THE ORGANISATIONAL NATURE OF UNION CAREERS: The touchstone of equality policies? Comparing France and the UK

The paradox of the contrast between the continuous feminisation of the workforce and the constant under-representation of women in the structures, cultures, and agendas of unions has been confirmed by recent national (Briskin 2002; Cobble and Bielski Michal 2002; Contrepois 2006; Kirton 2006a) and comparative studies (Colgan and Ledwith 2002; Silvera 2006).Although women contribute to the stabilization of union membership,¹ they remain under-represented among branch officials and paid national officers. The creation of women trade union leaders has been identified as a key issue for unions, both as a way to challenge the oligarchic nature of union leadership and to instigate organizational changes (Healy and Kirton 2000), and as a lever to bring the necessary amount of equality and democracy for union renewal (Foley and Baker 2009). Among the classic putative causes of the under-representation of women in the labour movement, research papers refer to different structural factors that could explain women's lesser propensity to participate, such as the historical under-organization of female workplaces, male trade unionists’ unwillingness to represent women's interests, work–life balance issues, or women's specific working patterns – notably part-time and temporary contracts. Following recent developments in the sociology of gender and organisation (Wajcman 1998; Halford and Leonard 2000), other articles have focused their analyses on the embeddedness of gendered inequalities within trade unions (Cockburn 1991; Colgan and Ledwith 2002; Kirton 2006a). This theoretical approach acknowledges the internal organisational structures, practices, and cultures that result in and maintain gender inequalities within unions, providing a new understanding of the permanence of the gendered nature of union careers.

Because they are widely envisaged as democratic organisations, especially in the UK literature, unions are rarely scrutinized as employers (Watson 1988; Kelly and Heery 1994). However, unions select and train their officers/officials, promote them and occasionally dismiss them or help them to get back to their former occupations. Despite many controversies about what should be the democratic nature of union careers, a large majority of union leaders (even when they are elected, e.g. senior national officers in the UK) embark on atypical ‘organisational careers’ (Glaser 1968). The different career routes, based on election or appointment, are less distinctive in practice than in theory within the two countries being considered. In the UK, lay reps sometimes apply for full-time paid positions, and almost all paid officers have previously been activists (Heery 2006). In France, these two routes are even more interconnected as full-time paid officers are elected by members (or other elected officials) most of the time, and very often are paid by their employers² (see Table 1). Most officials are full-time, partly elected, long serving paid officers, but most of them have refused, until recently, to view their union commitment as a ‘career’. Informal processes, such as pre-selection of motions of support during congresses or interpersonal arbitration with potential rivals, are often organised to secure the career of top union leaders and to shut out opposing views, taking advantage of the power of incumbency and bureaucracy (Michels 1962). The organisation of union careers is also built on ‘masculine-neutral’ norms – the culture of long working hours (Franzway 2000) and available geographical mobility. These norms create difficult work–life balance issues, especially for mothers/parents and workers with specific employment patterns. As employers, unions also tend to develop internal equality policies to tackle these issues. Depending on periods of time and contexts, these policies are more or less sophisticated and effective, while producing their own side effects and limitations.

In this article, we explore the selective and unequal representation of women within the labour movement from a comparative perspective. By choosing two European countries (France and the UK) and three unions displaying different approaches to equality and different levels of feminisation, we want to draw attention to the role of organisational contexts and periods in the under-representation of women in union government. To understand the feminisation of union leadership – its shape and dynamics – we suggest that unions could be considered as a specific category of ‘internal labour markets’ (Doeringer and Piore 1971), with distinctive characteristics: ports of entry and exit, typical/atypical career routes, internal job segregation and hierarchy, norms of job evaluation, human resources management policies and gender/race equality policies. We also argue that the presence of women is linked to the variations in unions’ ‘repertoire of action’ (Tilly 2008). The transformation of union practices/agendas and the model of unionism are critical in understanding the kind of profiles of those unionists selected and hired to work for the union. In France and in the UK, the recent emphasis put on collective bargaining and/or social partnership has offered new opportunities for men and women with professional expertise and higher qualifications (Heery and Kelly 2001), yet with no activist background. Comparing the three unions, we will highlight the main organisational and institutional processes that help or prevent women from accessing leadership union positions.

