure incorporates the basic element of ‘delegation from Member States to the Commission’ and the comitology oversight 212

Politics Beyond the State.book Page 213 Wednesday, December 5, 2007 8:41 AM

Delegation Beyond the State?

instrument that goes with it. Hence, the New Approach is a double delegation exercise (public-public and public-private) and in this regard it provides a double control. At first sight New Approach delegation only concerns technical activities and no political decisions are made. Nonetheless, in some cases mandates and essential requirements are too vague and as a result some politically sensitive decisions are shifted to the standardization bodies. The risk of such a shift was one of the main concerns in the Packaging case we examined: the mandates remained fairly vague and it was left to the standardization bodies to define clear-cut thresholds. Vague mandates eventually resulted in equally vague standard: some specifications were poorly developed and flexible management control systems were used rather than clear thresholds. Vague mandates can be the result of deliberate blame or burden shifting tactics by the principal or it can be a reflection of the principal’s inability to achieve a political consensus. In the packaging case, the Directive itself was heavily discussed and some participants even suggested a fundamental review of the New Approach elements of the directive to achieve a satisfactory solution. However, this option was not withheld by the Committee as they choose to create a second mandate. In addition to this, control over the standardization procedure is fairly weak and mostly ex post. This was obvious in the Packaging case: the first standards that were delivered lacked technical specification, an extra nonmandated umbrella standard was added on initiative of the standardization bodies and the specific legal terminology was not respected. The ex post safeguard procedure had some effect, but the second set of standards were published despite the persistent Member States concerns. The combination of all the above factors leaves this type of delegation prone to agency loss, which may easily occur when the agent’s interests do not correspond with the public interests defended by its principal. In the Packaging case, the agents have put marketing and presentation as priority criteria for packaging reduction and favored flexible management control system above fixed thresholds. We can consider delegation in the New Approach in terms of accountability.6 Accountability, as the means to control the agent, is rather weak: low level involvement of the principal, information deficit, poor 6.

For a thorough discussion on the different types of accountability in delegation see Lupia, 2003: 35-36

213

Politics Beyond the State.book Page 214 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

involvement of public interest groups which could serve as third party control agents and strong reliance on ex post control mechanisms with mixed results. However, if we interpret accountability as an outcome which reflects the principal’s interest we must conclude that accountability is present in ‘consumer safety related areas’. In these areas the agent’s preferences and the principal’s preferences are more likely to overlap as consumer safety is not only of public importance but also of economic importance. As a result health and safety issues delegated under the New Approach procedure, did not reveal any major agency loss. Furthermore, delegation could be considered a success as it resulted in an efficient and flexible policy adapted to modern technologies. Unfortunately, this kind of match is much less self-evident in the area of environmental policy, as environmental issues are often considered by producers and consumers, as a financial burden in the short term. Standardization for environmental purposes is burdened by both information deficit and preference heterogeneity, but no extra control instruments are introduced. This makes the delegation exercise prone to agency loss and underpins the importance of a thorough cost-benefit calculation before transferring delegation from one area to the other, as this might alter the outcome. High levels of independence, concentration of information within the European standardization bodies, and low levels of control provide ideal circumstances for agency loss. Nevertheless, delegation to private actors fits nicely into the governance approach the EU pursues. New Approach standardization reflects the call for increased involvement and participation of stakeholders in each step of the policy (expressed in the White paper on Governance CEC 2001e). The New Approach has even been cited as an example of co-regulation to be used to create better policies and regulation. This adds an institutional argument to the rational choice arguments mentioned above. However, this should not distract us from the real complexity of the delegation exercise at work in the New Approach directives. The New approach may well fit in the new governance approach promulgated by the European Commission, but it also reflects the potential downsides of governance-like tactics: loss of control, agency loss and dependency on private initiatives. It also underlines the importance of clear agreements between agents and principals, and of at least a minimal alignment of their respective preferences. These elements should be put into the cost-benefit equation before even starting delegation, as both control instruments and preferences could negatively affect the balance.

214

Politics Beyond the State.book Page 215 Wednesday, December 5, 2007 8:41 AM

Delegation Beyond the State?

Although delegation can be seen as the only practical means to achieve the governance objective, which was the central reason for the development of the New Approach, it is important to make a full assessment of costs, benefits and outcome before expanding its use to other policy areas.

215

Politics Beyond the State.book Page 216 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State.book Page 217 Wednesday, December 5, 2007 8:41 AM

Involving Private Actors in Policy-making

Involving Private Actors in Policy-making: Reflections on Gender Mainstreaming and Governance Petra Meier

Since the fourth UN women’s conference in Beijing in 1995, gender equality policies have increasingly been oriented towards a broader and more structural approach of gender inequalities. Next to specific or targeted gender equality policies, other policy areas are also meant to promote gender equality. By recognising how previous and existing policies contain biases (re)producing gender inequality and by assessing how these can be overcome, gender mainstreaming seeks to embed a gender perspective in all policy areas. Gender mainstreaming is in sum meant to gender the mainstream of public policies. From the outset, there were high expectations about gender mainstreaming and the possibilities it raised for achieving gender equality. Disappointment about the results of former policies to promote gender equality increased hopes that the new structural approach of gender mainstreaming might be more successful. But there were also critical voices underlining that gender mainstreaming would not be the magic spell. A decade after the introduction of gender mainstreaming feelings are still mixed. Whereas many underline its ‘revolutionary potential’ (Bustelo, 2003; Mazey, 2000; Verloo, 2001), results cannot be qualified as an overwhelming success. A number of comparative research projects conducted under the EU fifth framework programme confirm that results are limited (Mosesdottir and Erlingsdottir, 2005) and that only Sweden applies a genuine gender mainstreaming approach (Daly, 2005; Rubery et al., 2004). Explanations for the lack of results refer to the ‘revolutionary’ potential, arguing that barriers to gender mainstreaming are due to the patriarchal opposition to the feminist goals implied in the strategy (Stratigaki, 2005). Others point to the lack of necessary and facilitating conditions or prerequisites (Rees, 2005; Woodward, 2003), or to the high expectations raised by the definition of gender mainstreaming itself (Meier, 2006). Still other explanations refer to the difficulty in assimilating the concept and to different gender (in)equality policy frames and their impact on the 217

Politics Beyond the State.book Page 218 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

implementation of gender mainstreaming (Behning and Serrano Pascual, 2001; Beveridge, Nott and Stephen, 2000; Pollack and Hafner-Burton, 2000; Lombardo and Meier, 2006). The literature focused on explanations for the success story failing to materialise, and attributed the apparent lack of success mainly to unfavourable external conditions. Gender mainstreaming as a strategy or concept was not fundamentally questioned. Only recently have some authors focussed on shortcomings or flaws in the concept itself, arguing that there is need for a more theoretical elaboration of gender mainstreaming (Daly, 2005; Squires, 2005; Verloo, 2005). The aim of this chapter is to analyse the challenges of gender mainstreaming from a governance perspective. The motivation for this governance orientation is twofold. First, there is an expanding role of private actors in politics and in the delivery of ‘public’ goods, and the question is to what extent gender equality policies follow this trend. Secondly, while feminist actors tend to underline the empowering potential of the recent shift towards gender mainstreaming, the concept is not necessarily permeable to the idea of governance and the latter is not necessarily an empowering strategy. In this respect the chapter joins the new academic focus on the concept of gender mainstreaming itself. We explore how both the concept of gender mainstreaming and the practice of gender equality policies relate to the governance trend. To start, we discuss the concept of gender mainstreaming and its relation to governance, understanding the latter mainly as a move away from statebased institutions towards private actors, as set out by Jans in this volume. We then draw on empirical evidence. The chapter will discuss some major gender mainstreaming initiatives undertaken by the federal Belgian government. In the conclusions we address the added value of involving private actors in policy-making.

The concept of gender mainstreaming Gender mainstreaming as a policy strategy has witnessed a rapid global diffusion since the 1995 UN women’s conference in Beijing. Originating in the field of development cooperation after the Nairobi Forward Looking Strategies, it was promoted as the general approach to tackle problems of gender inequality in the Beijing Platform for Action. The EU endorsed gender mainstreaming as its official policy approach to gender equality in

218

Politics Beyond the State.book Page 219 Wednesday, December 5, 2007 8:41 AM

Involving Private Actors in Policy-making

the Amsterdam Treaty (1997) and made its adoption a condition for the member states that were to join in 2004 and 2007. Notwithstanding its diffusion through UN events and networks (True and Mintrom, 2001), the most common definition of gender mainstreaming has been formulated by an expert group on behalf of the Council of Europe: Gender mainstreaming is the (re)organization, improvement, development and evaluation of policy processes so that a gender equality perspective is incorporated in all policies at all levels and at all stages, by the actors normally involved in policy-making. (Council of Europe, 1998: 15). This definition and the accompanying conceptual framework are without doubt the most widely used by both practitioners and scholars of gender mainstreaming. Therefore it will also be used as the basis for the analysis in this chapter. What is gender mainstreaming? By far the most important feature of gender mainstreaming is the underlying concept of equality. Although gender equality policies and women’s policy agencies carry many different meanings, the concept of gender equality has broadly speaking evolved from a legalistic approach based on equal rights, to equal opportunities supported by positive action, and finally to what is called a more structural conceptualisation of gender equality. In the feminist literature gender mainstreaming represents this latest approach. It implies an underlying model of equality that is based on assigning new standards for gender equality not only for women but also for men (Rees, 1998). Although not all scholars agree on this far-reaching definition as being the sole model of gender equality in gender mainstreaming (see, for instance, Booth and Bennett, 2002), there is a tendency to argue that equality should not be defined as sameness, whereby male standards are accepted without discussion and the aim consists of upgrading the societal position of women to equal that of men. Neither should equality be defined as an equal valuation of difference between the sexes whereby women (and men) might still be stuck in traditional role patterns. Gender mainstreaming rather implies that new standards are defined for both sexes that are neither male nor necessarily confirm traditional role patterns (Squires, 2005). Verloo (2001) argues that gender mainstreaming stands for the recognition of the impact of gender biases on the reproduction of gender inequality. Her description of this dynamic suggests that both gender inequality itself and its mechanisms of reproduction have a structural basis. Similarly, Liebert (2002) argues that gender mainstreaming widens the gender equality frame by expanding thinking about the structural and institutional causes of inequality. Neither the definition of equality as sameness, nor its definition as the equal valuation of differences between the sexes is considered to fully recognise the structural causes of gender equality. A 219

Politics Beyond the State.book Page 220 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

similar reflection on equality can be found in the “shifts” that Lombardo (2005) has defined in order to measure the occurrence of gender mainstreaming as opposed to more traditional approaches to gender equality (see below). Gender mainstreaming seeks to be more than a broader definition of gender equality. The definition of what has to be achieved will influence the organisation and the course of the policy-making process. Gender mainstreaming is a particular conceptualisation of how to approach problems of gender inequality in order to achieve substantive equality. This is reflected in the shifts defined by Lombardo. A first shift to be triggered by gender mainstreaming implies that the mainstream policy agenda is reoriented in order to give priority to gender equality. Policy objectives and measures meant to achieve substantive gender equality have to receive a relevant place in meaningful policy initiatives and fields. Gender equality objectives and targeted policies of special relevance for women should get priority. Another shift builds up on the previous one and requires that a gender perspective be built into the mainstream political agenda. Policy measures have to be screened regarding their effects on both sexes. Policy ends and means have to be evaluated and re-articulated from a gender perspective. A related shift concerns the institutional and organisational cultures of political decision-making. Acquiring the necessary gender expertise and knowledge on the mechanisms causing and reproducing gender inequality as well as on the necessary remedies implies shifts in the policy-process, in policy mechanisms and regarding policy actors.1 Gender mainstreaming is meant to intervene in the existing policy-making process. The minimal result is that the policy process is challenged, for instance by the addition of new elements or by a minor reorganisation. The maximum is that the policy-making process becomes completely revised and remodelled, adding major new dimensions to it. The different impacts

1.

Lombardo also mentions a shift that implies parity between men and women in decision-making bodies and processes. It is a move towards the inclusion and participation of a higher number of women in political decision-making. Lombardo does not clarify the precise link between an equal participation of both sexes in political decision-making and gender mainstreaming. Parity could actually be seen as one of the outcomes of gender mainstreaming, the latter being a tool to achieve substantive equality. Substantive equality implies an equal share of power between men and women. It is however arguable whether there needs to be a shift from a minimum presence to an equal share of women in political decision-making in order to be able to speak of gender mainstreaming. But Lombardo is not alone in underlining that a sharing of power between the sexes is a condition for rather than a consequence of gender mainstreaming (see for instance Council of Europe, 1998).

220

Politics Beyond the State.book Page 221 Wednesday, December 5, 2007 8:41 AM

Involving Private Actors in Policy-making

of gender mainstreaming on the policy-making process are reflected in two approaches labelled as the agenda-setting approach (Jahan, 1995) and the integrationist approach. Gender mainstreaming is supposed to reorient the mainstream political agenda because it fundamentally rearticulates policy paradigms, ends and means from a gender perspective and it prioritises gender objectives. While the agenda-setting approach transforms mainstream policies, an integrationist approach does not challenge the mainstream itself. It would merely limit itself to inserting a gender perspective, to adding women to the mainstream, without questioning the underlying policy paradigms. Feminists argue that gender mainstreaming is precisely about challenging and transforming the mainstream. A fundamental difference with former gender equality policies is that gender mainstreaming requires the various actors involved in policy-making to adopt a gender equality perspective. Previously gender equality policies were confined – and limited – to the women’s policy agency. The fact that both new types of actors are involved and that the goal is framed in a broader way requires innovative practices when it comes to policy making. It is this request for innovation that opens the door to governance. In the following sections we will discuss the relation between gender mainstreaming and governance.

Gender mainstreaming and the role of private actors in policymaking Where Lombardo addresses the shift in institutional and organisational cultures of political decision-making that gender mainstreaming would provoke, she underlines the need for new actors in policy making, including civil society. Similar to others (cf. Council of Europe, 1998) she underlines that the furthering of gender equality in various policy areas requires knowledge of the current gender relations. Gender mainstreaming puts special emphasis on the availability of statistics and data segregated by sex, information that is often lacking or not up to date. Next to quantitative data more meaningful insights in the mechanisms shaping gender regimes and guiding gender positions in every day life are needed. While the women’s policy agency is already lacking part of such knowledge, it goes all the more for other policy-makers involved in gender mainstreaming. Making an appeal to external expertise – on a temporary or long term basis – is seen as a solution, in order to bridge the lack of in house expertise. Involving external and in this respect private actors in gender

221

Politics Beyond the State.book Page 222 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

mainstreaming typically has implied the participation of both the women’s movements and feminist academics in policy development. The Council of Europe’s handbook on gender mainstreaming refers to different actors and their role in gender mainstreaming. Next to typical policy makers it underlines the importance of researchers and experts in gender studies, of NGOs and of interest groups. They are considered to play a role in the detection of policy issues, in setting the agenda, in developing policy tools and in performing a watchdog function. The involvement of external actors also follows from the policy instruments suggested by the Council of Europe to achieve gender mainstreaming, namely, analytical tools (eg. gender impact assessments, gender budgeting, research, data); educational tools (awareness-raising campaigns, training, expert speakers, manuals) and consultative or participatory tools (eg. think tanks, hearings, working groups). These policy instruments all require significant involvement and the contribution of third parties outside the governmental structures to bring about policy change. Similar references to consultative practices, to the need for expertise and to the corresponding external actors can be found elsewhere (Rees, 2005). Non-governmental parties portrayed as crucial actors for gender mainstreaming can also be found in Woodward’s frequently cited ‘velvet triangles’. These triangles include policy makers, the women’s movement and academic gender experts (Woodward, 2004). According to her, one of the key strengths of gender mainstreaming is its effectiveness in providing gender experts with an important role in policy preparation and in policymaking. Given the lack of high level gender expertise available inside policy-making circles, most of it comes from outside. For instance Mazey (2002) underlines the empowering potential of gender mainstreaming for the women’s movement. Given the prevailing lack of expertise, the European Women’s Lobby (EWL) became an important player in the European policy-making process. The EWL participated in effective technical policy-making and became active in policy areas formerly closed to it. According to Mazey this did not only change the role of the EWL but also enhanced its standing and legitimacy in the policy community. Other scholars underline that gender mainstreaming only has a limited appeal to gender experts from the academia and from the women’s movement. An analysis of implications of the integration of the gender mainstreaming strategy into the European Employment strategy in Austria, Denmark, Finland, the Netherlands, Spain, Hungary and Iceland reveals that the social partners and women’s movements played but a marginal role in the design and implementation of gender equality policies (Mosesdottir

222

Politics Beyond the State.book Page 223 Wednesday, December 5, 2007 8:41 AM

Involving Private Actors in Policy-making

and Erlingsdottir, 2005). The authors argue that the adoption of gender mainstreaming has induced national authorities to introduce a technical approach to gender equality, whereby public officials mostly manage the work in the field of gender mainstreaming while external actors play a limited role. Furthermore, the lack of a stable system of consultation leads to variation in who is selected as an actor to shape gender policies. The temporary character of external, non-governmental involvement in combination with the narrowly defined tasks constrains the opportunities to influence policy making. Scholars argue that a distinction should be made between expert bureaucratic models on the one hand and participatory democratic ones on the other hand (Walby, 2005; referring to Beveridge et al., 2000). While the former involve experts and specialists the latter call upon a broad range of individuals and organisations. Gender mainstreaming is often presented as primarily a participatory model giving various actors, previously outside the privileged policy arena, a voice. It is meant to enhance the participatory features of democracy and in this respect to contain an empowering potential. But it can also be conceived as a technical process, whereby mainstream policy actors promote the gender equality agenda with the help of toolkits such as gender impact assessment. Verloo (2005) points out that the Council of Europe remains vague when it comes to the involvement of experts and the women’s movement. Also, the role ascribed to them is purely supportive. Feminist expertise and voices are deemed to be neutral and void of normative orientations. They are meant to support the development of gender mainstreaming and to provide ‘objective’ expertise. The overall conception of gender mainstreaming as a technocratic process excludes political debates and dissonant voices. Daly (2005) confirms this consensual character of gender mainstreaming. Even though gender mainstreaming entails both a critique of the state, as well as a reform agenda, the dominant conviction is that public officials just miss the right perception and knowledge of gender issues. All problems are deemed to be solved once public officials acquire the necessary frame. Putting gender mainstreaming in practice – and achieving gender equality – is an issue of providing for the appropriate consciousness and skills. There is no analysis of power interests intrinsic to gender relations and state institutions are not dealt with as sites where conflicts over gender relations take place. Benschop and Verloo (2000) report on the consensual dynamics of gender mainstreaming when describing their personal experiences as academics involved in the introduction of a gender perspective in the human

223

Politics Beyond the State.book Page 224 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

resources management of the Flemish administration. They describe how their dependence on the Flemish public officials made them strive for consensus rather than to allow divergent opinions to result in conflict. This search for consensus on behalf of external experts seems to be inherent to gender mainstreaming because the strategy depends on the approval of policy makers organised in non-hierarchical settings. The experts who are brought into the policy-making process in order to promote gender mainstreaming have no other authority than their academic status or eventually their (temporary) mandate but they are not fully integrated in the existing hierarchy. Being called upon to foster new ideas, they depend on the goodwill of the political and administrative hierarchy to accept these innovations. External experts might come with the theoretical and technical knowledge, but they do not dispose of the tools to translate this expertise into concrete policy initiatives. For this they depend upon the approval of the political and/or administrative hierarchy. In order to ensure themselves of the cooperation of the latter, external experts might be tempted to look for compromise, or even self-censorship and the strategic framing (Verloo, 2001) of gender equality goals, all of them meant to anticipate and avoid possible conflict and negative reactions. There is evidence that the more gender mainstreaming is strategically framed in accordance with dominant policy frames, the more the political system is receptive to it (Pollack and Hafner-Burton, 2000). But strategic framing and preventive self-censorship can have a boomerang effect, in that the message is not passed, relevant details on how gender relations should change or even the basic assumptions of gender mainstreaming do not get across. In the end, strategic framing and self-censorship might undermine implementation of gender mainstreaming. From the outset, the definition of gender mainstreaming underlines the importance of mainstream public officials. The main feature of the strategy is to have regular public officials conceive and implement gender equality policies. Regular public officials are used for mainstreaming purposes as opposed to femocrats, to whom gender equality policies were formerly confined. However, in both cases public officials and not private actors are supposed to be the main actors. In this respect gender mainstreaming refers to government rather than to governance. But it keeps up with the governance trend in that private actors are called upon to achieve and deliver what the state can no longer do in its own. Feminist actors are meant to join the process in order to close the knowledge and expertise gap, be it in a short or long term capacity. The technocratic perception of feminist actors as gender experts tallies well with the governance discourse. But gender mainstreaming should be 224

Politics Beyond the State.book Page 225 Wednesday, December 5, 2007 8:41 AM

Involving Private Actors in Policy-making

treated with caution when it is presented as a means to provide feminists as feminists with access to policy-making processes. The empowering feminist potential of the strategy resides in the results it is meant to achieve rather than in the process that could lead to the desired results. Gender mainstreaming empowers to the extent that it should profoundly change the everyday living conditions of men and women but it does not necessarily open the policy-making process to feminist actors as understood by state feminism (Lovenduski et al., 2005; Mazur, 2001; Outshoorn, 2004; Stetson, 2001; Stetson and Mazur, 1995). Feminist concerns are not necessarily put on the agenda as such but wrapped up in generally acceptable claims. Gender mainstreaming, as set out in its definition, is not meant to provide for ‘presence’ as a meaningful participatory act to articulate and further feminist claims. Furthermore, the literature does not explain how then, a participation of gender experts striving for consensus will help bring about the revolution gender mainstreaming is meant to provoke. It should be noted that the governance approach itself operates a technocratic model when it comes to involving private actors in policy-making and the delivery of public goods. While issues of legitimacy or accountability are addressed, the focus is less on questions of empowerment. In this respect gender mainstreaming fits into the governance trend when it comes to gender experts being added to a process of policy-making. But the governance perspective relates poorly to the revolutionary potential that gender mainstreaming is meant to contain. In the following sections we will draw on some empirical evidence to further analyse the role of private actors in policy-making, more precisely when it comes to initiating gender mainstreaming processes in Belgium.

