Abstract
International trade union organisations, like unions at national level, commonly affirm their commitment to internal democracy. But what does this mean? There exists a vast literature on union democracy, addressing the questions whether democracy in trade unions is desirable; whether it is possible; and if so, how it can be achieved. However, the focus of analysis is almost exclusively at the national (or sub-national) level, with the premise that union members are individual workers. But international unions (like many national confederations indeed) do not have individual workers as members: they are organisations of organisations. What does this imply for our understanding of union democracy? We begin our article by summarising the broader literature on union democracy, then develop an interpretation of international unions as ‘meta-organisations’. We next explore some of the implications for debates on democracy at international level, and end by asking whether theories of deliberative democracy can help in understanding the options for international union democracy.
Introduction
International trade union organisations operate in many forms. At global level, there have long existed cross-sectoral bodies comprising national confederations, with the International Trade Union Confederation (ITUC) now the only significant example. Sectoral or industrial federations have an even longer heritage, originally known as International Trade Secretariats (ITSs) but reconstituted as Global Union Federations (GUFs). The global organisations have their regional counterparts, often in a subordinate role. Exceptionally, the European Trade Union Confederation (ETUC) is an autonomous body which however maintains close relations with the ITUC; and the sectoral European Trade Union Federations (ETUFs) vary between a high level of integration within their associated GUFs and relative independence. Together these bodies constitute ‘a functionally and spatially complex structure of transnational trade unionism’ (Plater and Müller, 2011: 8).
Like their national trade union affiliates, international unions commonly insist that their representative legitimacy derives from their character as democratic membership organisations: this provides a mandate lacking in the case of many social NGOs. According to its Constitution, the ITUC ‘is open to affiliation by democratic, independent, and representative trade union centres, respecting their autonomy and the diversity of their sources of inspiration, and their organisational forms. Its rules are to guarantee internal democracy, full participation of affiliates, and that the composition of the Confederation’s governing bodies and its representation respect its pluralist character’ (ITUC, 2018: 6). IndustriALL explains in more detail (2016: 2–4) that strong democratic unions are essential to social equality and democracy […]. Through organizational development, we build strong, united, democratic, independent, representative and self-sustaining trade unions throughout the world […]. Unity must be founded on democratic principles […]. IndustriALL is democratic and transparent in its policies and practices throughout its structures at global, regional, sectoral and national levels.
However, a crucial difference between the international organisations and individual trade unions at national level (though a characteristic which they share with many national confederations) is that they are ‘associations of associations’ (Platzer and Müller, 2011: 864): their members are not individual workers but collective organisations. One of the resulting challenges for democratic governance is that international unions are further removed from their constituents than are national unions. More crucially, though, the fundamental principle of most conceptions of union democracy – ‘one member, one vote’ – simply does not apply at international level. Global unions are an example of what have been termed ‘meta-organisations’, and one of our aims in this article is to show how the literature on this theme can help in understanding the dynamics of international trade unionism.
We argue that ideas of union democracy formulated in the context of national trade unions must be radically reinterpreted in the content of international unionism: many of the challenges are distinctive, and so must be the solutions. We seek to contribute to this reinterpretation by examining how far ideas of ‘deliberative democracy’ – often proposed as a potential solution to a perceived ‘democratic deficit’ in the European Union (EU) – may offer a distinctive route to strengthening internal democracy in international trade unionism.
The contested meanings of union democracy
There exists a vast literature on union democracy, dating back more than a century. However, the focus of analysis is almost exclusively at the national level, with the premise that union members are individual workers. There are conflicting views on whether democracy in trade unions is desirable and whether it is possible; and if so, what union democracy actually means and how it can be achieved. When Sidney and Beatrice Webb published their ‘scientific analysis’ of British trade unionism, they gave it the title Industrial Democracy. They argued (Webb and Webb, 1897: xix) that ‘Trade Unions are democracies: that is to say, their internal constitutions are all based on the principle of “government of the people by the people for the people”’. And indeed, as we noted above, most trade unions insist, with reason, that they are democratic organisations; or at least, to use contemporary management-speak, they are ‘striving to be democratic’.
Only a minority of analysts have questioned the need for union democracy. For some, effectiveness and democracy in trade unions are incompatible. In the mid-20th century, Allen (1954: 15) insisted (in an argument he later disavowed) that trade-union organization is not based on theoretical concepts prior to it, that is on some concept of democracy, but on the end it serves. In other words, the end of trade-union activity is to protect and improve the general living standards of its members and not to provide workers with an exercise in self-government.
