http://eiop.or.at/eiop/texte/2004–007a.htm
: 6.7.2004| This paper's comments page | Send your comment! to this paper |
| The paradoxical effects of institutional change for the legitimacy of European governance: the case of EU Social Policy |
| Frank Wendler |
| European Integration online Papers (EIoP) Vol. 8
(2004) N° 7; http://eiop.or.at/eiop/texte/2004–007a.htm |
| Date of Publication in
: 6.7.2004 |
| | Abstract |
Back to homepage |
PDF | | This paper's comments page | Send your comment! to this paper | |
1
For some time now, scholars of European integration have become interested in the interrelation between the institutional development of the EU polity and its democratic legitimacy. In connection to this, a conceptual turn can be observed with regard to the analytical perspective on institutions and their development: By turning around the focus of ‚classical‘ integration theories on the causes of institutional evolution, especially the broad literature associated with the ‚governance‘ approach to European integration has been influential in concentrating on the effects of institutional change for policy-making in the European Union – thus departing from the point of view that „polity determines politics and policy“ (Jachtenfuchs/Kohler-Koch 2004: 101, cp. Jachtenfuchs 2001). Whereas a great deal of attention in this literature has been dedicated especially to the different variants of ‚network governance‘ and the decentralised, heterarchical, and discursive forms of policy-making most recently epitomised by the Open Method of Coordination, it appears that two aspects in particular have been given too little attention: First, too few attempts have been made to locate the analysis of these institutional arrangements in a systematical comparative perspective, especially one that distinguishes changes and continuities in the evolution of institutional arrangements over time and relates these to changes in the effects of EU policy-making. Second, whereas different terms of normative democratic theory – especially participation and deliberation – are frequently used in the ‚governance‘ literature, frequently no explicit distinction is drawn between two levels of analysis:
2
the analytical reconstruction of processes that are related to the legitimization of European governance – especially through the adaptation of institutional arrangements to the context of their political support and contestation within a policy-making arena –, and the normative evaluation of the corresponding development in the terms of democratic theory that remains external to the policy process. However, separating these perspectives could keep two questions more distinct – how and by what principles institutional arrangements are legitimized, and as how legitimate they appear from the viewpoint of democratic theory.
This paper will make the attempt of elaborating and contrasting these two perspectives. This will be done by focusing on EU Social Policy, an area in which a high degree of institutional experimentation and a rather strong contestation of the legitimacy of Community governance is present. Departing from this background, the question discussed in this paper is twofold: First, how has institutional change affected key framework conditions of policy-making that are related to the legitimization of European governance in this particular field of governance? And second, how can these changes be judged from the viewpoint of democratic theory? The argument to be made from the discussion of these questions is twofold: First, a reconstruction of the main developments in the policy-field shows that key parameters of the institutional setup – with regard to the distribution of power, the setting of institutional frameworks for the interaction of political actors, the definition of policy ideas and the mechanisms of institutional evolution itself – appear to have adapted to the problematic and fragile legitimacy base of European governance in this field. Second, however, a normative assessment of the resulting institutional arrangements reveals legitimacy deficits that can be related not just to disequilibria of the input- and output- dimensions of legitimacy, but also to tensions between different normative standards of legitimate governance.
The argument of the paper proceeds in three steps: First, the point of departure is developed through the juxtaposition of the developments pushing for the extension of social policy competences of the European level and the main arguments concerning its legitimacy base (2.). The two subsequent steps discuss the main tendencies of institutional change (3.) and their evaluation with regard to democratic legitimacy (4.), before the argument is summarized in a conclusion (5.).
It has become almost commonplace to state that demands for a stronger role of the supranational level in social policy have to be made against the backdrop of a „constitutional asymmetry“ of economic and social – protection issues institutionalized in the basic structure of the Community Treaties (Scharpf 2002: 647). From this point of departure, however, arguments for the transfer of social policy competences to the Community level have been made both on analytical and normative grounds: Departing from the assumption of the prevalence of the mechanism of „negative integration“ over the realisation of means of „positive integration“ (Scharpf 1999: 47ff.), the main concern in the academic debate with regard to the European dimension of social policy has been focused on the possible destructive effects that the dynamic of market integration might have on national welfare states. Especially the corresponding notion of a „semi-sovereign welfare state“ that would result from the process of European integration makes a strong point for the compensating transfer of competences to the European level in the social field (Leibfried/Pierson 2000: 284ff.).
