Since at least Maastricht, the European Union has been in search of novel mechanisms and arguments to ground its democratic legitimacy. An increasingly influential view, which came to prominence with the debate following the Commission's White Paper on governance,Footnote 2 holds that the European Union has evolved new modes of governance (NMG) that either compensate for its lack of democratic legitimacy or offer more participatory and deliberative styles of democratic politics than the traditional electoral and representative forms of democracy associated with the nation-state. This view involves two distinct but interrelated arguments. On the one hand, the policy problems dealt with at the European level are said to be mainly regulatory rather than redistributive, and so can be more appropriately handled by the ‘delegation’ of powers to specialized, and largely expert (or at least, non-majoritarian) institutions.Footnote 3 On the other hand, the associated NMG involve innovative, less hierarchical, and soft-law based decision-making processes that purportedly widen democratic involvement at various levels.Footnote 4 Some commentators even argue that they form part of an emergent and experimental architecture of governance whose principles – though mainly instantiated through informal channels and practices – reflect those underpinning democracy more generally.Footnote 5
Both these arguments are usually presented in terms of their ability to secure democratic ‘outputs’ notwithstanding – and possibly because of – their lack of conventional democratic ‘input’.Footnote 6 Nevertheless, these output arguments invoke implicit claims to satisfy certain ‘input’ criteria for democratic legitimacy, albeit in unconventional ways. These claims rest on delegated bodies and NMG supposedly providing the means to represent social actors, general interests or even an overarching European interest that the conventional democratic channels of political parties, electoral majorities and parliamentary representatives fail to register. As we shall show, their alleged superiority in achieving better democratic outputs largely assume these representative inputs.
We start by outlining the nature of European governance and the role that delegation and NMG play in it. We then assess the representative claims that are made for each of these mechanisms. We argue that lack of effective formal channels for authorizing representatives and holding them to account, undermines the substantive representative claims of these agents and agencies to ‘stand’ or ‘act’ for the European public.
EUROPEAN GOVERNANCE AND NEW MODES OF GOVERNANCE
Despite disagreement regarding the scope and nature of European governance, most analysts agree that at its heart lies the so-called community method (CM). Majone has characterized this approach in normative terms as involving three constitutional principles – independence, sanctioning and the offering of guarantees – to regulate the interaction between the main Union institutions. Independence underlies the Commission's exclusive prerogative to initiate proposals, execute policies, act as guardian of the Treaties and represent the Community internationally. The Councils of Ministers and the European Parliament (EP) possess the power to sanction the proposals made by the Commission. Finally, the European Court of Justice (ECJ) guarantees a balance between the institutions while upholding the integrity of the European legal system.Footnote 7 More descriptive accounts of the CM point to how various institutions have assumed particular responsibilities in relation to different areas of policy-making. For instance, they distinguish between a stricter application of the CM with regard to areas such as agricultural and fishery policies, and slightly modified sets of rules of engagement and institutional responsibility in areas such as competition policy, regulation and distributional issues.Footnote 8 These descriptive accounts imply a greater role for social, sectoral and regional actors, and in some cases for the mechanisms and logic of the market.
From an institutional perspective, the Commission provides the most innovative aspect of this structure of governance. Neither its bureaucratic nor its executive function operate on traditional lines, while its de facto veto power, deriving from its agenda-setting prerogative and role as the guardian of the Treaties, lend it important legislative and quasi-judicial functions. Majone suggests that the ‘organizing principle of the Community is not the separation of powers but the representation of (national and international) interests’.Footnote 9 The interlocking of competences and the procedures followed in the decision-making process make the Commission a bearer of political interests, which are balanced with those represented by the other institutions that make up the CM. In Majone's view, the CM offers a form of ‘mixed government’. However, as he also notes, this arrangement is characterized by the extensive delegation of powers from the member states (MSs) to the Commission, which exercises the role of a supranational non-majoritarian regulator.Footnote 10 In this capacity, the Commission acts as the agent for the MSs. As we shall see, the centrality of this principal–agent relationship in the CM has important consequences for the conceptualization of the Commission's representative function.