As in politics more generally, it is more common to accept that union democracy is in principle desirable, but to define its meaning narrowly: in particular, to advocate ‘passive’ democracy, whereby members have mechanisms through which to assent to (or dissent from) decisions taken in their name, rather than ‘active’ democracy, whereby they can actually initiate policy collectively. This approach can also be traced back to the Webbs. They opened their analysis of British trade unionism (1897: Ch. 1) with a somewhat patronising account of ‘primitive democracy’: the earliest unions were committed to the active involvement of the whole of the membership, but this was feasible only in small, local unions with a homogeneous membership who formed a tight-knit occupational community. Such a model was later described by Turner (1964) as ‘exclusive democracy’. For the Webbs, the only viable model of democracy in modern trade unionism was a parliamentary form, in which the members elected representatives who shaped overall policy and who oversaw a cadre of specialist officers who undertook the day-to-day management of union business. The latter, however, came to constitute an official governing class, more and more marked off by character, training, and duties from the bulk of the members. The annual election of the general secretary by a popular vote, far from leading to frequent rotation of office and equal service by all the members, has, in fact, invariably resulted in permanence of tenure exceeding even that of the English civil servant. (Webb and Webb, 1897: 16)
Undeniably, effective trade union action requires some degree of overall coordination together with the definition of strategic priorities which unite the membership, so that energies are not dissipated in a multiplicity of disparate and perhaps mutually contradictory initiatives. Certainly any large, complex trade union requires professional leadership competences. But in any context, the ethos of ‘managerial unionism’ contradicts the appeal of trade unionism as ‘sword of justice’ (Flanders, 1970) directed to the self-emancipation of the weak, vulnerable and oppressed. When the traditional structural and associational power resources of trade unions are almost everywhere diminished, and their institutional power resources are increasingly precarious, effectiveness requires active commitment and openness to mobilisation among the members (and potential members), together with a capacity to inspire broader societal support through a social vision. ‘The managerial emphasis on administrative rationality risks diminishing the representative rationality of trade unionism, which is grounded in multiple spaces of representation and participation, as well as in exchange opportunities between different organizational levels’ (Thomas, 2013: 33). In Britain, Flynn et al. (2004: 328) note a similar trend towards centralisation, meaning that ‘union branch, group and regional committees have become dependent on the centre for resources, significantly curtailing their autonomy’. Hence ‘the redefinition of union democracy around centralized decision making conceived as answers to the decline of activist participation and membership losses, in turn, may feed these very phenomena’ (Thomas, 2017: 667).
One counter-argument to the ‘efficiency trumps democracy’ thesis is that union advocacy of ‘democracy at work’ lacks legitimacy unless unions themselves can demonstrate their democratic credentials (Gumbrell-McCormick and Hyman, 2019). Another is that to be effective, trade unions must be able to mobilise collective action, which implies a ‘willingness to act’ among the members (Offe and Wiesenthal, 1985). Case studies of US unions have demonstrated that active participation and deliberation can be a source of strength. An analysis of the Longshoremen by Levi et al. concluded (2009: 223) that ‘democracy may sustain and facilitate hard bargaining [and] it may even facilitate organizing’; while in his account of the successful strike by the Teamsters against UPS, Parker (1998: 57) insisted that ‘democracy, it turned out, was not just the icing on the cake, but the very foundation of union power in a critical struggle against a corporation’. As Lévesque and Murray suggest (2003: 16), organisational power requires effective processes of internal democracy, together with ‘a culture favouring discussion between rank-and-file and officials and educational work to ensure that policies are well understood and reflect the conditions experienced on the ground’; and we may regard modern information and communications technologies as potentially valuable instruments to this end – even if this potential is often unfulfilled (Geelan and Hodder, 2017). As we suggest below, such considerations are also relevant in the context of international trade unionism.
The diversity of national models
‘The meaning of “trade union democracy” is different in different countries and indeed within countries’ (Edwards, 2005: 265). There is great diversity in formal decision-making structures, both within and between national union movements. The implication of the debates discussed above is that unions need both membership involvement and strategic leadership, and the tension between these two requirements creates an enduring dilemma for trade union democracy. How can unions be actively democratic, not only relying on the passive consent of members who have other priorities? Answers to this question, embedded in the prevailing constitutional arrangements, differ greatly between countries; these differences are linked to historical traditions, which also shape variations in the practical understanding of union democracy and the structures adopted to achieve it, across and often also within countries (Carew, 1976; Martin, 1989; Von Beyme, 1980).