3
On the political level, the advocacy of a “social dimension” of the Common Market since the 1980s has also become strongly connected to the normative aspiration of giving broader legitimacy to the project of European integration through a social component, although even in the rhetoric of its most famous proponent, Jacques Delors, its justification always rested firmly embedded in the project of the realisation of the Single Market (Geyer 2000: 45f.). Through this link, the political development that has followed on the adoption of the Single European Act in 1986 has not resulted in a complete sidelining of social policies, but has led to the ongoing expansion of social policy competences on the European level: Following on the intensification of EU regulation due to the introduction of qualified majority voting in the field of health and safety in the SEA (cp. Rhodes 1995), especially the adoption of the Social Protocol in the Maastricht Treaty made a big step through its extension of legislation to important areas of labor law and the establishment of the Social Dialogue, which has helped to adopt EU directives in the field of parental leave and „atypical“ work (Falkner 1998, 2002). Second, the role of the EU in the ongoing debate on the contours and changes of the „European Social Model“ has intensified along with the establishment of the Luxembourg Employment Strategy and the application of the Open Method of Coordination to the fields of social inclusion and pension reform. Through the adoption of these broad and systematic approaches of non-legislative coordination, the EU has established its own role not just as the mediator of welfare reform, but as an actor capable of defining priorities and making assumptions about the necessity and direction of change: Apart from setting a number of qualitative and quantitative „benchmarks“ that are set as standards for the scope and direction of employment and welfare reforms in the member states, the Commission`s recent Social Policy Agenda stresses the need for an „integrated European Union approach aimed at economic and social renewal“, aimed not just at preserving, but „modernising the European social model“ (Commission 2000: 7). Furthermore, the process of defining a set of common European social rights and values is now close to its final institutionalisation through the adoption of the Charter of Fundamental Rights in the future Constitutional Treaty (de Burca 2003: 14ff.). To sum up, although „Social Europe“ is certainly far from being a federal welfare state, it has developed into a supranational regime with three elements that start resembling ‚stateness‘ – the ability to adopt a significant amount of legislation, a developed and differentiated set of institutional arrangements on the supranational level, and the ability to define a rather far – reaching set of social rights and political priorities for the member states.
With regard to its effects for democratic legitimacy, however, it would certainly be wrong to assume an unequivocal relationship between this growth of EU competences in social policy and the legitimacy of its governance. In fact, both structural and normative arguments are brought forward in the academic debate that assume a more problematic, or even adverse relation: First, a strong structural obstacle to the harmonisation of the social protection systems certainly consists of the large differences in the amount of spending, but also the institutional diversity of welfare states according to their adherence to one of the four basic „families“ of welfare (Esping – Andersen 1990, Ebbinghaus 1999: 14ff.). Second, and with a stronger emphasis on the normative requirements of political integration, a number of authors have referred to the inability of the European system of governance to proceed with political issues as strongly concerned with questions of political and normative conflict as in social policy. In this vein, strong limitations of the European polity with regard to the legitimization of decisions in social policy are assumed due to the lack of adequate institutional arrangements for the reconciliation of the relevant social interests, a sufficiently integrated public sphere, and not least, the absence of a European form of citizenship with relations of solidarity comparable to those in the nation state (Majone 1998: 13f.). Put in terms of democratic theory, the degree of „input“ – legitimation that EU institutions can mobilize is considered too feeble to sustain an extension of EU competences to areas of distributive social policy and any political questions that have an implication for basic ideas of solidarity (Scharpf 1999: 16ff., 2002: 651ff.).
4
From this point of view, Social Europe would not enhance, but massively put at risk the legitimacy of the project of European integration, furthering distrust and alienation of citizens from its institutions: „For all these reasons, the development of welfare policies at European level would actually aggravate the legitimacy problem, reinforcing the popular image of a highly centralised and bureaucratised Community“ (Majone 1998: 14). It is basically from this tension – between the strong push towards an extension and intensification of EU activitites in the social field on the one hand, and the normative reservations against the transfer of genuine social policy competences to the supranational level on the other –, from which normative questions about the appropriateness and legitimacy of the institutional arrangements in EU Social Policy arise.
Departing from this point, the first question to be discussed is the following: How has the institutional evolution of EU Social Policy adapted to the need to legitimize the competences of the supranational level against their political contestation within the policy-making arena? In order to develop this question, the development of EU Social Policy will be discussed from a historical perspective along four lines that correspond to the main features of historical institutionalism identified by Peter Hall and Rosemary Taylor (1996: 937ff.) – the role of institutions in structuring power relations (3.1.), the way institutions affect actors` behaviour and interactions (3.2.), the legitimizing role of ideas and beliefs (3.3.), and the intensity and prevailing mechanisms of institutional change itself (3.4.).
One of the most striking institutional developments of European Social Policy from its modest beginnings after the Social Action Programme 1974 to the present state is the strong tendency of horizontal and vertical decentralisation of policy making. Reference was made to a first variant of this tendency a few years ago by the observation of a trend „towards multi-actor and multi-level governance“ (Falkner 1999: 83ff.). This assumption referred mainly to the introduction of the bipartite and tripartite Social Dialogue between trade unions and employers` federations: After being entitled to negotiate binding framework agreements through the adoption of the Social Protocol in the Maastricht Treaty, the European-level social partners contributed to the legislative activity of the Community by the conclusion of agreements on parental leave and various forms of ‚atypical‘ work (Falkner 1998: 97ff., Falkner 1999: 83ff.). More recently, this tendency has been taken further especially through the inclusion of the Social Partners in tripartite consultations with EU institutions: First, as a complement to the coordination of employment (‚Luxembourg process‘) and structural economic reform (‚Cardiff process‘), a macro-economic dialogue was set up in 1999, which brings together representatives of the ECB, the ministers of finance and the social partners in a regular, twice-yearly exchange (Dufresne 2002: 87ff., Deppe et al. 2003: 184). In addition to this, the informal summits of the social partners with the heads of government before the meetings of the European Council were formalized through the introduction of a tripartite summit for the preparation of the annual Spring Summit on the Union`s social and economic policies (EIRO 2003). While leading to the dispersal of decision-making processes into policy networks that go beyond the formal boundaries of the EU organisations, this tendency appears to aim at the support of the supranational institutions by exposing intended pieces of legislation and the coordination of economic and social policies to the scrutiny of the social partners, rather than assigning an independent role to civil society actors:
5
Whereas the tripartite concertation with the social partners is confined to a non-binding, and rather informal exchange of views, the bipartite Social Dialogue has so far remained under the firm control of the Commission, which originated all negotiated framework agreements through its intitial proposals. The function of this part of decentralisation is therefore mainly supportive of the Community institutions, and of policy-making at the supranational level.
