Towards a proactive role for local government in EU decision-making? Exploring bottom-up Europeanization in Flemish cities

Transcription

1 Centre for Local Politics Department of Political Science Ghent University - Belgium Towards a proactive role for local government in EU decision-making? Exploring bottom-up Europeanization in Flemish cities dr. Tom Verhelst (tom.verhelst@ugent.be) Working paper submitted for presentation at the 64th PSA Annual International Conference (please do not cite without the author s permission) April 2014 Manchester, UK

2 1. Introduction In October 2013, the European Court of Justice fined the Belgian state for its failure to meet the European regulation on the purification of wastewater. This fine was remarkable for two reasons. First, it was the first time ever that the Belgian state received a European fine as such. Second, the fine stood out for its particular jurisdiction as municipalities are the basic governments responsible for wastewater treatment in Belgium. Accordingly, the case exemplifies how the EU, in its capacity of regulatory state, also (and increasingly) impacts upon the local level (Goldsmith, 1993; John, 2000; Fleurke & Willemse, 2006; Kern, 2009; Hamedinger & Wolffhardt, 2010; Greenwood, 2011) 1. In policy fields as broad and varied as environment, energy, transport, agriculture, public procurement, state aid, social and health policy, tourism, infrastructure and public services, the impact of EU regulation on municipalities is substantial nowadays (Schultze, 2003: 123; John in Callanan, 2012: 754; RoB, 2013: 9). According to Callanan (2012: 754), this implies that EU legislation affects local authorities in their role as service-providers, as employers, as monitoring, enforcing and licensing bodies, and as agents for the development of their area. In the case of the purification of waste water, for instance, the EU directive obliged Belgian municipalities to drastically rearrange their sewer system an investment which still puts a heavy strain on the budgets of many municipalities (De Wachter, 2009; Leroy, 2012). However, with the obligation to comply with European legislation came new opportunities for local governments to defend or promote their interests and influence the European decision-making process alike (De Rooij, 2002; Heinelt & Niederhafner, 2008). According to this point of view, the EU is not just the overarching regulatory state which commands its subnational authorities (SNAs). It also functions as an additional policy arena which offers the SNAs new room and possibilities to advocate their policy issues and concerns (Fleurke & Willemse, 2006). Such dynamic and proactive processes led by SNAs are captured by the concept of regulatory mobilisation (Callanan & Tatham, 2013). The rationale behind this type of mobilisation seems straightforward: the huge impact of Europe upon domestic actors propels these actors to engage in the policy-making process that will affect them in a later stage. Obviously, this requires domestic actors to develop a corresponding political strategy (Beyers, 2002). In this way local governments should evolve from being mere passive policy-takers to becoming more proactive policy-makers (Schultze, 2003). Callanan (2012: 757) summarizes this argument as follows: It is clear ( ) that there is an incentive for local government in different countries to mobilise to influence new legislative and policy initiatives at EU level (vis-à-vis the EU institutions and/or national government), given the considerable administrative or financial consequences these imply for local authorities. In spite of this clear argument, however, one could wonder to what extent local governments actually develop strategies to influence European decision-making in practice let alone that these efforts would be successful. The discussion on the regulatory mobilisation of local governments in EU decision-making is situated at the intersection of studies on interest representation and 1 According to Annemarie Jorritsma, chair of the CEMR, some 70 to 80 per cent of the regulation local governments have to deal with has been enacted at the European level (Van Bouwel, 2013). Considering this relation the other way around, the Committee of the Regions reports that some 70 per cent of EU legislation has to be implemented at the sub-state level (RoB, 2013). 2

3 Europeanization (with a focus on the local level). Regarding the former, Dür (2008: 560) underlines the importance of analysing interest group influence to the understanding of the political process as a whole. This is particularly valuable in the context of European decision-making, where interest representation is inevitably connected to issues of democracy and legitimacy (Greenwood, 2011). Yet while academic attention for interest representation in the EU is shifting from the analysis of interest representation as an element of European integration towards an attempt to understand its logic in se, few studies have actually tried to measure influence and its determinants (Woll, 2007; also Vass, 2008; Klüver, 2011). After all, the participation in European decision-making is no guarantee for concrete influence (Marks et al., 2002). Also within the field of local level Europeanization, and despite the apparent impact of EU legislation on SNAs, regulatory mobilisation and its conditions for success have not been examined thoroughly or systematically yet (Callanan, 2012; Callanan & Tatham, 2013). Most academic attention is still devoted to the analysis of the financial mobilisation (i.e. the mobilisation aimed at obtaining European funding) of SNAs (Goldsmith, 2005; Callanan, 2012). This prejudice probably stems from the fact that SNAs themselves are predominantly occupied with financial mobilisation in the first place (see Goldsmith, 1993; John, 2000; Callanan & Tatham, 2013 in general; Van Bever et al., 2011 for Belgian local authorities). Meanwhile, studies who do address interest representation of SNAs tend to direct their attention to the regional level, often neglecting its local counterpart (Marks et al., 2002; Schultze, 2003). The few exceptions that have scrutinised the influence of local governments on EU decisions-making have not produced unanimous conclusions. Some particular cases seem to hint at possible success in very specific settings (e.g. De Rooij, 2003; Tosics, 2010; Kiers, 2014), yet on a more general level there is still scepticism about whether local governments are really able to make a tangible impact on their own (John, 1994; Greenwood, 2011: ). The goal of our research is therefore to analyse regulatory mobilisation towards the EU in the specific context of Flemish local government 2. In doing so, we take a broad perspective on local government, including for instance local networks, associations, etc. in addition to the individual local authorities. Two basic questions guide our analysis: (a) to what extent are local governments in Flanders able to exert real influence on European decision-making, and (b) if so: which ingredients form their recipe for success? In this paper we present the theoretical foundation of our research project. The paper continues by introducing the theoretical and conceptual framework of the project, which integrates theory of interest representation in the context of the Europeanization of local government. Section three outlines the opportunity structure of both EU decision-making and Belgian governance. Indeed at least from a theoretical point of view, both provide clear opportunities for the regulatory mobilisation of Flemish local governments in the EU. Section four then sets out the analytical framework by combining the theory of access with a specific approach to analyse influence on EU decision-making. Finally, the ensuing research plan is discussed is section five. 2 Flanders is one of the three Belgian regions. 3