The relative powers of national officers, executive committees and conferences, the degree to which middle-range officials are elected from below or appointed from above, all vary. Though the mechanics of its implementation differ widely across (and to a lesser extent, within) countries, trade union movements tend to embrace a two-way conception of democratic policy-making. Members at the grass-roots level meet to discuss policy questions, not least in respect of collective bargaining, elect their own local officers and also choose representatives to participate in higher-level structures (district, regional and ultimately national). In most unions, organisational structures exist at workplace level, but patterns of authority between such structures and the national, regional or local union are complex and shifting; an added complexity in most European countries is the relationship between workplace union representation and works councils.
While cross-national differences reflect diverse understandings of the meaning of union democracy, they also derive from relatively contingent decisions made a century or more ago (for example, unions subject to state repression often adopted highly centralised, almost military methods) which have persisted despite changed circumstances. Unions in some countries (such as Germany) have a high ratio of paid officials to members, others depend heavily on ‘lay’ activists (as in Britain and France); such differences have evident implications for the internal distribution of power. There is also a problematic relationship between formal decision-making structures and the complex and elusive dynamics of real intra-union politics.
It is often suggested that where unions emerged in semi-legality, often under the tutelage of a social-democratic party, they tended to adopt centralised structures – as in Germany over a century ago (Crouch, 1993; Marks, 1989; Taylor, 1989). Conversely, where unions originated within a laissez-faire political framework, as in Britain, their development was more spontaneous and fragmented and principles of ‘primitive democracy’ could more easily take root.
How are paid officials chosen? While there is a general principle that top officials are either directly elected or are chosen by a representative conference or congress, these two mechanisms have very different implications. In some countries, there is a strong tradition of election of lower-level paid officials as well. In the other direction, the democratic credentials of top leaders and executive committees give them the authority to prescribe a policy framework for the lower levels of the union. Declining membership participation is a widespread problem, and the trend in many countries to create conglomerate ‘super-unions’ through mergers can generate ‘crises of interest aggregation and representation’ and a weakening of internal democracy (Thomas, 2017).
Though this is not the main focus of our analysis, we should also note that there are major differences in the balance of authority between national confederations and their affiliated unions. Where confederations were created ‘from below’ by autonomously functioning craft or sectoral unions, their powers and resources are normally limited; where they were created ‘from above’, typically as part of a social-democratic project, the relative powers of confederation and affiliates are very different. Here we should note that in most countries, individual trade union members typically join, and pay their dues to, the affiliated unions rather than the confederation. In this respect, most confederations constitute ‘meta-organisations’, as we discuss below.
Famously (or notoriously), Michels (1915), echoing the Webbs, argued that trade unions and socialist parties were subject to an ‘iron law of oligarchy’. Most members lacked the knowledge or motivation to engage actively in the democratic processes of policy-making; officials had the skills and the personal interests to dominate the decision-making and electoral processes; creating a vicious circle. Even if many of these arguments were one-sided and exaggerated, they contained an important core of truth. Even more today than when Michels wrote, unions struggle to attract a significant attendance at membership meetings. The trend towards ‘mega-unions’ increases internal heterogeneity and the distance between leadership and rank-and-file. Workers have many more exciting ways of spending their leisure time than attending union meetings. Those who do participate tend to possess strong ideological commitments which may incline them to lines of policy which most members fail to endorse. However, over the years trade unionists have developed a variety of institutional practices designed to counteract the tendency to oligarchic rule, and where formal structures have failed to guarantee democracy, informal practices have been used to considerable effect to ensure that rank and file members maintained some control over their own destiny. (Carew, 1976: 21)
A related debate concerns centralisation as against decentralisation in union policy-making. Are decentralised structures more democratic, in that they provide greater scope for membership involvement in decision-making? Such an argument is consistent with participative theories of democracy. But a counter-argument is that (at least beyond a certain point) decentralisation precludes overall strategic direction, a particular problem when key employer strategies are increasingly centralised (Streeck, 1988). Other approaches investigate how the democratic vitality of decentralisation and the strategic coherence of centralisation might be reconciled; for example, Kjellberg (1983) has argued that Swedish unions combine both authoritative national decision-making and workplace-level autonomy over key issues, with close articulation between the two levels providing a source of strength and democracy. Much earlier, Cook (1963: 187) demonstrated that scope for meaningful participation in local structures enhanced the democratic quality of American unions: ‘a natural source of union vitality is the members’ realization that they do indeed have a reason for being, i.e., for coming to meetings, for making decisions, and, in fact, for making the union their union’. As Carew argues (1976: 195), ‘the union must find a balance between the decentralisation of control which leaves decision-making in the hands of ordinary members and thereby serves as a force for education in the democratic process, and the centralisation necessary for effective operation’.