This distinguishes it from a second, quite distinct variant of vertical decentralisation in recent years that is also tied to the transfer of policy – making functions to the EU level, but affects more directly the allocation of competences between the Union and member states. Particularly prominent in this regard are the relatively new forms of non-legislative coordination of member states`s employment and social policies that were developed after the inclusion of an employment chapter into the Treaty of Amsterdam and the ensuing European Employment Strategy (Goetschy 1999, 2003, Foden/Magnusson 2003). After the introduction of the quite similar, but less formalized Open Method of Coordination through the Lisbon European Council in March 2000, the cycle of setting guidelines, drawing up of National Action Plans and a joint evaluation process led by the Community institutions has been applied to a number of social policy areas, including the reduction of social exclusion and poverty, the reform of pension systems, and in the future also the reform of health care systems (de la Porte/Pochet 2002). On the level of the vertical distribution of powers between the supranational level and the member states, the new procedures combine a strongly extended function of the EU in the setting of political goals and the definition of guidelines with a framework of action that does not create the conditions for the achievement of „positive integration“, but lays a particularly strong stress on the diversity and autonomy of member states (Scharpf 2002: 652ff., Mosher/Trubek 2003: 68ff.). Politically speaking, this sharing of functions offers potential gains to both political levels: Whereas allowing for the establishment of a broad and differentiated „legitimizing discourse“ of the EU that is framed by the „Lisbon strategy“ and covers a whole range of social, economic, and structural policies through the Open Method of Coordination (Radaelli 2003: 7), the member states do not have to cede a large amount of competences and might actually use the new framework for strategies of selective adaptation or „blame avoidance“. Second, with regard to the balance of powers within the institutional framework of the EU, the new procedures have strengthened primarily intergovernmental, and executive institutions: Apart from stressing the goal-setting function of the European Council (especially its „spring“ formation responsible for the coordination of social and economic policies of the Union), all relevant decision-making processes of the OMC are taken through the interaction of the Council of Ministers and the Commission, whereas the role of the European Parliament is reduced to the issuing of recommendations, and the traditionally strong role of the ECJ in social policy is completely sidelined due to the legally non-binding character of all measures (Radaelli 2003: 14ff., Göbel 2002: 145ff.). Furthermore, the structure of inter-institutional relations takes on a more informal shape, giving rise to a growing relevance of expert bodies like the Social Protection Committee, the Employment Committee, and the Economic and Financial Committee, and their interaction with the responsible Commission directorates and Council formations (Göbel 2002: 59ff., de la Porte/Pochet 2003: 35). In short, whereas a strong rationale exists for the adoption of the non-legislative forms of coordination on the grounds of member state diversity and autonomy, these do not just amount to more effective or cooperative styles of policy-making. Instead of leading to a dispersal of power relations, the decentralisation dynamic in the institutional development of EU Social Policy appears primarily as an instrument for the deliberate splitting of competences between different political levels, the strengthening of certain focal points in the policy-making process and the safeguarding of the autonomy of the most relevant political actors. Within this development, a shift of focus is discernible in the allocation of decision-making competences: In the sequence of events from the Maastricht treaty to the end-1990s, it appears that the main dynamic of institutional decentralisation of EU governance has turned from a first variant, which appears largely supportive of the allocation of competences on the supranational level, to the much more pronounced emphasis on the autonomy of the member states of the second variant.
6
Concerning the second dimension of analysis – the way institutions influence the behaviour, preference formation and interaction of actors – two main theoretical currents have developed in the institutionalist literature: First, the rationalist or ‚calculus‘ approach that focuses on instrumental behaviour and strategic interaction of actors, treats preferences as exogenous from institutional contexts, and concentrates on the function of institutions as devices for the setting of rules for the attainment of decisions and their enforcement; second, the sociological-constructivist or ‚cultural‘ approach that includes values, norms and symbols in its definition of institutions and extends their function to the setting of normative and cognitive orientations that affect the worldviews and identities of actors, thus treating preferences as endogenous to institutional contexts. In correspondence to these differing conceptualisation of institutions, the behaviour of actors is explained and interpreted according to a ‚logic of consequentiality‘ that is aimed at the realisation of given preferences through strategic behaviour on the one hand, and to a ‚logic of appropriateness‘ through which actors adapt their preferences and behaviour to the rules and values that are given in a particular situation, on the other (cp. Hall/Taylor 1996: 939f., March/Olsen 1989: 21ff., Pollack 2004: 138f., Risse 2004: 160ff.). The eclectic use of parts of both theories in historical institutionalist analyses (Hall/Taylor 1996: 940) is mirrored in different analyses of European social policy that lean on this approach: A good example are the differences between the rationalist premises in Pierson`s analysis (Pierson 1996: 148ff.), and the reconstruction of a process of „joint preference formation“ in Falkner`s analysis of the negotiations leading to the adoption of the Social Protocol, which leans much more on elements of sociological institutionalism, such as the relevance of ideas and learning processes (Falkner 1998: 200ff., Falkner 2002: 112ff.). Beyond the theoretical debate, the opposition of the two logics of explanation is also a valuable tool to show how the functions of institutions have changed over time: As a temporal perspective shows, the function of the institutional arrangements in EU Social Policy has shifted from the formal and transparent setting of rules for processes of decision-making to the institutionalisation of procedural mechanisms that foster communicative interaction, the exchange of information, and the development of preferences. Politically, this change is linked to the pursuit of a tendency of proceduralisation that aims not just at the overcoming of blocks to the process of policy-making, but also at the deliberate reinforcement of the preference-formation function of institutions in the evolution of EU Social Policy, along with the increasing dilution of their rule-setting function.
