In the 1980s and early 1990s, several authors used ‘policy networks’ to explore EU policy-making, often tentatively.[1] Helen Wallace (1984: 141) talks of ‘the emergence of horizontal policy networks which cut across national boundaries’. Simon Bulmer (1994: 14) claims that subsystem government is ‘well established in integration studies’ ever since the publication of Wallace, Wallace, and Webb (1977). Dudley Coates’s (1984: 158) account of the foods standards regulatory policy similarly concludes that implementation is ‘heavily dependent on a rather small policy community’ (see also Lewis and Wallace 1984). He notes that ‘the complexity of the institutional arrangements does not seem to inhibit the process’. Rhodes (1986c) examines the relationship between the EU and local governments, points to the triadic links between central, local, and supranational governments, and suggests there are ‘emergent’ policy networks in some policy areas. These early studies have four features in common. First, they focus on policy implementation and not policy initiation. Second, they stress the large differences between policy areas. Third, they talk of incipient or emergent networks, not settled policy communities. Finally, they all recognize the need—in a complex, intergovernmental, policymaking system—to aggregate and coordinate the many affected public and private interests. Networks do not recognize institutional boundaries. Policy emerges from the struggle between government and non-governmental organizations (see Atkinson and Coleman 1992; Rhodes 1995b).[2]

The 1990s saw more interest in ‘policy networks’, mainly to describe EU policy-making. Thus, Peters (1992: 77) provides a broad account of EU policymaking, arguing that it is ‘best understood as bureaucratic politics’ and such decision-making takes place in ‘policy communities’. The EU executive is fragmented and ministers of functional departments are ‘involved in games over particular policy interests’ (Peters 1992: 79). EU policy-making is both differentiated and specialized and ‘many policy communities or networks appear to exert great influence, if not control, over public policy, more than in most national governments in Europe’ (Peters 1992: 81). Three interlocking games recur in EU policy-making: the national game in which each member state tries to maximize its return from EU membership; the institutional game in which the institutions ‘seek to gain more power relative to others’; and the

bureaucratic game in which the Directorate-Generals of the Commission (DGs) ‘compete for policy space’ (Peters 1992: 106-7). For Peters, differentiation, specialization, interdependence, and bureaucratic politics characterize EU policy-making. The problem is that his account is as brief as it is broad; for example, he does not explain the links between the policy communities and his three games.

Mazey and Richardson (1993: 4) are both more focused and more cautious. They explore interest groups in EU policy-making and suggest that, although ‘some interests have managed to become part of a cohesive policy community at the EU level’, the majority ‘are involved in less integrated types of policy networks’. There is ‘no dominant model or style of EU-group relations’ and the ‘procedural ambition’ of the Commission to set up ‘stable and regularised relationship with the affected interests’ (Mazey and Richardson 1993: 9) remains just that, an ambition.[3] They conclude:

at the EU level there are indeed quite significant variations in the nature of policy networks (and... in some specific policy areas no network may exist) but there is at least a case to be made that the network concept is quite useful

(Mazey and Richardson 1993: 253).

Marks (1992, 1993, 1996) focuses on EU structural policy, arguing that since 1988 the European Regional Development Fund (ERDF) has been ‘creating policy networks that encompass sub-national governments and private interests in individual regions’ (Marks 1992: 192). In effect, the reforms are an exercise in ‘institution building that strengthens the Commission’ and attempts to ‘technocratize—and in a narrow sense depoliticize—a key growing area’. Also, the reforms create direct links between the funds and regional political institutions, and so challenge ‘centralised decision making within member states’ (Marks 1992: 212). The several layers of government are ‘enmeshed in territorially overarching policy networks’ (Marks 1993: 392 and 402). Marks (1993: 407) generalizes his conclusions about structural policy to cover intergovernmental relations in Europe.[4] He claims:

we are witnessing the emergence of multi-level governance in the European Community, characterised by co-decision-making across several nested tiers of government, ill-defined and shifting spheres of competence (creating a consequential potential for conflicts about competencies), and an ongoing search for principles of decisional distribution that might be applied to this emerging polity.

Even in this initial formulation, there were two problems with Marks’s analysis (and see the Afterword at the end of this chapter). First, although the links between levels of government multiply, they are not necessarily an effective challenge to centralized decision-making. Marks’s evidence most commonly refers to the participation of local and regional governments in decisionmaking and not to their effect on the outcomes of decision-making. Yet, there is great variation between member states in the effectiveness of subnational participation.