Other analyses have drawn on pluralist theories which were dominant in the USA in the post-war years to argue that the key criterion of democracy in trade unions is the ability to challenge the incumbent leadership. Most notably, Lipset et al. (1956) studied the American printers’ union, in which a highly structured internal party system brought an electoral alternation of leadership. While this case showed that there was no ‘iron law of oligarchy’, its exceptional character was hardly a recipe for union democracy more generally. In Britain, Martin presented a weaker version of the same argument, suggesting that the widespread existence of organised factions within unions was a sufficient condition of democracy: ‘faction is an indispensable sanction against leadership failure to respond to membership opinion’ (1968: 207). Van de Vall, drawing primarily but not exclusively on Dutch experience, developed a similar analysis with his theory of ‘polyarchy’: ‘the polyarchic organization differs from the oligarchic in that, in addition to the powerful leaders and the passive membership, there is a third group, the active participants. By their two-way communication within the organization (controlling from members to leaders and informing from leaders to members), they act as its democratic core’ (1970: 153).
Union effectiveness requires ‘the capacity to interpret, decipher, sustain, and redefine the demands of the represented, so as to evoke the broadest possible consensus and approval’ (Regalia, 1988: 351). This is one of the functions of leadership, which is therefore a prerequisite for participative democracy to deliver beneficial results. As Barker et al. insist (2001: 15–17), it is crucial to differentiate between authoritarian and authoritative leadership, and between leadership as hierarchy and as process or function: ‘leadership is exercised at all manner of levels and locations…and not only by those obviously designated as “leaders”’. Gramsci’s notion of the ‘organic intellectual’ is relevant here: grass-roots activists may develop a breadth of information and analytical capacity which distinguish without distancing them from their colleagues. Hence there can, and must, be a complex dialectic between leadership and democracy. For example, two German studies (Schoefer, 2000; Von Alemann and Schmid, 1993) identify a virtuous circle (or ‘magic triangle’): a transparent process of strategic leadership enables and encourages rank-and-file participation in debates and decisions; more informed and engaged members display greater willingness to act, enhancing union effectiveness; this in turn provides the organisational capacities and leadership confidence that facilitate transparency and participation. In the final section we develop these themes further.
Political theorists commonly argue that a prerequisite for genuine democracy is the existence of a demos: a constituency with a sense of shared identity and interests in common, and such a consciousness has to be socially constructed. Hence the absence of a European demos is sometimes regarded as an irrevocable obstacle to democracy at European level (Innerarity, 2014). How is such a collectivity – a demos – formed? Richards argues (2001: 35–36) that ‘labour solidarity has always been a constructed and contingent phenomenon built on local foundations’; while Dufour and Hege (2002) show that ‘representative capacity’ within union organisations depends to an important degree on the quality of the interrelationships between representatives and their constituents, on the responsiveness of representatives to the often individualised everyday concerns of workers, indeed their readiness to deal with issues arising outside of work itself. Since networks of sociability precede formal collective organisation, they can provide the springboard for unionisation and the resource for effective intra-union dialogue.
The distinctive challenges of democracy in meta-organisations
As we noted at the outset, the premise of most literature on democracy in trade unions is that these are organisations with individual workers as members, where the principle of ‘one member, one vote’ can meaningfully apply. Yet even at national level, this is not necessarily the case. In most countries, as we indicated above, central confederations do not directly recruit individuals as members; it is their affiliated sectoral or occupational unions which do so. Such confederations, as Hartmann and Lau (1980: 366) note, are ‘organizations of organizations’. Here, problems of union democracy are compounded; as they suggest (1980: 367–368), because of the additional level of remoteness from individual worker-members, one might expect Michelsian tendencies to be reinforced. Yet in practice, they argue, confederations are anything but the super-bureaucracies which some of their critics maintain. They are short of formally trained experts, the division of labor between staff and line functions tends to be blurred, the formal hierarchy is somewhat subverted by councils, committees, and task forces cutting across hierarchical levels, and the purposive rationality of professionals for the most part ranks second to political exigencies and an ideological sense of mission. (Hartmann and Lau, 1980: 369)
In their later, more extended analysis of meta-organisations, Ahrne and Brunsson point to a number of problems of governance, asking (2008: 3) ‘how is it possible to lead organizations that already have leaders of their own?’. They add (2011: 61, 108) that ‘a meta-organization and its members compete for identity, autonomy, and authority’ and that variation in membership/resources is a source of potential conflict. Issues of central authority as against subsidiarity generate other conflicts. These problems may be intensified in what they call (2011: 75) ‘meta-meta-organizations’: bodies whose affiliates are themselves meta-organisations. International trade union bodies whose affiliates are national confederation obviously fall within this category.