Gender mainstreaming in Belgian gender equality policies What is generally considered to be the first real Belgian women’s policy agency was set up in the wake of the 1985 Nairobi UN World Conference on Women. Miet Smet, a former president of the Flemish Christian Democrats’ political women’s organisation (Vrouw en Maatschappij) became State Secretary of Environmental Affairs and also negotiated an Equal Opportunities portfolio. Formally, the State Secretary’s competence covered everything related to social emancipation, but throughout her time in office Miet Smet focused on violence against women, their economic position and their participation in political decision-making, issues that are still on the agenda twenty years later. When Miet Smet became Minister of Labour and Employment in 1991 she retained what became the Equal 225

Politics Beyond the State.book Page 226 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Opportunities portfolio. The women’s policy agency attained the level of a fully-fledged Minister, including the necessary structural political and administrative support. Over time the federal women’s policy agency witnessed a steady upgrading in its autonomy, in its scope and to a certain extent in the government hierarchy. However, in terms of resources such as budget and personnel it has always been marginal compared to other Ministers and State Secretaries. Furthermore the affinity with gender equality of the Ministers in charge of equal opportunities declined after 1999. Subsequent ministers developed a less pronounced stance on gender equality and none had such direct ties with the women’s movement as minister Miet Smet did. The development of a federal structure also triggered new gender equality policy infrastructures at the regional level. Since the first Flemish regional elections in 1995, an Equal Opportunities portfolio was created in the Flemish regional government. With some delay most of the other regional authorities also developed gender equality policies.2 In Flanders the women’s policy agency became an important player, but for the rest gender equality policies are mainly initiated by the federal policy structure. From the outset these federal policies were defined as a separate policy area and the focus was on equal opportunities in the strict sense of the term. A first formal step towards gender mainstreaming was undertaken in 1996, when parliament voted an act regarding the follow-up of the Beijing Platform for Action3. The government has to report annually to parliament on the achievement of the resolutions in the Beijing Platform for Action. In principle, this reporting prompts the various Ministers to develop a gender mainstreaming approach, but the compulsory character is limited to the reporting and does not command progress. In 2006 the government agreed upon a bill inciting the various Ministers to define a series of strategic objectives on gender equality at the beginning of a legislature. Progress will be monitored and a final report has to follow at the end of the legislature. Also, the share of the budget earmarked for the furthering of gender equality policies has to be identified.4 This bill

2.

3.

Since the 2004 regional elections, the Walloon and Brussels governments have a Minister or State Secretary of Equal Opportunities. No Minister of the Francophone Community Government is in charge of equal opportunities but there is a unit of equal opportunities in the Ministry of the Francophone Community. The Germanspeaking Community has no women’s policy agency. Wet van 6 maart 1996 strekkende tot controle op de toepassing van de resoluties van de Wereldvrouwenconferentie te Peking, van 4 tot 14 september 1995.

226

Politics Beyond the State.book Page 227 Wednesday, December 5, 2007 8:41 AM

Involving Private Actors in Policy-making

contains a more direct impetus for gender mainstreaming than does the 1996 act. The bill is actually based on the most concrete attempt so far to introduce gender mainstreaming in federal Belgian politics. When the new Minister of gender equality policies took over in 1999, members of her staff pushed for concrete initiatives to develop a gender mainstreaming approach. In 2000 the Minister of Equal Opportunities convinced her colleague Ministers to define a strategic objective with respect to the furthering of gender equality within their policy competencies. In all twenty policy areas strategic objectives were articulated. The project was initially meant to last for one year (2001), but was extended until the end of 2002, since the objectives had not been attained at the end of the first year.5 For the entire period of two years a Cell Gender Mainstreaming was set up, composed of academics from different Belgian universities. Their task consisted in providing the necessary knowledge on gender mainstreaming and in supporting the regular policy makers within the cabinet and the administration.6 The Flemish authorities stimulated the development of a concrete tool for gender mainstreaming. Drawing from the Dutch example, a gender impact assessment tool was developed but up to date its application is limited to a couple of individual initiatives. The Federal Cell Gender Mainstreaming launched a gender budgeting project but on the whole no specific policy tools were set in place to achieve gender mainstreaming at the federal level. Other incentives for gender mainstreaming tend to come from the supranational level, more particularly from the EU. However, EU initiatives do not always foster far-reaching new policy initiatives. Collectively negotiated objectives are vague so that they can be adopted to different policy environments. This vagueness allows for the inclusion of a broad set of policy initiatives, including existing ones. In many respects, objectives are negotiated so as to allow for a maximum output with a minimal investment and objectives are adapted to existing policies rather than clearing the path for new ones. Also, the more concrete policy objectives 4.

5.

6.

Wet van 12 januari 2007 strekkende tot controle op de toepassing van de resoluties van de wereldvrouwenconferentie die in september 1995 in Peking heeft plaatsgehad en tot integratie van de genderdimensie in het geheel van de federale beleidslijnen. Many objectives were not reached at the end of 2002 either, but the project was not prolonged for several reasons, amongst which the fact that elections were due in May 2003. Per policy area one civil servant and one member of the cabinet were appointed by the Ministers to follow up the gender mainstreaming project. This charge was a supplementary one, often assigned to those already responsible for positive actions.

227

Politics Beyond the State.book Page 228 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

become the less far-reaching they tend to be. Furthermore, objectives are set without the provision of the necessary means to achieve them. Finally, these measures are seldom evaluated. In many cases, their achievement is even left to the discretion of other actors involved in the policy-making process, such as the social partners (see for instance Booth and Bennett, 2002; Grosser and Moon, 2005; Liebert, 2002; Mazey, 2002; Mosesdottir and Erlingsdottir, 2005; Perrons, 2005; Pollack and Hafner-Burton 2000; Rees, 2001; Rubery, 2002). The subsequent section mainly draws upon the activities of the Cell Gender Mainstreaming because of its extensive involvement of external partners in the policy-making process.

The added value of private actors in policy-making: the Cell Gender Mainstreaming The Cell Gender Mainstreaming is an interesting case as it provides information on the promises and pitfalls of public private partnerships in gender equality policies – and beyond. The Belgian experience fits with the idea of private partners delivering missing means, in this case knowledge and expertise on gender equality issues. In most policy departments the expertise on gender equality policies was absent or poor, and the Cell Gender Mainstreaming was a first attempt to put gender mainstreaming into practice. An appeal was made to academics for their knowledge, not for their academic status. Belgian politicians and public officials also tend to call upon consultants in order to promote gender equality policies, but at that time academic actors were thought to be the most suitable private partner for the implementation of gender mainstreaming. The academic partners remained private partners throughout the policy-making process. At no stage were they integrated into the existing decision-making structures. Their voice was that of an external, advising partner. Few other external actors were involved in the project. The Minister of the Self-employed, seeking to adress the position of wives helping in small and medium sized family enterprises (SME), repeatedly consulted with interest groups and urged these groups to formulate an advice on how to improve the fiscal and social status of women working informally in these familyowned SMEs. The State Secretary for Energy and Sustainable Development asked the Council for Equality of Men and Women for advice on gender mainstreaming. Another initiative was taken by the Minister of Metropolitan Affairs who organised a conference on ‘Women and Cities’.

228

Politics Beyond the State.book Page 229 Wednesday, December 5, 2007 8:41 AM

Involving Private Actors in Policy-making

Finally, the Minister of Scientific Research pleaded for an increased involvement of women’s organisations in the steering committees for research projects, so that they could provide input on gender issues and stimulate researchers to pay more attention to the issue. The main motivation for contacting external partners, both academics and women’s movements, was to access the expertise available ‘in the field’. On the whole, the input of private actors in the Cell Gender Mainstreaming was marginal rather than small. In fact, the external actors invested considerable time and resources. In this respect their input was not small. But it was marginal, because the involvement did not necessarily have an impact on the result. The ambiguity of the position of private partners involved in gender mainstreaming can be clearly illustrated with the example of the delineation of their activities and responsibilities.7 At the start of the project, the various tasks and responsibilities had been roughly defined, but at the time of the project prolongation the tasks/ responsibilities were summed up in detail in a protocol approved by all parties. The initiative for this protocol came from the external experts. The external experts sought to limit their workload and to stress the responsibility of the cabinets and the administrations to undertake action. The external experts unanimously stressed that too much had been put on their shoulders during the first year. In the second year the cabinets and administrations were responsible for drawing up a plan of action, including a time schedule and budget, and for defining monitoring indicators. They had to take the lead in describing the methodology, the tools and the resources (including the required external expertise). They were responsible for setting up a working schedule defining the responsibilities and tasks of all actors concerned. They were also in charge of the delivering the progress reports. The experts were meant to contribute to these activities but were not mean to be the primary drivers of the process. The responsibility of the experts concerned the initiation and the follow-up of the cooperation among the various actors (eg. establishing contact with the cabinets and administrations or organising working groups). Even though the various tasks and responsibilities were defined during the second half of the project, matters worked out differently. In some policy areas no working schedules were drawn up, nor was there any other clear 7.

With respect to factual information on the Cell Gender Mainstreaming, the remainder of this section draws on the final report of the Cell Gender Mainstreaming, of which the present author has been one of the contributors and coordinators: Cel mainstreaming (2003) Eindrapport en evaluatie van de door de federale regering opgerichte cel gender mainstreaming’. Brussels: FOD Werkgelegenheid, Arbeid, en Sociaal Overleg.

229

Politics Beyond the State.book Page 230 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

delineation of tasks and responsibilities among the various actors. In some cases this went hand in hand with a lack of any initiative on behalf of the public officials to achieve the strategic objectives. In others it led to confusion about the individual tasks and responsibilities, while the initial idea behind the working schedule was to prevent such misinterpretations about tasks and responsibilities. The vacuum hindered the functioning of certain actors and allowed others for remaining passive. Where a clear delineation of the various tasks and responsibilities existed, it had often been initiated and drawn up by the external expert and not by the public official meant to do so. This not only reflects the fact that many of the experts saw themselves compelled to go beyond their tasks and responsibilities, even though a protocol had been signed in order to clearly delineate what to expect of the experts. This unilateral action on behalf of the expert also hampered the establishment a common platform from which to start. Eventually the established working schedule was not always been respected, at least not by all parties involved8. In a couple of policy areas the working schedule was developed by the cabinet or the administration, but always with extensive input of the expert. The issue of the working schedule reflects a major problem in respect to the role of the external expert. Most experts did more than they were obliged to do. The reasons for this were ample, including lack of importance attached to the issue, lack of time, lack of knowledge. Experts were seen as a welcome additional workforce, able to fulfil certain tasks that public officials would then not have to accomplish. Part of this attitude stems from the fact that the gender mainstreaming project implied a substantial additional workload for the public officials. The experts adopted a different perspective and underlined their supportive role, generating information and ideas, and answering questions. While the external experts perceived a reactive role for themselves, many of the public officials saw the role of the experts as an opportunity to offload and delegate certain tasks. Experts also wanted to delineate their tasks because they were concerned about their official status. Their competence concerned the delivery of content, not of formal decisions. They did not dispose of the necessary authority or decision-making power, which had to be delivered by the administration or especially by the cabinet staff. The experts were simply 8.

It should be noted that the external experts respected all their obligations. This can be explained by a high interest in the topic and the requirement to deliver value for money. The latter also reflects the dependency of gender equality research on public money and the anxiety not to loose credibility for future research.

230

Politics Beyond the State.book Page 231 Wednesday, December 5, 2007 8:41 AM

Involving Private Actors in Policy-making

not the right persons to accomplish certain tasks. However, it were often these tasks that required expertise in gender equality issues, such as the definition of strategic objectives and their translation into concrete policy measures, which led the cabinet officials and/or the administration to rely on the experts. Defining strategic objectives is more than setting a target on, for instance, the percentage of women in office. This goal-setting has to be informed by knowledge on mechanisms that (re)produce inequality and by expertise on how to intervene in these mechanisms. Defining a feasible strategic objective requires that the broad framework of the various actors and their responsibilities, of the budget and time line have been determined. In this respect cabinet officials cannot simply set a goal and leave the rest to the administration or to external experts. Since the definition of strategic goals needs to tally with the definition of all that is needed to achieve them, an extensive deliberation between the cabinet officials, the administration and possibly the external expert is required. The working sequence is one of simultaneous interactive processes rather than one of a chronological completion of steps. In this context, participants of both the administration and of the universities underlined that cabinet officials – and their Ministers – need to be willing to invest in a gender mainstreaming process. Finally, the experts saw another reason not to position themselves inside the policy-making process. In order not to hamper the future development of a gender mainstreaming process within the cabinets and administration, they thought it was important to avoid the impression of a close association or political partnerships between the Minister and the external experts. The process should be able to continue without them once the Cell Gender Mainstreaming ceased to exist.9 Most policy makers were nonetheless convinced that the external experts were a crucial element in the process. They mostly brought new insights to issues that had generally not been considered before. Their presence was also seen as a gain of time, the existing information was available and had not to be sought for. They were a welcome additional workforce. Furthermore, active support by external experts was considered to be motivating for those who perceived a policy making environment that was 9.

The extent, to which the experts took on roles and tasks they were not supposed to carry out, is an indicator for the continuity of the gender mainstreaming activities once the Cell Gender Mainstreaming ceased to exist. At the end of 2002 most projects had not got any further than the conceptual phase, except for initiatives meant to gather data (research) or to sensitise (conferences, work shops). These initiatives are important in order to gather the necessary basis for gender mainstreaming, but as such they do not further gender equality.

231

Politics Beyond the State.book Page 232 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

not intrinsically concerned by or hostile to the issue of gender equality. The project and the presence of the expert legitimised action and the external expertise was also a moral support for the actors in favour of gender equality policies. In a couple of cases, the structure of the Cell Gender Mainstreaming and the role of the expert lead to a more intense cooperation between the cabinet officials and the administration. The expert was considered to be an objective actor, being able to mediate and remedy. In some cases, therefore, the expert was not only a source of knowledge and a workforce, but became a tool in mediating professional and interpersonal relations. This did not facilitate their task. It also explains why many actors involved did not appreciate the high turnover in cabinet staff, civil servants and experts. The frequent changes of participants led to a constant need to reinvest in interpersonal relations10. Governance involves contracting private expertise and know-how but it also stands for public-private partnerships in collective and public decision-making. In the case of the Belgian Cell Gender Mainstreaming, we can definitely speak of the reliance on private knowledge and expertise. The participation of private actors was important in terms of the delivered knowledge and expertise. The role of the external experts was much more limited when effective decision-making was concerned. Joint decisionmaking between the cabinet, the administration and the external experts did not occur systematically. Under specific conditions (eg. proactive experts, good interpersonal relationships, etc.) experts could influence certain decisions but there was no systematic inclusion of the external experts in the decision-making process.

Conclusion The Belgian case illustrates some promises but also some pitfalls inherent to the involvement of private actors in gender equality policy-making, and, more broadly, in the governance approach. Gender mainstreaming facilitates the access of private actors, such as academic or social movement gender experts, to processes of policy-making. By its explicit reference to the role of private partners in policy development the concept of gender mainstreaming matches the idea of governance. Gender mainstreaming initiatives seem to put this idea of a public private partnership into practice. Recent initiatives such as the Belgian Cell Gender Mainstreaming match 10. Staff mobility in cabinets also hampered the delivery and the achievement of the goals set out at the beginning of the project.

232

Politics Beyond the State.book Page 233 Wednesday, December 5, 2007 8:41 AM

Involving Private Actors in Policy-making

the governance trend and the expanding role of private actors in politics and the delivery of ‘public’ goods. However, while feminist actors tend to underline the empowering potential of gender mainstreaming, it does not seem to be an empowering strategy for the external actors involved in policy-making. External gender experts do not necessarily play an important role in the policy-making process. Their opportunities to influence the agenda and the outcome are constrained and limited. The strength of the involvement of external experts, or of a governance approach in general, resides, amongst others, in the quick access to knowledge or experience not available in the policy-making environment. The weakness lies in the fact that these experts are not fully integrated into the decision-making structures, which allows decision-makers to pick and choose what to accept. In a governance mode of decision-making, external actors are not necessarily equal partners. Rather the external experts are accepted to the extent that they deliver what is needed and/or wanted. In the case of gender equality policies, their position might be particularly weak. Even though gender experts might provide for knowledge on problems of gender inequality that is lacking in policy-making circles, the gender equality objective is often not a high priority issue. This lack of importance attached to the objective undermines the position of external experts, which leads to the paradoxical situation that those who are meant to introduce and add what is missing do not dispose of the necessary status to do so.

233

Politics Beyond the State.book Page 234 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State.book Page 235 Wednesday, December 5, 2007 8:41 AM

EU Policies on Violence against Women. Contested Policy Areas and the Potential of a Governance Approach Marjolein Paantjens

Introduction After more than 20 years of struggle, feminists all over Europe have succeeded in changing the idea that domestic violence is a private matter, into the view that it is a state responsibility. When we look across Europe, there is no country where domestic violence is not framed as a public policy concern1. Some states already recognized violence against women as a policy issue in the 1970s (e.g. United Kingdom, the Netherlands), others followed in the 1980’s (e.g. Spain) (Roggeband, 2002), and over the last decade, we also saw the new Eastern European member states developing policies in this field (Kriszan et. al., 2005). However, it is necessary to put this success into perspective. Data on domestic violence are still inconsistent (if they exist at all) and not comparable (Carlshamre report, A6-0404/05). On top of this, violence against women2, still remains a problem that is endemic to any country. The European Union (EU) is often mentioned as a ‘policy entrepreneur’ or leader when it comes to gender equality policies. The emphasis of gender (equality) policies in EU policy making has been on equal opportunities and equal treatment mainly in the field of labour and economy. The directive on equal pay for example, was more progressive for many member states than their own equal opportunities provisions, confronting them with costly implementation processes (Van der Vleuten, 2001). 1.

2.

Large parts of the material in this chapter have been collected under the MAGEEQ project (see www.mageeq.net). One of the main findings of the comparative analysis on domestic violence was that the issue was articulated as a public policy concern in all the countries studied. These were, next to the EU level: Austria, Greece, Hungary, Netherlands, Slovenia, Spain. I would like to thank the whole MAGEEQ team, but especially Mieke Verloo for taking me on board and Petra Meier for her extensive help and support in ‘reframing’ this chapter. How to define violence against women and domestic violence is an issue as such that will be dealt with in the next section.