Although the structure of EU governance is fairly innovative compared to decision-making processes within the MSs, the CM retains certain traditional elements of governance: notably, a hierarchical division of competences, binding decisions and the more or less uniform and strict implementation of rules through the use of sanctions for non-compliers. However, alongside the CM, there have emerged other modes of governance aimed at policy coordination between different institutional and national actors, and forms of selective transgovernmental cooperation. As Scott and Trubeck note, these alternative modes depart from both the CM and traditional governance in two important respects.Footnote 11
The first departure consists of a series of specific variations in how the CM operates, such as the introduction of more flexible and non-binding legislation and the substitution of procedural prescriptiveness for substantive uniformity. Other similar departures comprise the introduction in the policy-making process of new institutional actors – in the form of comitology for instance, partly to direct and control the CommissionFootnote 12 – and the more frequent recourse to consultation with civil society organizations (CSOs) through ad hoc initiatives or more institutionalized fora and procedures.Footnote 13 This kind of departure corrects and transforms the institutional equilibrium and competences of the CM while giving greater leverage to national and sub-national actors in the implementation of policies and in the application of directives and other legislation, without changing the basic principles of traditional governance. Scott and Trubeck call this ‘new, old governance’.Footnote 14
The second kind of departure from the CM is presented as more radical, amounting to a wholesale alternative to traditional models of governance. As argued by Citi and Rhodes, the open method of coordination (OMC) provides the most innovative of these new and alternative instruments;Footnote 15 Scott and Trubeck list a series of characteristics that in their view set NMG apart. These include:
-
the valorization of forms of participatory governance involving CSOs;
-
the full acknowledgement of the multilevel nature of EU governance;
-
the recognition that legislation needs to adapt to diversity and subsidiarity;
-
the centrality of deliberation in policy-making, both as an instrument for problem-solving and as a form of legitimation;
-
the adoption of soft-law measures, and flexibility in implementation;
-
policy-making processes and mechanisms favouring experimentation and knowledge creation.Footnote 16
However, none of these characteristics is entirely new either. In one form or another, they have been integrated into traditional governance, be it as part of the process of policy formulation, or as second-best options, or as a growing trend in international policy coordination.Footnote 17 From this perspective, what distinguishes ‘new modes’ from ‘new, old governance’ is the more systematic application of these features, with some people claiming that they are embedded in a new architecture or ecology of European governance.Footnote 18
Whether we see NMG as supplementing or substituting for the older forms, with differences a matter of degree or kind, three core elements stand out. First, NMG are heterarchic, Second, they opt for soft-law and flexible instruments, embracing a weak conception of authority and uniformity in organization and policy-making. Finally, they privilege deliberative, consensus-based and reciprocal learning forms of policy-making and problem-solving. It remains to be seen whether either the old governance of the CM or these NMG can sustain forms of representation that go beyond those traditionally associated with standard democratic processes.
REPRESENTATIVE CLAIMS IN DELEGATION AND NEW MODES OF GOVERNANCE
To represent is either to ‘act for’ or ‘stand in place of’ someone (or something) else. Representation involves a paradox: to make present what is absent.Footnote 19 This paradox suggests that representing is a mental construction of a complex relationship in which both the means of representation and the nature of what is represented are continuously negotiated. This negotiation is particularly true of political representation, and especially of democratic politics, where representatives and represented tend to influence and reflexively redefine their respective roles, perceptions and behaviour.Footnote 20
Political representation is often equated with modern democracy and the ways the institutions of representative democracy translate the will (or preferences, depending on the approach) of the people into political decisions and action. However, as was evident from the tensions between elites and masses when representative democracies first developed in the eighteenth and nineteenth centuries,Footnote 21 democratic representation does not exhaust political representation, or representation more broadly. With the spread of universal suffrage and the emergence of political parties, this tension has become less marked. Democracy and representation now appear as complementary and almost synonymous, rather than as alternative forms of government.Footnote 22 Yet, that they need not coincide has become increasingly evident with the growth of informal and non-electoral forms of political representation.
Analyses of representation in Europe usually focus on the more traditional forms of democratic and electoral representation, and concentrate on either the EP as the representative of European citizens, or the Council as the indirect representative of the European peoples through their governments.Footnote 23 However, these traditional forms of democratic legitimacy are supplemented by informal and semi-formal non-traditional forms of representation provided by the Commission and NMG. Moreover, it is these bodies that exercise the main executive and legislative functions of determining European policy. It is to the description and assessment of the degree to which these alternative types of representation can also lay claim to democratic legitimacy that we now turn.
Representation and Delegation in the Community Method
Although the CM employs the hierarchical and authoritative structures of decision-making typical of old modes of governance, it lacks the classical features of representative and democratic government found in national and federal states. As many commentators have observed, neither the construct of ‘people’ nor that of ‘government’– both central to the idea of representative government – apply easily in the EU context. The existence and feasibility of a European demos is famously contested,Footnote 24 while what Hix calls the ‘double executive’ arrangement of the CM can hardly be described as a government in the traditional sense.Footnote 25 We noted above how the EP and Council can claim to be respectively the direct and indirect representatives of European citizens and peoples and to have a responsibility towards their electorates and, in a loose sense, be held accountable for what they do. But, as we have also seen, the CM does not formally rest, as representative and democratic governments do, on its capacity to fulfil the mandate that comes through the formal channels of political representation. Though the Council and the EP can sanction decisions taken at a European level, they have no monopoly over legislative and executive matters. Rather, it is the Commission – a non-elected and non-majoritarian institution – that plays the crucial role of initiating policy. Yet it lays claim to be representative in a different way in virtue of its independent status.