A first indicator of this development is the change in the logic of decision-making through the reduction of veto options and the setting of incentives for negotiations that has occurred in the subsequent phases of institutional evolution since the mid-1970s: Being confined to unanimous voting in the Council, decision-making after the Social Action Programme in 1974 allowed for the taking of veto-positions by any of the actors involved, thus maximising the incentives for the strategic pursuit of interests and practically excluding mechanisms of bargaining (Geyer 2000: 36ff.). A first change to this was introduced through the adoption of qualified majority voting in the area of health and safety of workers (Art. 118a) in the Single European Act in 1986: Apart from diluting the veto option for each government, this change gave the Commission a certain leeway in building coalitions and stretching its mandate by the extensive interpretation of the Treaty base (such as in the case of the 1993 Working Time Directive, cp. Rhodes 1995: 99ff.). A further strong incentive for a negotiation-oriented style of policy-making was set through the adoption of the Social Protocol, which added the Social Dialogue as a second track of negotiation about the content of future directives, apart from the conventional route through the Community institutions.
7
It is especially through the interconnection of both tracks – as proposals from failed negotiations between the Social Partners can be taken back to the institutional route – that a strong incentive has been set for both social partner organisations to work for a consensus in the negotiation of framework agreements. Through the application of this subtle pressure on the relevant actors, a procedural exchange was realized that both fosters, and could build on on the gradual evolution of shared beliefs and ideas between the social partners (Falkner 1998: 200f., Falkner 2002: 113). The main result of this process of institution building was to overcome the structural barriers that had impeded the attainment of consensus, expressed in the dilemma of the „joint decision trap“ and the difficulties of reaching the approval of the employers` European federation of entering into binding agreements with the trade union side (Streeck 1995: 415f.).
Again, in contrast to this supportive approach of trying to support the EU`s capacity to accommodate the conflicting interests in a consensus and reach binding decisions, a second, more recent strategy to address the „decision – making problem“ of the European level has been the introduction of various mechanisms of „soft governance“. It appears as the characteristic approach of these mechanisms that conflicts between differing actors and interests are not solved, but circumvented: A first instance of this approach is the increasing tendency towards the adoption of „neo – voluntarism“, facilitating the political agreement on pieces of EU regulation by granting flexibility clauses, exemptions, and reducing interventions to recommendations (Streeck 1995: 423ff.). A still more consequent step to shift the function of EU Social Policy from the attainment of binding decisions to the construction of a discourse that aims mainly at the shaping of preferences is the adoption of the Employment Strategy and the OMC. Both of these procedures are also strong efforts at de-blocking the policy process in the social field by reducing the need to integrate all interests in the reaching of binding decisions. In this vein, the approach of non-legislative coordination has reached a high degree of elaboration, especially through the systematic and iterative methodology of the coordination cycles, its foundation on an increasingly differentiated set of guidelines, indicators, and the statistical data developed by the Commission and the responsible expert committees, and its promotion on the highest political level of the European Council in the framework of an overarching political strategy. These qualify the OMC as an approach that goes far beyond its precursors of non – legislative coordination in European Social Policy, such as the drafting of the ‚social budgets‘ in the 1970/80s, the 1993 recommendation on the convergence of social protection systems, and the early attempts of building an employment strategy until the Essen monitoring procedure in 1994 and the Luxembourg summit in 1997 (Goetschy 1999: 118ff., Göbel 2002: 161ff., Geyer 2000: 157ff.). A remarkable feature of this overall shift of the institutional framework to functions stressed in the ‚cultural‘ variant of new institutionalism is that it has not just occurred as a side – effect of the ongoing process of integration, but as the result of a quite deliberate political strategy, aiming at a re-definition of EU Social Policy as a field of governance focused on the development and dissemination of ideas and knowledge, rather than the setting of transparent and enforceable rules (Mosher/Trubek 2003: 79ff.).