235

Politics Beyond the State.book Page 236 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Addressing issues of gender outside the economic sphere, has proved to be difficult. Also, whenever they are addressed, they tend to get reformulated in economic terms3. On top of this, gender equality initiatives outside the economic sphere largely rely on soft measures of a non-binding character. However, this picture of gender policies might be a bit too sceptical, as the fact remains that issues outside the economic sphere, such as the position of women in political decision-making, did get on the agenda and were translated into policy programs at the EU-level. Explanations for the absence of binding measures or budget priorities regarding gender equality policies might be more subtle than the narrow EU focus on economic matters, as it is not only in the field of gender equality that we find a strong reliance on soft policy instruments. EU social policy at large, can be viewed as an area characterized by a so called governance approach associated with an arsenal of ‘soft’ policy instruments of a voluntary or non-binding character, lacking clear sanctions and having no specified procedure nor clear pre-defined goals (see Jans in this volume). The same goes for EU gender equality policies. This soft approach might be logical when seen in the light of the essentially contested character of the concept of gender equality. It has different meanings, norms and values attached to it, and the specific interpretation of it varies across Europe (Verloo, 2005; Walby, 2005)4. The use of ‘soft’ and flexible policy instruments allows to downplay the politicised nature of gender issues. The non-binding character of many gender equality policies can also avoid risks of contestation, protest or veto (Benschop and Verloo, 2002). Especially because gender equality can have many different meanings across member states it is sometimes necessary to strategically frame it in such a way that it can be linked with existing policy frames. It can be a strategic choice to ‘sell’ gender related issues under a different label to get them on the agenda and to create the necessary political will for action. An example of this strategic framing is the highlighting of the high costs of violence against women on society. Policies on violence against women also rely on soft measures, and a reason for this is that the EU has no official competencies for dealing with the 3.

4.

When ‘reconciliation of work and family life’ for example, became an issue, the debate quickly shifted in emphasis from having men and women share tasks to labour supply side arguments (Stratigaki, 2005). Verloo points out that debates on the action and strategies necessary to reach the goal of gender equality hide the often highly politicised question of what gender equality is. The MAGEEQ project revealed that the answer to this question was not unambiguously answered across Europe, and that gender equality has many different interpretations attached to it.

236

Politics Beyond the State.book Page 237 Wednesday, December 5, 2007 8:41 AM

EU Policies on Violence against Women

issue, except on the basis of ratified human rights provisions. There is the Charter of Fundamental Rights, the Convention on the Elimination of all forms of Discrimination Against Women (CEDAW treaty) and, for instance, the resolution on all forms of violence against women (A2-44/ 86). On top of that, the EU was one of the main initiators and signatories of the Beijing Platform for Action, which raised fighting violence against women to the main issues. Judicial or criminal legal aspects of violence against women fall outside the scope of EU policy making. Nonetheless it remains a fact that the EU does take initiatives in the field of violence against women. In sum, while there are openings for dealing with the issue, violence against women does not explicitly fall under the EU remit, which might explain the soft governance approach to the issue. Although such a soft approach is often contested for not being efficient and for not producing policy output that will help furthering gender equality, the present chapter will examine the approach from a different angle. Looking at the case of EU initiatives in the field of violence against women, it will explore the potential of a soft governance approach to contested or complex policy areas like gender (equality) policies.

Problem definition and central questions In this chapter EU policy documents on violence against women are analyzed by focusing on the problem diagnosis, the suggested prognosis and the actors meant to be involved. Comparing these findings and their evolution over time, it is possible to notice some clear shifts in the problem definition and analysis, some of them occurring chronologically, others simultaneously. Instead of pointing at the problems this might involve for policy output, this chapter focuses on its potential for policy making in the field of violence against women. The underlying assumption is that the sometimes sudden and large shifts in framing the issue can also be interpreted more positively; namely in the sense that this conceptual ambiguity, the variety of problem definitions, offers an opening for the advancement of policies in this field. That it is exactly this variety that creates windows of opportunities for furthering gender equality. The analysis of the policy documents is based on a critical frame analysis as it was developed under the MAGEEQ project, on the basis of a combination of social movement theory, discourse analysis and gender theory. A frame is an interpretation scheme that structures the meaning of reality (Verloo, 2005). A policy frame then, can be conceptualized as an organizing principle that turns fragmentary information into a structured 237

Politics Beyond the State.book Page 238 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

and meaningful policy problem, in which a solution is enclosed (Verloo, 2004, p.6). Frame analysis assumes that the way in which a policy problem is articulated is not necessarily neutral or objective, rather that it is structured or framed in a specific way. The definition of what the problem is, and why, is not an objective fact, it is highly politicized, although this is not necessarily done in a conscious way (Meier et al., 2005). For example, violence against women, and more specifically domestic violence, was for long viewed as a private matter, and was therefore not recognized as a policy problem. At best it was an individual problem for the women who faced it, but it was not seen as an issue of concern to policy-makers. Furthermore, violence against women might be viewed as a problem because of the gendered nature of it; namely that its root causes lies in an unequal balance of power between the sexes. It might however, also be conceptualized as a problem, merely because it involves high economic and social costs. In other words, a problem definition contains a certain number of assumptions. The same can be said of the way a problem is meant to be solved, and of who is responsible for doing so. The diagnosis, prognosis and actors taken together are a policy frame, although they need not form a coherent entity in terms of content (Meier, 2006). Studying policy documents on their content – the problem definition and why this is thought to be a problem, how it should be solved, what gender dimension (if any) is attached to it, and who is thought to be responsible for causing and for solving the problem – reveals (hidden) assumptions. Critical frame analysis focuses on the diagnosis contained in a policy document (what is the problem and by what is it caused), the prognosis (what to do about it) and the actors (who is going to do something about it). By closely studying the ways in which violence against women is framed as a policy problem and what solutions are meant to solve them, we can reveal the underlying argumentations and differences in analysis and articulation of the problem. In doing so, this chapter will demonstrate the ways in which the framing of violence against women has shifted in the period studied. It will give an idea about how different interpretations can exist of one policy problem, from depoliticized and compromising, to rather radical or contested interpretations. Before analyzing EU policies on violence against women, it is necessary to clarify the concept a little further, as it is not always clear what is covered by this term. Violence against women is used as an umbrella term in the EU, covering problems ranging from rape, prostitution and domestic violence, to sexual harassment or female genital mutilation. These problems have different causes and require a broad range of solutions, all of which are hidden by the umbrella label. Also, there is a widespread 238

Politics Beyond the State.book Page 239 Wednesday, December 5, 2007 8:41 AM

EU Policies on Violence against Women

discussion among women’s movements and feminists, on whether prostitution and trafficking should always be viewed in terms of violence against women. In this chapter the term violence against women is employed, but in a critical sense if there are no separate problem analyses presented in the policy documents. Next to the fact that violence against women is used as an umbrella concept, EU policies do not address the issue in a consistent manner (which might actually be due to the fact that they employ an umbrella concept). The EU policy programs and campaigns focus on different aspects of violence against women. At some points the term violence against women is used, at other points attention is directed specifically to domestic violence, a term which can actually cover particular ‘subproblems’ and does not necessarily only focus on gendered aspects of the problem. However, issues like sexual harassment in the workplace or trafficking are for instance dealt with in separate directives5. The documents analyzed in this chapter all focused on violence against women. The specific analyses and descriptions that are to follow will clearly show the exact problem articulation, make clear whether or not separate analyses for separate problems are provided etc., so that it is possible to use the term ‘violence against women’, when at the same time keeping in mind the different aspects covered by it. The next sections will give a detailed overview of the development of the issue of violence against women as a public policy concern in the EU. The analysis starts with the first resolution on violence against women that was adopted in 1986 (A2-44/86), and ends with the latest report on violence against women (Final A6-0404/2005) adopted by the European Parliament in December 2005. The emphasis will be on the way in which the issue is analyzed, focusing on diagnosis, prognosis and actors involved. The various documents will be compared on these dimensions in order to show policy shifts and the potential that lies within these shifts.

5.

Due to their specificity, these directives are not separately studied in this chapter. (Sexual Harassment: Commission communication of 24 July 1996 concerning the consultation of management and labour on the prevention of sexual harassment at work, COM(96) 373; SEC(97) 568; Commission Recommendation 92/131/EEC of 27 November 1991, C 27 of 04.02.1992, L 269 of 05.10.2002; Commission code of practice on sexual harassment, Official Journal L 49 of 24.02.1992) (Trafficking: Commission communication of 20 November 1996 to the Council and the European Parliament on trafficking in women for the purpose of sexual exploitation, COM(96) 567; COM(98) 726, COM(2000)854)

239

Politics Beyond the State.book Page 240 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Getting violence against women on the agenda As mentioned in the introduction, the EU is often considered as an important policy entrepreneur or a leading force in shaping gender equality policies. However, the preoccupation of the EU with issues linked to labour and economy, has led to a relative late recognition of violence against women as a public policy concern. Late, in comparison to many of its member states. At the EU level, the issue of violence against women was translated into public policy programs after the 1995 Beijing World Conference on Women6. However, violence against women already incidentally appeared on the agenda of the European Parliament (EP) in the 1980’s, in response to an own initiative report on violence against women, produced by the Committee on Women’s Rights and Equal Opportunities. Though the EU’s preoccupation with economic issues might have played a role in the late recognition of the problem, policies in the economic sphere, more specifically the equal opportunities regulations, provided a springboard for the introduction of the first resolution on violence against women. The EP Committee on Women’s Rights asked for European action on the problem of violence against women, by pointing out that the EU could not implement effective equal opportunities policies in the field of labour and employment, when women were victims of violence in their homes and in society at large. As long as violence against women would exist, women could never ‘really’ enjoy their equal opportunities. By framing the problem in these terms, the committee was able to establish a link between the problem and the economic and labour policies, namely through the EU’s equal opportunities policies in this area. In line with the European member states, the issue of violence against women was put on the political agenda under the influence of feminist actors (Van Lamoen, 2004). In most European countries feminist groups strived for recognition of the issue as a public policy concern, leading to discussion and development of policies at the national level. In its 1986 6.

It is difficult to point out exactly which events or actions by which actors led to the introduction of domestic violence as a policy problem at the EU level. The entrance of the topic in EU policy making might be partly related to the rise of women’s NGOs at EU level and the extension of the operational scope of the EU with the adoption of the Treaty of Maastricht (1992), but in the complex European arena many other actors and feminist voices might have played a role. At any rate, NGO’s like the European Women’s Lobby (EWL) (established in 1990) and Women Against Violence Europe (WAVE) (1994) have been actively engaged in the agenda setting process as well as in criticising the policy programs established in a later stage.

240

Politics Beyond the State.book Page 241 Wednesday, December 5, 2007 8:41 AM

EU Policies on Violence against Women

report the EP Committee on Women’s Rights presented violence against women as both the cause and the consequence of the unequal division of power between men and women (Hoskyns, 1996). The report ventilated a strong power analysis, stating that violence against women was a symptom of the unequal power balance between women and men, rooted in patriarchal structures. The problem was seen as universal, and the problems defined, ranged from sexual violence, harassment and rape to domestic violence. This power analysis of the problem, tracing the causes of violence against women back to the unequal power distributions between women and men, is typically associated with a feminist conceptualization of violence against women (Lombardo & Meier, 2006). The report was followed by a resolution on Violence against Women (A244/86), which did not take over the whole power analysis of the problem, but kept some of its key elements (Paantjens, 2004). After that resolution, it took several years before the issue was dealt with again.

Development of policies after 1995 The Fourth World Conference on Women in Beijing in 1995 seems to have given a major impetus to putting the issue on the European agenda again. Violence against women was one of the 12 areas of concern of the Platform of Action that was also endorsed by the EU. Between 1995 and 2000 the EU developed two main initiatives in the field of violence against women. These two programs already show different ways of articulating violence against women as a public policy concern: as a ‘human rights’ problem or as a threat to ‘public health’. These frames do not seem to originate from any of the member states, but seem to have developed at EU and international level (Kantola, 2004). Whereas the human rights discourse seems to have a long history in the EU, the Public Health frame has become the most dominant frame around 2000, while the human rights discourse re-appeared in 2005. The roots of conceptualizing violence against women as a breach of human rights can be traced back to the 1993 Vienna Declaration that recognized women’s rights as human rights, and before that to the Convention on the Elimination of all forms of Discrimination against Women (CEDAW)7. With the Amsterdam Treaty in 1997 and the adoption of the European 7.

In the international political arena, women’s rights movements have often tried to integrate women’s rights in human rights declarations. It is an often used strategy to frame violence against women as a human rights issue. This way of framing can be traced back more than a century. See e.g. Verucci; in F. Tabak, 1997.

241

Politics Beyond the State.book Page 242 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Charter of Fundamental Rights in 2000, human rights gained an even stronger position in the EU. The rise of the Public Health frame in the European political arena can be traced back to the period in which the Daphne Program – adopted in 1999 – was prepared. The strategy to approach violence against women as a health problem is derived from the World Health Organization. Its definition of health as a ‘state of complete physical, mental and social well-being and not merely the absence of disease or infirmity’ (24 Jan. 2000)8 was explained as to encompass the issue of violence against women as well. The main EU policy initiatives in which violence against women has been addressed, the ‘Zero Tolerance’ campaign, the Daphne Program and the Carlshamre report (A6-0404/2005), correspond with the two main ways of framing violence against women as a public policy problem: a gendered human rights approach in the Zero Tolerance documents and a degendered public health approach in Daphne documents. The latest report on violence against women seems to go back to the original framing of the problem in terms of gender and human rights. The following sections describe these three stages in the development of violence against women on the EU policy agenda.

The ‘Zero Tolerance’ Campaign (1997-2000) In 1997, the European Commission launched the campaign Zero Tolerance addressing violence against women, accompanied by a brochure specifically focusing on domestic violence. It followed the EP Resolution to establish a European wide campaign for zero tolerance of violence against women (A4-0250/1997). The campaign was legitimised mainly on the ground that violence against women has to be seen as a violation of human rights. Violence against women was to be a public policy concern arguing that according to CEDAW it is a ‘crime’. The problem was said to be endemic to any society. Arguments claiming that domestic violence is a ‘private matter’ or that women are to blame were labelled as ‘myths’. The campaign was based on the idea that violence against women, seen as violence exercised by male perpetrators towards female victims, “not only reflects unequal gender power relations in our society, but also forms a formidable barrier to efforts to overcome inequality between women and

8.

This definition is literally quoted in the article 3 of the decision to establish the Daphne Program.

242

Politics Beyond the State.book Page 243 Wednesday, December 5, 2007 8:41 AM

EU Policies on Violence against Women

men” (EP resolution A4-0250/1997)9. The fact that domestic violence is often hidden and surrounded by taboos causes low awareness on the issue. Another mechanism said to be reproducing the problem relates to the stereotypical ideas concerning women and attitudes in society that regard women’s bodies as commodities, as reflected for instance in pornography and prostitution. Also the lack of proper legal, social and economic instruments to protect victims are named as part of the problem. As a part of the campaign, a Eurobarometer survey on violence against women was held and a brochure entitled ‘breaking the silence’ was published. The measures proposed to fight domestic violence are mainly located on the level of awareness raising, aiming to tackle the low awareness and taboos surrounding the problem. The brochure called upon male perpetrators to seek help, and upon victims and witnesses to break the silence. Moreover, the campaign also recognized that there was a need to enhance the protection of victims and witnesses, and to improve legislation, but the brochure itself was of course mainly an awareness raising instrument. By portraying women as survivors and not as victims, for example, it tried to contribute to the ‘empowerment’ of women. It can be said that during the years 1997 – 2000, there was a clearly gendered power analysis of the problem of violence against women. The causes were located in the imbalance of power between women and men. Domestic violence was seen as the ‘most graphic symptom of the imbalance of power in the relationship between women and men’ (Brochure ‘Breaking the Silence’, 2000). Solutions were not only sought in victim protection and empowerment of women, but also in offering help programs for male perpetrators to end their violent behavior. The attribution of roles in the problem analysis was dichotomous and gendered; explicitly referring to female victims and male perpetrators. The EU assigned itself an active role in combating violence against women, mainly through a mechanism of ‘awareness raising’. There is a strong reliance on soft policy instruments, and apart from an awareness raising campaign and the dissemination of information, this policy initiative was not accompanied by any legally binding measures or obligations for member states. At several conferences and meetings the issue was addressed, a brochure was published, and a Eurobarometer was organized. It is clear that the EU committed itself to fighting violence against women, 9.

Referring to violence against women as both the cause and the symptom of unequal power relations between women and men, could already be found in the 1986 report of the Committee of Women’s Rights, and is a characteristic of a feminist articulation of the problem.

243

Politics Beyond the State.book Page 244 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

although there are boundaries to this fight on the EU level. The criminal law aspects of the problem, e.g. punishment of perpetrators, fall within the authority of the member states, which also explains the strong reliance on ‘soft’ measures. The EU thus mainly takes a coordinating role, but is active with publicity and information. On paper, the analysis of the problem is thorough, with a strong emphasis on the unequal distribution of power between men and women as the main root of the problem. This considerably outspoken framing of the problem might have been facilitated by the fact that the EU did not assign a large budget to measures combating violence against women, and apart from the costs of the campaign and the Eurobarometer survey, no financial burdens had to be bared. In addition to this, the publishing details in the brochure state that the opinions expressed do not necessarily correspond with the official standpoint of the Commission (breaking the silence brochure, 2000).

The Daphne Program (2000-2004) With the introduction of the Daphne Program (2000), a shift took place from framing violence against women as a human rights issue to a health issue. Although reference was made to some human rights and UN documents, in the final ‘decision establishing the Daphne Program (293/ 2000/EC, adopted by the EP and Council) violence is not seen as problem because it reflects a violation of human rights, but because it forms ‘a major health scourge’ and involves ‘high social and economic costs’. The problem definition was also enlarged from addressing ‘violence against women’ to ‘violence against children, young persons and women.’ In spite of criticism by the Committee of the Regions that a distinction should be made between the way in which women are addressed and the way in which children are addressed, a gender-sensitive analysis of violence against women or a reference to structural causes of the problem was not integrated in the first nor the second phase of the program. The World Health Organisations (WHO) broad definition of health forms the basis for the definition of violence in the Daphne Program. Originally Daphne was supposed to have Article 23510 as its legal basis, but after discussion in a Council working group, Article 129, which deals with 10. Article 235 states that: “If action by the Community should prove necessary to attain, in the course of the operation of the common market, one of the objectives of the Community and this Treaty has not provided the necessary powers, the Council shall, acting unanimously on a proposal from the Commission and after consulting the European Parliament, take the appropriate measures.”

244

Politics Beyond the State.book Page 245 Wednesday, December 5, 2007 8:41 AM

EU Policies on Violence against Women

Public Health. Many MEP’s were afraid the Article 129 alternative would be too narrow for addressing the issue, arguing that human rights issues should not be reduced to health issues. In the end the change to public health was accepted (1999, parliamentary sitting of 8 March) because of the danger that the Program would be blocked in case it would rely on article 235 (EP, 1999, sitting of march 8th). The aims of the Daphne program were targeted at the protection of victims and the prevention of violence. These goals were to be reached through awareness raising and research. All measures were aimed at victims. Perpetrators were not mentioned, neither in the problem analysis, nor in the adopted decisions, although the Committee of the Regions and the Social and Economic Committee underscored the need to pay attention to perpetrators, in the form of preventive measures and punishment or help programs. The Daphne program contains no real diagnosis of the problem. The emphasis lies very much on the symptoms, taking the existence of violence as a given. Also the emphasis lies on the health implications for the victims of violence, and not on e.g. the criminal aspects of it. The existence of violence and the fact that there are victims of this violence is framed as the problem. Women are, among other ‘vulnerable’ groups in society, seen as the problem holders. Men or persons committing violence are absent altogether. The feeble diagnosis might explain the strong focus on the victims of violence, and the lack of attention for the perpetrators. It is nonetheless a contradiction that the goal of the program is ‘prevention’ while perpetrators are left out of the picture. Unlike the ‘Zero Tolerance’ campaign, no references are made to inequality or to an unequal distribution of power between men and women as an underlying cause of the problem. The causes of violence are left unaddressed. No analysis of the problem is provided, and no causal relations or mechanisms are mentioned. The framing of the problem as a health issue, in combination with the extended problem definition including children and young persons, appears to lead to a ‘de-gendered’ approach of the problem. The Daphne Program identifies many forms of violence against women, but does not provide a separate problem analysis on this topic. It tends to put more emphasis on other forms of violence such as sexual exploitation and trafficking. The EU is addressed as a responsible actor only to the extent that the Commission should install the framework program and make available some funding. Hence, the EU is only responsible for the provision of resources, the work itself is to be executed by NGO’s, national or regional governments. The coordinating role of the EU, mainly providing funding for other actors to combat violence against women, is also a clear example 245

Politics Beyond the State.book Page 246 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

of EU governance. The Daphne program merely focuses on providing the framework and the funding for NGO’s, private and public actors, who should combat the problem of violence against women. The adoption process of the Daphne program shows how soft instruments can avoid deadlock or veto. As mentioned, the program was based on public health, as there was a real threat of veto from some member states (Kantola, 2004), who were reluctant to have a policy program on violence against women taken up into community law11. The non-binding and weaker implications of a program based on public health, helped to get the program adopted, something that remained uncertain if based on article 235 (Kriszan et.al., 2005). This shift to public health in the legal basis, coinciding with a broader problem articulation (with a large focus on children and many different forms of violence) led to a ‘de-gendered’ problem articulation. In contrast to the previous policy initiatives and resolutions, the Daphne program was accompanied by a (modest) budget, which might have also played a role in the de-gendered framing of the issue. Dropping the power analysis entirely, and bringing in children and young persons as target groups, made the loss of attention for women and issues of gender possible.