As already remarked, Majone has characterized the resulting institutional structure of EP, Council and Commission as a modern version of mixed government. The key aspect of this arrangement is that each of the three main institutions is ‘bearer of a particular interest that it strives to protect and promote’.Footnote 26 Unlike the separation of powers, within mixed government the separate institutions are politically and not merely functionally distinct: they are separate political centres, and not separate organs of the state.Footnote 27 They do not operate in distinct spheres of competence but rather cooperate in the decision-making process, bringing to the table different interests whose valence and relative force depend on the nature of the issues at stake.
At the core of the CM's mixed government is a basic dualism of power between the MSs represented in the Council and the other European institutions. The Commission and the EP act as the bearers of supranational interests; but whereas the EP is elected by European citizens, and can claim to represent them and their interests (at least formally) at the EU level, the Commission is not, nor is it accountable to the citizens directly – only very indirectly via the EP, which approves its members. If anything, the Commission is an expression of the Council, and its composition reflects the composite nature of the EU as an organization of separate MSs.Footnote 28 Is there, therefore, another way in which the Commission can be said to ‘represent’ European supranational interests? We contend that the Commission's representative claim results from three separate processes, which can be categorized as functional, societal and delegatory forms of representation, whose combination sets it aside from other European institutions.
Part of the Commission's claim to represent European interests, and thereby justify its role as the formal agenda-setting institution, is its functional responsibility to act as the guardian of the Treaties. This responsibility produces a norm-orientation to act ‘on behalf of the abstract “European interest” as defined in the Treaties’.Footnote 29 Although the Commission is neither specifically nor personally accountable for the way in which it interprets this task, its position as a kind of representative of the European interest is formally sanctioned by the EU Treaties, and the allocation of powers and responsibilities within the CM. The point is made explicitly in the White Paper on governance, which describes the CM as providing two filters through which the policy-making process arbitrates between different interests. One filter is provided by ‘democratic representation’ via the Council and EP; the other is that of ‘the general interest at the level of the Commission’.Footnote 30 In this regard, the Commission's role as a functional representative is not dissimilar to that attributed to national constitutional courts with respect to their constitutions; with the important difference that the Commission operates more directly as a political and policy-making actor, so there is no pretence of its operating as a purely judicial power.
From another perspective, the functional representation claimed by the Commission approximates what Dryzek and Niemeyer have recently called ‘discursive representation’. They argue that, given that no definite ontological ground exists to identify the precise entities deserving political representation, discourses – as well as individuals, particular aspects of a person, or groups – can be legitimately represented politically. Indeed, they contend that discursive representation is particularly ‘feasible when the representation of persons is not so feasible (especially in transnational settings lacking a well-defined demos)’.Footnote 31 From this perspective, the Commission qua guarantor of the EU Treaties represents the discourse of European interests. Theoretically, this approach limits the degree to which the Commission can act as the bearer of the European interests to the way these are defined by the Treaties. In practice, though, that leaves ample latitude for the Commission to interpret Europe's interests as it sees fit. The Commission de facto constitutes as well as represents the discourse of Europe's interests by virtue of its power to promote policies at the European level. This blank claim to representation gives the Commission a privileged position, at least in principle, even if historically the political conditions in which the Commission operates drastically limit its ability to shape the European agenda and its capacity for autonomous action.Footnote 32
However, such functional representation is rather abstract, and provides the Commission with a weak and ill-defined basis for its claim to represent the general European interest. Moreover, the claim can appear self-serving, while its validity rests more on an output rather than an input perspective. As a result, the Commission has placed increasing emphasis on a second type of representative claim typically couched in the language of democratic participation. This second claim, perhaps best characterized as societal representation, consists in the development of procedures and institutional settings for the consultation of CSOs and social partners as a more integral part of European governance. Such initiatives have long existed, going as far back, for example, as the network-building with social non-governmental organizations that the Commission promoted with the first anti-poverty programme of 1974. But they were foregrounded in the official discourse of the Commission and became more formalized with the White Paper on governance and its commitment to create a ‘culture of consultation and dialogue’.Footnote 33
Rather than establishing legal rules and procedures that risked slowing down considerably the process of policy-initiation and decision-making, the White Paper suggested tightening up standards for consultation, making it an essential part of the policy process with less ad hoc criteria and procedures for selecting and involving the relevant civil and social organizations. Additionally, the White Paper suggested these organizations be required to ‘tighten up their internal structures, furnish guarantees of openness and representativity, and prove their capacity to relay information or lead debates in the MS’.Footnote 34 This opening-up was especially necessary in those key policy areas where there were already established histories and channels of consultation, for which the White Paper envisaged a structured dialogue through ‘partnership arrangements’.Footnote 35
This new emphasis on civil society participation and consultation has been described as heralding a new regime in European governance that introduces a participatory model alongside the established expert-based and partnership-based models.Footnote 36 The attempt to increase civil society's involvement in European governance also seeks to enhance the representativeness of the CSOs and the European institutions that consult them, above all, the Commission. In the words of the White Paper, a more participatory and consultative regime will make the Commission ‘better placed to act in the general European interest’Footnote 37 – not least because it can claim to have listened to a representative sample of relevant European opinion on the issues it tackles, thereby enhancing its standing vis-à-vis those EU institutions with an electoral mandate.