At the level of policy ideas, the development of EU Social Policy is certainly first of all a story of expansion, having grown from very confined activities of the EC in the area of workers` freedom of movement to the establishment of full-fledged policy programs that include core welfare policies such as social inclusion, pensions, and employment. What is striking, however, is that this broadening of scope is accompanied by a somewhat evasive strategy of legitimization:
8
Especially in the more recent phases, the definition of European Social Policy, and its relation to the idea of a “European Social Model” is developed not in contrast, but as a complement to goals of market integration, and shows a profile that is strongly in favour of elements of diversity and includes many elements that are ambiguous or left open to interpretation. The first of these elements is present in the definition of social policy as a „productive factor“ in the Commission`s current agenda for the years 2000-2006, which is further underlined through the realisation of studies directed at the proof of the „costs of non-social policy“ (cp. Fouarge 2003, Rubery et al. 2003). Furthermore, by integrating social policy into the „Lisbon strategy“ as one component in a synergetic triangle with economic and employment policies (Commission 2000: 8), an effort is made to build a connection of the social agenda to the famous legitimizing formula for growth, flexibility, and competitiveness propagated through the strategy. At the level of policy programs, the general tendency of European welfare policy towards the paradigm of the „activating welfare state“ has been adopted, with strong emphasis on the employability and adaptability of workers and the adoption of a supply-side approach to economic and social policies. This certainly marks a contrast to the more regulatory, interventionist and state-oriented approaches of earlier phases, especially the one marked by Delors`s program for a „structured economic area“ (Deppe et al. 2003: 186ff.). However, even within this shift of focus, the EU programs work with a considerable amount of ambiguity, taking up a number of different influences and being adaptable to a variety of contexts across the different „families“ of welfare states and labor markets. Thus, the Employment Strategy shows equally strong affinities with typically Scandinavian and social democratic conceptions of employment policy as with the Blairite „third way“ and more strongly neo-liberal conceptions, therefore rather reflecting and compromising between a number of different approaches from current discussions in the member states than setting up an entirely new one (Trubek/Mosher 2003: 71ff., Goetschy 1999: 124ff.). This is mirrored by the fact that with regard to the recommendations set out in the evaluation of the National Action Plans by the Commission and Council, no single ‚type‘ of labor market emerges as the clear favorite, although the Scandinavian countries receive a much smaller, and especially the continental and Southern European countries a bigger number of recommendations (de la Porte/Pochet 2002: 37). Still, the principles and guidelines set up by the EU strategy appear flexible enough to be endorsed by all member states and be integrated quite smoothly into their labor market programs: As the first 5-year evaluation of the EES has shown, quite different labor market regimes can identify with the European guidelines, with the three Scandinavian countries and the UK reporting almost ‚total compatibility‘ of their labor market strategies with the EU requirements, and other (more critically evaluated) countries such as Italy and Spain insisting that the guidelines have reinforced, rather than introduced, pre-existing tendencies and processes (de la Porte/Pochet 2003: 43ff.). Concerning the actual effects of the European strategy in the member states, empirical evidence so far suggests some EU-induced changes in specific sectors – such as the coordination of labor market policies between different ministries of the realization of equal opportunity programmes – rather than a change of paradigms in member states or the convergence towards a single type of labor market policy (de la Porte/Pochet 2003: 45, Goetschy 2003: 82ff.). At least for the case of employment, this gives evidence to the view that EU governance in the social field has added a rather broad and multifaceted, not actually very innovative, and quite permissive framework of guidelines to the activities of the member states. With regard to the legitimization of policy programs, this corresponds to the interest of reconciling different models of welfare represented by the member states in the definition of the EU`s social agenda, which supplements and differentiates, but does not question the overall – and more deeply institutionalised – agenda for liberalisation and competitiveness.
9
Concerning the form and causality of institutional change itself, it appears that processes of informal and experimental institution-building have been employed to overcome the restrictions and path-dependencies of institutional change that are emphasised by historical institutionalist analyses (Hall/Taylor 1996: 941). For the case of EC Social Policy, impressive evidence of such effects has been given especially by Pierson`s analysis, which concentrates on the restrictions of „member state control“ over the Community`s institutions due to the restricted time horizons of policy makers, the unintended consequences of Treaty Articles, and the strong barriers to any encompassing institutional reform (Pierson 1996: 148ff.). However, in order to explain the strong elements of dynamism that are evident in this field – including a whole set of procedures and institutional arrangements that were introduced in the roughly fifteen years after the adoption of the Single European Act –, it does not appear satisfactory to refer just to unintended consequences and gaps in member state control, which are left as practically the only means for the explanation for change (ibid.: 156). More conclusive insights into the evolution of institutional arrangements can be taken from the distinction of the level of formal institutional creation, led by the enacting coalition of actors responsible for Treaty changes in the framework of Intergovernmental Conferences (IGCs), and the level of institutional operation in the interregnum phases between those ‚grand bargains‘. On this second level, more informal, but equally relevant institutional changes are effected by the executing coalition of actors responsible for the unfolding and application of those provisions in everyday policy-making (cp. Lindner/Rittberger 2003: 445ff., Farrell/Heritier 2003: 3ff.). Concerning the case of social policy, it appears that while the creation of institutional innovations has become increasingly less confined to the formal institutional creation phase during the IGCs, the executing coalition of political actors involved in the institutional operation phases has carved out a decisive role in creating, negotiating and also implementing new informal institutional arrangements, which where then ratified through Treaty changes. Again, a first example is the enactment of the Social Protocol in the Maastricht Treaty, which was not only preceded by a longer phase of negotiations between the member states and Commission, but could also build on a social partner agreement on the provisions of the Social Dialogue. The enactment of new provisions in the Treaties was thus based on processes which went far beyond the confined (and very controversial) bargaining situation at the Maastricht conference: „It is thus essential not to restrict the analysis to the 1991 IGC itself“ (Falkner 1998: 88). Again, this development has been taken further in the following phases, in which the role of the IGC was practically reduced to the ratification of innovations that had been prepared over various years in the institutional operation phase: The main innovation in EU Social Policy brought by the 1996 IGC in Amsterdam, the inclusion of an employment chapter, was mainly a formalization of an ongoing process following on the Commission`s 1993 White Paper and the Essen Council in 1994. Through these steps, practically all relevant parts of the Employment Strategy (such as action plans, guidelines and indicators) had already been developed under the auspices of the Commission before they became institutionalised in the Treaty (Goetschy 1999: 120ff.). This form of a more informal and experimental introduction of changes, which is less bound to the restrictions of the IGCs, is still intensified through the introduction of the Open Method of Coordination. Here, the European Council has confined its role to the setting of general guidelines, which are then adapted to different policy fields even without a formal treaty base for the method as such (Radaelli 2003: 11). Therefore, whereas arguments about the barriers to the reform of existing institutions are important for the form and direction of change, they do not preclude the possibility of encompassing institutional changes as such: Through the ‚layering‘ of institutions in the ongoing addition of new procedures and instruments to the existing ones, which was mostly coupled to steps of further economic integration (especially in the cases of the Social Protocol and the Employment Strategy), rather fundamental changes have become institutionalized even in absence of an encompassing revision of all social policy provisions.