Drawing on biographical matching and comparative analysis (Crompton 2001), we have used career narratives to explore the organisational processes and career norms required for the formation of trade union leaders: the different sequences that a ‘union career’ follows (Ledwith et al.1990; Kaminski and Yakura 2008), the necessary steps, main tracks, atypical routes and turning points of such sequences, and also the role of ‘significant others’ (Hugues 1958). This approach to union careers (Kirton 2006b) emphasizes the articulation and interferences between different spheres, and the role of the identities of trade unions in the promotion of specific profiles of trade unionists (in terms of age, sex, etc.), depending on periods of time considered. This cross-national comparison based on trade unionist biographies allows us to embed these objective and subjective trajectories of women and men in complex and evolutionary organisations; trajectories that would be difficult to analyse through questionnaires. All these interviews have been analysed entirely using ‘biographical tables,’ describing in three separate columns the different sequences of occupational life (appointment, promotion, change of job), family/residential life (marriage, births, relocation), and union/political life (union participation, union roles, political commitment). Particular attention was given to dating individual events in order to establish a link with the corresponding trade union history. This approach can be classified as a ‘societal’ qualitative cross-national comparison, which integrates class, gender, and ethnicity³ (O'Reilly 2006). But between the micro-level of individual life experience and the macro-level of employment regimes and equal opportunity policies, we have placed an emphasis on the meso-level of each trade union, paying attention to period and generational variations.

In 2007, we asked two major UK trade unions, UNISON (Britain's, and also Europe's, largest public sector union) – well known for its high commitment to equality and diversity – and GMB (a general union resulting from many amalgamations) – that has supported such an agenda more recently – to allow us to interview visible activists (20 each). These 40 trade unionists with national, regional, and local responsibilities, comprising 28 women and 12 men, are almost all working full-time for the labour movement, and are either paid by their union or on time release (main features presented in Table 1). In France, we asked the CFDT (Confédération Française Démocratique du Travail), the second major representative French union confederation, with a pioneer commitment to internal equality (‘mixité’), to allow us to interview 40 visible trade unionists, with national, regional and local responsibilities, most of them being elected paid officials. To be able to make a comparison with the two UK unions, we have selected CFDT unionists from different industrial federations covering sectors that are more or less feminised (health, education/energy, metalwork). These 24 women and 16 men range from a female member of the national executive committee (three women out of eight) to local rep. Although union density is not at the same level (28 percent in England but only 8 percent in France) and the industrial relations regime differs strongly within each country (Hyman 2001), we have defined functional equivalents in terms of union roles in order to use a ‘union career’ methodology when comparing France and the UK. We have distinguished three stages in a typical ‘rank and file’ career: entry into positions of union responsibilities at local level (rep, shop steward in the UK/elected ‘délégué du personnel’ or nominated ‘délégué syndical’ in France); consolidation of a power base outside the workplace (branch level in the UK/‘syndicat’ level⁴ or ‘union départementale’ or ‘union régionale interprofessionnelle’ in France); development of a directing role (paid official at regional or national level/office holders and paid officials at national level in the UK, ‘fédération’ or ‘confédération’ in France). These biographical interviews were then completed using archives and documents on equality policies.

To understand the selective and unequal representation of women in the structure and agenda of unions, formal equality policies first need to be considered. Over the years, unions in Europe, and especially in the UK, have tried to develop new forms of a more inclusive trade unionism.

In the UK, the issue of delivering union equality was identified more than 20 years ago, pushed by second wave feminism movements and vanguard equality activists (Heery and Kelly 1988; Colgan and Ledwith 1996; Heery and Conley 2007).In a changing political and social environment, it seemed untenable for unions to fail to properly represent the interests of a large and increasing proportion of their members while at the same time claiming to be democratic organisations. However, the drive to diminish this democratic deficit for women (and other minority groups) has also been analysed as ‘an instrumental rather than a moral or ethical concern’ (Dorgan and Grieco 1993: 152), in the context of a huge membership decline (union density has fallen from 54 percent in 1979 to 29 percent in 2005), especially among male blue-collar workers, and of a growing level in the employment rate for women (73 percent in 2005). In 1979, the equality agenda was translated into recommendations by the Trade Union Congress (TUC) in its charter for Equality for Women within Trade Unions. This leading role of the TUC, also encouraged by key unions (UNISON, MSF, CWU), has never ceased. The TUC has also made constitutional changes to increase diversity in its own structures. Following this example, in order to encourage the representation of varied constituencies within trade unions, most large unions have switched from liberal (women-only courses, equality officers, women's conference) to radical measures⁵ : proportionality, ‘fair representation’ (McBride 2001) for elections and conference delegations within the union, and self-organised groups in UNISON.