The Commission has no common systematic consultation regime, with practices differing across policies and directorates, although these differences can be justified as reflecting the nature of the policy good concerned and the type of group that needs to be consulted. While decision-making remains technocratic, formal arrangements for involving social partners, such as the tripartite Advisory Council of representatives from governments, employers and unions used for occupational health and safety policy or consultations with the European Economic and Social Committee, have worked reasonably well.Footnote 38 However, the evidence is more mixed with regard to the broader consultation with civil society. The selection of CSOs remains biased towards Brussels-based umbrella organizations, remote from the constituencies they purport to represent, while lobbying of the Commission favours business and professional organizations over public interest groups by the order of 76 per cent to 20 per cent, and the older and larger over the newer and smaller MSs.Footnote 39 Moreover, CSOs are often financed by the EU and have typically been employed by the Commission in the preparatory phases of policy-making to legitimate the extension of its influence into areas with a weak or non-existent Treaty base.Footnote 40 As recent studies have emphasized, the Commission manages lobbying by both firms and societal interest groups to create its own insider organizations that foster trust between elite groups and Brussels officials and improve the flow of information from relevant parties to the Commission and the making of credible regulatory commitments, but also allow collusion with and capture by groups with the best organizational and resource advantages.Footnote 41 Participation tends to be limited to the early stages of the policy process, continues to be through informal and semi-formal rather than formal channels, offers little scope for feedbackFootnote 42 and excludes critical voices unwilling to exchange the possibility of initial consultation for subsequent passive compliance.Footnote 43
Finally, the third sense in which the Commission aims to represent Europe-wide interests derives from its character as a non-majoritarian, supranational regulatory agency to which extensive powers have been delegated. Although, conceptually, delegated and functional forms of representation are not neatly distinguishable, the way the Commission acts as a delegate is specific enough to be considered as a separate category. Its role as a regulatory agent follows from the introduction of the Single European Act and the growth of its competences in the area of competition policy so as to facilitate the working of the single market.Footnote 44 Using Franchino's categories, one can locate the Commission at the receiving end of two processes of delegation: one, which Franchino calls ‘Treaty delegation’, has the EU MSs as the direct principals; the other, ‘executive delegation’, has the EU legislators, mainly the Council, but occasionally the EP, in the role of principal.Footnote 45
Looked at in a stylized form, regulatory agencies can be considered as the representatives for the principals who select them, give them their regulatory powers and may dismiss them. Indeed, the formal structure of representation, as identified by Pitkin, can be applied easily to delegation:
-
Representation (delegation) involves a representative (agent) X being authorized by constituency (principal) Y to act with regard to good Z.
-
Representation (delegation) involves a representative (agent) X being held accountable to constituency (principal) Y with regard to good Z.Footnote 46
From the perspective of principal–agent theory, democratic representation and regulatory delegation look rather similar, since they can both be nested in an overall chain of delegation of powers from citizens to non-majoritarian institutions, passing through legislative and executive bodies, and occasionally public bureaucracies.Footnote 47 However, as the literature on regulatory delegation shows, at a more substantive level some of the operations, as well as the mechanisms of role-formation and agent's motivation, are distinctive, following dynamics of their own that are not entirely reducible to political, or even bureaucratic forms of representation.