10
Rather than the classical historical institutionalist argument that „like Frankenstein`s monster, structure takes on a life on its own“ (Aspinwall/Schneider 2000: 10), it is the change to more flexible and dynamic relations between the levels of institutional creation and institutional operation that account for a great part of the dynamism in the recent evolution of EU Social Policy. What matters about this argument is that it shows how a strategy was found to circumvent the restrictions on attempts of deliberate institutional design in the bargaining situation of the IGCs – and thus to encounter a more flexible, and dynamic way of adjusting the EU`s institutional structure to demands for an extension of EU competences in the social field.
To sum up: With regard to the four dimensions of institutional analysis discussed above, it is possible to identify tendencies of institutional development that have contributed to the adaptation of EU Social Policy to its fragile legitimacy base especially by avoiding a „strong“ form of supranationalisation, and by attempting to connect and reconcile apparently contradictory principles and elements:
How can these institutional developments be judged from the viewpoint of democratic legitimacy? Whereas an elaborate investigation of this question would require a separate paper, the argument proposed in this article will confine itself to outlining two theoretical approaches to legitimacy deficits that follow from the institutional developments discussed above:
11
First, in an institution-centered perspective, an imbalance appears between the development of the input- and output-dimensions of legitimacy (Scharpf 1999: 16ff.): Through the general increase in the „decision-making capacity“ of the EU in the social field, and the extension of Community policy-making to core areas of social and employment policies, the overall ‚output‘ of European Social Policy has globally increased in terms of acts of legislation, but also in its more subtle presence in the definition of social and employment guidelines. While being more actively involved in a wide range of social policy questions, however, one standard of ‚output‘-legitimacy that the EU has not been able to fulfil is a sufficient degree of „accountability of results“ (Majone 1998: 24). The main deficit addressed here concerns the clear specification of political goals and their visible and transparent realisation: As stated before, not just the ambivalence of many of the EU`s policy guidelines in the framework of Open Coordination, but also its tendency towards „neo-voluntarist“ regulation and ‚soft‘ mechanisms of governance are elements that put especially this standard of output-legitimacy into question. Given the overall increase in the scope and political weight of the ‘output’ of EU Social Policy, a point needs to be made for stronger mechanisms of input-legitimation – that is, a higher degree of participation, a better procedural accountability and possibly the approximation towards deliberative forms of decision-making. However, an evaluation of developments at the input-side shows equally strong deficiencies: Although the introduction of co-decision by the European Parliament in important areas of social legislation and the significant role of the social partners have undoubtedly led to a greater openness of decision-making in comparison to the intergovernmental model that prevailed until the 1980s, the Commission and Council of Ministers still remain in firm control of the policy process – with an actually more pronounced role for intergovernmental and executive institutions in the recent trend towards the Open Method of Coordination. Apart from that, deliberation is certainly present in many elements of the present institutional arrangement with its emphasis on expert-led discourse, the gathering of information and the dissemination of “best practice” in the framework of Open Coordination (Teague 2001). However, it is much less present in the sense of trying to create a normative consensus of what values should hold the „European Social Model“ together, and to mediate a consistent idea of Social Europe to a wider public sphere as a part of the project of European integration. Comparing the procedural and substantive aspects of deliberation, the definition of the contents of EU Social Policy not just seems too politicized, but also by far too ambivalent and open to interpretation to be actually based on the idea of an argumentation-based, deliberative consensus – especially in the absence of a wider parliamentary or public debate. Furthermore, it appears as the most problematic consequence of the tendencies of decentralisation and proceduralisation that accountability is also put into question at the ‚input‘-side of policy-making. In this regard, not just the dispersal of political responsibilities into policy-networks that are both horizontally differentiated and spread over the vertical axis between the EU and member states, but also the informalization of policy-making and the omission of parliamentary and judicial control appear as principle deficits especially of non-legislative coordination. From this point of view, therefore, an imbalance exists primarily between the growing political output of the EU level in the social field, and the lack of mechanisms to provide for sufficient mechanisms of legitimation at the ‘input’-side of policy-making.