4 2. Theoretical framework: the regulatory mobilisation of local government, a particular case of EU interest representation 2.1. The Europeanization of local government Basically, our research project is concerned with interest representation as one particular aspect of the Europeanization of local government. In this section we outline the latter concept while the general theory on interest representation is introduced in the following section. The concept of Europeanization has a wide variation of meanings and possible applications (Olsen, 2002; Hamedinger & Wolffhardt, 2010). In our research context it refers to the domestic impact of the development of European institutions, policies and identities (Olsen, 2002: 932). Radaelli (2000: 4) broadly defines Europeanization as processes of a) construction, b) diffusion and c) institutionalization of formal and informal rules, procedures, policy paradigms, styles, ways of doing things and shared beliefs and norms which are first defined and consolidated in the making of EU decisions and then incorporated in the logic of domestic discourse, political structures and public policies. 3 In order to apply the Europeanization concept in empirical analysis the researcher has to make explicit choices on two dimensions: (a) the objects of Europeanization (i.e. what is exactly Europeanised?) and (b) the dimension of change (i.e. to what extent is this happening?) (Radaelli, 2000). Regarding the former choice, we explicitly confine our research design to the regulatory mobilisation of Flemish local governments. The academic attention for the local perspective flourished from the 1990s onwards as the implementation of new EU regional policies drastically increased the interactions between cities and the EU (John, 2000; Hamedinger & Wolffhardt, 2010). The concept of Europeanization consequently emerged in the literature as one of the global trends in local government across the European continent (John, 2001; Denters & Rose, 2005). John (2000: 182) argues that Europeanization entails a fundamental transformation in which local decision-making becomes an integral part of the EU, while European ideas and practices find their way to the centre of local decision-making the other way around. As such, local or urban Europeanization is considered as a circular, dynamic and interactive process of interdependence between Europe and local government (Hamedinger & Wolffhardt, 2010). Hamedinger and Wolffhardt (2010: 28) accordingly define urban Europeanization as the interplay between actors and institutions on the European and the city level, which leads to changes in local politics, policies, institutional arrangements, discourse, actors preferences, values, norms and belief systems on both levels. The process of local (or urban) Europeanization comprises three broad dimensions (Kern, 2009; Hamedinger & Wolffhardt, 2010). The top-down, hierarchical or download dimension is probably the most visible and tangible dimension. It refers to the European legislation that has to be implemented 3 According to Hamedinger and Wolffhardt (2010), the definition of Radaelli forms one of the best starting points to understand and analyse the concept of Europeanization. The definition bears a close resemblance to the definition of Börzel and Risse (2000: 6), who conceive Europeanization as a process of change at the domestic level in which the member states adapt their processes, policies and institutions to new practices, norms, rules and procedures that emanate from the emergence of a European system of governance. 4

5 at the local level, as well as to the funding opportunities for local government which are created by the EU structural funds and other programmes. Following from this logic, it is the EU which takes the initiative to engage with local governments. The latter then have to comply with the European legislative output and/or meet the eligibility requirements set by the EU in order to receive funding. Second, a bottom-up, cooperative or upload dimension of local level Europeanization originated in response to the pressure of the EU on local government. As EU regulations increasingly impact on the local level, it is argued that local authorities better try to have a finger in the pie when these regulations are elaborated and negotiated in the first place (John, 2000). Consequently, the bottomup and top-down dimensions are part of the same vertical dynamics in local Europeanization. The bottom-up perspective is also strongly connected to the idea of multi-level governance, according to which the EU has provided a multiplication of extra-national channels of sub-national political activity (Hooghe & Marks, 1996: 73). Indeed, the multi-level setting implies that the EU operates as a dynamic decision-making system that involves different actors from different government levels (Hamedinger & Wolffhardt, 2010). For local governments the EU thus creates an additional arena and opportunity to proactively influence decision-making processes and promote their interests (De Rooij, 2002; Fleurke & Willemse, 2006; Kern, 2009). In bottom-up processes of Europeanization, it is the local authority which acts as the initiator of the interaction with the EU, either individually or organised in a network or partnership arrangement. This final point brings us to the third dimension of local Europeanization: the horizontal dimension. Ad hoc or formal organisations, networks or partnerships of local governments might not only offer promising opportunities to qualify for European funding, keep track of European legislation or aggregate local interests in the European arena. They could also form an asset for local governments per se. Via the exchange of best practices and the sharing of information, experience and resources, local authorities might enhance their individual capacity and produce collective innovative solutions for universal problems (Kern & Bulkeley, 2009) 4. From this overview we can conclude that Europeanization is basically a story of both opportunities and constraints for local government (Fleurke & Willemse, 2006). Two major opportunities have been singled out by Callanan and Tatham (2013: 4) under the label of financial mobilisation and regulatory mobilisation. Financial mobilisation is defined as the tracking and gathering of information with a view to accessing EU funding. Whilst this rather reactive strategy forms a part of the top-down dimension of local Europeanization, regulatory mobilisation is part of the bottom-up equivalent, being the proactive, dynamic process where ( ) local government interests seek to influence EU policy and legislative outcomes. Obviously, both processes can co-exist in practice. Not seldom is the financial mobilisation of municipalities a precursor or even catalyst of more proactive Europeanization strategies such as regulatory mobilisation (Goldsmith, 1993; De Rooij, 2003; Zerbinati, 2004). It is this regulatory mobilisation of Flemish local government we select as the object of Europeanization in our research 5. 4 This is why horizontal Europeanization is related to other general trends in local government such as urbanization and globalization (Bontenbal & Van Lindert, 2009). 5 As we have indicated in the introduction, this perspective has not often been taken in (local) Europeanization research yet. 5