Regalia (1988: 353), in discussing Italian trade unionism, refers to a ‘precariousness of internal control, or of the ability to make the represented respect decisions taken. From this point of view unions are, so to speak, disarmed organizations: they lack efficacious sanctions for dissenters’. This may be a reasonable representation of the Italian situation, and more generally of countries where rival unions compete for membership, particularly if unions provide mainly ‘public goods’ (Olson, 1965). But typically, the individual member is far more dependent on the union than the reverse, so that expulsion constitutes a genuine sanction. In international union bodies, the balance is far more in line with Regalia’s argument: the international needs its affiliates – or at least, the larger, wealthier and more powerful ones – more than these need the international. In a study of the national confederation in Australia, Brigden (2007: 490) remarks that with affiliation remaining voluntary, any large-scale withdrawal by key affiliates would compromise the capacity of the peak union to be seen to be able to exert organization power over affiliates or, more seriously, to pursue meaningful collective movement power for those remaining. undermined the authority of labour’s voice in international affairs, diminishing the standing of the organization that sought, on behalf of the largest group of organized workers in the non-communist world, to represent the values of trade unionism to the international community at large. (Carew, 1996: 177–178)
The relative resource poverty of international unions by comparison with their wealthiest affiliates can lead to dependence on alternative sources of funding. Harrod (1972: 399) wrote (in relation to the overseas activities of the British TUC) of ‘the absorption of the trade union foreign policy decision-makers into a foreign policy-making elite at the national level […]. They did not act as trade unionists but rather as quasi-government policy-makers’. For officials at supranational (global or European) level, external dependence is even more evident. This relates closely to the requirements of international diplomacy. The predecessors of the ITUC internalised the tripartism of the ILO, founded in 1919; and became its main defenders, in the face of at best limited enthusiasm from most governments and employers’ organisations for a trade union role. The priority of international diplomacy could also result in a ‘revolving door’ between office in the international trade union movement and in the ILO itself, a consideration which might perhaps influence the behaviour of international union officials (Gumbrell-McCormick, 2013b). The ‘social partnership’ which underlies tripartism is likewise a fundamental element in the relationship between the ETUC and the institutions of the EU.
Gläser (2009) has suggested that the ETUC – and the same could be said of the ITUC – faces two dilemmas which are a source of weakness. The first is between representativeness and capacity to act – a tension between the logics of membership and of influence, as Dølvik (1997) puts it, or between broad representativity and homogeneity (Braud, 2000). Mitchell (2014: 418) makes a similar point: ‘for some time, the ETUC has attempted to combine three distinct tactics – negotiation, lobbying, and protest – into a multipronged European strategy. But […] the three roles are ultimately not compatible’. The second tension identified by Gläser is between political independence and financial dependence on the European institutions; or in the words of Martin and Ross (2001), ‘the dilemma of borrowed resources’. The ETUC is to a significant extent subsidised by the Commission, and acquires much of its raison d’être from its role as ‘social partner’ at European level (a status enshrined in the EU Treaties). Hence much of its work is a response to an agenda defined by the EU institutions – ‘un travail en réaction’ (Wagner, 2005: 44). In the case of the ITUC, ‘borrowed resources’ comprise primarily funding from individual affiliates, often indirectly derived from national governments; as Cotton and Gumbrell-McCormick put it (2012: 715), in addition to affiliation fees, the ITUC receives roughly €1m a year in voluntary contributions to its Solidarity Fund, just over half coming from its German and Japanese affiliates. Far more substantial – about €7m a year – are the project-oriented Development Aid Funds. Almost half this funding is provided by the Dutch government and trade unions, with other substantial contributions from the Swedish unions and the ILO. These are advanced and well-organised funding sources but largely depend on the political support of national governments. Given the political shifts in western Europe, the funds are increasingly dependent on the ability to show concrete outputs and benefits for the donor countries.
International trade unionism is a specialist arena of union action, still relatively under-researched. Ford and Gillan (2015: 459, 463) note how little industrial relations literature ‘is concerned specifically with understanding the global unions – a category that includes the GUFs but also the ITUC – as institutions and as social, political and industrial relations actors’. They also stress the diversity of international unions, noting that ‘the GUFs are quite different in ideological and political orientation, their governance arrangements and their capacity for engagement. In terms of governance, most remain hierarchical’.