Second, much of what has been said above must be interpreted not just along the divide of procedural and substantial standards of legitimacy, but rather between different normative standards of democratically legitimate governance. In this vein, a normative evaluation of the present institutional arrangement by the standard of responsive governance – that corresponds to the standard of participation on the input-side, and the decision-making capacity on the output-side of politics – would actually establish a rather positive view on many of the institutional developments in EU Social Policy.
12
It would also endorse its multi-level and network character and the principle of ‚soft‘ governance in areas where more binding and interventionist approaches can not be established without the risk of frequent blockades in decision-making. A reverse perspective, however, is established from the standard of accountable governance. From this point of view, in which the present governance arrangement in EU Social Policy appears deficient with regard to both the input- and the output-side of politics, a normative recommendation for the future development of European Social policy would have to point in the way of a simpler, more formalized and more clear-cut system of governance. This would have to make a more restrained use of ‚flexible‘, and a more consequent use of binding instruments, the actual purpose and outcome of which could be shown more transparently and thoroughly. Finally, a third perspective on European governance is established by approaches of deliberative democracy: From this point of departure, the normative critique of the present governance arrangement in EU Social Policy would have to focus on the fit between processes of arguing, reason-giving, and the collection of knowledge on the input-side of politics, and the actual ability of EU policy-makers of basing the contents of Community policies on an argumentative agreement that can be made transparent to a wider public on its output-side. These three normative standards of democratically legitimate governance are certainly demanding – but as the argument shows, they actually cut across the divide between the input- and output-side of politics. As such, the distinction of these criteria might prove useful in providing different normative standards that arguments about the legitimacy of EU governance might be angled towards.
To sum up, the purpose of this paper has been to point to the paradoxical relationship between the institutional development of EU governance in the social field and its democratic legitimacy: On the one hand, a historical reconstruction reveals a set of tendencies in the evolution of institutions that helped to legitimize the assumption of social policy competences by the EU level against the backdrop of its rather weak and contested legitimacy base. On the other hand, a normative evaluation of the resulting institutional arrangements show legitimacy deficits that can be identified with regard to the input- and output-dimensions of legitimacy, but also in comparison between different normative standards of legitimate governance. The main conclusion of this paper, therefore, is that a paradox exists between the institutional adaptation of EU governance to demands for its legitimization within the policy process, and a normative evaluation of these changes from the external perspective of democratic theory. To discuss the contrast between these two perspectives, and to locate them in a systematic comparative perspective is a question that the discussion about institutional change in the EU should more thoroughly address.
Aspinwall, Mark/Schneider, Gerald, 2000: Same menu, separate tables: The institutionalist turn in political science and the study of European integration, in: European Journal of Political Research, Vol. 38 (1), pp. 1 – 36
Commission of the European Communities, 2000: Social Policy Agenda, published by the Directorate-General for Employment, Industrial Relations and Social Affairs, Brussels
De Burca, Grainne, 2003: Fundamental Rights and Citizenship, in: de Witte, Bruno (ed.): Ten Reflections on the Constitutional Treaty, European University Institute, Florence, pp. 11 – 44
De la Porte, Caroline/Pochet, Philippe (eds), 2002: Building Social Europe through the Open Method of Co-ordination, P.I.E., Frankfurt a.M.
De la Porte, Caroline/Pochet, Philippe, 2003: A twofold assessment of employment policy coordination in light of economic policy coordination, in: Foden/Magnusson (eds), pp. 13 – 68
Deppe, Frank/Felder, Michael/Tidow, Stefan, 2003: Structuring the State – The Case of European Employment Policy, in: Kohler-Koch, Beate (ed.), Linking EU and National Governance, Oxford, pp. 175 – 200
Dufresne, Anne, 2002: Oskar Lafontaine`s dream: an opportunity for economic policy co-ordination?, in: Degryse, Christophe/Pochet, Philippe (eds), Social Developments in the European Union, edited by the European Trade Union Institute, Brussels, pp. 85 – 113
Ebbinghaus, Bernhard, 1999: Does a European Social Model Exist and Can it Survive?, in: Huemer, G/Mesch, M./Traxler, F. (eds): The role of employer associations and labor unions in the EMU, Ashgate, pp. 1 – 26
EIRO, 2003: Tripartite social summit discusses growth and employment, 18 December 2003, available at http://www.eiro.eurofound.ie/2003/12/feature/eu0312207f.html.
Eriksen, Erik Oddvar/Fossum, John Erik (eds), 2000: Democracy in the European Union. Integration through deliberation?, Routledge, London
Esping-Andersen, Gøsta, 1990: Three worlds of welfare capitalism, Cambridge
Falkner, Gerda, 1998: EU Social Policy in the 1990s. Towards a corporatist policy community, Routledge, London
Falkner, Gerda, 1999: European social policy: Towards multi-level and multi-actor governance, in: Kohler-Koch, Beate/Eising, Rainer (eds): The Transformation of Governance in the European Union, Routledge, pp. 83 – 98
Falkner, Gerda, 2002: How intergovernmental are intergovernmental Conferences? An example from the Maastricht Treaty Reform, in: Journal of European Public Policy 9:1, pp. 98 – 119
Farrell, Henry/Héritier, Adrienne, 2002: Formal and Informal Institutions under Codecision: Continuous Constitution Building in Europe, in: European Integration online Papers, Vol. 6, No.3, available online at http://eiop.or.at/eiop/texte/2002–003a.htm.