6 The second fundamental choice has to do with the dimension of change in the process of Europeanization (Radaelli, 2000). Several authors have classified local governments according to their degree of Europeanization. Goldsmith and Klausen (1997; see also John, 2000), for instance, distinguished between counteractive, passive, reactive and proactive local governments. The first category refers to those authorities which are sceptical about EU affairs and deliberatively choose not to participate in Europeanization processes. Second, if passive authorities engage with Europe, this is more often a coincidence than a deliberative policy choice. Reactive governments, on the other hand, do have an interest in Europeanization dossiers and are keen on learning, even though their engagement usually builds further upon the initiatives from other actors. Finally, the proactive group is a fairly small one. It consists of those authorities which have a clear vision on their place in Europe and consequently take the lead in Europeanization processes. These four categories largely coincide with the four general outcomes of Europeanization that Radaelli (2000) has described: retrenchment, inertia, absorption and transformation. John (2001) classified the different aspects of local Europeanization as ascending steps on a ladder. In the minimal phase, local authorities merely seek to meet the EU s legal requirements. The second phase is financially orientated as the authorities aim to maximize their efforts in the competition for European funding. When local governments enter the third phase, they engage in networking with other authorities in the EU. In the final phase local governments are fully Europeanized: they incorporate EU policies in their internal policy agenda and enter in the process of EU decisionmaking. Even though one can question the strict sequential and cumulative logic of these steps, the ladder clearly indicates that the regulatory mobilisation of local government is one of the most exclusive processes of local Europeanization. In summary, our research will deal with those local governments (and/or their associations) which we can classify as proactive. These are the authorities which have evolved from passive policytakers to proactive policy-makers (Schultze, 2003). In the literature such authorities are still perceived as exceptions though (e.g. RoB, 2013 for the Netherlands; De Rynck, 1997 and Van Bever et al., 2011 for Belgium). They are pioneers, the bright stars that illuminate the EU galaxy (Balme & Le Galès, 1997; also Kern & Bulkeley, 2009). The determinants of proactive local governments are for instance local government size and structural position, available resources (e.g. finance, personnel, expertise and knowledge), but also motivations, perceptions and political and administrative policy entrepreneurs (De Rooij, 2003; Van Bever et al., 2009; Hamedinger & Wolffhardt, 2010; RoB, 2013) Interest representation in the EU The process of the regulatory mobilisation of local governments in Europe (i.e. local governments proactively seeking to influence EU decision-making) is an example of what the literature describes as interest representation. Many concepts are used as synonyms for interest representation, such as interest intermediation (Heinelt & Niederhafner, 2008), interest group action/politics (Coen, 2007) or lobbying (Hauser, 2011) 6. The EU (COM, 2008: 3) defines interest representation in its official Transparency Initiative as the activities carried out with the objective of influencing the policy 6 Some authors, however, apply the concept lobbying to the specific actions of private actors (e.g. Woll, 2007; Vass, 2008). 6

7 formulation and decision-making processes of the European institutions. Even though this practice contains both attempted and successful activities (Hauser, 2011), the ultimate goal of interest representation is obviously to achieve an outcome which is situated as close as possible to the own ideal position (Klüver, 2011). This goal can be reached via several activities in practice. For instance, interest representation may include agenda setting, issue framing, contributory thinking, detailed input to draft legislation, legitimacy and political support, opposition and consideration of policy measures and implementation (Greenwood, 2003: 27). This wide field of possible applications makes it is necessary to further confine the concept for our empirical analysis. Bouwen and McCown (2007) discern two general strategies in interest representation: legal and political strategies. The former are known as litigation and concern the appeal to the European Court of Justice (ECJ), seeking to have a court rule on the unconstitutionality or otherwise improper nature of legislative provisions in order to change policy (Bouwen & McCown, 2007: 526). Taking EU policies and regulations to court is a valuable strategy to defend one s interests at the end of the policy cycle (Bouwen, 2004; Coen, 2007; Greenwood, 2011). Nevertheless, we do not take these actions into account in our analysis. Instead, we solely focus on the political strategies of local governments. In this category Beyers (2004: ) makes a further distinction between voice and access. Voice denotes the public political strategies to influence EU policy and decisions, for instance via public protest or media campaigning. These strategies are directed towards a broader audience in order to indirectly draw the attention of the decision-makers 7. Voice strategies differ with regard to their primary target audience. Information politics (e.g. a press conference on a strategic moment in the policy cycle) is mainly aimed at signalling information at particular audiences (e.g. certain public constituencies or key decision-makers). Protest politics (e.g. public demonstrations), on the other hand, deliberatively stage an event to enhance public conflict and raise more general awareness amongst decision-makers about a policy issue. Access then refers to the exchange of policy-relevant information with public officials through formal or informal networks (Beyers, 2004: 213). This form of interest representation is also known as inside lobbying in the EU decision-making arena. In this process, interest groups deliver information to EU decision-makers with the explicit intention of influencing EU policies and/or regulations (Bouwen & McCown, 2007). Usually, this takes place behind the scenes (e.g. in political or administrative committees, advisory bodies, etc.) through contacts between interest groups and policy-makers, thus far away from public attention. Albeit these different forms of interest representation are often part of a more comprehensive approach, and a combination of different strategies probably augments the overall effectiveness of the lobby action (Beyers, 2004), we confine our scope of research to the access strategies of Flemish local governments. According to Bouwen and McCown (2007), this is also (still) the most common and preferred form of interest representation in practice. In section four of the paper, we further elaborate on the access-theory in order to construct our analytical framework. Moreover, our aim is to mainly focus on legislative lobbying (i.e. the making of legislation in a policy area at the European level ; Bouwen, 2002: 366) as part of the regulatory mobilisation of local 7 As such, voice strategies strongly resemble the concept of advocacy, understood as any awareness-raising activity addressing the public or decision-makers (Vass, 2008: 72). 7