Recent initiatives 2005-2006 In 2005 an EP own initiative report on violence against women was produced by MEP Carlschamre. The document was a motion for a European Parliament resolution on the current situation in combating violence against women and on future action, and was adopted in the European Parliament at the end of 2005 (Final A6-0404/2005). This report stands in strong contrast with the simultaneously running Daphne program, but followed the line of argumentation present in the Zero tolerance campaign. It ventilated a strongly gendered discourse, with a recognisable power analysis, referring to violence against women as a universal problem rooted in the unequal distribution of power between women and men.

11. The member states opposing were Germany, Britain, France, Denmark and Belgium. The legal base for the program would have made the program a law. Article 129 was suggested as the alternative. The more narrow powers of the Article 129 were accepted but many female MEPs and the Committee on Women’s Rights were disappointed.

246

Politics Beyond the State.book Page 247 Wednesday, December 5, 2007 8:41 AM

EU Policies on Violence against Women

In this report, the problem is rearticulated as violence against women, with an emphasis on domestic violence, or violence in the intimate sphere, thereby also including forms of violence like genital mutilation of girls or honour killing. Violence against women is regarded as a violation of human rights, reflecting unequal power relations. The unequal power distribution between women and men is seen as the root of all violence against women, and also as an obstacle to equality and democracy. The report almost literally copies the articulation as found in the Zero Tolerance campaign, which itself was based on the 1986 report by the EP Committee of Women’s Rights. The report however also includes some new elements in the form of explicit attention for female genital mutilation and honour killing, explicitly rejecting any cultural legitimation of violence against women. Male perpetrators are deemed responsible for causing the problem. Female victims and children are the problem holders, as they are at the receiving end of the stick, but also perpetrators are seen as problem holders and in need of treatment and punishment. The main goals are penalisation and legislation. Preventive strategies and proportionate penalties for perpetrators and victim protection are considered equally important. A huge emphasis is put on the coordination and the development of a methodology to gather comparable statistics and data on the occurrence of violence against women across member states. The problem diagnosis and attribution of roles is explicitly gendered. Men are seen as perpetrators, women as victims, only ‘children’ are not treated as gendered. Also society as a whole is seen as a problem holder because of the social costs of violence for this society. As mentioned, the central cause of the problem is inequality between men and women. It is striking that the problem diagnosis is so explicitly gendered, as the articulation of gender is completely absent in the Daphne program, which was up and running during the adoption of the report. Another striking element is that the problem diagnosis is rather elaborated and shows many resemblances to gendered power articulations of the problem. There are several mechanisms mentioned to be (re)producing the problem, one of the most important being the public-private dichotomy12 which is said to have a continuing influence on perceptions of state and of non-state responsibilities.

12. Culturally and historically rooted norms and ideas about the roles of the family, society and the state continue to exist and reinforce this dichotomy. This can be seen for example by the way in which the state still assigns large budgets to security in the public sphere (e.g. terrorism) and not in the private sphere.

247

Politics Beyond the State.book Page 248 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

The report presents a long list of different responsibilities, policy instruments and policy goals. Despite it being an EU initiative, most policy goals and measures to be introduced are aimed at member state level, and formulated mainly as a member state responsibility. However, no sanctions or legally binding measures are set in place to ensure observance of these member state responsibilities. The report does point out that the treatment of women should also be part of the accession criteria for candidate member states. It can be said that the EU does, on paper, accept a more active stance in combating violence than it did in the Daphne program, on the other hand however, the implementation is almost entirely left to member states and other actors. Also, there is an absence of strong legally binding goals for member states and most goals are formulated along the lines of what ‘should and could be done’ in theory, rather than what ‘has to be done’ in practice. So the rise of a gendered articulation of the problem of violence against women, the sometimes highly politicized and at points far-reaching wording of the report, does not come with equally strong measures, sanctions and a matching high budget.

Conclusions Looking at the evolution of the issue of violence against women in the EU over the last two decades, we can notice large shifts in its articulation as a policy problem. At first sight, the whole policy frame seems to be susceptible to change and inconsistency. Within short periods of time, the representation of the problem of violence against women seems to undergo major changes. Different ideas about the underlying problems and the link with gender can be found. In the first articulations of violence against women as a policy problem (1986), the zero tolerance campaign (1997), and the latest initiative (2005), one can clearly distinguish a gender power analysis of the problem, with gender inequality as the major underlying problem. The Daphne program stands in strong contrast with this. The absence of a (gendered) diagnosis of the problem and the broadening of the problem to other issues are at some points contradictory to the latest report on violence against women. Taking all the programs together, it seems that the initiatives offering the most binding measures are also the least farreaching in terms of their problem analyses and their proposed measures. This could lead to the impression that gender equality policies clash with dominant policy discourses, or at least, are not given priority. Even though this might be true, the present overview of EU policies on violence against women shows that a soft governance approach contains policy potential for complex and contested policy issues like gender equality. 248

Politics Beyond the State.book Page 249 Wednesday, December 5, 2007 8:41 AM

EU Policies on Violence against Women

The absence of strong, legally binding measures and the reliance on soft measures and voluntary action allows actors to frame and reframe an issue more freely, without being bound to the prevailing dominant common denominator. A governance approach allows for undertaking action simply because it is not binding. Actors who want to act can do so, as is evidenced by the Carlshamre report. They can even take a very different stance than what is embedded in the official discourse of the Daphne program. A governance approach leaves the possibility to go further than the common denominator, which is determined by the lowest standards, one of the reasons why Nordic and Scandinavian feminists tend to be hesitant on what comes from Brussels. Binding measures on violence against women, to which all governments agree, will definitely lead to a very weak compromise, if not, a vetoed text. This threat was witnessed in the first discussions on the Daphne Program. In this sense non-binding measures are a good alternative. They leave the possibility to articulate a given problem in a variety of ways and to determine several even contradicting objectives. In a multi-level setting such as is the EU, such frames can nonetheless serve as an example to follow, even in member states that would not be willing to agree on binding measures. What comes from Brussels can have a certain attractive and persuasive aura, as show experiences from the Southern and new member states (Verloo and Pantelidou, 2005). Although many documents might not be binding, they can provide guidelines and examples for aspiring member states, even if it is only integrated into their gender equality policies to ‘make a good impression’. Finally, the recurrent appearance of an issue on the political agenda and especially its particular framing might in the end lead to an erosion of opposition to it. EU gender equality policies have regularly been threatened over the last decade, which is very well illustrated by the case of the general action programs on gender equality policies, but in the end they are still part of EU policies. The recurrent appearance of the gendered human rights frame in the field of violence against women, and its underlying power analysis of gender relations, is a good illustration of the fact that governance – at least in this particular case – allows for the continued reemergence of the issue on the policy agenda. Through this recurrence the frame might in the long run be (partly) integrated into the main way of framing this policy issue and also lead to more substantial policy measures. It is true that this frame is on the agenda since two decades, but it is still there. It was not simply replaced by the public health discourse.

249

Politics Beyond the State.book Page 250 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

To come back to the potential of a governance approach in complex policy areas like gender equality, it seems fair to conclude that, next to the various shortcomings or contradictions found in the policies scrutinized, it does offer possibilities and openings for the articulation of gendered issues from a gender equality perspective. Although not all policies in the field are equally thorough, clear-cut or efficient, and the development of the issue is at points haphazard and inconsistent, the sheer possibility of the existence of two completely different articulations of the same policy problem at the same time is hopeful from a gender equality perspective. It could constitute a fertile soil for further policy development.

250

Politics Beyond the State.book Page 251 Wednesday, December 5, 2007 8:41 AM

Multi-level Governance at Work: The Case of the European Integrated Product Policy (IPP) Isabelle Bédoyan, Theo Jans, Irina Tanasescu

Public-private cooperation is an essential dimension to the governance approach. In fact, without the public-private dimension the concept of “governance” itself cannot be distinguished from “government” or from other complex systems of intergovernmental relations in federal systems. The formats of public-private cooperation may vary substantially (self-, coregulation, implementation) but it always entails a significant private actor involvement in the formulation and execution of public policy (Héritier, 2001). In practice, governance entails public authorities outsourcing certain public functions to private actors in one form or another. The “governance” approach can also be associated with specific policy instruments, which differ from coercive instruments or prescriptive and enforceable legislation (normally associated with “government”). The “soft” policy instruments associated with governance can (a) be voluntary or non binding, (b) have no clear sanctions in case of non-compliance, (c) have no specified procedure, or (d) have no clear goals. The soft policy instruments will lack at least one and often a combination of the features of classic regulation and, instead, will provide room for private actor involvement in different phases of policy formulation and implementation. The prescriptive and coercive nature of the instrument is thus replaced by stakeholder involvement, while compliance is sought through direct private actor involvement in the policy formulation rather than through imposition and sanctioning. The reasons for the shift from traditional forms of legislation to soft instruments are multiple and well detailed in the literature (Collier, 1998; Scott&Trubek, 2002; Beetham&Lord, 2001), but the most important one seems to be the fact that stakeholders provide both legitimacy and information, both much needed assets in multi-level political systems. Soft instruments can constitute a solution to the problem of incomplete information that public authorities face, while reducing conflict at the same time. An illustration of the features and dynamics of governance can be found in the EU’s Integrated Product Policy (IPP). IPP as conceived by the European 251

Politics Beyond the State.book Page 252 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Commission covers in fact many features associated with the governance approach. The policy requires the mobilization of resources, capabilities and prerogatives at different levels of government. It is heavily dependent on information, expertise and eventually on the cooperation of private actors. Both the scope of the policy and the administrative and legislative constrains (length and veto-sensitivity of the EU legislative process) push the European Commission to apply existing voluntary instruments to implement IPP. Nevertheless, the current track record of such instruments shows that they cannot work in the absence of accompanying measures being put in place by national or regional public authorities. What this reliance on voluntary instruments and on stakeholder involvement means for an EU-level public policy such as the IPP will be the focus of this chapter. The conclusions can be relevant for other policy areas and point to possible failures in the overall architecture of governance in general. This chapter will develop as follows: the first part of the chapter presents the concept of IPP as seen and interpreted by academics and policymakers at national and European levels. Following the conceptual discussion, the policy events and steps leading up to the IPP Communication (CEC 2003a) are briefly described together with the content of the policy. IPP entails a complex and comprehensive implementation process. The policy measures developed so far are presented and discussed, followed by a brief overview of the voluntary instruments mentioned in the IPP Communication as possible implementation tools for IPP. The chapter concludes with an assessment of the Commission implementation strategy and its heavy reliance on voluntary and market-based instruments. In addition to using official documents and the relevant academic literature, we also base our assessments on the results of a series of interviews carried out between April and June 2005 with EU, national and regional policy-makers and stakeholders involved in IPP and in the different voluntary instruments.

A product-oriented environmental policy The first environmental policies focused on emissions polluting particular media such as air or water and on the handling of waste. The object of these approaches were the production and consumption processes that caused these emissions. Policies consisted of command and control instruments and detailed technical provisions concerning the process (Oosterhuis et al., 1996). Such approaches did not include life cycle thinking therefore environmental impacts could be moved from one phase of the life cycle to another. The possible environmental impacts on different media or on 252

Politics Beyond the State.book Page 253 Wednesday, December 5, 2007 8:41 AM

Multi-level Governance at Work: The Case of the European Integrated Product Policy (IPP)

different phases of the life cycle were not scrutinized or evaluated. Policies were developed from a top down perspective and the emphasis was on direct governmental action to address pollution in a particular medium or at a specific phase. Product-oriented environmental policy changed the policy focus from processes to products: considering the product along its entire life cycle rather than at only one phase (for example, production or waste). This comprehensive approach provides a vertical integration of issues, incorporating the input and output streams of each phase in the product’s life cycle. It also provides a horizontal integration of issues, dealing with the interaction between different environmental problems related to a product. Therefore, the focus is not on one substance or issue but on a system of issues and groups of substances. Such an inclusive approach entails the active involvement of more actors than in the process approach, and calls for a strong cooperation between public and private partners. Finally, the policy is designed to be proactive, anticipating future problems rather than waiting for problems to materialize (Oosterhuis et al., 1996). The introduction of an integrated life cycle approach and the inclusion of a wide variety of actors are the defining characteristics of product-oriented environmental policy and constitute the foundations of the Integrated Product Policy concept. Several definitions of IPP have been developed in the literature. Berkhout and Smith (1999) for instance define IPP as “Public policy that explicitly aims to modify and improve the environmental performance of product systems.” The authors further qualify IPP as a systematic concern for the whole life cycle of the product. Measures to implement IPP should focus on waste, innovation, information exchange along the product chain, and on the allocation of responsibilities. Finally, the measures should be developed to be compatible and even reinforce each other thus achieving true integration. Tufet-Opi (2002) underlines the sustainable development dimension of IPP, describing it as a ‘hybrid policy’ with elements from industry, consumer protection and environmental policy. In this context, the setting of sustainable development goals integrating environmental and socio-economic targets is central to the policy. However, to make the policy work, long term sustainable development goals should be complemented by concrete objectives developed in cooperation with stakeholders. Concrete objectives must then be used as a yardstick against which the effectiveness of the policy can be measured. To define policy choices (identification and prioritization of policies), the use of Life Cycle Assessment conducted on particular products or product groups is

253

Politics Beyond the State.book Page 254 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

suggested. For the implementation of the IPP objectives and goals voluntary instruments must be favored, as the author does not believe any additional environmental benefits can be obtained by further environmental legislation at EU level. Finally, in order to succeed, IPP must be developed with respect for market forces and all stakeholders should have a clear idea of the concrete aims of the IPP policy. Charter and Belmane (1999) provide yet another definition of IPP: “Integrated Product Policy is a government policy toolbox aimed at greening markets that incorporates tools to green consumption (demandside) and tools to green product development (supply-side). […] It is a policy concept that takes a lifecycle perspective (cradle to grave), includes all relevant stakeholder viewpoints and considers the product development process from idea generation to product management and reverse logistic.” This description distinguishes itself from the others through the explicit inclusion of the consumption aspect and the design phase of the product life cycle. Moreover, it describes IPP as a toolbox for government policy. These definitions all share a series of common elements. The life cycle perspective of the policy is the most obvious one, followed by the involvement of stakeholders at all phases of the development of IPP. The different definitions call for the involvement of stakeholders and the incorporation of their views concerning the development, implementation and target setting activities. Stakeholder involvement can range from consultation to the promotion of voluntary instruments. The market element, present in most of the definitions, can vary from an acknowledgement of the market forces to the creation of a green market. Other elements, such as the integration of services in addition to products, or the incorporation of the three pillars of sustainable development (environmental, economic and social concerns), may vary from definition to definition. The scope of IPP itself can vary from a fully developed public policy to a mere toolbox of instruments. The academic definitions of IPP reflect the variety of defining elements that can be used as well as the different weights that can be attributed to these defining elements. Even a basic feature such as the goal of IPP seems to vary from the improvement of environmental performances to the greening of markets. The different ways in which defining elements can be selected, highlighted and combined will ultimately impact on the scope of the Integrated Product Policy. It can be conceived as an inspirational policy concept, as a toolbox of policy instruments or it can be interpreted as a fully fledged and comprehensive policy. The conceptual ambiguity and vagueness of the concept is particularly ironic as the need for a clear 254

Politics Beyond the State.book Page 255 Wednesday, December 5, 2007 8:41 AM

Multi-level Governance at Work: The Case of the European Integrated Product Policy (IPP)

definition and a uniform understanding of IPP by all stakeholders has been raised by several sources and is considered to be one of the main stumbling blocks for building a successful policy (Tufet-Opi, 2002). In addition to the academic definitions, national experiences with IPP can also further our understanding of the policy concept. National and regional IPP policies, especially in the Nordic countries, were developed before, or even in parallel with, the European IPP. A Nordic IPP group, involving Denmark, Finland, Norway, Sweden and Iceland, was created by the Nordic Council and its activities started in 1997 with the first Nordic IPP workshop. The Nordic group defines IPP as a ‘product-oriented environmental strategy’ whose key elements are the involvement of market forces and the need for dialogue and cooperation. Public authorities must not only set limits concerning environmental impacts but must also function as the ‘catalyst’ of a market driven process. IPP is seen as a complement to existing legislation. It is a commitment of government and stakeholders to achieve better environmental performances than those legally required. The government has a clear role as a facilitator: setting the framework, motivating the market, and offering credible tools to the different actors to improve environmental performances (Nordic Council of Ministers, 2000:15). A clear choice is made to concentrate on the environmental properties of products in cooperation with market forces and stakeholders. In line with this supportive approach, governmental incentives for voluntary instruments must provide results that go beyond the legal minima. In Sweden, IPP activities started as early as 1993 and are characterized by a clear market dimension. The Swedish IPP is intended to provide better conditions for (green) market actors by laying down clear rules to build a level playing field, by increasing the supply and demand of green products and by the internalization of environmental costs using for example tools like taxation. Moreover, the need for stakeholder involvement and cooperation between government, industry, NGO’s and the scientific community is stressed (e.g. use of product panels, voluntary agreements and dialogue groups). Emphasis is put on improving the information flow regarding the environmental impacts of products throughout their lifecycle: making this information available to all the stakeholders concerned, making public life cycle data more accessible, coordinating different type of information tools (e.g. eco-labels and environmental product declarations), and harmonizing the different eco labels (Swedish Environmental Protection agency, 2003).. The Swedish IPP is based on the idea that the government has to establish the minimum criteria on which

255

Politics Beyond the State.book Page 256 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

market driven incentives can build to achieve levels of environmental protection which go beyond these minima. As a consequence, Sweden has concentrated on developing a legal basis for the extension of producer responsibility, to be complemented by other instruments such as green taxation, subsidies and public procurement. Voluntary instruments are also suggested to further develop IPP and explicit reference is made to Ecomanagement and Audit system (EMAS), eco-labeling, environmental declaration and eco-design. A second Nordic country, Denmark, started the development of a national IPP in 1996. The Danish definition emphasizes the use of market forces. It aims to include environmental properties of products in the competition on the market (Danish Environmental protection Agency, 2002). The policy provides incentives for the supply and demand of cleaner products and the strengthening of enterprises willing to produce these products. Helping businesses to improve the environmental performances of their products, and to use and promote their environmental profile on the market. With this strong market orientation, the Danish IPP policy shows a clear interest in preparing their businesses for a future market in need of cleaner products, thus combining environmental and trade objective (Danish Environmental Protection Agency, 1998, 13). IPP is considered to be “a market based policy: a framework and incentives for voluntary initiatives building upon a solid foundation of legislation defining the minimum acceptable market performance.” (Kristensen, 2004). This definition states the need to combine legislation and voluntary instruments to achieve a successful IPP. The main implementation instrument in the Danish IPP is product panels. Product panels join all stakeholders involved in the lifecycle of a specific product or product group (industry, retailers, consumers, NGOs, etc.) in a joint effort to improve and reduce the environmental impact of a product or product group throughout its life cycle (from production to waste) (Knudsen, Jensen and Chabert, 2003). Such panels have been set up for different sectors ranging from agriculture to electronics, textiles, retail and transportation. Other voluntary instruments which focused on information and product design such as Life Cycle Analysis (LCA), Environmental product declaration, eco-labels and environmental design were also used to improve the environmental performance of products (Danish Environmental Protection Agency, 2005). Stakeholder involvement has been ensured in these implementation tools through participation in the respective Boards (Eco-labeling Board, LCA central Board), working groups and advisory groups. In addition to these voluntary instruments, legislative instruments such as legislation of green public procurement and taxation were introduced.

256

Politics Beyond the State.book Page 257 Wednesday, December 5, 2007 8:41 AM

Multi-level Governance at Work: The Case of the European Integrated Product Policy (IPP)

Even though each of the above-mentioned national approaches emphasized different elements, the involvement of stakeholder, the market perspective and the use of voluntary instruments seemed to be recurrent features in the national IPP approaches. These national initiatives were a source of inspiration for the development of the European IPP and the choices that had to be made there. In addition, existing diversity of IPP approaches was considered to be a strong argument for the development of a coordinating policy at the European level.