Majone characterizes the reasons for delegation as being primarily of two kinds: the reduction of decision-making costs, and the enhancement of commitment and long-term credibility. These two reasons align the principal's and agent's preferences in different ways to produce two divergent accounts of delegation.Footnote 48 Delegating to reduce decision-making costs assumes that principal and agent share similar preferences. Indeed, the main problem for this kind of delegation is to ensure there are no agency losses. As a result, principals need to design selection procedures and post-delegation mechanisms that avoid dangers such as shirking (when agents follow their own preferences irrespective of their principals'), slippage (perverse institutional mechanisms that make agents' preferences diverge from their principals'), and capture (when agents collude with the actors whose behaviour they are meant to regulate).Footnote 49 This form of delegatory representation parallels that of mandated political representatives, for whom electoral mechanisms serve to guard against these risks. However, it is unclear that appointed, non-majoritarian bodies have anything as effective as electoral accountability to keep them on their toes. Indeed, their main representative claim rests on the second reason which, by contrast, requires that delegates be insulated against the need for undue responsiveness to their principals' preferences.
The rationale for delegation to guarantee market credibility and maintain commitments assumes principals suffer from akrasia and act for short-term personal advantages at the expense of long-term collective benefits, even if they ultimately stand to gain from them. Principals can avoid this dilemma by adopting a pre-commitment strategy and selecting agents whose incentive structure coincides with the long-term commitments required by markets rather than the short-term popularity that politicians typically need to court. One consequence of this de-alignment of preferences between principal and agent is to increase agents' discretion and relax considerably, if not completely, the accountability and control conditions to which they are subjected. Agents no longer represent their principals' own short-term understanding of their interests and preferences, but rather respond to their principals' supposed second-order preference. This involves their agents acting according to their own independent judgement as to where their principals' first-order interests and preferences lie in the long term, so as to produce the commitment and credibility required by the market. Thus, when the Commission acts as a non-majoritarian regulatory agent of the European market, it does not directly represent the MSs (as in Treaty delegation) or the EU legislative bodies (as in executive delegation), but rather represents the long-term interests of the European Union and its MSs, even if its reading of these interests and preferences diverges from how the other institutional actors perceive them.
The implicit view of the Commission's representative role within this conceptualization of delegation reinforces other characteristics of how non-majoritarian regulators operate. These features proceed from the way these institutions have emerged as distinct and semi-autonomous actors from governments. Many have developed what Coen and Thatcher call ‘relational distance’, with their own modus operandi and internal and self-referential organization.Footnote 50 Thus, the European Commission ‘has created its own network of national competition regulators’, thereby moving delegation further along the line, and making the decisions of these regulators even more distant from their principals. Moreover, the delegation of power to non-majoritarian regulators works as a kind of incomplete contract.Footnote 51 The kind of actions and intervention they may need to undertake cannot be fully predicted when power gets delegated, creating a considerable area for them to exercise their discretion in potentially arbitrary ways.
The account of representation implied by the Commission's activity as a delegated regulator in this second sense is that of ‘trusteeship’. As Majone observes, it acts as a ‘fiduciary’.Footnote 52 This fiduciary role is further reinforced by the functional and societal modalities of representation discussed above. Acting in each of these modalities, the Commission claims autonomously to interpret and express the European public interest. It does so in accordance with the various tasks it performs within the CM: as the guarantor of the Treaties, as the mediator with CSOs of European society's interests, and as an expert-based, non-majoritarian regulatory institution. However, as we shall see in the next section, it remains unclear on what basis the Commission can claim to be actually representing Europe and Europe-wide interests, and whether the mechanisms through which the Commission interprets the public interest are effective and have democratic legitimacy. For, though Burke famously championed trusteeship as the prime responsibility of political representatives, it has generally come to be seen as being at odds with their democratic status as elected servants of the popular mandate.Footnote 53
Representation in the New Modes of Governance
Most of the representative claims made in support of NMG parallel those invoked for the Commission. Indeed, the three modalities of representation – functional, societal and delegatory – apply here too. We shall start by briefly examining delegatory representation since it operates on the same principles, and through very similar mechanisms to those described for the Commission. We shall then explore the more distinctive forms that societal and functional representation take in this case.
Once again, the delegatory modality of representation follows from the delegation of regulatory tasks to independent and non-majoritarian agencies, whose main task is creating markets or correcting their behaviour. Much of the discussion of the fiduciary role played by the Commission can therefore be transposed to NMG more broadly. However, in this case the principal is often the Commission, which, as we have seen, acts in its turn as the agent for other principals. Consequently, with NMG the chain of delegation gets extended even further, increasing the scope for private organizations and self-regulation and widening the relational distance identified by Coen and Thatcher as one of the ways in which regulatory agents acquire more autonomy and discretion in the decision-making process.Footnote 54 Most importantly, regulation is seen from a sectoral perspective, and so is more dispersed and self-referential.