8 government 8. Pedler (2002: 322) in fact confirms that it is above all the regulatory nature of the EU which promotes EU interest representation. Yet not every legislative lobby has the same objective in mind. Some are positive lobby efforts, seeking to convince policy-makers to include or enact a favourable passage or regulation/directive. Others are negative lobby efforts: they mainly try to block unfavourable passages or regulations/directives (Hauser, 2011: 682). Furthermore, a lobby strategy might be proactive or rather reactive (RoB, 2013). Whereas the former tactics imply that local governments take the lead in placing local issues on the EU agenda, the latter entails that they go along with the possibilities offered (or granted) by the EU in the decision-making process. Another distinguishing feature in the process of interest representation is the nature of the interest at stake. Michalowitz (2007; also Greenwood, 2003) opposes cases of high politics (which are much harder to influence) against cases of low politics or day-to-day interest representation. Second, Beyers (2002 and 2004) makes a distinction between specific interests and diffuse interests according to the constituencies the interest groups represent. Specific interests represent a well-defined constituency and are therefore able to put forward more clear and uniform demands. Diffuse interests, on the other hand, represent a broader constituency. They find it more difficult to formulate such clear, to the point positions, hence make a strong impact on the decision-making process 9. Third, Vass (2008: 74) classifies the interests in EU lobbying on two axes. On the one hand, European policies have to weigh up the national interests and the European interests against each other. On the other hand, public interests often oppose to their private counterparts. Finally and in addition to the above, Coen (2007) argues that EU interest representation is also determined by the nature of the policy domain under consideration. For instance, the importance of Enterprise, Environment, Agriculture and Health in EU policy is reflected in the higher degree of lobby efforts these domains are subject to. The number of interest groups drastically increased from the late 1980s and early 1990s as a result of the growing EU competencies in the wake of the creation of the EU s single market (Woll, 2007; Greenwood; 2011; Hauser, 2011) 10. Woll (2007: 465) defines these groups as formally organized groups who are united by specific political objectives and who try to influence the policy process in the pursuit of these goals. The most numerous and influential groups in the EU lobby arena are private business groups (Bouwen & McCown, 2007; Hauser, 2011; Klüver, 2012). Territorial interest groups such as local government, on the other hand, are strictly speaking not even interest groups, but rather non-group actors (Greenwood, 2011) 11. However, many authors agree that when local governments engage in explicit actions of regulatory mobilisation, they are not any different from the traditional interest/lobby groups (John, 2000; Marks et al., 2002; Lobbying in the European Union, 2005; Heinelt & Niederhafner, 2008). Bulmer (2007: 801) sums up this argument as follows: 8 Most authors do not distinguish explicitly between influence attempts towards EU regulation and EU policy (see Bouwen & McCown, 2007; COM, 2008; Callanan, 2012; Callanan & Tatham, 2013). 9 We find a similar classification in the work of Greenwood (2011) who discerns broad horizontal interests from sectoral and even specialist interests. 10 The Commission s 2001 White Paper on Governance (and its attempt to enhance democratic legitimacy in EU decision-making) further boosted the activities of interest groups in the EU (Greenwood, 2011). The rapid increase in interest groups was matched by analogous academic attention (Woll, 2007), even though Pedler (2002) notes a strong Anglo-Dutch bias in this field of inquiry (as there is in lobby practice as well). 11 Specific interest associations of local government, such as Eurocities, however, do qualify as genuine interest groups (see further in section four). 8

9 Of course, local authorities are not normally regarded as interest groups. However, in their relation with the EU authorities, this is the best kind of parallel to make for their activities. Usually, the interest representation of local government is framed under the wider umbrella of the sub-national authorities (SNAs; Hooghe & Marks, 1996; Bomberg & Peterson, 1998; Fleurke & Willemse, 2006; Callanan & Tatham, 2013) or sub-state entities (SSEs; Tatham, 2010). The actions of local government and their associations or networks are then put on a par with those of their larger, regional equivalents 12. Nevertheless in practice, there is often an obvious cleavage between local and regional interests (Marshall, 2005; Greenwood, 2011; Callanan & Tatham, 2013) 13. This is why academics stress the need to consider interest representation of local government as an end in itself (Marks et al., 2002; Schultze, 2003; Marshall, 2005; Heinelt & Niederhafner, 2008). However when it comes down to assessing the actual effectiveness of territorial interest representation in the EU, Callanan and Tatham (2013) argue that it probably makes more sense to differentiatie between strong regions and strong local government associations on the one hand, and weak regions, weak local government associations and individual local authorities on the other. 3. Contextual setting: the European and Belgian opportunity structure 3.1. Decision-making in the EU: in search for knowledge and support The process of interest representation is largely determined by the political system in which it unfolds (Greenwood, 2011). As institutions drive this process, interest groups should adjust their agenda to the opportunity structure in place 14. And even though the system of EU interest representation has certain characteristics in common with other political systems, it is certainly different for some particular aspects as well (Woll, 2007; Greenwood, 2011). For Bouwen and McCown (2007) the distinguishing feature of the system of EU interest representation is the sheer number of opportunities it presents to interest groups which seek to influence the decision-making process. This makes EU interest representation at the same time very diverse and complex, hence difficult to generalise (Bouwen, 2002). Decision-making in the EU is generally considered as very open, accessible and impressible (Vos in Van Moerkerke, 2009; Callanan, 2012). Due to its multi-level character, the EU is a political system without fixed centre of authority, hierarchy and decision-making (Orbie, 2009; Greenwood, 2011). Consequently, each dossier is up to negotiation in different institutions and possibilities for interest representation increase accordingly. Meanwhile, power in this process is dispersed over different institutions and actors whilst shifting according to the particular phase in the policy process. In the phase of agenda setting and policy design, the European Commission is the primary institution to turn to. Subsequently, interest groups can seek to influence the policy discussions and adjustments in 12 This narrative is also present in the EU system, which for instance clusters both local and regional interests in its Committee of the Regions. 13 Over and above this distinction, there is a difference in local interests according to the local system and the type of local authority as well (Callanan & Tatham, 2013). 14 Hauser (2011) presents a good example of this logic: in the aftermath of the Lisbon treaty, he predicted a strong increase in lobbying activities because the expansion of the system of qualified majority voting would increase the EU s legislative output. 9