The EU-level IPP scheme Two periods must be distinguished for the analysis of the European IPP. The first period covers the development of the IPP concept and the consultation process leading to the European Commission’s Communication of June 2003 (CEC 2003a). The second period can be qualified as the implementation phase and covers the different initiatives undertaken by the Commission to implement the policy after the adoption of the IPP communication. In retracing the evolution of IPP at the EU level, references to it can be found as early as the 5th Environmental Action Plan (EAP) of 1992 and the 5th EAP Progress Report from 1996. 1996 was also the year the European Commission ordered its first study on the topic, carried out by Ernst & Young and the University of Sussex between 1996 and 1998. In this study, IPP is seen as a public policy with an explicit aim to improve the environmental performance of product systems. The main building blocks of the policy are considered to be waste management, creation of markets, green product innovation, allocation of responsibility and the transmission of environmental information (Ernst and Young, 1998:11). This first initiative towards the development of a European IPP policy concept started a long period of stakeholder consultations, which eventually led to the Commission’s Green Paper. Between the first study and the Green Paper, IPP was also discussed by the Council of ministers during a 1999 Informal Meeting of EU Environmental Ministers in Weimar under the German presidency. In the conclusions of this meeting the importance of a Community approach to IPP and the need for a life cycle approach towards products and their environmental impact to achieve sustainable development in Europe were recognized (Federal Ministry, 1999). 257

Politics Beyond the State.book Page 258 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

In February 2001 the next important step was taken with the publication of the Commission’s Green Paper on IPP (CEC 2001d). A wide array of instruments were suggested: differentiated taxation, environmental labeling, green public procurement, Life Cycle Analysis, eco-design, standardization, product panels, Environmental Management and Audit Systems (EMAS).The paper suggested command-and-control instruments as well as voluntary instruments but emphasis was placed on the latter. In this document, IPP is defined as “an approach which seeks to reduce the life cycle environmental impacts of products from the mining of raw materials to production, distribution, use, and waste management” (CEC 2001d: p.5). This approach should also be reflected in the decisions of the stakeholders. Product design, consumer choice and the pricing of products are some of the important decision points (pinpointed in the policy paper) where environmental improvement of products can occur. The introduction of eco-design, informed consumer choice and the integration of pollution cost in the pricing are suggested in combination with instruments focusing on the life cycle as a whole. Finally, in June 2003, the Commission published its Communication on IPP (CEC 2003a), which contained the first steps towards the implementation of IPP. In this document, no definition of IPP is given, rather a collection of guiding principles and building blocks for IPP (CEC 2003a: p.5): Life cycle thinking: “[considering] a product’s life-cycle and aims for a reduction of its cumulative environmental impacts”. Working with the market: “setting incentives so that the market moves in a more sustainable direction by encouraging the supply and demand of greener products.” Stakeholder involvement: “[IPP] aims to encourage all those who come into contact with the product (i.e. industry, consumers and government) to act on their sphere of influence and to encourage co-operation between the different stakeholders.” Continious improvement: “IPP aims for a continuous improvement in these rather than setting a precise threshold to be attained.” A variety of instruments: “…[ranging from] voluntary initiatives to regulations and from the local to the international scale. Within IPP, the tendency is clearly to work with voluntary approaches, although mandatory measures might also be required. The determining factor is the effectiveness of the tool to achieve the desired result with regard to sustainable development.”

258

Politics Beyond the State.book Page 259 Wednesday, December 5, 2007 8:41 AM

Multi-level Governance at Work: The Case of the European Integrated Product Policy (IPP)

Life-cycle thinking, as the overarching concept of IPP, is meant to prevent burden shifting from one phase to the other. However, this does not mean that all phases will or should be actively addressed as the 2003 Communication foresees to concentrate its IPP activities to the phases where its policy resources can be most efficiently used. This pragmatic approach contrasts with the emphasis in the Green paper on the life cycle and the integration of environmental impacts at each phase of the life cycle (CEC 2001d: p. 5). “Involvement of all stakeholders at all possible levels of action”, as stated in the Green Paper, is also reflected in the policy making phase. All Communications were accompanied by consultation rounds and stakeholder meetings cf. IPP workshop after the E&Y study in 1998, a written consultation round in 2001 combined with a stakeholder meeting kick starting the public debate on the Green Paper (March 2001) and seven expert workshop on the different instrument (May to June 2001). Stakeholders are the pivot of the policy and the role public authorities is that of a facilitator. The public authorities are there to enable stakeholders to contribute and to facilitate the effective use of their input. The portrayal of public authorities as facilitators is explicitly stated in the Communication and is reflected in the strong emphasis on voluntary instruments like eco-labeling, EMAS, or standardization and in the market driven character of the policy. The initiatives announced in the Communication of 2003 can be divided into initiatives related to co-ordination and integration efforts on the one hand, and specific product-oriented initiatives on the other hand. Two important projects were started in 2004 by the European Commission concerning product-oriented initiatives. The first project is a two-phase study designed to identify the products with the greatest potential for environmental improvement. The first phase, launched in January 2004, is the development of a methodology to identify the products with the greatest environmental impact. During the methodology development the different stakeholders were consulted and two scientific expert workshops were held in May and September 2004. The intermediate results were discussed during the Regular Meeting of September 2004. A first draft of the final report was published in May 2005 and formed the basis for consultation activities (May and June 2005) and for an expert stakeholder meeting in July 2005. The final report was completed and published in November 2005. The second phase will build on the results of the first and aims to identify the products with the greatest potential for environmental improvement by 2007. 259

Politics Beyond the State.book Page 260 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

The second product-oriented initiative is the establishment of Voluntary Pilot Projects. This initiative was announced in the Commission’s Communication and a comparable tool was suggested in the Green Paper. In the Danish IPP a similar initiative was used under the label ‘products panels’. The Danish product panels were developed by the Danish EPA as a forum, gathering actors from different parts of the product life cycle (producers, consumers, retailers etc.), to incorporate environmental considerations in the product development process. The Danish EPA financed the secretariat of the panels. Its representatives, however, participated to the panels on an equal footing with the other members. To select the participant for the panels, the Danish EPA focused on frontrunners and motivated individual, rather than branch organisations (Traberg, 2003). The panels could largely determine their own framework and design their own action plan (Knudsen et al, 2003). The discussions within the product panel resulted in action plans, which got an extra incentive as they could solicit subsidies for their implementation via the Danish EPA’s program for cleaner products. The European voluntary pilot projects, launched by the European Commission in 2004, have a similar goal as the Danish product panels: to evaluate the environmental impacts of a specific product and design possible tactics for improvement through deliberation amongst the different stakeholders. These deliberations should ideally materialize in an action plan. The Commission started the initiative through a call for projects and two pilot projects were selected out of 22 stakeholder’s applications. The selection was based on their coverage of the supply chain and on the familiarity of the products to the wider public (CEC, 2004a). For the purpose of each pilot project, a small group of stakeholders were selected for direct participation. These groups were selected by the Commission and the project initiators: Nokia (project on mobile phones) and Carrefour (project on teak garden furniture). Although these small groups make up the core of the project a wider stakeholder consultation is foreseen after each important step in the process. The pilot project on mobile phone largely consists of representatives of the major ICT industrial community (phone manufacturers, component manufacturers and network operators) These actors are complemented by several EC officials, consumer’s and environmental NGOs (BEUC, WWF), and representatives of government agencies (DEFRA and SYKE). A similar mix of actors can also be found in the pilot project on teak garden furniture. The Commission chairs both projects and adopts the role of facilitator. However, the Commission does not finance the participation of the stakeholders. The strong stakeholder 260

Politics Beyond the State.book Page 261 Wednesday, December 5, 2007 8:41 AM

Multi-level Governance at Work: The Case of the European Integrated Product Policy (IPP)

dimension in the pilot project reflects both the Danish product panels and the basic principles of IPP. The voluntary pilot projects developed at EU level could bring clarification to the concept of IPP for the general public and the stakeholders concerned. In addition, these projects can help overcome skepticism regarding the viability of IPP by demonstrating how it can work in practice. Moreover, the experience with the voluntary pilot projects could lead to further use of product panels as an addition to the IPP toolkit. These product panels could then serve as a means to integrate the different instruments in the toolkit. As an initiative to strengthen the co-ordination and integration Regular Meetings were set up where member states of both the EU and the EEA and stakeholders could meet. This forum, organized twice a year, monitors and promotes the implementation of the Commission’s Communication on IPP. The members states’ representatives to the forum are typically from a national Environmental Ministry or Agency, while the stakeholder community is represented by both public interest groups, such as the EEB and BEUC, and by industry associations like UNICE, Eurocommerce and UEAPME. The Regular Meetings provide a fixed framework in which stakeholder consultation can be conducted. The first Regular Meeting was held in February 2004 and was followed by meetings in September 2004, March 2005, November 2005 and June 2006. During the first meetings, the decision was taken to set up two specific working groups in addition to the regular meetings. The first working group was to concentrate on one of the crucial areas of IPP, namely the product information needs. The second working group was given the task of defining the format for the IPP report. Subsequent meetings have been used to monitor IPP activities like the working groups, the IPP pilot projects and the studies on IPP indicators. In addition to this, the Regular Meetings have been functioning as a forum to discuss IPP related policies (ETAP and EuP), instruments of the IPP toolbox (e.g. EMAS and Ecolabels) and IPP implementation efforts at national level.

Voluntary instruments The mix of instruments that the European Commission suggested as possible means to implement IPP is based on a series of governance mechanisms. Many voluntary instruments have already been developed with respect to EU environmental or product policy and are now being 261

Politics Beyond the State.book Page 262 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

reused within the new IPP framework. It is the case, for instance, of the EU Eco-label, New Approach standardization, the European Eco-Management and Audit Scheme (EMAS), or Voluntary Environmental Agreements (VEAs). Given that the first two have been analyzed in detail in chapters by Bédoyan and Tanasescu of this volume, we will now focus on the last two1. EMAS uptake offers a mixed picture across the EU, with wide difference among Member States and among types of industry/organizations. Nevertheless, one thing seems to be clear and that is the fact that its success, or lack thereof, depends mostly on the attitude of national or regional public authorities, on their proactiveness and on their willingness to provide regulatory relief for participating companies. Within IPP, EMAS is a potentially useful instrument, as it focuses on production processes and management both in companies and organizations, thus virtually covering the entire life-cycle of a product. Nevertheless, a clear link should be made between the EU-level IPP and a nationally/regionally-applied EMAS, as this integration is not obvious or easy to make, at least for the public authorities we interviewed (interview EMAS Belgium, April 2005). VEAs have clearly had a longer tradition in national contexts (317 VEAs in the EU countries in 1997) than at the EU level (about a dozen), but their efficacy and legitimacy has been contested in both contexts. At the level of the EU, both NGOs and the European Parliament have been extremely critical of the use of this instrument by the Commission for several reasons. It is argued that there is little evidence that environmental agreements are effective, that the existing control systems are based on self-monitoring and hence subjective and that negotiated agreements (when industry commits itself to a certain environmental performance while the government in return agrees to refrain from direct regulation in this issue area while the agreement is in effect) severely restrict the range of policy options for future governments. Besides, these agreements are often concluded in the absence of parliamentary control and without the participation of the public in the negotiations. Industry on the other hand, wants the agreements to retain as much flexibility as possible and is not too happy about publicizing negotiations and results. The lack of a clear and workable set of criteria for using and assessing VEAs make this instrument even more subject to criticism and, therefore, not an uncontested choice for policymakers.

1.

For a more detailed discussion of voluntary instruments in the context of IPP see Pallemaerts et al. (2006).

262

Politics Beyond the State.book Page 263 Wednesday, December 5, 2007 8:41 AM

Multi-level Governance at Work: The Case of the European Integrated Product Policy (IPP)

IPP – a failed experiment? Integrated product policy has been long in the making. The earliest references to IPP can be found in EU policy documents in the mid-nineties. The preparatory phases of policy formulation have been long and encompassing. Technical expertise, stakeholder and member state views were included in these preparatory phases. The consultation covered broad public hearings, expert workshops, oral and written submissions and a sustained dialogue between public authorities and interested parties. Nevertheless, this extensive phase of policy formulation that came to a temporary close with the publication of the Communication on IPP offered little precision on the problems that are targeted by IPP. The policy problem is thus a very broad and comprehensive phenomenon that lacks borders or delimitations. Policy problems with such a massive and broad scope come across as hard to manage and are unlikely to mobilize the necessary attention, resources, and support of policy entrepreneurs. The problem that IPP appears to tackle will need to be translated into directly identifiable and clear problems of a more manageable scale if the IPP is to draw and maintain the attention of policymakers. The intended policy outcome is no less than a market transformation triggered by a mentality change of producers and consumers. The Communication does not set any quantifiable targets instead it proposes an effort to continuously improve the environmental performance of products. This approach encourages companies to set their own pace and to focus on their own most cost efficient improvements. To attain such a process of continuous improvement, the need for incentives towards the producers and consumers to produce and buy greener products is acknowledged. In addition to certain public incentives the policy is premised on the assumption that forerunners in the market will lead a transformation that will eventually encompass the entire market. The analysis of target groups probes the question to whom the policy is directed. Policy documents point at three target groups: (1) producers (2) public authorities (3) consumers. The policy targets markets, and seeks to influence the key actors in the market, namely, suppliers, buyers and governments regulating markets. Although three groups are mentioned the emphasis clearly lies on producers. Producers are coined as the driving forces to achieve the reduced impact of products on the environment. The market transformation, which IPP seeks to achieve, needs to be carried out by producers. If producers are the policy’s main target, it will be necessary to base it on a set of incentives that clearly matter to and impact on producers. Policy documents offer little information on what these 263

Politics Beyond the State.book Page 264 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

incentives to alter producer behavior are likely to be. It is suggested that especially market leaders will be expected to vanguard the market transformation and subsequently reap the benefits of the transformed market. If producers are the main target, two conclusions can be drawn: (1) IPP policy instruments must directly impact on producers to achieve the desired behavioral change and market transformation, (2) there are no reasons or explanations provided in the policy documents why producers are expected to lead the transformation and why they are likely re-evaluate and change their products to reduce environmental impacts. On the demand side, consumers are mentioned as targets. Consumers should be convinced to buy greener product through the diffusion of better, more reliable and clear product information. Eco-labels stand out as a possible means to alter consumer behavior. Despite being targeted at consumers, each of the consumer-oriented measures shift the burden of adopting and delivering ‘greened’ products to the voluntary commitments of producers. Public authorities are mentioned as another group targeted by the policy. Policy documents clearly state that direct market intervention is unlikely to be the main role of public authorities. Governments are to facilitate the process (of continuous improvement). In practice, the facilitation has up until now entailed: (1) the organization of policy debates and stakeholder consultations, e.g. Green Paper (2) the development, the structuring, and promotion of information and techniques to be taken up by producers and other levels of government at their discretion, e.g., handbooks, databases, reports, regular meetings, etc. (3) reviewing and developing concrete formats for direct policy implementation, e.g., identifying products with greatest impact and greatest room for improvement, product panels, etc. Finally, public interest groups are targeted as partners to identify relevant issues and develop practical solution. Tasks relating to the promotion of IPP instruments, e.g. Eco-label or EMAS, the greening of standardization at national level, and educational and awareness-raising activities were suggested by the European Commission. The identified target groups and their respective roles present a division of labor in which producers are to drive and deliver the policy program. Public authorities are to generate a framework and the conditions for producers to come forward and to achieve the policy goals. Organizing persuasion and policy learning seems to be the main function of the European public authority in the IPP, rather than achievement of behavioral changes through constraining legislation or tax incentives. Confining the role of public authorities to that of a facilitator is not without 264

Politics Beyond the State.book Page 265 Wednesday, December 5, 2007 8:41 AM

Multi-level Governance at Work: The Case of the European Integrated Product Policy (IPP)

consequence. Member states facing regulatory competition in a common market and producer resistance to measures which could entail costs and competitive disadvantages, point to the EU as the relevant policy level for market correction. At the supranational level direct and coercive intervention is largely discarded as an adequate or viable policy option. Instead IPP posits facilitation as the main governmental responsibility and it singles out producers as the actors that must achieve and deliver the market correction. The shift from state to markets to establish collective decisions combined with the presence of different policy levels can easily give rise to extensive responsibility and burden shifting and eventually to governance failures. The analysis of the policy design draws the attention to the concrete measures and decisions that are being developed under the IPP label. The Communication specifies the actions, the budget and the instruments of IPP. Two types of actions can be discerned. First, the Commission has set up two pilot projects and has launched studies to determine the products with the greatest potential for environmental improvement. Furthermore a working group was set up to determine the information needs of different stakeholders and to develop means to compensate for possible information gaps. These initiatives can be qualified as information gathering efforts to identify feasible formats and appropriate objects for the implementation of IPP. Second, the Communication proposes a strengthening of the instruments included in the IPP toolbox: green public procurement, EMAS, LCA, standardization, taxes, product design and eco-labeling. If the toolkit of instruments is to be used for IPP purposes, it is crucial that the policy communities working on or with these instruments are adequately informed and knowledgeable on IPP. The interview results indicate that few policy actors involved in the instruments know or appreciate the relevance of IPP. Those policy makers who are informed on IPP interpret the policy approach in differing ways and there is little shared understanding of the meaning and nature of the policy. The policy instruments in the toolkit are information-based and/or voluntary instruments. Regulatory measures are mentioned in the policy documents and not entirely excluded but the emphasis is on voluntary instruments. These instruments are only viable when the interests of the policy makers and those of the producers coincide. There is little or no indication that such parallelism of interests exists, rather the contrary seems to be true. The Commission’s reliance on a voluntary approach puts private actors and producers at the centre phase of the policy (formulation and 265

Politics Beyond the State.book Page 266 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

implementation). The suggested instruments (eco-label, EMAS, VEA, standardisation) all rely heavily on the input and the commitment of private actors to achieve outcomes. The experience with EU-based voluntary instruments, as discussed earlier, does not seem to be very promising when it comes to private actor involvement and participation. The limited number of eco-label applications and voluntary environmental agreements indicates that industry is not very forthcoming and that an IPP policy based on these instruments is likely to suffer the same fate. The use of standardization as an instrument to achieve IPP goals points at a principal-agent problem. New Approach standardization worked well when it involved product safety standards, which have direct consumer and producer utility. The system proved much less successful when it came to setting environmental standards which entailed costs rather than benefits for the producers. In those cases where the interests of the agent or standardisation body do not match the underlying goals and interests of the public authority or principal, delegations will either fail entirely or trigger high monitoring and control costs. Moreover, principal-agent relations tend to be reversed when the principal has few or no other alternatives than to resort to the services and expertise of the agent. Information dependence or lack of administrative capacity on behalf of public authorities can create a situation where the agent steers and guides the principal rather than the other way around. The reliance on private standardization bodies to achieve collective goals may bring certain advantages (e.g. tapping into private resources at limited administrative expense) but also entails dangers of goal displacement and loss of public control. It should be highlighted that it is possible for public authorities to create effective incentives for producers to adopt voluntary instruments. The literature and our own findings point at the importance of flanking or supportive measures to stimulate the adoption of voluntary instruments by producers. Regulatory relief, tax breaks and public procurement stand out as effective means to provide producers with incentives that would otherwise be lacking if the markets and producers are left to their own devices. IPP documents suggest a toolkit of instruments but offer little as to how these different instruments are to be mixed. Since IPP can not be achieved through isolated instruments, a mix of instruments will be required. The actual integration of instruments into a coherent and effective instrument mix is not tackled by the relevant policy documents nor is it convincingly broached in the subsequent policy initiatives. IPP offers building blocks

266

Politics Beyond the State.book Page 267 Wednesday, December 5, 2007 8:41 AM

Multi-level Governance at Work: The Case of the European Integrated Product Policy (IPP)

but without construction plan and without “contractors” to build the policy. The failure to come clear on the instrument-mix is crucial omission in the policy design for it allows public authorities and producers to pickup on certain instruments without contributing to IPP goals. Presenting policy instruments without establishing how they are to contribute jointly to IPP invites policy-stretching. The policy concept is stretched to include such a vast scope of activities and instruments that the policy looses meaning, practical relevance, and its capacity to guide policy development. Product panels could constitute a possible alternative to the formulation of pre-defined policy mixes. Based on Scandinavian experiences the European Commission has put in place panels of stakeholders to develop an implementation plan to reduce the life cycle impact of a product on the environment. Stakeholders rather than public authorities determine which instruments from the toolkit can be applied for a given product or product group. Although such product panels could turn out to be a useful format to achieve the much needed integration of policy instruments, it is currently unclear what the precise status and prerogatives of the EU product panels are. The pilot projects can be interpreted as a feasibility study and a means to convince producers that IPP is a viable and worthwhile policy approach. The pilot projects can also be an experiment to test the option of turning stakeholders into effective policymakers that decide on the goals to be achieved and the instruments to be adopted for a given product. Both options have their downside. Product panels as decision-makers shift very significant amounts of power from public officials to private actors. Such delegations of power are typically subject to principal-agent problems. If product panels are merely to be interpreted as a feasibility study, the exercise may convince certain producers but may fail to achieve a coherent approach to reduce product impacts. The extensive and balanced consultation efforts in the initial and preparatory phases of the policy (before the Communication) allowed for ample and decisive input of public interest groups. The conditions for a systematic and meaningful involvement of public interest groups are less if at all present in the policy instruments. Many of the voluntary instruments (e.g. standardization, VEA’s, EMAS) do not constitute a level playing field. Whether for lack of resources, expertise or formalized prerogatives, public interest groups struggle to have a significant impact on the development and outcomes of voluntary measures. In view of the increased reliance on and the importance of measures to be crafted through voluntary instruments, it seems essential to not limit equitable participation to the formative phases of policymaking but to ensure even more forcefully during the implementation process. 267