Foden, David/Magnusson, Lars (eds), 2003: Five Years` experience of the Luxembourg Employment Strategy, edited by the European Trade Union Institute, Brussels
Fouarge, Didier, 2003: Costs of non-social policy: towards an economic framework of quality social policies – and the costs of not having them, Report for the Employment and Social Affairs DG, Brussels
Geyer, Robert, 2000: Exploring European Social Policy, Cambridge
Göbel, Marcus, 2002: Von der Konvergenzstrategie zur offenen Methode der Koordinierung. EG-Verfahren zur Annäherung der Ziele und Politiken im Bereich des sozialen Schutzes, Baden-Baden
Goetschy, Janine, 1999: The European Employment Strategy: Genesis and Development, in: European Journal of Industrial Relations, Vol. 5, No. 2, pp. 117 – 137
Goetschy, Janine, 2003: The Employment Strategy and European integration, in: Foden/Magnusson (eds), pp. 69 – 110
Hall, Peter/Taylor, Rosemary, 1996: Political Science and the Three New Institutionalisms, in: Political Studies 44: pp. 936 – 57
Hemerijck, Anton, 2002: The Self-Transformation of the European Social Model(s), in: Esping-Andersen, Gøsta/Gallie, Duncan/Hemerijck, Anton/Myles, John (eds): Why we need a new welfare state, Oxford, pp. 173 – 213
Jachtenfuchs, Markus, 2001: The Governance Approach to European Integration, in: Journal of Common Market Studies, Vol. 39, No.2, pp. 245 – 64
Jachtenfuchs, Markus/Kohler-Koch, Beate, 2004: Governance and institutional development, in: Jachtenfuchs/Kohler-Koch (eds), European Integration Theory, Oxford, pp. 97 – 115
Jacobsson, Kerstin/Schmid, Herman, 2003: The European Employment Strategy at the crossroads: contribution to the evaluation, in: Foden/Magnusson (eds), pp. 111 – 140
Leibfried, Stephan/Pierson, Paul, 2000: Social Policy. Left to Courts and Markets?, in: Wallace, William/Wallace, Helen (eds), Policy-Making in the European Union, 4th ed., Oxford, pp. 267 – 292
Lindner, Johannes/Rittberger, Berthold, 2003: The Creation, Interpretation and Contestation of Institutions – Revisiting Historical Institutionalism, in: Journal of Common Market Studies, Vol. 41, No. 3, pp. 445 – 73
Majone, Giandomenico, 1998: Europe`s ‚Democratic Deficit‘: The Question of Standards, in: European Law Journal, Vol. 4, No.1, pp. 5 – 28
March, James/Olsen, Johan, 1989: Rediscovering Institutions. The Organizational Basis of Politics, Free Press, New York
Mosher, James S./Trubek, David, 2003: Alternative Approaches to Governance in the EU: EU Social Policy and the European Employment Strategy, in: Journal of Common Market Studies, Vol. 41, No. 1, pp. 63 – 88
Pierson, Paul, 1996: The Path to European Integration. A Historical Institutionalist Analysis, in: Comparative Political Studies, Vol. 29, No.2, pp. 123 – 63
Pollack, Mark, 2004: The New Institutionalisms and European Integration, in: Diez, Thomas/Wiener, Antje (eds), European Integration Theory, Oxford, pp. 137 – 156
Radaelli, Claudio, 2003: The Open Method of Coordination: A new governance architecture for the European Union? Swedish Institute for European Policy Studies, Report Nr. 1/2003, Stockholm (available at http://www.sieps.su.se).
Rhodes, Martin, 1995: A regulatory Conundrum: Industrial Relations and the Social Dimension, in: Leibfried; Stephan/Pierson, Paul (eds), European Social Policy. Between Fragmentation and Integration, Washington, pp. 78 – 121
Risse, Thomas, 2004: Social Constructivism and European integration, in: Diez, Thomas/Wiener, Antje (eds), European Integration Theory, Oxford, pp. 159 – 176
Rubery, Jill/Humphries, Jane/Fagan, Colette/Grimshaw, Damian/Smith, Mark, 2003: Equal Opportunities as a Productive Factor, European Work and Employment Centre, Manchester School of Management, UMIST
Scharpf, Fritz W., 1999: Regieren in Europa. Effektiv und demokratisch?, Frankfurt/NY
Scharpf, Fritz, 2002: The European Social Model: Coping with the Challenges of Diversity, in: Journal of Common Market Studies, Vol. 40, pp. 645 – 70
Streeck, Wolfgang, 1995: From Market Making to State Building? Reflection son the Political Economy of European Social Policy, in: Leibfried, Stephan/Pierson, Paul (eds), European Social Policy. Between Fragmentation and Integration, Washington, pp. 389 – 430
Teague, Paul, 2001: Deliberative Governance and EU Social Policy, in: European Journal of Industrial Relations, Vol. 7, No. 1, pp. 7 – 26