10 both the European Parliament and the Council. Finally, interest groups might influence the implementation and monitoring phase by addressing the member states and the Commission (Coen, 2007; Hauser, 2011; Greenwood, 2011; Callanan, 2012). As we have already mentioned in the introduction of the paper, the inclusion of interest group activity in European decision-making bears directly upon the democratic legitimacy of the EU system (Greenwood, 2011; Hauser, 2011). A legitimate political system combines democratic decisionmaking and participation on the input side with effective governance on the output side (Scharpf, 1999). In terms of output legitimacy, the relative small size of the Commission and Parliament makes these institutions hugely dependent on outside resources such as information, expertise and technical knowledge in order to draft and amend legislation (Coen, 2007; Heinelt & Niederhafner, 2008; Vos in Van Moerkerke, 2009; Hauser, 2011; Klüver, 2012). This is why EU institutions actively call for policy input from domestic stakeholders, and stimulate and support a wide range of interest groups (Greenwood, 2011). For interest groups, this resource dependency presents an outstanding opportunity to influence European decision-making by providing expert knowledge in certain dossiers (Orbie, 2009). There is little doubt that SNAs can provide such information concerning policy issues with a local dimension. Similarly, cities can give feedback to the Commission about the implementation and application of EU regulations in practice (i.e. assuming the watchdog function; Heinelt & Niederhafner, 2008; also Greenwood, 2011). According to Bomberg and Peterson (1998), the concern to deliver efficient policies also underpinned the introduction of the subsidiarity principle, which implies that power should be exercised at the lowest possible level. Yet simultaneously, subsidiarity was meant to augment legitimacy on the input side of the EU system by reducing the democratic deficit in decision-making. The latter refers to the gap between the remote EU institutions and EU citizens which can be reduced by incorporating a wide range of citizens, stakeholders and organisations in EU decision-making to compensate for the EU s lack of direct democracy (Schultze, 2003; Woll, 2007; Greenwood, 2011; Hauser, 2011). For the EU, interest groups are therefore the natural constituencies of the system, a proxy for civil society (Greenwood, 2011). One of the paramount examples of this logic are exactly the territorial interests or SNAs (i.e. regions and cities). In fact, no interest group is better placed to meet the legitimacy requirements on the input side of the EU system on the basis of the SNAs status of elected (hence accountable) government closest to the people (Heinelt & Niederhafner, 2008). The involvement of SNAs in EU decision-making was echoed in policy initiatives such as the White Paper on European Governance, the structured dialogue with regional and local government and the subsidiarity definition in the Lisbon Treaty (Callanan, 2012) 15. The above overview demonstrates that the EU keeps an open mind on (territorial) interest representation. We could even claim that interest representation is both a necessary and healthy asset for the quality of decision-making in the EU (Lobbying in the European Union, 2005: ix). However, the systematic inclusion of interest groups might also pose a threat for democracy when particular groups become overly influential (Michalowitz, 2007: 132; also Lobbying in the European Union, 2005 and VASS, 2008). In this way, the heavy resource dependency of the EU, the great distance between the EU and the member states and the complex EU system which requires a huge 15 Another, more strategic political reason why the EU might welcome territorial interests in decision-making processes is the intention of reducing the power of nation states in order to enforce the position of central European institutions (Greenwood, 2011). 10

11 investment from interest groups to play their game could contribute to such unequal access and influence (Hauser, 2011). However, the EU tries to prevent this from happening by combining a pluralistic system of checks and balances and fragmentation of power with the greatest possible openness on the process of decision-making and interest representation (Greenwood, 2011). The concern for pluralism is also the reason why the EU prefers the aggregated opinion of local government associations over individual positions of cities and regions (Heinelt & NIederhafner, 2008; Callanan, 2012; this claim applies to other interest groups as well, see Greenwood, 2011). Taking these elements into consideration, the conclusion of some observers on EU interest representation might even be that the very nature and dynamics of EU decision-making produces more democratic quality than traditional political systems (Orbie, 2009). Yet also irrespective of the normative dimension of this discussion, one thing seems for sure: the EU has created a political system in which interest groups can have a tangible impact on the decision-making process (Greenwood, 2011) The Belgian context: fertile ground, shifting cultivation? In addition to the European context in which policy input from domestic stakeholders is welcomed or even actively pursued, the national tradition matters as well for subnational interest representation (Hooghe & Marks, 1996; Bomberg & Peterson, 1998; Woll, 2007). In this respect, some elements of the Belgian political culture and structure could undoubtedly favour local interest representation in the EU. Beyers (2002: 585) calls this the fertile ground for possible successful Europeanisation. He sums up a number of elements that comprise this fertile ground (2002: ). One is the profederalist ideology towards Europe which is shared by the majority of the Belgian political elite 16. In combination with less pronounced feelings of national sovereignty, Belgian politicians thus should have an open mind towards Europe. Also the Belgian politico-administrative system has developed noticeable familiarity with EU policy-making, partly because of the longstanding involvement in EU matters as one of the EU s founding members and the experience with internal multi-level governance and multinational/linguistic politics. In addition to these apparent political and administrative connections, we could add the (reasonably) positive attitude of the Belgian population towards the EU as an underpinning or legitimising factor. Indeed from a comparative perspective (see results from Eurobarometer), the Belgian population tends to display higher levels of awareness, trust and attachment towards the EU. Beyers (2002: 589) further mentions two practical elements which could encourage Belgian interest representation in the EU. The first and most obvious one is the political but also geographical reality that Brussels is at the heart of EU decision-making. It is no overstatement that the centres of national/regional and European decision-making are literally within walking distance. The second practical element applies to many smaller member states in the EU. Due to their limited impact in dossiers of qualified voting, these states might be less empowered to defend particular domestic 16 Not surprisingly, many Belgian politicians have played a leading role in the history of European integration. Still some parties have developed a more Euro-critical discourse in recent years (even though we could qualify them as fairly moderate compared with the paramount examples of true Eurosceptic parties in the EU). 11