Politics Beyond the State.book Page 268 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Conclusions Our analysis of IPP was largely inspired by the assumption that policies are sets of clear and formalised measures, programmed to achieve precise and pre-defined goals. When we analyse IPP from that classic perspective, the policy is found wanting on numerous accounts. First, the policy confronts the European Commission with rather extreme interdependencies. Carrying out the policy requires coordinated and coherent action within EU institutions (e.g. between DGs), between different levels of government (EU, national, regional), between governments and agencies, between public and private actors. The resulting interdependency renders imposition and hierarchical control difficult if not impossible. Mobilizing resources, awareness, support, capabilities, etc. among so many and diverse actors over which EU institutions have no direct or only partial control seems difficult to achieve with the resources made available for the policy. Second, IPP is characterised by a long process of policy formulation in which policy concepts, approaches and instruments were discussed at length by both public authorities and a broad range of stakeholders. The actual decision making should have occurred with the adoption of the Communication. However, the Communication included few decisions. The document provides orientations and suggestions but offers little in terms of concrete measures or decisions. In effect, actual decision making was shifted to the implementation phase. Essential decisions regarding the scope (which products), the format (how) and the expected outcomes (what) were to be determined during the implementation of the policy. Implementation was to shape the policy rather than the inverse. This implementation strategy embedded in the Communication and in subsequent policy measures clearly displays the characteristics of what Richard Matland (1995) coined as “experimental implementation”. The implementation process is less geared towards successful outcomes but seeks to trigger a learning process whereby policy measures are seen as natural experiments to enhance policy maker’s understanding of policy effects and formats. Policy ambiguity is seen as opportunity rather than as obstacle in this type of implementation. Attempts at policy implementation occur in varying contexts based on differing understandings of the policy and thus give rise to varied experiences from which policy makers can learn and achieve clarity on more viable and desirable policy measures. This approach does require significant monitoring, evaluation and feedback on the different attempts at implementation. Without the sustained scrutiny of policy experiences at different policy levels and in 268

Politics Beyond the State.book Page 269 Wednesday, December 5, 2007 8:41 AM

Multi-level Governance at Work: The Case of the European Integrated Product Policy (IPP)

different policy domains, policy learning is unlikely to ensue. Furthermore, as contextual conditions (domestic, policy domain) will largely determine the implementation process, outcomes will differ accordingly and show high levels of variation. Following this reasoning it seems likely that EU member states will forward their existing productrelated environmental policies as cases of IPP. This would indeed lead to highly context-dependent outcomes and to a mere re-branding of existing policies on an IPP ‘logo’. Although policy ambiguity may hamper the delivery of concrete results it may also engender advantages. Ambiguity may mitigate potential conflicts. Clear goals, measures and outcomes can trigger clear and effective opposition. In absence of clear policy statements, the perceptions of what the policy is and what it entails will widely vary and allow different actors to interpret the policy as they see fit without arousing contestation and opposition which could stall the policy approach before it can take hold. However, the relevance of IPP should maybe not merely be assessed on the basis of direct and concrete policy outcomes. The policy introduces an important and potentially useful policy concept or idea. Although, it remains unclear whether and how the policy concept can be put into practice, it may be important to allow the concept to mature and ‘travel’ in different domestic contexts and in different policy domains. Evidence of the IPP concept emerging in other but directly related policy domains can already be found. It is the case for instance of the Environmental Technology Action Plan (ETAP) and of the directive on the eco-design of energy-using products (EuP). Only time will tell if these policy initiative will deliver and will be able to transpose IPP-like objectives and principles in practice. For the moment, given the fact that they also rely on voluntary instruments, they face the same dangers as those identified in this chapter.

269

Politics Beyond the State.book Page 270 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State.book Page 271 Wednesday, December 5, 2007 8:41 AM

Bibliography

ANEC (2000), CEN standards in the field of packaging and the environment – an inadequate complement to the Packaging Directive, Brussels, ANEC ANEC and ECOS (2005), ANEC-ECOS Position paper on the revised Packaging Standards prepared under the second Standardisation Mandate M317 ANEC2005/ENV/001, Brussels, ANEC. Armstrong, K. (2006). Inclusive governance? Civil Society and the Open Method of Co-ordination in Smismans, S. (ed.) Civil Society and Legitimate European Governance, London, Edward Elgar Arnold, Douglas, R. (1990), The Logic of Congressional Action, New Haven: Yale University Press. Bacchi, C. L. (1999), Women, Policy and Politics. The Construction of Policy Problems. Sage Publications, London. Bacheller, J. (1977), Lobbyists and the Legislative Process: The Impact of Environmental Constraints, American Political Science Review, vol 71, p. 242-63. Bachmann, K. (2005). Deliberating or Bargaining – the New Members and the Convention on the Future of Europe in Lane, Th., Stadtmüller, E. (eds.) Europe on the Move. The Impact of Eastern Enlargement on the European Union. Munster: LIT. Badiello, L. (1998), Regional Offices in Brussels: Lobbying from the Inside, in Paul-Henri Claeys, Corinne Gobin, Isabelle Smets & Pascaline Winand (eds), Lobbying, pluralisme et intégration européenne, Bruxelles, Presses Interuniversitaires Européennes, p. 328-344. Badiello, L. (2000), Ruolo e funzionamento degli uffici regionali europei a Bruxelles, Le istituzioni del federalismo. Regione e governo locale, n° 1, p. 89-119. Ballmann, A., Epstein, D. & O’Halloran, S. (2002), Delegation, Comitology, and the Separation of Powers in the European Union, International Organization, 56 (3), p. 551-574. Balme, R., Chabanet, D. & Wright, V. (2002), L’action collective en Europe, Paris, Presses de Sciences Po. Baumgartner, F. & Jones, B.D. (1993), Agendas and Instability in American Politics. Chicago, University of Chicago Press. Baumgartner, F. & Jones, B.D. (eds) (2002), Policy Dynamics, Chicago, University of Chicago Press.

271

Politics Beyond the State.book Page 272 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Baumgartner, F. & Leech, B. (1998), Basic Interests: The Importance of Groups in Politics and in Political Science, Princeton, Princeton University Press Baumgartner, F. & Leech, B. (2001), Issue Niches and Policy Bandwagons: Patterns of Interest Group Involvement in National Politics, Journal of Politics, vol 63, p.1191-1213. Beetham, D., Lord, C. (2001), Legitimizing the EU: Is There a ‘post-parliamentary basis’ for its legitimation?, Journal of Common Market Studies, 39, 3, 2001, 443-462. Behning, Ute and Amparo Serrano Pascual (2001) Gender Mainstreaming in the European Employment Strategy. Brussels: ETUI. Benhabib, S. (1996), Toward a Deliberative Model of Democratic Legitimacy in Benhabib, S. (ed.) Democracy and Difference. Contesting the Boundaries of the Political. Princeton, Princeton University Press Benschop, Y. & M. Verloo (2000), Geen roos zonder doornen: reflecties op gender mainstreaming. Tijdschrift voor Genderstudies, 3(4), p. 23-32. Benz, A. & Burkard E. (1999). The Europeanization of Regional Policies: Patterns of Multi-Level Governance, Journal of European Public Policy, vol. 6, no 2, p. 329-348. Berkhout, F. and Smith, D. (1999), Products and the environment: an approach to policy, in European Environment, 9, 174-185. Berry, M., Portney, K.E. & Thomson, L. (1993), The Rebirth of Urban Democracy. Washington, DC, Brookings. Beveridge, F., Nott, S. and Kylie, St. (2000) ‘Mainstreaming and the engendering of policy-making: a means to an end?’ in Journal of European Public Policy 7 pp. 385-405. Beyers, J. (2002), Gaining and seeking access: The European adaptation of domestic interest associations, European Journal of Political Research, vol 41, no 5, p. 585-612. Beyers, J. (2004), Voice and Access: Political Practices of European Interest Associations. European Union Politics. 5: 211-240 Blichner, L. (2000). The anonymous hand of public reason: interparliamentary discourse and the quest for legitimacy in Eriksen, E. O., Fossum, J. E. (eds.) Democracy in the European Union. Integration through Deliberation, Routledge. London. Booth, C. and Bennett, C. (2002) ‘Gender Mainstreaming in the European Union. Towards a New Conception and Practice of Equal Opportunities?’ in European Journal of Women’s Studies 9 pp. 430-446. Borchardt, K.D. (1999), The ABC of Community law, Fifth edition, http://www.europa.eu.int/eur-lex/en/about/abc/index.html

272

Politics Beyond the State.book Page 273 Wednesday, December 5, 2007 8:41 AM

Bibliography

Börzel, T.A. & Risse, T. (2005), Public-Private Partnerships: Effective and Legitimate Tools of International Governance? in Grande, E. & Pauly, L.W. (eds.), Complex Sovereignty: On the Reconstruction of Political Authority in the 21st Century, Toronto: University of Toronto Press, 195-216. Börzel, T.A. (2002). States and Regions in the European Union. Institutional Adaptation in Germany and Spain, Cambridge, Cambridge University Press. Bouveroux, J. (1996), Van zwarte zondag tot zwarte zondag, Icarus, Antwerpen Bouveroux, J. (2003), Van zwarte zondag tot paars groen, Houtekiet, Antwerpen Bouwen, P. (2002), Corporate lobbying in the European Union: the logic of access, Journal of European Public Policy, vol. 9, no. 3: 365-390 Bouwen, P. (2003), The democratic legitimacy of business interest representation in the European Union: Normative implications of the logic of access, Preprints of the Max Planck Institute for Research on Collective Goods, Bonn 2003 (8). Browne, W.P. (1990), Organized Interests and their Issue Niches: A Search for Pluralism in a Policy Domain, Journal of Politics, vol 52: 477–509. Buelens, J. & Deschouwer, K. (2002), The Belgian Greens in government, in F. Müller-Rommel & T. Poguntke, Green Parties in National Governments, Environmental Politics, vol 11, no 1, p. 112-132 Bullmann, U. (1996). The Politics of the Third Level, Regional and Federal Studies, vol. 6, no 2, p. 3-19. Bursens, P. (1998), Het (Belgisch) middenveld en de Europese beleidsvorming: institutionele factoren binnen een meerlagig systeem, Res Publica, vol 40, no 2, p. 231-246. Bustelo, M. (2003) ‘Evaluation of Gender Mainstreaming’ in Evaluation 9 pp. 383-403. Bustelo, M. et. al. (2005), Comparative Frame Analysis: Domestic Violence & Sexual Violence against Women; The European Union, Austria, Greece, Hungary, Slovenia, Spain and The Netherlands. Vienna, IWM Caciagli, M. (2003). Regioni d’Europa. Devoluzioni, regionalismi, integrazione europea, Bologna, Il Mulino. Cambré, B., J. Billiet, et al. (1996). De kennis van de Vlamingen en hun houding tegenover de Vlaamse overheid, Leuven, ISPO Caramani, D. (1996), The nationalisation of electoral politics: a conceptual reconstruction and review of the literature, West European Politics, vol 19, no 2, p. 205-224 Caramani, D. (2000), The societies of Europe. Elections in western Europe since 1815. Electoral results by constituencies, London, Macmillan 273

Politics Beyond the State.book Page 274 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Caramani, D. (2004), The nationalization of politics. The formation of national electorates and party systems in Western Europe, Cambridge UP Carlshamre, M. (2005), Own initiative report ‘on the current situation in combating violence against women and any future action’ (2004/ 2220(INI), Committee on Women’s Rights and Gender Equality, Maria Carlshamre, Final (A6-0404/2005) CEC (2001a) Committee for the adaptation to scientific and technical progress of directive 94/62/EC on packaging and packaging waste. Summary record of the 20th meeting held on 3 October 200., Brussels, European Commission. CEC (2001b), Committee for the adaptation to scientific and technical progress of directive 94/62/EC on packaging and packaging waste. Summary record of the 19th meeting held on 6 June 2001, Brussels, European Commission. CEC (2001c), Committee for the adaptation to scientific and technical progress of directive 94/62/EC on packaging and packaging waste. Summary record of the 18th meeting held on 16 January 2001, Brussels, European Commission. CEC (2001d), Green paper integrated product policy, COM (2001) 68 final. CEC (2001e). European Governance: A white paper (COM(2001)428final). Brussels, Commission of the European Communities. CEC (2002a). Communication from the Commission: Towards a reinforced culture of consultation and dialogue – General principles and minimum standards for consultation of interested parties by the Commission (COM(2002)704final). Brussels, Commission of the European Communities. CEC (2002b). Report from the Commission on European governance (COM(2002)705final). Brussels, Commission of the European Communities. CEC (2003a), Communication from the Commission to the Council and the European parliament. Integrated Product Policy, COM (2003) 302 final. CEC (2003b), Vademecum on European Standardisation. PartII: European Standardisation in Support of European Policies. Chapter 6: Procedure for Formal Objection against standards, 15 November 2003 online (http:// europa.eu.int/comm/enterprise/standards_policy/vademecum/doc/ formal_objections_procedure_en.pdf) CEC (2004a), Commission and industry initiate projects for greening products, Brussels, 25 June 2004 online http://europa.eu.int/comm/environment/ipp/pdf/040625pressrelease.pdf

274

Politics Beyond the State.book Page 275 Wednesday, December 5, 2007 8:41 AM

Bibliography

CEC (2004b), Communication from the Commission to the Council, the European parliament and the European economic and social committee. Integration of Environmental Aspects into European Standardisation, COM (2004)130 final. CEC (2005a), Committee for the adaptation to scientific and technical progress of directive 94/62/EC on packaging and packaging waste. Summary record of the 26th meeting held on 2 February 2005, Brussels, European Commission CEC (2005b), Commission communication in the framework of implementation of the European Parliament and Council Directive 94/62/EC of 20 December 1994 on packaging and packaging waste OJ C 44/23 of 19.2.2005, Brussels, European Commission. CEC (2006). Special interest groups. On-line available: www.europa.eu.int/ comm/secretariat_general/sgc/lobbies/communication/groupint_en.htm (07.02.2006) CEDAW (2002), Concluding Remarks on Hungary’s Combined fourth and fifth periodic report to the CEDAW. CEEG (2002). Logon Report 2002. Lobbying in Brussels. A Challenge for Local and Regional Governments. Charter, M. and Belmane, I. (1999), ‘Integrated Product Policy (IPP) and eco-product development (EPD), in The Journal of Sustainable Product Design, July 1999.17-29. Christensen, T., Laegreid, P. (2003) New Public Management. The transformation of ideas and practice. Ashgate Publishing Company, Burlington USA. Clark, P.B. & Wilson, J.Q. (1961), Incentive Systems: A Theory of Organizations, Administrative Science Quarterly, vol 6: 129-66. Cohen, J. & Sabel, C. (1997). Directly-Deliberative Polyarchy, European Law Journal, 3: 4: 313-342. Cohen, J. (1998), Democracy and Liberty in Elster, J. (ed.) Deliberative Democracy. Cambridge. Cambridge University Press. Collier, U (1998). The environmental dimension of deregulation, in Collier, U. (ed.), Deregulation in the European Union. Environmental Perspectives, London, Routledge, 1998, p. 3-22 Committee of the Regions (1999), Opinion on the amended proposal for a European Parliament and Council decision adopting a program of community action (the Daphne Program 2000-2004) on measures to prevent violence against children, young persons and women, 1999/C 198/63 Corporate Europe Observatory (2004), Curbing Corporate Lobbying Power, 25/10/04. http://www.corporateeurope.org/barroso.html

275

Politics Beyond the State.book Page 276 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Council of Europe (1998) Gender Mainstreaming. Conceptual framework, methodology and presentation of good practice, Strasbourg: Council of Europe. Cram, L. (1998), The EU institutions and collective action. Constructing a European interest? Greenwood J. & Aspinwall M., Collective action in the European Union. Interests and the new politics of associability, London, Routledge. Cullen, P. P. (2004). Obstacles to Transnational Cooperation in the European Social Policy Platform, in Smith, J. & Bandy, J., Negotiating Differences and Unity in Transnational Struggles Against Neo-Liberalism, Lanham, Rowman and Littlefields. Daly, Mary (2005) ‘Gender Mainstreaming in Theory and Practice’ in Social Politics 12:3 pp. 433-450. Danish Environmental Protection Agency (1998), Statement on A Productorientated Environmental Initiative, online http://www.mst.dk/indu/ word/Redeg%C3%B8relsen,%20engelsk%20version.doc. Danish Environmental Protection Agency (2002), ‘New paths to competitive advantage. Cleaner products – a key issue in Denmark’, online http://mst.dk/udgiv/Publications/2002/87-7972-314-4/html/ kap01_eng.htm Danish Environmental Protection Agency (2005), ‘The Integrated Product Policy (IPP) in Denmark’, updated 24/11/2005, online http:// www.mst.dk/homepage/ Dastoli, V. (2002), Les limites de la société civile, Transnational associations – associations transnationales, vol 2: 33-35. De Bast, C. (2002). ECOLO au pouvoir. Bruxelles, Luc Pire. De la Porte, C. & Nanz, P. (2003), OMC – a deliberative-democratic mode of governance ? The cases of employment and pensions, Paper presented at the ESPAnet Conference “Changing European Societies – The Role for Social Policy”, 13-15 November 2003. De Winter L. & Türsan, H. (1998), Regionalist parties in Western Europe, Routledge, London De Winter, L. (2002), Parties and government formation, portfolio allocation, and policy definition, K.R. Luther & F. Müller-Rommel, Political parties in the new Europe. Political and analytical challenges, Oxford University Press, p. 171-206 Deschouwer, K. & Verdonck, M. (2003), Patterns and principles of fiscal and financial federalism in Belgium, Regional and Federal Studies, vol 13, no 4, 2003, p. 91-110 Deschouwer, K. (2000), Belgium’s quasi-regional elections of June 1999, Regional and Federal Studies, vol 10, no 1, 125-132

276

Politics Beyond the State.book Page 277 Wednesday, December 5, 2007 8:41 AM

Bibliography

Deschouwer, K. (2005), Kingdom of Belgium, in J. Kincaid & A. Tarr, Constitutional Origins, Structure, and Change in Federal Countries, McGillQueen’s University Press, Montreal & Kingston, p. 48-75 Deschouwer, K. (2006), Political parties as multi-level organizations, in Richard Katz & William Crotty (eds), Handbook of party politics, Sage Publications, 2006, p. 291-300 Diamantopoulou, A. (2000),Diamantopoulou speaks out against domestic violence, Speech held at the closing of the international conference on violence against women, Portugal, 4 may Diamantopoulou, A. (2003), Tackling Domestic Violence and Trafficking in Human Beings – the Role of the European Union, Member States and Civil Society, speech held at expert meeting by Anna Diamantopoulou, 30 may 2003 Dinkel, R. (1977), Der Zusammenhang zwischen Bundes- und Landeswahlergebnissen, Politische Vierteljahrschrift, vol 18, p. 348360 Dryzek, J. (2000). Deliberative Democracy and Beyond. Liberals, Critics, Contestations. Oxford. Oxford University Press. Dür, A. (2005), Interest Group Power in the EU: Access, Influence and Resources, Paper presented at the workshop CONNEX WP1 – The State of the Art in EU Interest Representation. Leiden, the Netherlands, 14-16 April 2005 Durant, I. (2003). A ciel ouvert. Bruxelles, Luc Pire Economic and Social Committee (1999), Opinion on the amended proposal for a European Parliament and Council decision adopting a program of community action (the Daphne Program 2000-2004) on measures to prevent violence against children, young persons and women Economic and Social Committee (2000). Opinion of the Economic and Social Committee on the Commission discussion paper 'The Commisssion and non-governmental organizations: building a stronger partnership (COM (2000)11final)’ (CES 811/2000). Brussels, Economic and Social Committee. EEB (2000), CEN at work: How the requirements of the European packaging and packaging waste directive (94/62) are bypassed by CEN standards. EEB publication 2000/15, Brussels, EEB. Egan, M. (1998), Regulatory strategies, delegation and European market integration, Journal of European Public Policy, 5(3): 485-506. Egan, M. (2001), Constructing a European Market. Standards, Regulation, and Governance, Oxford: Oxford University Press Eiderström, E. (1998). Ecolabels in EU Environmental Policy in Golub, J. (ed.) New Instruments for Environmental Policy in the EU. Routledge. London. 277