However, progress in the unions affiliated to the TUC has been described as uneven and incremental (Kirton and Healy 1999), very dependent on the level of activism of women, black, lesbian and gay, disabled, and young trade unionists, on the action of feminist senior activists, and on the recognition by the existing leadership of the importance, to the revitalization of the union, of emerging activists/leaders representing particular social groups (Healy and Kirton 2000). In our study, UNISON and GMB give a good illustration of this variety. UNISON is always cited as an example because of its long-established equality policy, not only in terms of the amount of resources made available, but also in the actual representation of women in the governing body members and in the executive team (43 percent of full-time national officers in 2008; cf. Table 2). After years of limited resources being allocated to the existing equality structures, the GMB has recently conducted an equality review, under the supervision of its first female deputy general secretary, recognising that ‘major steps towards making the GMB a more inclusive union needed to be taken’(CEC 2007: 6): women were only 30 percent of full-time national officers in 2008.

In the 1970s, propelled by strong feminist movements, vanguard female equality officers, and numerous women's strikes (Maruani 1979), the CFDT took a leading role in the adoption of an agenda dealing with the condition of women (1976 congress) that included a wide range of women's issues (free abortion, contraception, working time, domestic work), and strong links were built with feminist associations. From 1978, the CFDT took a new direction and started to distance itself from political parties, social movements, and ‘non-work’ issues. The huge membership decline that started at the end of the 1970s (union density in France was 28 percent in 1950 and 8 percent in 2010) led to the redefinition of union goals and doctrine. The CFDT dispensed with class analysis and socialist values to convert to ‘reformism’ – an acceptance of capitalism and the use of a paramount strategy of negotiation with employers (Defaud 2009). This conversion was linked to the promotion of a ‘mass unionism’ able to address issues for everyone, including women, and the promotion of a participative model of unionism, called ‘syndicalisme d'adhérents.’⁶ Specific campaigns were launched in different sectors to attract women members, and specific training was designed to promote the new political line and to train future female leaders. In 1982, the CFDT adopted a mix of liberal measures (dedicated financial resources, specific training courses, equality officers, equality conferences) and radical measures (reserved seats for the national executive committee and minimum representation for congress delegations). The CFDT was the first confederation to have a woman general secretary, Nicole Notat, who served for more than 10 years (1992–2002). These measures were quite controversial at the time. They were seen as a tool to promote a new leader profile, more in line with the reformist approach promoted by the confederation (Le Brouster 2009). Shortly after the election of François Mitterand in 1981, this internal policy was weakened by the subsequent departure of feminist union leaders, including Jeannette Laot who had been in charge of equality in the CFDT for 20 years (1961–1981). These feminists were absorbed by ‘State Feminism’ (Mazur 1995) upheld by the Socialist Party, and they consequently ceased to stimulate unions.

Since then, the CFDT has shown reluctance toward any association with feminist movements, especially any described as ‘radical’ (e.g. near to left or extreme-left ideology). Overall, equality measures and feminisation of membership have led to a certain level of feminisation at the upper levels of the confederation and of major industrial federations (32 percent of federation executives in 2004, cf. Table 2). However, this equality policy has had diminished results at the lower levels of the hierarchy (24 percent of local union secretaries in 2004) and advances remain difficult in the less feminised federations (blue-collar sectors). The CFDT is now caught between their determination to achieve better female representation at all levels and the difficulty in taking new enforcing actions. Because the confédération feared being unable to fill more reserved seats, complete parity for the Bureau National (equivalent of the NEC) was not decided on at the last 2010 congress, in contrast to the actions of the CGT (Confédération Générale du Travail, the first general trade union confederation in France) who set an example with the ‘parity law’ adopted in the political field (Bereni 2007). Similarly, the CFDT seems to hesitate between promoting mainstreaming actions and maintaining separate structures and actions. The principle of ‘self-organisation’ for women or any other type of minority group is not easy to implement within a general union tradition.