12 stakeholder issues. As a result, the stakeholders could be more encouraged to actively pursue participation in the EU decision-making process. To conclude this overview of contextual opportunities, we add two further archetypical elements of the Belgian political culture. First, the institutionalised and widespread system of multiple officeholding creates strong, direct and often personal links between the different government levels in Belgium. Politicians at the national, regional or provincial level regularly combine their mandate with a local office 17. Even European members of parliament are allowed to combine their office with a seat in the council of their municipality. Such linkages offer local government the most direct possibility to promote their interests in EU policy-making. Second, this feature is complemented and supported by another earmark of the Belgian political system: the extensive degree of party government. The latter entails that many policies are (at least partly) determined in the party headquarters; i.e. in the corridors of the formal circles of decision-making. For local politicians, it might be a fruitful strategy to raise their matter of concern with politicians involved in European decision-making, or politicians connected to its inner circle, via the central party channels. Yet in spite of this fertile ground, empirical evidence still labels Belgium as a slow starter in terms of the Europeanization (and particularly the regulatory mobilisation) of its individual local governments (De Rynck, 1997; Van Bever et al., 2011) 18. A possible reason for this hesitation might lie in other facets of the structure and culture of Belgian local government. With an average population of nearly inhabitants, Belgian municipalities are quite small to act as individual authorities in the European arena. The large majority of them do not have a full-fledged strategy, or even the intention and knowledge, to try to influence European decision-making in a systematic way either. And thirdly, municipalities room for political manoeuvre might be curtailed by the ambition and assertiveness of the strong Belgian regions. This hypothesis has also been raised by Greenwood (2011): regions with strong autonomy tend to use the EU-scene for intra-state profiling. This often happens at the expense of the local government level below the surface. A final line of explanation could be more cultural. Many insiders mention the poor lobby culture in Belgian local government as one of the most important reasons for the lack of organised regulatory mobilisation in Europe. In this sense the close geographical distance to the centre of EU decision-making might generate average effects. When the informal lobbying behind the scenes really gets going after a EU-meeting, many Belgian representatives already went back home again Summary: the balance of probabilities To summarise the opportunity structure of EU interest representation for (Flemish) local government, we draw on the balance of probabilities that has been sketched by John (1994: ). John singles out five positive factors and seven potential drawbacks (in the UK context). The five positive factors seem valid in the Belgian/Flemish setting as well. They include (a) the democratic surplus of local government as elected institution with corresponding legitimacy (which fits well into the Commission s quest for legitimacy in order to reduce the democratic deficit in EU decision- 17 It is prohibited, however, to combine two executive offices across different government levels. 18 John (2000) for instance notes how such conclusions are remarkable in the Belgian setting, particularly from a comparative European perspective. 12

13 making); (b) the opportunistic motive of the Commission to involve local stakeholders in the decision-making process as a means to bypass central states; (c) the considerable local knowledge and expertise needed to enhance the quality of EU policies on the output side of the system; (d) the strong involvement of local authorities in local, supra-local and transnational networks (in fact, the EU considers this networking principle of paramount importance); and (e) the inclusion of local politicians in political networks (fostered by the Belgian culture of multiple-office holding and party government). John, however, also mentions seven potential drawbacks for successful local government lobbying. The first four elements can be copy-pasted to the Belgian/Flemish context. They refer to: (a) the nature of the interests of local authorities (i.e. often involving general interests due to the multifunctional character of local government), which hampers competition with more specialised interest groups; (b) the strong role of national government in EU decision-making, which often makes local governments dependent on their relation with national politicians and institutions for gaining access to the EU lobby arena; (c) the political culture of local government, in which the large majority of the municipalities are foremost concerned with obtaining European funding instead of proactive interest representation; and (d) the poor financial situation of the local governments which constrains proactive strategies (e.g. establishing a lobby office in Brussels). The three other elements in John s balance of probabilities are not entirely in tune with the Belgian/Flemish context. First, John mentioned the UK local government structure at that time, i.e. lacking strong elected regional structures as emancipatory channels for local government. In the Belgian context, strong regions are important political actors. However, it is often claimed that their emancipation might exactly keep local government under the thumb. So also in our research context, the structure of local government, additionally marked by the small average size of the municipalities, might be an impeding factor for local interest representation. Second, he argued that the collegiate nature of local government (at the time of writing) impedes proactive bargaining and networking from a political leader in the city hall. Yet whilst Belgium formally still has such a collegiate local government system, entrepreneurial mayors tend to be capable of assuming a leading role in different kind of networks in practice. Finally, John referred to the poor links of the UK within EU institutions. Such claim seems less valid in the Belgian/Flemish context, since politicians and institutions have established strong ties and positive attitudes towards the European institutions. 4. Analytical framework: from access to influence Thus far, we have outlined the theoretical and conceptual framework of our research, as well as the overarching contextual setting. The goal of our research is to analyse the regulatory mobilisation of Flemish local governments, considered as the access strategies towards EU institutions in order to influence the decision-making process (i.e. excluding financial mobilisation and strategies of interest representation such as litigation and voice). In this section we further elaborate on the access-theory and influence as its ultimate goal. The integration of both perspectives comprises the analytical framework of our research design. 13