Second, Marks deliberately avoids the theory-laden notion of networks. The phrase ‘multi-level governance’—from now on MLG—describes the changing structure of EU government. It refers to the policy-making system as a whole and is used to draw attention to the common feature that many functional policy networks involve the participation of several levels of government. It is not used to explain either variations in that structure or why it has changed. Marks (1996: 399 and n. 10) eschews ‘a theory-impregnated conceptualisation’ of ‘network’, using the term to refer to ‘a more or less stable set of political relationships among actors’. Networks are central to his analysis of MLG because he compares the ‘diverse array of networks across individual member states’. Instead of looking to resource-dependence to explain this diversity he looks to variations in the national policy style and the different functional characteristics of each stage of decision-making to explain differences in local and regional participation.[5]

In the 1990s, there was an extensive literature on the links between local and regional government and the EU.[6] Peters (1992: 112) suggested the EU was developing ‘“picket-fence federalism” and intergovernmental relations similar to those in the United States’. The top-down conception of intergovernmental relations, in which central government proposes and local government disposes, gave way to multi-level negotiations. The intergovernmental game was one of several games in EU policy-making.[7] Such games are not limited to implementation, but can be found at all stages of policy-making. Bargaining and negotiation lie at the heart of policy networks that seek to integrate not only interest groups and government but also the several types and levels of governmental units that are the focus of MLG.

As a final example of policy networks and the intergovernmental game, Hooghe’s (1996) collection of essays on cohesion policy uses the network metaphor to describe variously the links between levels of government, crossborder regional cooperation, and transnational regional lobbies. Thus, Nanetti (1996: 86) concludes the Community Support Framework (CSF) planning process meant ‘the networks of territorial and institutional relations became more complex... more articulated and differentiated’; and saw the ‘steady emergence of the regional level as the new institutional partner of the Commission’. Rynck’s (1996: 158) account of Flemish regional policy-making talks of ‘multi-level governance processes’ and ‘EU cohesion policy networks’. Balme and Jouve (1996: 251) argue that policy networks are ‘crucial’ for understanding the linkages between state institutions and between the state and its environment They argue that the EU affects domestic policy networks by changing the distribution of resources. They conclude that, although the central state ‘clearly dominated all networks’ and ‘European public policy has helped to consolidate existing networks rather than to replace them or superimpose new networks on them’, nonetheless, the EU ‘appears as a driving force for the regionalisation of the state’ or ‘a regional polity grounded in the expansion of inter-organisational policy making’. Laffan (1996: 337 and 340) describes how EU regional policy brought local actors into the CSF policy network and enhanced their role in policy-making.

Such descriptions of EU policy-making are essential but it is more important to explain variations in that policy process. ‘Policy networks’ are used to analyse EU policy-making in specific policy areas, most notably by Jeffrey Anderson and John Peterson.

Anderson (1990) explicitly uses Rhodes’s (1988) notion of networks as clusters of public and private agencies exchanging resources. He focuses on domestic policy networks arguing:

The EU, which ‘commands resources, distributes benefits, allocates markets, and adjudicates between conflicting interests’ (Wallace 1982: 61), can alter domestic networks by affecting the inter-member distribution of resources, and thus their level of interdependence. Community initiatives can increase, decrease or leave unchanged the resource dependencies of network members (Anderson 1990:422).

He compares British and German responses to the reforms of the ERDF since 1979, concluding that the ERDF reforms ‘exposed resource dependencies previously of little consequence’, especially ‘the lack of administrative capacity to develop joint projects’ (Anderson 1990: 442). So, British subnational actors became more dependent on central government because the shift of emphasis to programmes instead of projects encouraged the regional offices of the Department of the Environment to play a proactive or ‘mothership’ role in mobilizing local authorities. The reforms reinforced the civil service’s position as a ‘gatekeeper’.

The ERDF reforms had a disruptive effect on German federal government, ‘exacerbating existing conflicts within the national joint policy-making framework’ (Anderson 1990: 443). Anderson argues there was a decrease in resource- dependency because the Bund could use the Commission as a scapegoat for checks on Lander regional programmes. The reforms of the ERDF since 1979 did not uniformly strengthen the regions at the expense of the centre because domestic policy networks have ‘different levels of resource interdependence’ that ‘vary across otherwise similar policy areas’ and so the EU has varying effects (Anderson 1990: 445). Anderson (1996: 188-9) updates this analysis of Germany and the structural fund. He concludes there was a bifurcation of the territorial network. For the West, the EU did ‘influence the content and structure’—the resource-dependencies—of its territorial policy network, although the Lander ‘remained tied to the federal policy network’ and ‘were not especially successful’ in fending off the Commission’s interventions. By contrast, the East benefited from EU resources, which gave ‘the new Lander ...the means to carve out a sphere of independence, albeit constrained, from the Federal government’.