This comparison shows that union equality policies are very diverse, not only when comparing one country with another, but also within the same country depending on various factors. The organisational gender composition of each union seems to have an influence on the career moves of men and women. If we look at the feminisation level of the union government within the CFDT and UNISON, compared with that of the GMB, we may agree that having higher proportions of women at the ‘focal job level’ improves the probability of the promotion of other women into that level (Cohen et al.1998), confirming the hypothesis of a ‘critical mass’ (Kanter 1977b) required to reduce gendered inequalities. However, this effect cannot be isolated from a broader analysis of the organisational context. As other studies have demonstrated, the presence of female managers has the strongest desegregating effects in larger and growing organisations (Huffman et al.2010) and may decrease over time. This assumption helps in the understanding of the relative stagnation of the increase in women leaders within the CFDT, as membership stagnates.

It also indicates that union gender desegregation challenges equality policies and the repertoire of action. As we can see in Table 2, the feminisation level can drop over time, even when it is initially above 40 percent, as in the example of UNISON at national level between 2005 and 2008. Radical measures, such as quotas or reserved seats implemented at the top level of the union hierarchy, can conceal remaining difficulties in the challenge that targets those male norms which contribute to the reproduction of gender inequalities. We can argue that the feminisation of the CFDT government is maintained thanks to the quotas implemented in 1982. This artificial stimulus conceals the fact that women remain under-represented at lower levels of the hierarchy, and that the advance of internal equality is encountering the maintenance of decentralised forms of organisations (Kelly and Heery 1994) and the weakening of a top-down leadership-driven strategy (Voss 2010), and that this advance has not been seen as a priority since the growing gap eventuated between the CFDT and feminist movements. Compared with UNISON, where the equality agenda has constantly been pushed by feminist national leaders and by coalitions with social movements encouraged by self-organised groups (McBride 2001), the CFDT case displays a more discontinuous story. Its conversion to ‘reformism’ has led the union to de-politicise its action, moving away from any type of social movement, including feminism. Over and above the legislative context which is quite supportive to equality in France (successive equality laws since 1982 and ‘parity’ laws since 2001), internal equality is mainly shaped by the changing identity and agenda of the CFDT. Although the CFDT has contributed to the adoption of equality legislation, as UK unions have done, its internal policy underlines a claimed union autonomy and/or a clear disconnection with political parties and broader social movements.

Even in favourable organisational contexts with supportive policies and a high rate of membership feminisation, women seem to have difficulties in accessing union leadership positions, especially at the local and regional levels where there are no radical measures. To understand the transmission of these difficulties from one generation of women to another, we have analysed the careers of our sample of women and men at three stages of their union career.

For men and women, union participation is highly dependent on job stability but also on union density and recognition at the workplace level. In France, more so than in the UK, the availability of union roles secured by the law and the huge membership decline have combined to open new opportunities for members to be elected as union rep or to be nominated as ‘délégué syndical.’ Many young representatives have been ‘sucked into the labour movement,’ becoming members and reps almost at the same time in the 1990s or 2000s. Union officers have approached them and offered them a role at the local or even at the ‘syndicat’ level. In the UK, this ‘vacancy’ effect is less noticeable. Union participation by new members is only triggered in newly organized workplaces. In both countries, it remains dependent on time-off agreements, thus explaining the strong difference in union density between public and private sector. The French legislation is more generous than the UK one, but time release is also linked to local bargaining, which can be very unequal depending on industrial sectors and employers. Union repression can be quite strong in the private sector but also in the public sector in the context of privatisation, as seen in the UK since the 1980s. Whatever the context, union participation depends on sponsoring by existing union officers. This selection can take different forms, from a quick detection to a long recognition process, from mentoring (by men or women officers) to rivalry (as women are often younger) and gendered favouritism. These processes of mentoring are even more crucial in contexts of membership decline or repression, where trade unionists are often alone and cannot rely on a group of peers. However, union participation also rests on specific circumstances. In stable union environments, dominated by long-established senior leaders, access to union responsibilities is aided and enhanced by unexpected dismissal or resignation. Within contexts of political or financial crises, the promotion/election of new profiles is favoured (Voss and Sherman 2000), with a higher probability of failure, however. Other factors influence the promotion of new entrants (including women), such as the development of new union practices. The focus put on recruitment strategies since the 1990s, in both countries, has allowed different profiles to become elected at the workplace level but also to gain visibility for future promotion, should they achieve good results in terms of membership growth, as was the case for Annick.