14 4.1. Access: the economic market of demand and supply Access is understood as the channelling or exchange of policy-relevant information through formal or informal networks with public actors (Beyers, 2002: 587). In this respect, it is important to note that access does not equal influence 19. Access is not the end in itself, but often a necessary condition for the ultimate goal of interest representation, i.e. exerting influence 20. Yet it is clear that there is a close connection between access and influence in practice (Beyers, 2002; Bouwen, 2002; Hauser, 2011). Hauser (2011; also Bouwen, 2002 and Bouwen & McCown, 2007) compares access to the EU as a market of supply and demand. On the one hand, interest groups wish to approach the EU institutions in order to obtain influence in EU legislation. The EU, on the other hand, is keen on technical information and support from interest groups in order to enhance its input and output legitimacy (i.e. resource dependency). Consequently, access rests on an exchange relation of (a) information as the central access good between interdependent actors on (b) the demand side and (c) the supply side of the system of interest representation (Bouwen, 2002). (a) The access good: information The transmission of information is central to the process of interest representation (Klüver, 2012). Indeed in a complex policy environment, EU decision-makers need detailed, real-life information to fully grasp policy issues and foresee the possible outcomes of regulation. Meanwhile, stakeholder input also serves to legitimise these decisions. Due to their daily involvement in certain policy issues, interest groups thus possess both the necessary expertise, affinity and legitimacy to transfer information to EU decision-makers. Bouwen (2002: ) discerns three basic types of information. The first type is expert knowledge (EK). Expert knowledge refers to the specific and technical know-how (i.e. expertise) which is needed to develop regulation in a given policy domain. Since expert knowledge mainly serves to enhance the quality of EU policies, it relates to the output legitimacy of EU decisionmaking. The two other types of information are labelled information about the European encompassing interest (IEEI) and information about the domestic encompassing interest (IDEI). They concern the aggregation of the individual interests of domestic or European types of actors (e.g. local governments). The encompassingness of an interest is determined by the representativeness of an interest group (based on the density of its membership) and its organisational domain (delineating which members are represented). Both the IDEI and the IEEI are important assets to augment support for EU decisions, hence bear upon the input legitimacy of EU decision-making. According to Klüver (2012), interest groups may gain influence in the decision-making process if they are able to provide and filter the information which is used by the EU to substantiate policies and regulations. The amount of information an interest group is able to provide is one of the possible 19 Bouwen and McCown (424: 425) for instance argue that while it is obviously possible to gain access to EU decision-makers without succeeding in changing the policy outcome in the end, there is also a possibility to gain influence by means of other lobby strategies such as voice. 20 This is why access can be regarded as a facilitating intermediate objective in interest representation (Truman in Bouwen, 2002: 366). 14

15 determinants of interest group influence (Klüver, 2012). It depends on the interest group s material resources (e.g. money, staff) and internal organizational structure (i.e. professionalization, decentralized decision-making structure and functional differentiation), as well as the complexity and salience of the policy issue (i.e. complex and salient issues generating more information supply). For Bouwen (2002: 382), the primary earmark of access is to provide the highest quantity and quality of the critical access good in the most efficient way. The critical resource or access good is the particular type of information which is of paramount importance for an organisation. In EU decision-making, the critical resource varies according to the access points on the demand side of the market system. (b) The demand side: access points On this demand side, the institutional structure of the EU and its decision-making process offer interest groups a broad and diverse range of access points (Beyers, 2004). The classic triad of EU access points comprise the European Commission, the European Parliament and the Council of Ministers. Since each institution plays its particular part in the legislative process, each is also dependent on and receptive to particular types of information (Bouwen, 2002 and 2004b; Hauser, 2011). Interest groups respond to this diversification by adjusting their strategies according to the relevant institution (Beyers, 2004). Nevertheless for a complete account of influence on EU decisionmaking, researchers have to consider the three institutions simultaneously (Bouwen, 2002). The Commission (i.e. commissioners, cabinets, directorates-general, expert groups, rapporteurs, committees, etc.) is the most important access point for interest groups (Pedler, 2002; Greenwood, 2011). In its capacity of initiator/designer of legislation and monitor of policy implementation, the Commission is very dependent upon (hence actively seeking) input from these groups. Greenwood (2011) even calls them the natural constituency of the Commission. Also cities and SNAs are explicitly targeted by the Commission as important stakeholders (Schultze, 2003). Furthermore, for interest groups the formative stage of the policy proposals is the best moment to make a considerable impact on the legislative process (Pedler, 2002). Given the nature of the Commission s competences, it tends to prefer expert knowledge and specific interests over more broad and general counterparts (Bouwen, 2002; Beyers, 2004) 21, albeit authors argue that the Commission is more receptive to aggregated input than individual interests in the case of SNAs (Heinelt & Niederhafner, 2008; Callanan, 2012). The Commission can be easily approached via online consultations, but also informal contacts are possible access strategies (Klüver, 2012). The second access point is the European Parliament (i.e. national MEPs, other MEPs, committees, intergroups, rapporteurs, etc.). Authors notice an increased importance of the parliament as an access point for interest representation following from the establishment (and extension) of the codecision procedure and the parliament s formal powers of policy amending and voting (Bouwen, 2002 and 2004; Pedler, 2002; Beyers, 2004; Heinelt & Niederhafner, 2008). The EP is highly 21 As the most supranational European institution (i.e. promoting European interests), however, and in order to secure the input legitimacy of its policy proposals, the Commission is also interested in information about the European encompassing interests (but less in information about the domestic encompassing interests) (Bouwen, 2002: ). 15

16 susceptible to outside influence because it has no in-built political majority and each majority thus has to be built from scratch (Vos in Van Moerkerke, 2008; Greenwood, 2011). As a forum for political discussion in the legislative process, the EP combines supranational and intergovernmental traits. This affects the preferred or requested type of information (Bouwen, 2002). On the one hand, the parliament is expected to evaluate Commission proposals through a European perspective while the decision-making procedure requires supranational coalitions as well. These traits make information about the European encompassing interest the critical resource for the parliament. Yet on the other hand, the MEPs are still elected in their home country. People therefore also expect them to promote national interests. This intergovernmental trait increases the importance of information about the national encompassing interest for MEPs (Bouwen, 2002 and 2004) 22. Arguably, cities are accurate channels to provide MEPs with information on the domestic policies and discourses (Heinelt & Niederhafner, 2008). The third access point of the triad of European interest representation is the Council of ministers (i.e. member states, national ministers, COREPER, the EU presidency, etc.). Lobbying the Council derives from its power in co-decision procedures and its position as the EU s supreme decision-maker (Bouwen, 2002). However, the Council is deemed as the least influential access point in terms of interest representation (Greenwood, 2011). Rather than the council as such, the individual member states are subject to lobby attempts. This is why national interest groups tend to have better access to the Council, and informal contacts and internal state relations play an important role in Council lobbying. Greenwood (2011) further mentions two possible reasons for the relative weakness of the Council as lobby subject. First, the potential scope of lobby influence is already strongly reduced when a policy issue reaches the Council. Second, the Council is the most intergovernmental institution in which member states have to find consensus decisions. This makes purposive lobby attempts less successful. In recent years the importance of the Council as access point has further decreased because of the partial power transfer to the EP and the increase in qualified majority voting which reduces individual state veto power (Bouwen & McCown, 2007). However, this does not imply that the council has no importance for interest representation at all. Indeed, the Council is often the last resort to exert pressure on the decision-making process (Beyers, 2004). Meanwhile, the political function of the Council as an arena to defend the national political positions makes it highly susceptible to information on the domestic encompassing interests (Bouwen, 2002). (c) The supply side: access routes Interest groups who seek to access EU decision-making processes can take different routes to transmit their information to the relevant institutions. Together, these constitute a comprehensive and varied lobby toolkit. Several ways exist to classify the possible routes to Europe (e.g. discerning between direct and indirect routes, extra-state and intra-state routes, individual and collective routes, institutional and non-institutional routes, etc.). In our research we apply the analytical framework of Beyers (2002: 595). Beyers distinguishes between the domestic level and the European level of interest representation. Together they form three different (but non-exclusive) routes for 22 Given its status as elected institution, the EP thus favours political arguments over technical input and expert knowledge albeit the latter is not without any importance either (Beyers, 2004). 16