Peterson (1992: 244) uses the policy network approach to explore European technology policy and concludes there was ‘a tightly integrated policy community’. Peterson (1994) employs the Rhodes model of policy networks because it ‘captures’ three key features of EU decision-making: ‘bargaining between multiple public institutions and private interests; sectoral variations in the policy process; and the resource dependencies which fundamentally shape relationships between different actors’ (Peterson 1994: 21). His emphasis on ‘meso-level decision making’ is an important qualification to the scope of the model’s application. He argues the concept is of limited use in analysing either the history-making decisions made at intergovernmental conferences or the major policy-setting decisions taken by the Council of Ministers. It is most useful in analysing ‘“second-order” decisions that address the question: how do we do it?’ (for example, technology policy). Again, the focus on coordination and implementation is paramount.

The distinctions between history-making, policy-setting, and policyshaping decisions pose problems, as does the emphasis on implementation. Rather like Hoffman’s distinction between high and low politics (Hoffman 1966), Peterson’s distinction has a common-sense appeal but is difficult to apply consistently because his criteria are broad. For example, history-making decisions cover revisions to the treaties, strategic decisions about agenda, priorities, and finance and legal decisions handed down by the European Court of Justice. The distinction between strategic decisions and policy-setting decisions is difficult, resurrecting the means and ends distinction with all its known ambiguities and problems. Also, although policy networks are central to implementation, they can play an important role in policy initiation. They can shape the policy agenda by excluding options of which they disapprove. The distinction between policy-making and its implementation is a useful analytical tool but separating the two can be difficult. It is a commonplace of policy analysis that the details of policy can decide outcomes and the ‘bottom- up’ approach to implementation has shown that street-level bureaucrats can reshape policy (see, for example, Sabatier 1986). We can recognize that networks are important in implementation without excluding their contribution at other stages in policy-making.

In sum, the idea of policy networks has several virtues as a tool for describing and analysing EU policy-making. It has the vocabulary and techniques for describing complex organizational linkages. The distribution of resources in a network explains the relative power of its members. Differences in the pattern and distribution of resources explain the differences between networks. Finally, the idea of networks and their management underpin the Commission’s strategy for managing MLG. However, although the policy network approach has some distinct advantages, it is not without its critics.

  • [1] The interest in networks is not confined to students of public administration and publicpolicy. Haas’s (1992: 3) notion of ‘epistemic communities’ or ‘network of professionals’ isequivalent to Rhodes (1988: 78) on professionalized networks. See also Keohane and Hoffman(1991).
  • [2] There were still relatively few accounts of policy networks in the EU by 1996. In addition tothe work cited in the text, see Grant, Paterson, and Whitson 1988; Josselin 1994; Preston 1984;Scharpf 1988; Smith 1990.
  • [3] The Commission’s ambitions became even more difficult to realize because of the increasingnumber of new interest groups in several policy areas; for example, ERDF policy in the UK inthe 1980s.
  • [4] The phrase intergovernmental relations (IGR) has several uses; for example, it can refer tothe links between the national governments of member states of the EU. This use is toorestrictive. In the study of federalism, it refers to interactions between governmental units ofall types and levels (see Rhodes 1981: 76). It is so used here and includes EU governmental units.I see no difference between IGR so defined and MLG. The quotation from Marks (1993: 407) inthe text applies equally to American federalism and the EU.
  • [5] This summary draws on Marks (1992, 1993, 1996). On his more recent work see theAfterword at the end of this chapter.
  • [6] See, for example, Batley and Stoker 1991; Keating and Jones 1985; Leonardi 1992; Mitchell1994; Rhodes 1986c; Sharpe 1993.
  • [7] Putnam’s (1988) influential article on domestic-international linkages characterizes themas a two-person game in which domestic groups lobby the national government to adoptfavourable policies and the government tries to satisfy these groups in international negotiationswithout incurring adverse foreign policy developments. However, this model assumes theprimacy of the national governments. It ignores the role of such supranational bureaucraciesas the Commission. It ignores linkages between interests and subnational governments whichbypass the national government. It ignores transnational interest groups. In short, it is a two-level model which does not deal with multi-level policy-making.
< Prev   CONTENTS   Source   Next >