At 58, Annick is the organizing and training officer of the steel and mining ‘fédération’ of the CFDT (FGMM). Because she left school at 16, she has always felt that she was not capable of getting promotion within the union. At 19, she entered a steel tube company as a non-qualified white-collar worker in the personnel department (punch card operator). Two years later, she joined the CFDT because they produced a leaflet concerning a migrant worker's case. She felt moved by that campaign, as she was giving literacy courses to migrant workers with her friends from a Catholic youth movement. During a big strike on pay issues (8 weeks in 1975), she achieved visibility by being on the picket line at night ‘with the men.’ Afterwards, the regional officer offered her the opportunity to go on a French–German unionists’ exchange. Within three years, she was nominated ‘déléguée syndicale’ after the dismissal of the former leader, following his financial negligence. Throughout her career, she has been sponsored by the (male) union treasurer, who said to her when she was asked to become the secretary of her union in 1986: ‘If they ask you, you have to trust them; they judged that you are capable, so you have to go.’ After six years of local responsibilities, and very good results in terms of recruitment of new members, Annick was offered two different positions at the regional and federal levels, as the previous leaders were about to retire, and the federation was striving to promote women. Because she was not sure of her political base at the regional level following political struggles over the election of Nicole Notat as General Secretary in 1992, she agreed to take a full-time paid (non-elected) position at the federation level in Paris. She then had to leave her elected seat on the governing board of the federation and became a paid officer, responsible for organizing/recruiting policy. As she was single with no children and had resigned from her job, she agreed to commute from Monday to Friday from her village, 180 miles away from Paris. She knows she will never be asked to become a national (elected) officer, not only because she is a woman in a very male-dominated federation (the one woman out of nine national officers has just resigned), but mainly because she is more confident in her organizing skills than her political ones, and is always acutely aware of her lack of formal credentials.

Although more women can now access these first levels of lay activism, especially for elected positions, they are over-represented in certain types of union roles, where they are responsible for social activities or daily representation (rep in UK, ‘délégué du personnel’ in France). The gendered job segregation is often reproduced within the union practices, sometimes justified by female ‘skills’ (altruism, empathy, etc.): skills that can be claimed by new entrants, men and women, to instil a new vision of trade unionism, emphasizing democratic and participative principles (Guillaume 2007), but that are inadequate for a direct path to union management, whilst very legitimate for election as senior lay rep in the UK or rep in governing bodies in France (‘bureau régional’, ‘fédéral’, ‘national’). But the real question is the effective power of these elected bodies compared with union management. Likewise, the development of new roles, such as those of equality or health and safety rep, has also allowed women to take side roads (alternative or parallel careers [Kirton 2006b] but maybe dead-end careers), leaving the most prestigious roles (grievance handling and contract negotiation) to men. This specialization has had an impact on union training that is offered, and this can have resulting effects on women's careers, at least in France. Although labour education is necessary to develop basic skills (Kaminski and Yakura 2008), the over-specialization characteristic of union courses on specific/technical topics will not contribute to improving the necessary political/general background leading to promotion in the union hierarchy. While union participation relies more on formal credentials, the traditional forms of ‘militant capital’ (Matonti and Poupeau 2004), acquired through experience, become devalued, and this can prevent oppressed social groups from gaining visibility and thus access to officials/officers roles.

In both countries, permanent union roles at the branch or ‘syndicat’ levels are key positions in building a union career, but they are less feminised than those experienced at the national level. In both countries, the union employment market is more or less open to new entrants, with strong competition to access the rare full-time positions available in the UK, and a slightly more open situation in France but with a higher turnover. The filling of these positions always depends on detection/co-option processes under the control of regional or local officials. Because of the decentralisation of structures and policies in each country, the equality agenda remains very dependent on the political will of the local executive team. In France, the absence of external publication of job vacancies increases the weight of co-option and social networks. In the UK, each union has its own HR practices, more or less structured. In the GMB, regional secretaries are responsible for the dissemination of internal adverts, whereas in UNISON, the use of external job publication through national newspapers (also for reserved seats at regional level) has increased the number of applicants and has also allowed a change in the candidates’ profiles (external women from the volunteer sector or women coming from administrative/secretarial union staff). Even in this context, internal applicants with an activist background are preferred, and they often can discover in advance that there will be a vacancy. In France, this co-option is also based on political control. Being perceived, or not, as being in favour of the reformist orientation chosen by the CFDT can boost the careers of young activists. Following numerous internal struggles, many leaders since the early 1990s have left the CFDT, sometimes joining more ‘radical’ unions such as the CGT. These departures have offered an opportunity for the reformist leaders to promote new leaders with less politicised views (and little feminist consciousness), including numerous young women. To counter the reluctance of male officers to promote junior women with a different background, senior women officers with feminist convictions (especially in the UK) sometimes opt for active mentoring, with activists or their own assistants (even in administrative roles). Women-only training and self-organisation structures in the UK (and especially in UNISON) also have the intention of helping women to build their self-confidence. Labour education also exists in France, but it is less gender-conscious than in the UK.