17 interest representation 23. In the context of local government, these routes are: (a) an indirect route via domestic supra-local actors; (b) an indirect route via European local interest associations; and (c) a direct route from local actors to the EU access points (i.e. the Euro-level supra-local actors) (see also Figure 1). The choice between the direct route and the indirect domestic route is strongly shaped by the intrastate relations amongst different levels and political parties (Hooghe & Marks, 1996; Tatham, 2010). Often, multi-level governance and local Europeanization are presented as opportunities for SNAs to bypass their nation state in order to promote local interests directly at the European level (Fleurke & Willemse, 2006; Heinelt & Niederhafner, 2008). Whereas such bypassing might simply go unnoticed (and is tolerated by national actors), there is also the possibility of plain conflict (Tatham 2010 and 2013). SNAs and national actors then work separately in their attempts to achieve opposing objectives (Callanan & Tatham, 2012). Moreover in addition to these forms of state bypassing, it is also possible, and sometimes even more rewarding, that SNAs cooperate with national actors to gain access to the EU (Bomberg & Peterson, 1998; Beyers & Kerremans, 2012; Callanan & Tatham, 2012). Both cooperation and the two forms of bypassing are part of what Tatham (2010: 77) calls paradiplomacy, diplomacy led by SSEs parallel to their member states. The indirect domestic route in the Belgian multi-level setting includes politicians and administrators at the provincial, regional and federal level. Due to the Belgian political culture of extensive multipleoffice holding and party government, these routes are also complemented with many informal contacts between actors from different levels. Moreover, Flemish government has also established VLEVA in 2005, a liaison office which facilitates and promotes contacts between domestic actors (such as local government) and EU institutions. The direct domestic route entails informal contacts between local actors and the EU, but also formal city lobby offices in Brussels. In Flanders, however, no individual local government has already established such office. Furthermore, municipalities might join forces in different sorts of networks as well. Formal networks of inter-municipal cooperation aim to enhance the governmental capacity of local authorities, and some also engage in local interest representation in the EU. Finally, the direct domestic route contains two formal institutions as well. The VVSG is the official association of Flemish local governments. The Knowledge Centre for Flemish Cities is explicitly directed towards the bigger cities. Both institutions aim to raise public awareness about EU affairs amongst their members whilst the VVSG also tries to play its part in terms of EU lobbying under the European umbrella of the CEMR. The latter is part of the third access route to Europe: the indirect European route of local actors. This category refers to the different sorts of networks and associations of local government interests organised at the European level. First, a wide range of (formal and less formal) cross-border networks has been developed. The 37 EGTCs (European Grouping of Territorial Cooperation) are a good example of such networks. Some are directed towards a specific programme, while others are more institutionalised as political actors. A second category consists of the sectoral organisations such as 23 In his model, Beyers analysed the interest representation of private actors. As we have indicated in the theoretical framework of EU interest representation, we can equate local governments with traditional private interest groups with regard to their lobby efforts. Therefore, we replace private actors with local actors and public actors with supra-local actors in our framework. 17

18 Eurocities (Greenwood, 2011). Eurocities is a transnational city network established in 1986 which represents the interests of larger cities in Europe. This specific perspective renders Eurocities very capable of delivering expert knowledge to the European access points. However on the other hand, its particular scope comes at the expense of input legitimacy since Eurocities is no interest representative of local government as a whole (Heinelt & Niederhafner, 2008). The latter criterion is less of a problem for the CEMR, a third possible actor in the European route for local actors. The CEMR was founded in As the international umbrella of the local government associations (such as VVSG), the CEMR is the peak organisation of local government in EU interest representation (Greenwood, 2011). In this respect, the CEMR aims to close the gap between the EU and individual municipalities in Europe and beyond by providing information about the European encompassing interest. This broad and general interest, however, prohibits the CEMR from taking strong and pronounced positions in the lobby game (Heinelt & Niederhafner, 2008). A similar critique applies to the Committee of the Regions (CoR), the EU s assembly of regional and local representatives 24. This committee has been set up by the European Commission in 1988 as an advisory body on local and regional policy issues in the decision-making process (Hooghe & Marks, 1996). Given its inherent problem of having to take lowest common denominator-positions, the Commission is not considered as very influential in practice (Vos in Van Moerkerke, 2008; Greenwood, 2011). Moreover, the Committee above all seems to take a regional perspective (Heinelt & Niederhafner, 2008) 25. Finally, local interests may be advocated by the CLRAE (Congress of Local and Regional Authorities in Europe) of the Council of Europe. This institution was established in 1957 and serves as an official advisory body for the Council of ministers and the European Parliament (Greenwood, 2011). Figure 1. Local government access routes to Europe (based on Beyers, 2002: 596) 24 The CoR is sometimes tabled as one of the EU access points because of its formal institutionalised position in EU decision-making. From the viewpoint of EU interest representation, however, the CoR also qualifies as a particular interest group. 25 In Belgium, for instance, the Committee members are chosen by regional government. No exclusively local representatives are consequently incorporated in the Committee. 18