Politics Beyond the State.book Page 278 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Elster, J. (1995). Strategic Uses of Argument, in Arrow, K. et al. (eds.), Barriers to the Negotiated Resolution of Conflict. New York. Norton. Elster, J. (1998). Deliberative Democracy. Cambridge. Cambridge University Press. EP (1986), Resolution on Violence against Women (A2-44/86) EP (1997), Resolution on the need to establish a European Union wide campaign for zero tolerance of violence against women. (A4-0250/1997) EP (1999), Parliamentary debate on Violence against Women: the Daphne Program, sitting of 8 march EP (2000), Decision of the European Parliament and of the Council of 24 January 2000, adopting a programme of Community action (the Daphne programme) (2000-2003) on preventive measures to fight violence against children, young persons and women. 293/2000/EC Epstein, D. & O’Halloran (1999), Delegating Powers. A Transaction Cost Politics Approach to Policy Making under Separate powers, Cambridge: Cambridge University Press. Ernst & Young and SPRU (1998), Integrated Product Policy. A study analyzing national and international developments with regard to Integrated Product Policy in the environment field and providing elements for an EC policy in this area. Executive Summary from the Final Report, March 1998. Euractive (2005), EU and US approaches to lobbying, 01/03/05, http:// www.euractiv.com/Article?tcmuri=tcm:29-13550916&type=LinksDossier European Commission (2003) Amended Proposal for a European Parliament and Council Decision adopting a program of Community action (the Daphne Program) on measures aimed to prevent violence against children, young persons and women, (COM(1999) 244 final) European Environmental Bureau (2004). EEB Evaluation of the European Eco-label Criteria and Scheme. What we wanted – what we got … July 2004. http://www.eeb.org/activities/product_policy/EEB-Ecolabelevaluation-What-we-wanted-what-we-got-July2004.pdf. Accessed 15/ 05/2005. European Social Observatory (2003). Rapport final “Dialogue social sectoriel”. http://europa.eu.int/comm/employment_social/social_dialogue/docs /rapport_final_dss_fr.pdf. Accessed 05/08/2005. EWL (2000) Unveiling the hidden data on domestic violence in the European Union. Brussels, EWL. Falkner, G. (2001), Policy networks in a multi-level system: convergence towards moderate diversity, in Goetz, K.H. & Hix, S. (eds), Europeanised politics? European integration and national political systems, London, Frank Cass: 94-118. 278

Politics Beyond the State.book Page 279 Wednesday, December 5, 2007 8:41 AM

Bibliography

Federal Ministry for the Environment, Nature Conservation and Nuclear Safety of Germany (1999), Informal Meeting of EU Environmental Ministers on Integrated Product Policy (IPP) Weimar 7th –9th May 1999. Background paper on Product Related Environmental Policy, Weimar, 1999, 20p. Fiers S. & Krouwel A. (2004), The Low Countries: From Prime Minister tot Minister-President? In Poguntke, T. & Webb, P. (eds.), The Presidentialisation of Politics in Democratic Societies? Oxford UP Flora, P., Kuhnle, S. et al., (eds) (1999), State formation, nation-building and mass politics in Europe. The theory of Stein Rokkan, Oxford, Oxford University Press. Follesdal, A. (2000), Subsidiarity and democratic deliberation, in Eriksen, E. O. & Fossum, J. E. (eds.) Democracy in the European Union. Integration through Deliberation. London, Routledge. Fossum, J.E. & Menéndez, A. (2004), The Constitution’s Gift: A Deliberative-democratic analysis of constitution-making in the European Union, in Closa, C. & Fossum, J.E. (eds.) Deliberative Constitutional Politics in the EU, ARENA Report No 5/04. Gargarella, R. (2000), Demanding public deliberation: the Council of Ministers – some lessons from the Anglo-American history, in Eriksen, E. O. & Fossum, J. E. (eds.), Democracy in the European Union. Integration through Deliberation, Routledge. London. Gerber, E.R. (1999), The Populist Paradox: Interest Group Influence and the Promise of Direct Legislation, Princeton, NJ: Princeton University Press. Geyer, R. (2001), Can European Union (EU) social NGOs co-operate to promote EU social policy?, Journal of Social Policy, vol 30: 477-493. Gilardi, F. (2001), Principal-Agent Models Go to Europe: Independent Regulatory Agencies as Ultimate Step of Delegation, paper presented at the ECPR General Conference, Canterbury (UK), 6-8 September 2001 Panel 11-8: Principal-Agency Institutional Relations. Goergen, P. (2004), Le lobbying des villes et des régions auprès de l’Union européenne. Göhring, R. (2002), Interest Representation and legitimacy in the European Union: The new quest for civil society formation, in Fairbrass, J. & Warleigh, A., Influence and interests in the European Union: the new politics of persuasion and advocacy, London, New York, Europa Publications/ Taylor&Francis Group: 118-137. Göhring, R. (2003), Shaping the New Europe – Interest Representation in the European Union, Politik-und Sozialwissenschaften, Otto-Suhr-Institut für Politiekwissenschaft. Berlin, Freien Universität Berlin: 257.

279

Politics Beyond the State.book Page 280 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Goldstein, K.M. (1999), Interest Groups, Lobbying, and Participation in America. New York: Cambridge University Press. Grande, E. (1996), The state and interest groups in a framework of multilevel decision-making: the case of the European Union, Journal of European Public Policy vol 3, no 3: 318-38. Greenwood, J. & Halpin, D. (2005), The public governance of interest groups in the European Union: does regulating groups for 'representativeness' strengthen input legitimacy? 3rd ECPR conference, Budapest 810 september 2005. Greenwood, J. (1997), Representing Interests in the European Union, New York: St. Martins Press. Greenwood, J. (1998), Corporatism, Pluralism and the capacities of Eurogroups, Claeys, P.-H., Gobin, C., Smets, I. & Winand, P., Lobbying, Pluralism and European Integration. Bruxelles, Presses Interuniversitaires Européennes. Greenwood, J. (2002), Inside the EU Business Associations, New York: Palgrave. Greenwood, J. (2002). Advocacy, influence and persuasion: Has it all been overdone? in Fairbrass, J. & Warleigh, A., Influence and interests in the European Union: the new politics of persuasion and advocacy, London, New York, Europa Publications/ Taylor&Francis Group: 19-34. Greenwood, J. (2003). Interest Representation in the European Union, London, Palgrave. Greenwood, J. (2005), The significance of citizen interest organizations for Europe’s economic and social model. Challenge Europe Online Journal, Issue 13, on http://www.theepc.net/en/ default.asp?TYP=CE&LV=177&see=y&t=42&PG=CE/EN/ detail&l=1&AI=418 Greenwood, J., Strangward, L. et al. (1999), The capacities of euro groups in the integration process, Political studies, XLVII: 127-138. Grosser, Kate and Jeremy Moon (2005) ‘The Role of Corporate Social Responsibility in Gender Mainstreaming’ in International Feminist Journal of Politics 7:4 pp. 532-554. Harvie, C. (1994), The rise of regional Europe, Routledge, London Heath, A., McLean, I. et al. (1999), Between first and second order: a comparison of voting behaviour in European and local elections in Britain, European Journal of Political Research, vol 35: 389-414. Heichlinger, A. (1999), A Regional Representation in Brussels: The Right Idea for Influencing EU Policy Making?, Maastricht, European Institute for Public Administration.

280

Politics Beyond the State.book Page 281 Wednesday, December 5, 2007 8:41 AM

Bibliography

Helfferich, B. & Kolb, F. (2001), Multi-level Action Coordination in European Contentious Politics- the Case of the European Women's Lobby, in Imig, D. & Tarrow, S.G. Contentious Europeans: protests and politics in an emerging polity, Lanham, Rowman and Littlefields. Héritier, A.(2001), New modes of governance in Europe: policy-making without legislating? Preprint 2001/14, Max-Planck-Projektgruppe: Recht der Gemeinschaftsgüter, Bonn. Hooghe, L. & Marks, G. (2001) Multi-level governance and European integration. Oxford: Rowman and Littlefield Publishers. Hooghe, M, Maddens, B. & Noppe, J (2006), Why parties adapt: Electoral reform, party finance and party strategy in Belgium, Electoral Studies, 2006 Hopkin, J. (2003), Political decentralization, electoral change and party organizational adaptation: a framework for analysis, European Urban and Regional Studies, vol 10, p. 227-238 Hoskyns, C. (1996) Integrating Gender. Women, Law and Politics in the European Union. Verso, London. Hoskyns, C. (2000). Deliberative Democracy and the European Union. Paper for the Political Studies Association-UK 50th Annual Conference, 1013 April 2000, London. Hough, D. & Jeffery, C. (2006), Devolution and electoral politics, Manchester University Press Hunter, R. & Molyneux, C.G., (2000), The Draft EC Directive on Electronic Equipment: An Imprudent and Illegal Proposal, 23, BNA International Environment Reporter 924 at 926. Cited in The New Approach in Setting Product Standards for Safety, Environmental Protection and Human Health, p.29. Imig, D. & Tarrow, S.G. (1999), The Europeanization of Movements? A New Approach to Transnational Contention, in Della Porta, D., Kriesi H. & D. Rucht, Social Movements in a Globalizing World. Basingstoke, Macmillan: 112-133. Jahan, R. (1995) The Elusive Agenda: Mainstreaming Women in Development. London: Zed Books. Jeffery, C. & Hough, D. (2003), Regional elections in multi-level systems, European Urban and Regional Studies, vol 10, p. 199-212 Jeffery, C. (1996a), Regional Information Offices in Brussels and MultiLevel Governance in the EU: A UK-German Comparison, Regional and Federal Studies, vol. 6, n° 2, p. 183-203. Jeffery, C. (1996b), Conclusions: Sub-National Authorities and ‘European Domestic Policy, Regional and Federal Studies, vol. 6, n° 2, pp. 204219.

281

Politics Beyond the State.book Page 282 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Joerges, C. & Everson, M. (2000), Challenging the bureaucratic challenge in Eriksen, E. O., Fossum, J. E. (eds.) Democracy in the European Union. Integration through Deliberation. London: Routledge. John, P. (2001), Local Governance in Western Europe. London, Sage. Jones, B.D., & Baumgartner, F.R. (2005), The Politics of Attention: How Government Prioritizes Problems. Chicago: University of Chicago Press. Jordan, A., Wurzel, R. K. W., Zito, A (2005) The rise of new policy instruments in comparative perspective: has governance eclipsed government? Political Studies, vol. 53, pp. 477-496. Jordan, G. & Maloney, W.A. (1996), How bumble-bees fly: Accounting for public interest participation, Political studies, vol 44: 668-685. Kantola, J. (2003), Transnational and National Gender Equality Politics: European Union’s Impact on Domestic Violence Debates in Britain and Finland. Paper ECPR conference, Marburg. Kantola, J. (2004), European Union and National Gender Equality Politics: the Case of Domestic Violence in Britain and Finland. Paper presented at the ECPR Second Pan-European Conference, Standing Group on EU Politics, Bologna, 24-26 June 2004. Keating, M. (1998), The new regionalism in Western Europe. Territorial restructuring and political change, Edward Elgar, London Keck, M. and Kikkink, K. (1998). Activists Beyond Borders: Advocacy Networks in International Politics. Ithaca, Cornell University Press. Kingdon, J. W. (1984), Agendas, Alternatives, and Public Policies. Boston: Little, Brown. Knudsen, P., M. Jensen and Chabert, A. (2003), Evaluating Product Panels An Evaluation Made in 2001 by Oxford Research A/S in Working Report Arbejdsrapport fra Miljostryrelsen: 14, 2003. Kohler-Koch, B. & Quittkat, C. (1999). Intermediation of Interests in the European Union. Mannheim, Universität Mannheim. Kohler-Koch, B. (1997). "Organized interests in EC and the European Parliament." European integration on-line papers (EIoP) 1(9). On-line available: http://eiop.or.at/eiop/texte/1997-009a.htm (12.09.2003). Kohler-Koch, B. (2002). "European Networks and Ideas: Changing National Policies." European integration on-line papers (EIoP) 6(6). Online available: http://eiop.or.at/eiop/texte/2002-006a.htm (12.09.2003). Kollman, K. (1998), Outside Lobbying: Public Opinion and Interest Group Strategies. Princeton, NJ: Princeton University Press. Koopmans, R. (2004). Integrated report WP2: Cross-national, cross-issue, cross-time. The transformation of political mobilisation and communication in European public spheres. On-line available: http://www.europub.com (21.10.2005).

282

Politics Beyond the State.book Page 283 Wednesday, December 5, 2007 8:41 AM

Bibliography

Kovziridze, T. (2002), Europeanization of Federal Institutional Relationships: Hierarchical and Interdependent Relationship Structures in Belgium, Germany and Austria, Regional and Federal Studies, vol. 12, n° 3, pp. 128-155. Kreppel, A. (2002), The European Parliament and Supranational Party System. Cambridge: Cambridge University Press. Kreppel, A. (2005), Understanding the European Parliament from a Federalist Perspective: The Legislatures of the USA and the EU Compared. Paper prepared for the EUSA 9th Biannual International Conference. Austin, TX. 31 March – 2 April 2005. Kristensen, P (2004)., ‘Product Panels in Denmark. Examples of market driven reduction in production and use of hazardous substances’, Presented during Workshop: Sustainable Chemistry -Integrated Management of Chemicals, Products and Processes, 27-29. January 2004, Dessau. http://www.sustainable-chemistry.com/download/IV_Kristensen.pdf Kriszan, A., M. Paantjens & I. Van Lamoen (2005), Domestic Violence: Who’s Problem? Greek Review of Social Research, vol. B’/2005, 117, pp. 63-92. Lancaster, T. D. (1999), Complex self-identification and compounded representation in federal systems, in Brzinski J., Lancaster, T. et.al (eds), Compounded representation in Western European federations, London, Frank Cass, 59-89 Larsson, T. (2000), The Internal Enlargement of the European Union and the Surplus of the Intermediate Level of Government, Eipascope, 2000/ 2 Le lobbying des bureaux de représentation, in La Gazette des Communes, n° 1698, 16/6/2003. Liebert, Ute (2002) ‘Europeanising Gender Mainstreaming: Constraints and Opportunities in the Multi-level Euro-Polity’ in Feminist Legal Studies 10 pp. 241-256. Linder, W. (1994). Swiss Democracy. Possible Solutions to Conflict in Multicultural Societies. New-York: St. Martin’s Press. Lombardo, E. & Meier, P. (2006), Gender Mainstreaming in the EU: Incorporating a feminist reading? European Journal of Women’s Studies, 13 (2), in print. Lombardo, E. (2005) ‘Mainstreaming Gender Equality in the European Constitution-Making Policy’ in Schonlau, J. et al. (eds.) The Making of the European Constitution. Houndmills: MacMillan. Loughlin, J. (2001), Subnational democracy in the European Union. Challenges and opportunities, Oxford UP Lovenduski, Joni et al. (2005) State Feminism and Political Representation. Cambridge: Cambridge University Press. 283

Politics Beyond the State.book Page 284 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Lupia, A. (2003), ‘Delegation and its Perils’ in Strøm, K., W.C. Muller and Bergman, T. (Eds.) (2003), Delegation and accountability in parliamentary democracies. Oxford: University Press: 33-54. Mahoney, C. (2003), Influential Institutions: The Demand Side of Lobbying in the European Union. Paper presented at the American Political Science Association annual meeting. Philadelphia, Pennsylvania. August 2003. Mahoney, C. (2004), The power of institutions. State and interest group activity in the European Union, European Union Politics, vol 5, no 4: 441-466. Majone, G. (1998), Europe's 'democratic deficit': the question of standards, European law journal, vol 4, no 1: 5-28. Majone, G. (2001), Two Logics of Delegation, European Union Politics, 2(1):103-122. Malone, M.M. (2004), Regulation of Lobbyists in Developing Countries, Current Rules and Practices. Institution of Public Administration. Dublin Ireland. Mansbridge, J. (1992). A Deliberative Theory of Interest Representation in Petracca, M. (ed.) The Politics of Interest. Boulder: Westview Press. March, J. & Olson, J. (1989). Rediscovering Institutions: The Organizational Basis of Politics. New York, Free Press. Marks, G. & McAdam, D. (1996), Social Movements and the Changing Structure of Political Opportunity in the European Union, West European Politics, vol 19, no 2: 249-278. Marks, G. & McAdam, D. (1999). On the Relationship of Political Opportunities to the Form of Collective Action: the Case of the European Union, in Della Porta, D., Kriesi, H. & Rucht, D. (eds), Social Movements in a Globalizing World. Basingstoke, Macmillan: 97-111. Marks, G., Haesly, R. & Heather A.D. Mbaye (2002), What Do Subnational Offices Think They Are Doing in Brussels?, Regional and Federal Studies, vol. 12, n° 3, p. 1-23. Marks, G., Hooghe, L. & Blank, K. (1996) European integration from the 1980s: state-centric v. multi-level governance, Journal of common marked studies, 34, 343-377 Matland, R. E. (1995) Synthesizing the implementation literature: the ambiguity-conflict model of policy implementation. Journal of Public Administration Research and Theory, 5, 2. Mazey, S. & Richardson, J. (1993), Lobbying in the European Community. Oxford, Oxford University Press. Mazey, S. & Richardson, J. (2001), Institutionalizing Promiscuity: Commission-Interest Group Relations in the European Union, in Stone Sweet, A., Sandholz, W. & Fligstein, N., The Institutionalization of Europe, Oxford, Oxford University Press, p. 71-93. 284

Politics Beyond the State.book Page 285 Wednesday, December 5, 2007 8:41 AM

Bibliography

Mazey, S. & Richardson, J. (2001), Interest groups and EU policy making: organizational logic and venue shopping, in Richardson, J. European Union: Power and policy-making,London, Routledge: 217-238. Mazey, Sonia (2000) ‘Introduction: Integrating gender – intellectual and ‘real world’ mainstreaming’ in Journal of European Public Policies 7 pp. 333-345. Mazey, Sonia (2002) ‘Gender Mainstreaming Strategies in the EU: Delivering on an Agenda?’ in Feminist Legal Studies 10 pp. 227-240. Mazur, Amy (ed.) (2001) State feminism, women’s movements, and job training: making democracies work in the global economy. New York: Routledge. McAteer, M. & Mitchell, D. (1996), Peripheral Lobbying! The Territorial Dimension of Euro Lobbying by Scottish and Welsh Sub-central Government”, Regional and Federal Studies, vol. 6, no. 3, pp. 1-27. McCarthy, J. D. & Mayer N.Z. (1978), Resource Mobilisation and Social Movements: A Partial Theory. American Journal of Sociology. 82: 1212– 41. McCubbins, M. & Schwartz, T. (1984), ‘Congressional Oversight Overlooked: Police Patrols versus Fire Alarms’, American Journal of Political Science, 28 (1), p. 165-179. McCubbins, M.D. & Page, T. (1987), A Theory of Congressional Delegation’, in McCubbins, M.D. and Sullivan, T. (eds.), Congress. Structure and Policy, Cambridge: Cambridge University Press, 409-425. McKissick, Gary J. (1995), Interests, Issues, and Emphases: Lobbying Congress and the Strategic Manipulation of Issue Dimensions, Presented at the annual meeting of the Midwest Political Science Association, Chicago. McLaughlin, A., Jordan, M.,G. et.al. (1993), Corporate lobbying in the European Community, Journal of Common Market Studies vol 31, no 2: 91-211. McNamara, K.R. (2002), Rational fictions: Central Bank Independence and the Social Logic of Delegation, West European Politics, 25(1): 47-76. Meier, P. (2006), Implementing gender equality: gender mainstreaming and the gap between theory and practice. In: S.K. Hellsten, A.M. Holli et K. Daskalova (red.), Women’s Citizenship and Political Rights. New York, Palgrave. Meier, P. et al. (2005), Gender mainstreaming and the bench marking fallacy of women in political decision making. Greek Review of Social Research, B’/2005, 117, pp 35-62.