At this level, as with the workplace level, the evolution of the union movement in the two countries has opened up new career routes for young activists. In the UK and, to a certain extent, in France (Guillaume and Pochic 2009b), the switch from a servicing model to an organising model (Heery 2006) has allowed the development of specialised positions (‘organizer’ in the UK and ‘développeur’ in France) that have been offered to new profiles to fit with the ‘like recruiting like’ principle. More attention has been given by the CFDT to sex, age, and a higher level of qualification (‘cadres’⁷ ), whereas UK unions, and especially UNISON, have extended their policies to reflect race and class (low-paid members [Thornley 2006]). These differences reveal different organizing strategies, with the CFDT becoming in the 1990s a union dominated by professionals and white-collar workers. Beside this membership dimension, several ‘expertise routes’ (legal action and advice, health and safety, economic analysis, social policy, international and European action) have been encouraged both by the evolution of union political action and by employer–union relations. Because of their present moderate leadership, the unions we studied have been involved as a negotiating partner or advisor on pension issues, unemployment benefits, working time, minimum wages, health and safety, equality, or redundancy. In both countries, this growing union participation in socio-economic regulation at the national or local level has led to the promotion of officers and activists possessing new types of qualifications and professional expertise. Dedicated training schemes, together with special targeted attention, are provided to retain these would-be senior officers. However, they face the risk of being directed into less valued positions. To reach more general positions/roles, these new entrants need to have strong personal resources (activist background, qualifications, and personality) and to benefit from equality policies to continue to progress at the national level, as Adishri demonstrates.

At 32, Adishri is very proud to be ‘the first Asian women national officer’ and considers that ‘the GMB has given me a lot in terms of my own confidence and skills’ through support and intense training. She became an activist for equality rights at 18 after refusing to get married to the husband chosen by her family. As a blue-collar worker in a female-dominated textile factory, she started as a shop steward for the GMB, vbeing particularly active in race equality and in youth campaigns. Very quickly, she became involved with the Northern region black workers’ committee, where she met her husband, an activist who returned to university and became a teacher upon finishing. At 26, she was offered the position of a 6-month placement as a recruiter, and she continued after this in a permanent position for 4 years. ‘Very committed and determined,’ she became the specialist on race equality, both for the GMB and for the TUC, as well as for the Labour Party in the Northern region, and felt ‘very empowered’ by the international campaigns she pursued through the Commonwealth TUC. When a national equalities officer job was advertised, she felt she ‘had nothing to lose and everything to gain.’ The fact that she has also assumed responsibility for the chemical and pharmaceutical industry, and has led the negotiations on behalf of the GMB in higher education, gives her the appropriate roles to ‘figure out the politics’ and feel confident in her job. She had to move to London during the first year, but then decided to commute from the Northern region to London the following year. She is now on maternity leave. Very conscious of work–life balance issues, Adishri is convinced that her partner will be supportive, and she wants to negotiate time off for equality reps, to encourage more women to become active.