Traditionally, the interaction between the global unions and their affiliates has mainly involved ‘international experts’ from among the national unions. Officials at this level require an atypical skill set, including extensive language competences which are relatively scarce within national unions. In most countries, it is still normal for top officials of individual unions to have risen from the ‘shop floor’, gaining long experience as front-line negotiators. This is less often the norm in national confederations, where it is common to commence as a researcher and to rise through the internal hierarchy. This is even more the case – though indeed not universally so – in international trade unions. Moreover, Scherrer (2019: 96), writing while ETUC deputy general secretary, has commented laconically that ‘the reasons why individuals are appointed are many and varied; a desire to maintain the political balance of power, a lack of viable alternatives […], the need to preserve an equilibrium between different regions, and sometimes pure dumb luck’.
Hyman (2005) has identified a historical trend in the role of international union leadership from ‘agitator’ to ‘diplomat’, primarily as a consequence of the development of intergovernmental regulation in the field of employment. As Windmuller commented (somewhat patronisingly) half a century ago (1967: 92–93), ‘it does not require a charismatic general secretary, a wealth of resources, or a vast array of enticing rewards to prepare position papers, draft telegrams of protest, urge more speed in upward social harmonization, speak up in support of a new ILO convention, and so forth’. Yet over time, as the challenges of globalisation have intensified, the functions and capacities of international leaderships have expanded markedly. Writing of the GUFs, Croucher and Cotton note (2009: 43) that ‘information is controlled and distributed by the senior officers’ whose ‘power in this respect probably exceeds that of union officers at national level, because of the linguistic barriers and wide range of difficult-to-interpret information involved internationally’, Platzer and Müller (2011: 39–42) suggest five different ‘modes’ of relationship between international unions and their affiliates, depending on their degree of relative authority. At one extreme the international is merely a letter-box for affiliates, at the other it is the driving force in policy formulation and implementation. Over time, most trade union internationals have shifted from the lower levels of central authority towards the higher, involving what Platzer and Müller (2011: 875) term the ‘management of interdependence’.
(How) can international unions be democratic?
How far are traditional conceptions of union democracy applicable to the governance of international unions? Ahrne and Brunsson suggest (2008: 169) that it is difficult to imagine how meta-organizations could be fully democratic and be similar to democratic states or democratic individual-based associations […]. In meta-organizations, it is difficult to define a democratic order. Is it more democratic that each member has a vote, or that each member’s votes be proportionate to its own number of members?
Nevertheless, the distinctive character of meta-organisations means that one of the foundations of Michels’ analysis of oligarchy – the subservience and apathy of the ‘masses’ and the superior expertise of the leaders – evidently does not apply in the case of international trade unionism. Here, the representatives of at least the largest and wealthiest affiliates deploy resources and enjoy a status often superior to those of the international leadership; though indeed, the individual union members are even further removed from the top of the meta-organisation.
The key problem for democratic governance is, rather, the diversity of affiliates, in terms of size and material resources but also of industrial relations traditions and indeed of understandings of the meanings and purposes of trade unionism itself. This is evident enough in the case of the ETUC (Seeliger, 2019), but even more obvious at global level given the vast differences in economic conditions and political traditions among affiliates. The formation of the ITUC in 2006, uniting confederations from previously competing ideological traditions, also intensified potential conflicts of perspective over goals and methods (Gumbrell-McCormick, 2013a). If diplomacy is one of the key skills required for international trade union leadership, this applies not only to relations with external interlocutors but also to the management of the internal diversity of the affiliates. As Ahrne and Brunsson note (2008: 123), internal conflicts can seriously weaken the cohesion and effectiveness of meta-organisations, encouraging ‘a strong preference for consensus’. The ITUC Constitution declares that ‘the endeavour of the Congress shall be to secure the widest possible measure of agreement on any decisions taken’. Likewise in the ETUC, the Constitution prescribes that ‘Congress shall endeavour to achieve the widest possible measure of agreement’. If a formal vote cannot be avoided, a two-thirds majority is required; otherwise the question is ‘referred back to the Executive Committee for examination and decision’.
One mechanism for reducing open conflict is what in the British TUC is known as the ‘composite resolution’: opposing propositions submitted for debate are welded into a single (but internally inconsistent) whole. This avoids overt disagreements on the conference floor or in executive meetings, but then allows considerable latitude for interpretation. As Ahrne and Brunsson put it (2008: 127–128), ‘if decisions are ambiguous enough, various members can interpret them as consistent with their own goals, values, and norms’. An example is provided by Mitchell (2014: 419): while broad opposition to deregulatory and austerity policies gives ETUC affiliates a shared objective, there is less agreement about what the ETUC is for, apart from defending an undefined European Social Model. Furthermore, there are deep and enduring disagreements among the ETUC’s affiliates about the best tactics for exerting influence at the European level.