A similar process of diffusion and segmentation applies to the societal modality of representation. Societal representation by CSOs has two rationales within European governance. The first is to provide one of the preconditions for political representativeness by constructing a social constituency for the European polity in formation.Footnote 55 It offers a social point of reference for a political system whose links with European citizens are tenuous at best. Kohler-Koch calls this process an exercise in ‘imaginary representation’, and considers it a ‘category that is supposed to help understand the formation of a “political system” and not to assess the democratic functioning of the EU’.Footnote 56 The second rationale for societal representation is linked to a perception of participatory governance, whereby policies are said to be more responsive and more likely to be regarded as legitimate through involving those affected in making them.
Both these rationales risk subsuming the European public as a whole into the plethora of civil and social interest-based organizations that EU institutions choose to consult. Nevertheless, we saw how, within the CM, the Commission plays a mediatory and filtering role that – however imperfectly and self-serving in nature – at least attempts to give some unity to the variety of concerns expressed by different autonomous forms of social organization. It seeks to reconcile different interests and in various ways to synthesize them to express an overarching European interest. By contrast, within NMG the representation of social interests and particular concerns is often even more haphazard. Different bodies and countries offer different degrees of access, participation tends to be informal and, as is invariably true within the CM, restricted to preliminary consultations. At the EU level, only those groups who can afford to be in Brussels are involved; political parties are excluded, and, as in the CM, experts and technocrats are favoured – especially in comitology, with only MS representatives and the Commission having formal participation rights on the relevant policy-making bodies.Footnote 57 Indeed, in some cases NMG arrangements for involving social partners and stakeholders have merely detracted from more inclusive mechanisms established within the CM.Footnote 58 As a result, NMG threaten to be still more partial and arbitrary in their representativeness of civil society as a whole.
In fact, the participation of stakeholders and the organizations of affected parties in NMG is usually justified less on the grounds that they give access to more groups and hence are more representative per se, and more because they are thought to produce better decision-making in a specific sector or activity. However, such partial involvement, and the fact that it is largely consultative, means that at best it gives these decisions a spurious legitimacy, at worst it fails to assess the impact of policies in a given sector on social interests not directly relevant to its operations or their relative importance compared to measures in other sectors. In this respect, societal representation merges into the functional modality of representation. As we noted, the main features of NMG are said to lie in their offering a heterarchical structure of authority, greater flexibility and enhanced problem-solving capacity. These qualities supposedly produce a more deliberative decision-making process aimed at achieving consensus on the best policy rather than bargaining to reconcile competing interests, while nevertheless remaining attuned to the diversity and multilevel nature of the EU. By bringing together state and non-state actors, experts and social representatives in a forum oriented towards the sharing of knowledge, mechanisms such as the OMC and network governance are held to foster ‘a reasoned discourse between expert and lay people’Footnote 59 that overcomes a ‘deficit of mutual awareness between civil society and public authorities’.Footnote 60 Strategies such as benchmarking and peer review supposedly encourage those involved to adopt the better rather than merely convenient or self-serving practices, while respecting relevant and legitimate differences stemming from the autonomy and distinctiveness of the MS.Footnote 61
However, one can just as easily imagine these structural constraints having precisely the opposite effect – of producing only mutually advantageous, Pareto-optimal improvements that benefit those already privileged within the status quo by leaving existing inequities and inefficiencies intact. Fair and equitable policies will only emerge from this process if the representation is itself fair and equitable between the main concerns that need to be aired, or if the representatives see themselves as serving public rather than sectional interests. Yet, few if any criteria exist to ensure that representation fulfils the requisite standards of either fairness or publicness. It might be argued that when dealing with the largely technical questions that form the bulk of the EU's business both can be met through appointing national experts. However, not only can experts often disagree on technicalities, but also most technical questions have broader social and economic effects. Again, such policy spillovers may go unaddressed without fair representation of non-technical parties from both within and, as we noted above, outside the sector. Health and safety standards for food production, for example, will have implications for a whole range of actors and policy areas – affecting the viability of different kinds of farming practice, consumer choice, regional development, environmental policies and so on. Quite how all these might be appropriately factored into the deliberative process to ensure all receive their due weighting within the discussion remains obscure. Indeed, the evidence thus far suggests little deliberation occurs within the NMG, be it due to time constraints, the absence of a plurality of actors, or a lack of commitment on the part of those involved.Footnote 62
As with the CM, the claim that NMG represent the European public interest appears more rhetorical than real. Moreover, with NMG this problem gets exacerbated by the fragmented and partial way social interests are represented. But the main issue concerning both modes of governance is whether, conceptually and normatively, clear mechanisms could be said to exist for ensuring that political representation occurs.