Although the effects of equality policies are more visible at the national level, women officers still have great difficulties in penetrating ‘prestigious’ domains such as negotiation, management, and political action (strategy, campaigning, lobbying). The exercise of power continues to be male-dominated, particularly in France and/or in less feminised unions in the UK such as the GMB. In UNISON, the introduction of management methods (targets, annual appraisals, equality audits) has helped in the promotion of experienced and graduate women. But this policy is facing a practical limit: senior male officers do not want to give up their seats. In the UK and in France, the policy of reserved seats for women or quotas at national level has noticeable side effects: suspicion about lack of skills and risk of ‘tokenism’ syndrome (Kanter 1977a) and of being labelled as ‘feminists’ (apart from UNISON). In the UK, many women have stayed in those reserved seats, finding it difficult to get elected onto a general seat. Union officers who access senior national officer positions still ‘complied with the male model of senior trade union official, that is, a long, unbroken record of active experience and offices held’ (Ledwith et al.1990: 121), and have thus managed to convince male-dominated, rather conservative panels, especially in the GMB. At that level, there is fierce competition among a small pool of long-serving loyal officers (mostly men), with repeated applications and numerous lateral moves, enabling them to stay in the union movement and to increase their visibility and their inclusion in networks inside the organisation. Unspecified hours and geographical mobility are key factors that favour climbing within the union organisation, just as they are in corporate careers (Guillaume and Pochic 2009a). For national positions, officers need to live in London or Paris, sometimes leaving their family during the week, as is the case with Clara.

At 42, Clara is a senior officer in charge of working conditions for the Health and Social Work federation of the CFDT. As a nurse, she became a member in 1995 because she agreed with the CFDT position in favour of the social security system reform proposed by the right-wing government, in the context of a huge national strike (3 weeks), but also in recognition of the assistance she received in a grievance about her third maternity leave. Following a restructuring project in her hospital in which the CFDT was involved, she then agreed to become full-time convenor and took a couple of union courses. Soon after, she was also offered an organizing role and was elected to the executive committee of the region, this result mainly being to prevent another dissident candidate from being elected; however, at that time she ‘was not aware of that political issue.’ She managed to have good results in terms of recruitment. She was then selected and offered a part-time job in Paris, in charge of the organizing section. She commuted every day from her home (2-and-a-half hours each day), becoming very tired as she also had to look after her three young children. In 2001, she obtained a full-time position in Paris, living apart from her family during the week. She finds that she is ‘more available for my children during the week-end and less stressed,’ and organising this is possible as her husband (train steward) has days off during the week. In 2005, in a development that surprised her, she was elected senior national officer. However, she is not very satisfied with her situation, as she has been away from home, Monday to Friday, for 7 years, and one of her teenagers is now having problems at school. His teacher has told her that it is her fault. She feels a lack of confidence in her job. She is obviously very frustrated by her union environment. She will not finish her mandate and is thinking about her next move. She does not want to return to her nursing job, for fear of appearing incompetent, but also because of her newly acquired organizing skills. But she has no firm ideas about possibilities for her career outside the union movement.

Sometimes, as Clara demonstrates, new entrants can benefit from internal political struggles, appearing as ‘neutral’ candidates, especially within the CFDT. In each country, as the officials are indirectly elected by the union members (through union/branch/region representatives at the congress – with a low participation rate), the result of elections depends less on objective skills than on ‘factionalism groupings’ (Ledwith et al.1990). These informal networks, ‘clans,’ are traditionally based on political orientations but also on shared experiences inside the union. The absence of a regional power base is particularly detrimental to the careers of experts, even if some of them have built their national visibility through innovative projects or media exposure. More recently, formal networks based on a specific generation (youth sections in France) or on gender (women's committees in both countries, or self-organised groups in UNISON) have tried to campaign for their candidates, generally upholding a different strategic orientation for the union. This networking by women within and across unions, ‘gendered factionalism’ (Healy and Kirton 2000), has a real impact from below when there is a critical mass, in female-dominated unions, as with UNISON. The election of women officials may also be supported from the top, depending on critical actions of senior executives with political will and feminist values. But in this case, these women have to deal with a ‘token’ stigma (Kanter 1977a) and have to fight to build their legitimacy ‘on the job.’ Although the introduction of low-paid seats in 1998 tried to address the class gap in the governing bodies of UNISON, it has been more difficult in the face of such fierce competition to develop trade union careers for women with working-class backgrounds. This selective union recruitment is really striking, although not really acknowledged by the unions investigated (Guillaume and Pochic 2009b). Most of the senior female unionists interviewed have a degree, sometimes obtained through continuous education. When they access national positions, working-class women tend to become and stay senior lay members, avoiding the ‘management’ route.