A characteristic feature of meta-organisations is that their members may themselves be key actors in the process of ‘diversity management’; but they perform this role in distinctive ways. Cotton and Gumbrell-McCormick (2012: 708) ‘argue that the existence of multilateral structures is necessary for genuinely global trade union activity to take place, but that what exists is an imperfect multilateralism which requires close working relationships between small groups of unions in order to function effectively’ and ‘that union capacity to carry out international solidarity action requires a robust relationship between members which can only be developed by working in a “minilateral” way with small groups of affiliates’. Hence the effective veto power of key national confederations entails that the search for public consensus is founded on micropolitics behind closed doors. To complicate matters further, the sectoral trade union organisations at both global and European levels are themselves meta-organisations which attempt to shape confederal policy in the light of their own interests, creating a horizontal field of internal contention (Seeliger and Wagner, 2016).
On the other hand, the growing activism of international unions in the face of intensified economic globalisation has accentuated the need for an explicit ‘willingness to act’ on the part of affiliates and their members (Croucher and Cotton, 2009), and this has evident implications for internal democracy. With limited material power resources, the leaders of international unions must rely primarily on mobilising ‘soft’ forms of authority by ‘defining an organizational identity and promoting values’ of cross-national solidarity, as Garaudel (2020: 9) puts it (in one of the very few studies to apply meta-organisation theories to international trade unionism).
As Fairbrother and Hammer (2005: 422) argue, in establishing a role in the new economic order, leaders of these international trade union bodies have both drawn upon familiar tools of representation, such as framework agreements, and on the rationales of representation, stressing accountability and participation. The GUFs have been involved in a long process of elaborating established methods and modes of representation in distinct and relevant ways. While negotiation and campaigning at an international level is not new, what is novel is the ways in which framework agreements, the negotiations that lead up to them, and associated campaigns are rooted in the day-to-day realities of members, and not the musings of remote international leaders.
Is ‘deliberative democracy’ a relevant principle for the theory and practice of international trade unionism?
So far we have discussed conceptions of union democracy which derive overwhelmingly from analysis at national level. We have shown how such conceptions cannot be fully mapped against the internal dynamics of international unions, which are ‘associations of associations’ rather than bodies which organise individual workers. We now consider the potential relevance of broader approaches to governance in supranational institutions which invoke the notion of ‘deliberative democracy’. In an early discussion of this concept, Miller (1992: 55) ‘starts from the premise that political preferences will conflict and that the purpose of democratic institutions must be to resolve this conflict’. But in complex political systems confronted with multiple choices, this needs to involve an open and uncoerced discussion of the issue at stake with the aim of arriving at an agreed judgement. The process of reaching a decision will also be a process whereby initial preferences are transformed to take account of the views of others (Miller, 1992: 55). a theory of democracy as deliberative democracy in which arguments, not individuals, are the basic units […]. The basic proposition is that political decisions gain their democratic legitimacy from the extent to which they are based on public deliberation. To be effective, such deliberation must include all relevant arguments and therefore be open to every citizen.
For Keohane et al. (2009: 8), democracy stands for governance on the basis of arguments and evidence that have been tested in public with a wide range of information. When policies are adopted deliberately – after sufficient discussion, debate, and the sifting of reasons and evidence, including from experts – they are more likely to be policies that people are prepared to live with. there is broad agreement between democratic theorists that the citizens’ preferences that do matter are those that have a chance of being created or modified within arenas of political contestation, and that what matters are institutions that reliably ensure that policies are responsive to these preferences, rather than matching by happy coincidence. Thus, one important challenge is to create institutions that provide such opportunities and responsiveness.
This implies that for effective ‘input’ it is necessary to sustain a sense of collective identity among the membership – maintaining a demos within the union constituency – and to foster the ‘social capital’ of members to facilitate their effective participation. Coherent ‘throughput’, in turn, requires a sophisticated structure of internal union institutions, procedures and networks that facilitates a multi-directional, interactive relationship among leaders, local officials, activists and ordinary members. Conversely, however, Delli Carpini et al. (2004: 321) identify some of the difficulties involved in implementing ‘deliberative democracy’ and evoke a strong and persistent suspicion that public deliberation is so infrequent, unrepresentative, subject to conscious manipulation and unconscious bias, and disconnected from actual decision making as to make it at best an impractical mechanism for determining the public will, and at worst misleading or dangerous.