CONCEPTUAL AND NORMATIVE ASSESSMENTS
Following Pitkin,Footnote 63 one can distinguish formal from substantive concepts of representation. Formal understandings contain two key elements, though different theories may focus more on one or the other: namely, authorization and accountability. These provide the formal processes through which representation takes place. We referred to these formal structures above in noting how representation consists in X authorizing Y (with regard to Z), and, at the same time, X being accountable to Y (with regard to Z). By contrast, substantive understandings concern the way in which the representative relationship works. Pitkin suggests that, broadly speaking, substantive concepts view representation as a way of either ‘standing for’ or ‘acting for’ someone or something else.
‘Standing for’ suggests a somewhat passive way of taking someone's place and ‘acting for’ a more independent way of doing so. However, such a sharp distinction would be overdrawn. Representatives who ‘stand for’ others may be obliged to act in a way that reflects their principals' preferences or spirit but they rarely have a precise imperative mandate from them detailing how they should act on all occasions. Thus, ‘standing for’, which can take descriptive or symbolic forms, allows representatives some lee-way for interpreting their role and a degree of independence in how they perform it. However, overall the idea of ‘standing for’ sees representation as involving either a correspondence of interests and views between the representative and those he or she represents, or a mirroring or reflection of those being represented in those that represent them, for example in their sharing a given quality such as gender or colour.
By contrast, ‘acting for’ focuses on the substance of the activity performed by the representative. As Pitkin says, ‘we are now interested in the nature of the activity itself, what goes on during representing, the substance or content of acting for others, as distinct from its external and formal trappings’.Footnote 64 Nevertheless, there is a range of ways of ‘acting for’ another person: be it as a substitute, a trustee, a deputed agent, a fiduciary, or an expert. As we shall see below, each of these ways of ‘acting for’ involves a different view of the relationship between the representative and the represented. Whereas some ways come close to the ‘standing for’ model, whereby an agent acts as their principals could be expected to, had they the ability or standing to do so; other ways involve agents acting for the benefit of their principals, even if their view of what would most benefit them conflicts with their principals' own reading of their best interests. However, all these cases differ from the ‘standing for’ model in involving some weaker or stronger sense in which representatives exercise their own judgement as to what is necessary to secure the best outcome for the represented, be that a view or knowledge of the most appropriate means or, in the strongest case, of what the better outcome would be.
Analysing the representative claims of the Commission and NMG through Pitkin's grid, it is clear that they emphasize the substantive over the formal aspects of representation. At most, particularly in the case of the Commission, their claims resemble those made by traditional structures of bureaucratic representation, whose representativeness in terms of authorization and accountability is nested within a more general structure of political and democratic representation involving some appeal to the electorate and public opinion more broadly. However, in the European case these appeals are extremely tenuous, because the chain between the general public and those who make decisions is such a long one. For example, the Commission may be appointed by the MS and subject to approval by the EP, but once in office they operate with a high degree of independence. Likewise, the delegated representatives in regulatory and other non-majoritarian bodies tend to be ‘authorized’ by governments or the Commission whose own formal authorization by, and accountability to, a European public is rather thin. Moreover, they then enjoy considerable discretion in view of their role as either experts or credible agents. Indeed, their credibility is often held to depend on criteria not only external to but deliberately insulated from any need to reflect or be accountable to the declared preferences of those they represent.
The representative claims of European governance are therefore mainly based on the substantive understandings of representation. What we termed the ‘societal’ modality of representation gestures towards the ‘standing for’ conception. As we saw, civil society organizations are treated as reflecting the diversity of European society at large. But in the absence of systematic formal mechanisms of authorization and accountability, such claims are more symbolic than descriptive. Whereas elections give all voters the opportunity to express their views on a range of policies on an equal basis, with majority voting in the context of a system of competing parties offering a fair means for aggregating their different preference schedules in a way that packages them so as to reflect roughly the preference schedule of the electorate,Footnote 65 no equivalent mechanisms exist to ensure that consultations with civil society fairly describe the balance of social views overall. Instead, there is a real danger that this system will overly respond to those with the organizational resources and commitment to gain access. In practice, civil society representatives no less than functional and delegated representatives end up making their representative claims on the grounds that they ‘act for’ a European society that has yet to develop the capacity to represent itself.
As Pitkin observed, the ‘acting for’ conception poses two main problems.Footnote 66 The first is the conceptual problem of whether we can have representation without formal authorization and accountability – how can we know that such a person is truly acting as a representative? The second is the normative problem of whether any independent purely substantive criteria exist that can enable us to ascertain what it means to act for the good of someone else (or, in our case, in the public interest).