285

Politics Beyond the State.book Page 286 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Meier, Petra (2006) ‘Implementing gender equality: gender mainstreaming or the gap between theory and practice’. In S.K. Hellsten, A.M. Holli et K. Daskalova (eds.) Women’s Citizenship and Political Rights. New York: Palgrave. Menon, A. (2005), The Limits of Comparative Politics: International Relaations in the European Union. Paper prepared for the EUSA 9th Biannual International Conference. Austin, TX. 31 March – 2 April 2005. Michalowitz, I. (2002), Beyond corporatism and pluralism: towards a new theoretical framework, in A. Warleigh, A. & Fairbrass, J., Influence and interests in the European Union: the new politics of persuasion and advocacy.. London and New York, Europa Publications Taylor & Francis Group: 35-53. Michalowitz, I. (2004), Analysing structured paths of lobbying behaviour: Why discussing the involvement of 'civil society' does not solve the EU's democratic deficit, European integration, vol 26 no 2: 145-170. Michalowitz, I. (2005), Service bureaux of decision makers or successful spindoctors: Assessing interest group influence in the EU and the US. Paper prepared for the EUSA 9th Biannual International Conference. Austin, TX. 31 March – 2 April 2005. Mintrom, M. (2001), Educational Governance and Democratic Practice, Educational Policy. 15 (5): 615-643. Nov 2001. Moravcsik, A (1994) Why the European Community strengthens the state: international co-operation and domestic politics. Cambridge, MA: Center for European Studies Working Paper Series No. 52, Harvard University. Moravcsik, A. (2002), In defence of the 'democratic deficit': Reassessing legitimacy in the European Union, Journal of Common Market Studies, vol 40, no 4: 603-24. Mosesdottir, Lilja and Rosa G. Erlingsdottir (2005) ‘Spreading the Word Across Europe. Gender Mainstreaming as a Political and Policy Project’ in International Feminist Journal of Politics 7:4 pp. 513-531. Müller, W. C. & Strøm, K. (eds) (1999), Policy, office, or votes? How political parties in Western Europe make hard decisions. Cambridge, Cambridge University Press. Muller, W. C., (2003), ‘Parliament Democracy’ in Strøm, K., W.C. Muller and Bergman, T. (eds.) Delegation and accountability in parliamentary democracies, Oxford: University Press, 3-32. Nielsen, F. & Salk, J. (1998), The Ecology of Collective Action and Regional Representation in the European Union, European Sociological Review, vol. 14, n° 3, pp. 231-254.

286

Politics Beyond the State.book Page 287 Wednesday, December 5, 2007 8:41 AM

Bibliography

Nordic Council of Ministers (2000), ‘A better environment through the market. New opportunities for sustainable development via an integrated product policy (IPP). Report from Nordic IPP Seminar, Saltsjöbaden, February 2000, vol.2. Oosterhuis, F., Rubik, F. and Scholl, F. (1996), Product Policy in Europe: New Environmental Perspectives, Kluwer Academic Publishers: Dordrecht. Outshoorn, Joyce (ed.) (2004) The Politics of Prostitution: Women’s Movements, Democratic States and the Globalisation of Sex Commerce. Cambridge: Cambridge University Press. Paantjens, M. (2004), Revised Preliminary Report, European Union, Domestic Violence. Report for MAGEEQ: Policy frames and implementation problems: the case of gender mainstreaming. Nijmegen, RUN. Pallarés F. & Keating, M. (2003), Multi-level electoral competition: regional elections and party systems in Spain, European Urban and Regional Studies, vol 10, p. p. 239-256 Pallemaerts, M. et al. (2006), The role of public authorities in Integrated Product Policy : regulators or coordinators ? Brussels: Belgian Science Policy, available on http://www.ies.be/papers/rappcp49_en.pdf Panebianco, S. (2000). Il lobbying europeo, Milano, Giuffrè editore. Papadopoulos, Y. (2002), Is “Governance” a Form of “Deliberative Democracy”?. Paper presented at the ECPR Joint Sessions of Workshops, 2227 March 2002, Turin. Parkinson, J. (2003), Legitimacy Problems in Deliberative Democracy in Political Studies 51:180-196. Pedler, R. (2002), European Union lobbying: changes in the arena, Palgrave, Basingstoke. Pedler, R.H. & van Schendelen, M.P.C.M (eds) (1994), Lobbying the European Union. Companies, Trade Associations and Issue Groups. Aldershot: Dartmouth. Perrons, Diane (2005) ‘Gender Mainstreaming and Gender Equality in the New (Market) Economy: An Analysis of Contradictions’ in Social Politics 12:3 pp.389-411. Peters, G. B., Pierre, J. (2004) Multi-level governance and democracy: a faustian bargain? in Bache, I, Flinders, M., (eds.) Multi-level governance. Oxford Scholarship Online Monographs, pp. 75-91. Pierre, J., Peters, G.B. (2000) Governance, politics and the state. New York, St. Martin’s Press. Platform of European Social NGOs (2003), About Us. On-line available: http://www.socialplatform.org (21.02.2003). Pollack, M.A. (1997), Delegation, Agency, and Agenda Setting in the European Community, International Organization, 51, 1: 99-134. 287

Politics Beyond the State.book Page 288 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Pollack, M.A. (2002), Learning from the Americanists (Again): Theory and Method in the Study of Delegation, West European Politics, 25(1), p.200-219. Pollack, M.A. (2003), The engines of European Integration. Delegation, Agency, and Agenda Setting in the EU, Oxford: Oxford University Press, 496p Pollack, Mark A. and Emilie Hafner-Burton (2000) ‘Mainstreaming Gender in the European Union’ in Journal of European Public Policy 7:3 pp. 432-56. Price, L.S. (2005), Feminist Frameworks: Building Theory on Violence Against Women. Fernwood Publishing, Halifax. Princen, S. & Kerremans, B. (2005), Opportunity Structures in the EU Multilevel System Reviewing the State of the Art.” Paper presented at the workshop CONNEX WP1 – The State of the Art in EU Interest Representation. Leiden, the Netherlands, 14-16 April 2005 Pullen, M. (2003), Playing the Lobbying Game, Legal Week. 16/10/03. Page 2. Putnam, R.D. (1993), Making Democracy Work. Princeton: Princeton University Press. Radaelli, C.M. (2003). The Europeanization of Public Policy, in Featherstone, K. & Radaelli, C.M. (eds.), The Politics of Europeanization. Oxford, Oxford University Press. Rawls, J. (2003). Political Liberalism. New York: Columbia University Press. Rees, T. (1998) Mainstreaming Equality in the European Union: Education, Training, and Labour Market Policies. New York: Routledge. Rees, T. (2001) ‘Mainstreaming Gender Equality in Science in the European Union: the ‘ETAN Report’’ in Gender and Education 13:3 pp. 243260. Rees, T. (2005) ‘Reflections on the Uneven Development of Gender Mainstreaming in Europe’ in International Feminist Journal of Politics 7:4 pp. 555-574. Reif, K.-H. & Schmitt, H. (1980) Nine second order national elections. A conceptual framework for the analysis of European election results, European Journal of Political Research, 8, 3-44 Reif, K.-H. (1984), National electoral cycles and European elections, Electoral Studies, 3, 3, 244-255 Rodriguez-Posé, A., Gill, N. (2003) The global trend towards devolution and its implications. Environment and Planning C: Government and Policy, 21, 3, pp. 333-51.

288

Politics Beyond the State.book Page 289 Wednesday, December 5, 2007 8:41 AM

Bibliography

Roggeband, C. (2002), Over de grenzen van de politiek. Een vergelijkende studies naar de opkomst en ontwikkeling van de vrouwenbeweging tegen seksueel geweld in Nederland en Spanje. Nijmegen, Katholieke Universiteit Nijmegen. Roggeband, C. (2004), ‘Immediately I Thought We Should Do the Same Thing’. European Journal of Women’s Studies, Vol. 11 (2), p. 159-175. Rohrschneider, R. & R. J. Dalton (2002). A Global Network? Transnational Cooperation among Environmental Groups, The Journal of Politics, vol 64, no 2: 510-533. Rosenberg, S. (2003). Reason, Communicative Competence and Democratic Deliberation: Do Citizens Have the Capacities to Effectively Participate in Deliberative Decision-Making? Paper presented at the 2003 Annual Meeting of the American Political Science Association. Philadelphia. Rosenberg, S. (2004). Reconstructing the Concept of Democratic Deliberation. Center for the Study of Democracy (University of California, Irvine). Paper 04’02. Rubery, Jill (2002) ‘Gender mainstreaming and gender equality in the EU: the impact of the EU employment strategy’ in Industrial Relations Journal 33:5 pp. 500-522. Rubery, Jill et al. (2004) ‘The ups and downs of European gender equality policy’ in Industrial Relations Journal 35:6 pp. 603-628. Rucht, D. (1999). The Transnationalization of Social Movements: Trends, Causes, Problems, in Della Porta, D.,Kriesi H. & Rucht, D.,Social Movements in a Globalizing World.. Basingstoke, Macmillan: 206-222. Rucht, D. (2001), Lobbying or Protest? Strategies to Influence EU Environmental Policies” in Imig, D. & Tarrow, S. (eds.), Contentious Europeans: Protest and Politics in an Emerging Polity. Lanham, MD: Rowman & Littlefield Publishers. Salisbury, R.H., Heinz, J.P., Laumann, E.O., Nelson, R.L. (1987), Who Works With Whom? Interest Group Alliances and Opposition, American Political Science Review. 81: 1211-34. Sandholtz W., Stone Sweet, A. (1998) European Integration and Supranational Governance. Oxford: Oxford University Press. Saurugger, S. (2005) Associative Democracy and the Democratic Legitimacy of the European Union. Paper presented at the workshop CONNEX WP1 – The State of the Art in EU Interest Representation. Leiden, the Netherlands, 14-16 April 2005 Sbragia, A. (2003), Key Policies, in Bomberg, E. & Stubb, A., The European Union: How Does it Work?, Oxford: Oxford University Press. Scharpf, F. W. (1988) The joint-decision trap: lessons from German federalism and European integration. Public Administration, vol. 66, pp. 239-278. 289

Politics Beyond the State.book Page 290 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Scharpf, F.W. (1999) Governing in Europe: effective and democratic. Oxford: Oxford University Press Schattschneider, E. E. (1960) The Semi-Sovereign People. New York: Holt, Rinehart and Winston. Schlozman, K. L., Tierney, J.T. (1986), Organized Interests and American Democracy. New York: Harper & Row Publishers. Schmitter, Ph. C. (1996), Imagining the Future of the Euro-Polity with the Help of New Concepts, in Marks, G., Scharpf, F.W., Schmitter, P.C. & Streeck, W., Governance in the European Union, London, Sage, pp. 121150. Schultze, C.J. (2003). Cities and EU Governance. Policy-takers or Policymakers?, Regional and Federal Studies, vol. 13, n° 1, pp. 121-147. Scott, J., Trubek, D.M (2002)., Mind the gap: law and new approaches to governance in the European Union, European Law Journal, 8, 1, 2002, 1-18. Smets, I., Winand, P. (2000), A la recherche d'un modèle européen de représentation des intérêts, in Magnette, P. & Remacle, E. Le nouveau Modèle européen, Bruxelles, Edition de l'Université Libre de Bruxelles. Smets, I. (1998)., Les régions se mobilisent – quel ‘lobby régional’ à Bruxelles?, in Claeys, P.H., Gobin, C., Smets, I. & Winand, P. (eds), Lobbying, pluralisme et intégration européenne, Bruxelles, Presses Interuniversitaires Européennes, pp. 303-327. Smismans, S. (2005), Reflexive law in support of directly deliberative polyarchy: Reflexive-deliberative polyarchy as a normative frame for the OMC in De Schutter, O., Deakin, S. (eds.) Social Rights and Market Forces: Is the Open Coordination of Employment and Social Policies the Future of Social Europe? Brussels: Bruylant. Smith, A. (1998a), Au-delà d’une ‘Europe du lobbying’ – l’exemple des rapports entre régions et Commission”, in Claeys, P.H., Gobin, C., Smets, I. & Winand, P. (eds), Lobbying, pluralisme et intégration européenne, Bruxelles, Presses Interuniversitaires Européennes, pp. 58-82. Smith, A. (1998b), The Sub-regional Level. Key Battleground for the Structural Funds?, in Le Galès, P. & Lequesne, C. (eds), Regions in Europe, London, Routledge, pp. 50-66. Squires, J. (2005) ‘Is Mainstreaming Transformative? Theorising Mainstreaming in the Context of Diversity and Deliberation’ in Social Politics 12:3 pp. 366-388. Stetson, D. and Amy Mazur (1995) Comparative State Feminism. London: Sage. Stetson, D., McBride (ed.) (2001) Abortion politics, women’s movements, and the democratic state: a comparative study of state feminism. Oxford: Oxford University Press. 290

Politics Beyond the State.book Page 291 Wednesday, December 5, 2007 8:41 AM

Bibliography

Stoker, G. (1998) Governance as theory: five propositions. International Social Science Journal, vol. 50, no. 155, pp. 17-28. Stratigaki, M. (2004), The Cooptation of Gender Concepts in EU politics: the Case of Reconciliation of Work and Family. Social Politics, 11(1), pp. 30-56. Stratigaki, M. (2005) ‘Gender Mainstreaming versus Positive Action: An on-going Conflict in EU Gender Equality Policy’ in European Journal of Women’s Studies 12 pp. 165-186. Streeck, W. (1992). From national corporatism to transnational pluralism: European interest politics and the single market, in Treu, T., Participation in public policy making. The role of trade unions and employers' associations, Berlin, New York, Walter de Gruyter: 97-126. Strøm, K. (1990). Minority government and majority rule. Cambridge, Cambridge University Press Strøm, K. (2003), Parliamentary Democracy and Delegation’ in Strøm, K., W.C. Muller & Bergman, T. (eds) Delegation and accountability in parliamentary democracies. Oxford: University Press,55-106. Strøm, K., Muller, W.C., Bergman, T. (eds.) (2003), Delegation and accountability in parliamentary democracies. Oxford: University Press Stubb, A., Wallace, H. & Peterson, J (2003), The Policy-Making Process, in Bomberg, E. & Stubb, A., The European Union: How Does it Work?, Oxford: Oxford University Press. Swedish Environmental Protection Agency (2003), ‘Towards greener products’, The Swedish EPA, Stockholm, http://www.naturvardsverket.se/ bokhandeln/pdf/620-5296-9.pdf Taylor Nelson Sofres Consulting (2001). Investigation of the market impacts and penetration of the European Eco-label over the years 1992-2000 and 2001-2004. Final Report. December 2001, http://europa.eu.int/comm/ environment/ecolabel/pdf/market_study/ sofresoutlookphase1_1201.pdf. Accessed 15/05/2005 Thatcher, M. & Stone Sweet, A. (2002), Theory and Practice of Delegation to Non-Majoritarian Institutions, West European Politics, 25(1): 1-22. The Gallup Organization (2003). Describing Regional Offices Representing Local Interests in Brussels: Similarities, Disparities, and Expectations. Report of Survey Results for Participants. Brussels, December 2003. Theorin, M. B. (2000), Speech MEP May Britt Theorin, held at the meeting on Violence against Women, Portugal, 4 May Thorlakson, L. (2001), One party or several? Party organization in federations, paper presented at the conference Multi-level electoral competition: devolution in comparative context, Institute of German studies, University of Birmingham, September 2001

291

Politics Beyond the State.book Page 292 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State

Traberg, R., Sharing responsibilities-the Danish Product Panels, presentation at the International Summer Academy “From Government to Governance: The Case of Integrated Product Policy”, September 7-11, 2003, Wuerzburg, Germany. True, Jacqui and Michael Mintrom (2001) ‘Transnational Networks and Policy Diffusion: The Case of Gender Mainstreaming’ in International Studies Quarterly 45:1 pp. 27-57. Tufet-Opi, E. (2002), ‘Life after end of life: the replacement of end of life product legislation by a European integrated product policy in the EC’, in Journal of Environmental Law, vol. 14, p.33-60. Union of International Associations, (2002). Yearbook plus of international organizations and biographies. München, K.G.Saur. Van den Heuvel, J.H.J. (1998) Beleidsinstrumentatie. Sturingsinstrumenten voor het overheidsbeleid. Uitgeverij Lemma, Utrecht, 1998. Van der Vleuten, A. (2001), Dure Vrouwen, Dwarse Staten. Een institutioneel-realistische visie op de totstandkoming en implementatie van Europees beleid. Nijmegen, Nijmegen University Press. Van Lamoen, I. (2004), Revised Preliminary Report, the Netherlands, Domestic Violence, report for the project MAGEEQ: Policy frames and implementation problems: the case of gender mainstreaming. Nijmegen, RUN. Van Peteghem, B. (2001), Paars met groen: de vorming van Verhofstadt I, Lannoo, Tielt Van Schendelen, M.P.C.M (ed.) (1993), National Public and Private EC Lobbying. Aldershot: Dartmouth Van Schendelen, M.P.C.M. (2003), Machiavelli in Brussels. The Art of Lobbying the EU. Amsterdam, Amsterdam University Press. Vara Arribas, G., Bourdin, D., eds. (2006) Tripartite arrangements. An effective tool for multi-level governance? European Institute of Public Administration: Maastricht. Verloo, M., Maloutas, M.P. (eds.) (2005), Special Issue: Differences in the Framing of Gender Inequality as a Policy Problem across Europe. Greek Review of Social Research, B’/2005, 117. Verloo, M. (2001) Another Velvet Revolution? Gender mainstreaming and the politics of implementation. IWM Working Paper No. 5/2001. Vienna: IWM. Verloo, M. (2004), Mainstreaming Gender Equality in Europe. A Frame Analysis Approach. Paper presented at the Conference of the Europeanists, Chicago, March, 11-13, 2004.

292

Politics Beyond the State.book Page 293 Wednesday, December 5, 2007 8:41 AM

Bibliography

Verloo, M. (2005) ‘Displacement and Empowerment: Reflections on the Concept and Practice of the Council of Europe Approach to Gender Mainstreaming and Gender Equality’ in Social Politics 12(3) pp.344365. Versmessen, E. (1995), In the kingdom of paradoxes: the Belgian regional and national elections of May 1995, Regional and Federal Studies, 5(2): 239-246. Vos, H. (1999). De eigenheid van regio’s in Europa – Een bijdrage tot het debat over regiovorming, Res Publica, vol. 41, n° 1, pp. 87-99. Walby, S. (2005) ‘Gender Mainstreaming: Productive Tensions in Theory and Practice’ in Social Politics 12:3 pp.321-343. Warleigh, A. (2001), Europeanizing' civil society: NGOs as agents of political socialization., Journal of Common Market Studies, vol 39, no 4: 619639. Watson, R. & Shackleton, M. (2003), Organized Interests and Lobbying in the EU, in Bomberg, E; & Stubb, A., The European Union: How Does it Work? Oxford: Oxford University Press. Watts, R. (1999) Comparing federal systems, McGill-Queens University Press, Montreal. WAVE, (2000), Good Practice Models; WAVE Training Program on Combating Violence against Women. www.wave-network.org. Wendler, F. (2003). Achieving democratic legitimacy through deliberative governance? The case of EU Social Policy. Paper presented at the 2nd ECPR General Conference, 18-21 September 2003, Marburg Wöll, C. (2005), Lobbying in a transatlantic comparison, Opportunity structures and lobbying content in the US and EU. Paper prepared for the EUSA 9th Biannual International Conference. Austin, TX. 31 March – 2 April 2005. Woodward, A. (2003) ‘European Gender Mainstreaming: Promises and Pitfalls of Transformative Policy’ in Review of Policy Research, 20 pp. 65-88. Woodward, A. (2004) ‘Building velvet triangles: gender and informal governance’, in T. Christiansen and S. Piattoni (eds.) Informal governance in the European Union. Cheltenham: Edward Elgar. World Health Organization (2002), World Report on Health. WHO, Geneva Wright, V. (1998) Intergovernmental relations and regional government in Europe: a skeptical view. in Le Galès,P., Lequesne, Ch., (eds.) Regions in Europe. Routledge: London, pp. 39-49.

293

Politics Beyond the State.book Page 294 Wednesday, December 5, 2007 8:41 AM

Politics Beyond the State.book Page 295 Wednesday, December 5, 2007 8:41 AM

List of Contributors

Isabelle Bedoyan is a researcher in the Department of Political Science of the Vrije Universiteit Brussel. Jo Buelens is a researcher in the Department of Political Science of the Vrije Universiteit Brussel. Kris Deschouwer is professor of politics at the Vrije Universiteit Brussel. Michel Huysseune is a post-doc researcher at the Department of Political Science of the Vrije Universiteit Brussel. Maarten Theo Jans is professor of politics at the Vrije Universiteit Brussel and Senior Fellow at the Institute for European Studies. Christine Mahoney is assistant professor in the Department of Political Science of Syracuse University, NY. Petra Meier is professor of politics at the Universiteit Antwerpen. Marjolein Paantjens is a researcher in the Department of Political Science of the Vrije Universiteit Brussel. Irina Tanasescu has been a researcher in the Department of Political Science of the Vrije Universiteit Brussel and is now working for the European Commission. Irina Stefuriuc is a researcher in the Department of Political Science of the Vrije Universiteit Brussel. Martine Van Assche was a teaching assistant and researcher in the Department of Political Science of the Vrije Universiteit Brussel. Joke Wiercx is a researcher in the Department of Political Science of the Vrije Universiteit Brussel.

295