This comparative work on the glass ceiling in unions draws attention to the making of inequality regimes (Acker 2006) depending on the characteristics and dynamics of union ‘internal labour markets’ and the rules that organize union careers. Following Acker, we can differentiate between various components of inequality regimes. By comparing the three unions, we argue that the degree of inequality is linked to the level of feminisation of the union hierarchy, but only to a certain extent. Even in feminised unions, a sharp horizontal and vertical job segregration prevails. When women are offered new routes to enter union careers, either as organizers or as experts, they seem to have difficulties in achieving more general positions, such as general secretary or senior national officer. This issue could be seen as a question to be resolved with time, with women leaders being younger than men leaders (at least in France), but this belief is rather optimistic given the internal organisational processes that result in maintaining gender inequalities. The effect of equality policies is also questionable. Radical measures such as quotas or reserved seats are useful in feminising upper levels of union hierarchy and union government bodies, but they may hide a wider difficulty with feminising the entire union hierarchy from bottom to top. This uneven feminisation lessens the possibility for the training and encouraging of new generations of future leaders. This situation can be explained by a lack of political will and feminist ideology, as in the French case, or can be seen as a side effect of ‘tracking’ women into less valued roles. Overall, the decentralised organisation of unions, the weakening of national strategies in the CFDT, and the isolation of self-organized groups in UNISON seem to be the main obstacles to the implementation of effective equality policies. This account confirms the centrality of ‘organisational repertoires’ in institutional changes (Clemens 1993).

If we look closer at the organisation of union careers and the nature of union work, the picture is rather ambivalent. Some aspects remain detrimental for women. Union work requires putting in long hours, particularly for full-time national positions, but also for elected roles at the workplace level. Union activity also implies frequent commuting or ‘militant trips,’ given the centralisation of union organisation in France (even for governing elected bodies at the ‘fédération’ level) and, to a certain extent, in the UK. Recruitment and hiring processes based on co-opting and informal practices tend to exclude atypical candidates, especially at the second stage of union careers, where equality policies fail to be properly implemented and controlled. The lack of social capital and internal networks is probably the main obstacle encountered by women when trying to reach the upper levels of the union hierarchy. Women leaders at the top continue to face ‘tokenism syndrome’ (Kanter 1977a) and often feel isolated. Their small number tends to create competition rather than solidarity, and very few of them feel entitled to promote other women, at least in the French context, probably because of their low feminist conscientiousness and/or the wish not to be identified as ‘feminists.’ In the UK, self-organized groups help and support elected female senior lay reps (Regional or National Executive Council), but their influence is less visible for full-time paid officers.

Other dimensions of union internal labour markets seem more favourable to women. The demography (and the number of available positions) is a factor that favours younger leaders, at least in France where a vast number of union leaders, at all levels of the hierarchy, are about to retire. The professionalization of union practices, the need for paid officers to possess a higher level of expertise, and the creation of more specialized union roles at the workplace level have opened opportunities for trade unionists with a different profile. This study shows that women leaders have a higher level of education than men, especially when they hold full-time paid positions. As in other male-dominated organisations, women have used the ‘qualification lever’ (Crompton and Sanderson 1990) to get hired and get a promotion. This is probably less true at the workplace level and for elected positions, where women can be chosen in the name of their supposed ability to promote (and identify with) a ‘tranformation leadership style’ (Heery and Kelly 2001). However, this rhetoric does not alter the internal job segregation, and may in fact reinforce it, while maintaining unspoken class (and race) inequalities. Despite their formal commitment to internal democracy through ‘fair representation’ or ‘syndicalisme d'adhérents,’ we wonder if UNISON and the CFDT really reflect this issue in their equality policies.

We gratefully acknowledge the support of the CFDT, GMB and UNISON. We would like to thank the anonymous reviewers for useful comments on an earlier draft.

Cécile Guillaume is a reader in HRM and employment relations at the University of Lille 1, France. Her research is located within the CLERSE (http://clerse.univ-lille1.fr/spip/). One strand of her research focuses on gender equality in the trade union context. In different publications, she has explored how and why women embark on trade union careers, women's union and gender identities, and the influence of equality structures on union women's participation.

Sophie Pochic is a researcher in work and employment at the CNRS and member of the Centre Maurice Halbwachs (EHESS). Her research interests focus on the making of ‘glass ceilings’ and the evolution of women and men's careers within different type of organisations (private companies, trade unions, public sector) or through unemployment experience.