The conceptual problem arises because ‘substantive acting for others,’ as Pitkin puts it, takes many different forms, not all of which can be categorized clearly as a form of representation. Pitkin distinguishes five major forms,Footnote 67 which we list here in descending order, from the weakest, which require a stricter adherence to the instructions of the represented, to the strongest, which give considerable autonomy to the person who acts:
-
those forms that refer to ‘acting for’ as the act of someone who is sent or delegated to do something specific (an ambassador, for instance);
-
those forms that refer to ‘acting for’ as a kind of ‘substitution’ (an attorney or someone acting in a vicarious way);
-
those forms that refer to ‘acting for’ as the action of an agent, who can, however, be considered as a ‘mere’ agent, or as a ‘free’ agent (this ambiguity is typical of the role played by elected representatives, even though Pitkin emphasizes that an important distinction remains between an ‘agent’ and a ‘representative’, a point to which we return);
-
those forms that refer to ‘acting for’ in the sense of taking care, or ‘acting in the interests’ of someone (a trustee and a guardian are the most common examples); and
-
those forms associated with the idea that experts and professionals are acting in the interest of others (a physician, for instance).
These five forms cover a semantic field according to which we can interpret ‘acting for’ in the two polar senses of either acting ‘instead of’ (mainly forms 1 and 2), or acting ‘to the benefit of’ (forms 4 and 5), with form 3 nicely poised in the middle, since it can be interpreted in either sense. For Pitkin, this ambiguity applies more generally to the very activity of representation.Footnote 68 But looking at these different forms within the context of our discussion of European governance, we can confidently assert that both the Commission and the various instruments of NMG are closest to the strongest forms 3 to 5. The implication is that their role as ‘representatives’ is mainly a function of their actions and decisions being beneficial to the European public, rather than expressing the public.
As Pitkin notes, it is unclear in what sense these latter claims are representative claims at all. What these strong senses of ‘acting for’ lack is the idea, central in political representation, that the represented is present (hence also responsible) in the action of their representatives. Pitkin distinguishes between a ‘representative’ and an ‘agent’.Footnote 69 The latter ‘does the actual work’ for someone else, so is a kind of tool or instrument; while the former is not a simple instrument, because he or she acts as if they were that person. The ‘presence’ of the represented in the action of the representative can only be understood if we take the substantive aspect of the idea of representation (‘acting for’ or ‘standing for’) in conjunction with the formal aspects of authorization and accountability. But, as we have seen, these formal aspects are lacking in NMG, and this weakens the sense in which we can say that they represent the European public. The suspicion arises that they merely represent their own view of what a European public would want. Yet such a public may not exist – it could be entirely ‘imagined’.
The normative problem enters here. For their own view need not reflect their own self-interest. It may represent a correct view of the interests of those they ‘act for’. The difficulty is how we can know whether this is the case or not. Again, formal mechanisms of authorization and accountability seem necessary. These exist for physicians and lawyers, say. There are professional standards that they have to meet, and bodies that authorize that they have met them and to whom they are accountable for continuing to do so. As a result, one can at least say they are agents qualified to act for others in pursuance of a given task. It is unclear whether anything similar exists for political representatives other than a democratic process. As we saw, it was claimed that NMG offer a process of public reasoning, yet we observed how the structural constraints in themselves were unlikely to produce such a result unless there was a fair representation of the public involved. No metric for ensuring reasons do reflect the public interest exists beyond their exposure to public challenge. Likewise, the qualities needed to represent the public lie in large part in the ability to take the public with you. The selection and sanctioning processes of elections serve both these purposes. Of course, the stronger senses of ‘acting for’, such as guardianship (form 4) or expertise (form 5), suggest that their principals lack the ability to see their own interests for themselves. By analogy, these cases suggest that we should view the European public as too immature, irrational or ignorant to perceive where their own interests lie. Yet that claim constitutes a somewhat paradoxical basis for democratic legitimacy as well as being deeply contentious.
CONCLUSION
We have argued that EU governance makes representative claims of a functional, delegative and societal nature.Footnote 70 However, these claims are of a substantive kind to ‘stand’ or ‘act’ for a European public. The formal mechanisms that might allow this putative public to ‘authorize’ these claims and hold those who make them to ‘account’ are residual and imperfect at best. Yet without such formal mechanisms it is unclear whether these claims can be regarded as representative in any meaningful sense. No European public is ‘present’ within the activities of their putative agents. We have only hinted at the likely consequences of such attempts to separate ‘substantive’ from ‘formal’ representation. However, a potential danger exists that those interests that are represented are so partial – being either expert or bureaucratic delegates easily captured by the governmental or commercial interests they are supposed to control, or CSOs that are the creatures of those who not only consult with, but largely finance them – that this system risks magnifying the disadvantages of pork barrel, pressure group politics often associated with conventional democracy, without the benefits of its compensating advantages of promoting political equality and responsiveness.Footnote 71