The idea of deliberative governance is often proposed as a solution to a commonly diagnosed ‘democratic deficit’ in transnational institutions such as the EU. As Follesdal and Hix comment (2006: 534), ‘there is no single meaning of the “democratic deficit”. Definitions are as varied as the nationality, intellectual positions and preferred solutions of the scholars or commentators who write on the subject.’ The whole thesis that there exists a democratic deficit in the EU has been contested, for example by Moravcsik (2004: 338): though centralized electoral control and collective deliberation remain relatively weak and diffuse, constitutional and material restrictions on the EU’s mandate, inter-institutional checks and balances, indirect democratic control via national governments, and the modest but increasing powers of the European Parliament are more than sufficient to assure that in most of what it does, EU policy-making is generally clean, transparent, effective and politically responsive to the demands of Europeans.
All this implies that in any meta-organisation (including international trade unions), for deliberation to be effective, strong institutional and normative foundations are required. Within international trade unionism, implicit norms of leadership behaviour are perhaps particularly important. These require transparency, reciprocal communication and a search for synthesis between competing positions which is more than just a lowest common denominator. Majone (1998: 14–15) has argued that ‘the expression “democratic deficit” can […] denote a set of problems – technocratic decision-making, lack of transparency, insufficient public participation, excessive use of administrative discretion, inadequate mechanisms of control and accountability’ which apply to all meta-organisations. But because norms of deliberative democracy in international trade unionism are predominantly implicit, their force may become apparent only when they are breached. Recent internal conflicts, particularly in the ITUC but to a lesser extent also in the ETUC, are evidence of this (Larsson and Andersen, 2019; Lepeytre, 2019; Rehfeldt, 2019).
In conclusion
Should international trade unions be democratic? Few would answer in the negative. But as with unions at national level, what democracy means in practice is ambiguous and contested. Indeed, we have shown how conceptions of union democracy developed in the context of national trade unions can be particularly problematic at international level. Partly this is for familiar reasons linked to geographical scale and scope: the diversity of languages, economic conditions, organisational capacities and industrial relations traditions encompassed among the memberships. But we have also deployed the literature on meta-organisations to explore how the very fact that international unions are ‘associations of associations’ in which individual workers are not directly members adds to the complexity of the challenges of democratic governance.
As the global economy becomes increasingly interconnected, international trade union action is in turn increasingly essential. As Platzer and Müller (2011) show clearly, international unions possess very limited hierarchical authority; and their affiliates, jealous of their own autonomy, are likely to ensure that this remains the case. For policies to be translated into practice, they need not just the passive consent of their members but their active commitment. This is obviously the case for sectoral internationals (GUFs and ETUFs), whose efforts to strengthen organisation, coordinate collective bargaining and present a united front to transnational employers necessarily require a willingness to act on the part of national affiliates and their members and workplace representatives. (In this context, it is important to note that sectoral internationals are not ‘meta-meta-organisations’, as are their cross-sectoral counterparts; in this sense, they are less far removed from workers on the ground.) Confederal internationals, to the extent that much of their activity is based on lobbying, diplomacy and ‘social partnership’, face a less immediate need to mobilise activism (though the intermittent efforts of the ETUC to organise demonstrations show that this too may be perceived as necessary). Commonly, the ‘logic of influence’ outweighs the ‘logic of membership’. Yet influence itself requires that international trade unionism can demonstrate its representative credentials, showing that its demands are understood and endorsed by its affiliates and their members.
In this context, our discussion of deliberative governance and deliberative democracy is clearly relevant. Sustaining representative legitimacy and fostering cross-national capacity for mobilisation require transparency, accountability and opportunities for effective participation. International trade unionists might well respond, perhaps with some irritation, that this is what their organisations have always practised: ‘communicative action’ (Habermas, 1981) has always been integral to the process of organising workers and integrating their interests. Yet particularly at international level, it has always been of critical importance to fashion, sustain and elaborate the deliberative procedures necessary for both horizontal and vertical communicative action. For this reason, there are grounds for concern in the extent to which, over recent years, international congresses have become less spaces for deliberation and more platforms for media-oriented public relations.
We do not presume to offer a blueprint for union democracy in the 21st century. However, we are convinced that just as unions stress the benefits of employee participation for social cohesion and economic participation, so it is necessary to protect and enlarge the spaces for deliberative democracy at every level of the trade union movement, and to strengthen the two-way communicative links between the different components of a complex, multi-level and multi-faceted system of worker representation. Modern communications technologies, whatever their limitations, clearly have an important potential role in engaging cohorts of workers for whom traditional modes of deliberation seem outmoded.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.