Article Generational transition in union employ- ment: the organisational implications of staff turnover in COSATU unions1 Sakhela Buhlungu South Africa's transition to democracy is having a profound effect on entrenched patterns of social relations in all spheres of society. Trade unions have not been immune to these changes which are reshaping the landscape of social reality. One such change is the de-racialisation of society which is resulting in the gradual breakdown of old social cleavages and the emergence of new ones. Notable in this regard is the disappearance of old patterns of solidarity and radical traditions among the pro-democracy social forces and the emergence of new processes of class formation which arc resulting in the emergence of a so-called new elite. One implication of these processes is that social and political movements which contributed to democratisation are finding that these changes are now impacting on the ways by which they have to operate. Current developments in the trade union movement suggest that it is itself undergoing a transition with regard to its traditions, practices and politics. The growth in the membership of many unions has made them large and complex and introduced the need for sophisticated strategies to sustain their influence and power. This has highlighted the centrality of full-time union officials in their day-to-day running. On the other hand, the de-racialisation of society has seen scores of union officials taking advantage of new employment opportunities in other fields. This paper examines the implications for trade unions of the so-called 'brain drain' of full-time officials,2 particularly those under the umbrella of the Congress of South African Trade Unions (COSATU). It is based on on- going research on the changing nature of trade unions with specific reference to practices and traditions which have been the hallmark of the TRANSFORMATION 39 (1999) ISSN 0258-7696 Sakhela Buhlungu post-1973 unions. The figures and tables presented here are based on a preliminary survey of union officials conducted in 1996 and 1997 (see Buhlungu 1997) by means of a 13-page questionnaire sent to full-time officials in all trade unions. It contained 59 questions organised into six sections, namely, personal information, the position and service of the official, details of the job, servicing workplaces (organisers only), resources and support, and attitudes. The results used in this paper are based on responses by 587 officials,3 approximately one-third of the 1681 full-time staff employed by COSATU and its 18 affiliates. Full-time union officials constitute a distinct stratum of union leadership whose job it is to ensure proper co-ordination and efficient administration. They are the public face of modern trade unionism and, as such, they lead the union in negotiations with management and the state. Their full-time availability to do union work allows them access to knowledge, skills and specialist information which shopfloor workers and shop stewards do not have. By contrast, union members and shop stewards spend most of their time on the shopfloor with the result that they are left with very little time to devote their attention to union business. In addition, most of them have neither the specialist skills and knowledge nor the public profile to match the union official. The dangers posed by this state of affairs for union democracy have been the subject of heated debates for close to a century. The seminal work of Robert Michels (1959) is one of the most prominent contributions to this debate. Michels' study of political parties and trade unions, which drew its inspiration from Max Weber's work on bureaucracy, took a pessimistic view of the question of internal democracy within political parties and other social movements, such as trade unions. His point of departure was his notion of an 'iron law of oligarchy' which is based on the view that there are 'immanent oligarchic tendencies' in every human organisation (1959:11). In a nutshell, Michels' iron law of oligarchy can be summarised thus: the growth in size and complexity of an organisation necessitates the employment of full-time officials which inevitably leads to a greater division of labour between the leaders and the led. This, in turn, leads to 'class' differentiation, goal displacement and conservatism by the leadership. In such circumstances democracy becomes an impossible dream while the rule of the oligarchy becomes an unavoidable reality. According to Michels' thesis, full-time union officials become, then, a stumbling block to democracy in trade unions. 48 Generational transition in union employment In South Africa, the unions which emerged in the wake of the 1973 strikes were not unaware of the dangers noted by Michels. Thus, from early in their existence they established a tradition of worker control which sought to diminish the power of full-time officials so as to make them more accountable to rank-and-file members, particularly through a system of mandated decision-making. This tradition which permeated every aspect of union organisation and functioning had a number of elements to it, for example: • the emphasis on shop floor structures led by shop stewards; • the creation of representative structures in which worker delegates were the majority; • the practice of mandated decision-making and regular report-backs to members; • factory-level bargaining which allowed workers and shop stewards to maintain control of the bargaining agenda and the conclusion of agreements; • the subordination of full-time officials to control by worker-dominated structures; • the involvement of workers, at all levels, in the employment of full-time officials. When COSATU was formed in 1985, the principle of worker control was built into the founding constitution. According to Baskin, the rationale for this was' to prevent union officials from dominating COSATU structures' (1991:57). However, the transition to a democratic political dispensation, the growing size of the post-1973 unions and the increasing complexity of the issues confronting the unions have put strains on practices which promote worker control and leadership accountability. Between 1985-1991, COSATU membership increased from 462,359 to 1,258,853 (Macun and Frost 1994). A recent study by Filita (1997) shows that today COSATU represents 1.8 million members, distributed across 19 affiliates ranging in size from 6,520 to 357,198. This phenomenal growth has transformed unions into highly complex organisations, with the result that their modes of operation and governance are undergoing rapid change. The nature of their complexity is related to the size of the unions, the gradual centralisation of collective bargaining at the top and the desire to achieve high levels of administrative efficiency. 49 Sakhela Buhlungu These trends have of course been observed in many other trade unions across the world. In their classic study of early British trade unionism, the Webbs observed that as trade unions developed from small local bodies into large national ones they left behind the ideal of'primitive democracy', a rudimentary form of participatory democracy (Webb and Webb 1920:8). They noted that union growth was accompanied by the predominance of representative democracy and an increase in the number of full-time officials. This 'civil service of the Trade Union world' became increasingly alienated from the rank-and-file (Webb and Webb 1926:466). Thus, despite the tradition of worker control in many of the post-1973 trade unions in South Africa, there is now growing evidence that power is in the process of shifting from union members, shop stewards and worker office-bearers to full-time officials within these unions. The implications of these trends for worker control have been noted by Baskin: Major union issues cannot be adequately tackled and grasped on a part- time basis. As a result, it is usually union officials who wield real power, with elected worker leaders and executive committees acting as a check on the abuse of that power ... Since the principle of worker leadership was originally intended to ensure hands-on leadership by workers, the system clearly no longer works effectively. While the principle is retained, officials wield more power than ever before, and effective worker leaders no longer spend much time at work. (1991:461, emphasis added.) This has led some to conclude that the notion of worker control is inappropriate for decision making in large organisations (Maree 1992). Some unions have employed more officials to deal with some of the complex organisational issues confronting them or by creating areas of specialisation such as research, legal advice and bargaining co-ordination. A recent study revealed that COSATU and its affiliates now employ at least 1,600 full-time officials in different capacities (Buhlungu 1997). With a few notable exceptions (such as Maree 1982 and 1992; Buhlungu 1994b and 1997), studies of unions in South Africa, particularly the post-1973 unions, have been concerned with documenting and analysing the mobilisation of workers and the development of trade unions. However, one unintended consequence of this pre-occupation has been the neglect of the characteristics, conditions and role of full-time officials. Where references to them have been made, these have been cursory and anecdotal rather than systematic attempts to understand their position within these organisations (see MacShane et al 1984; Webster 1985; Keet 1992; Baskin 1991, 1996a, 1996b; Buhlungu 1994a, 1994b; and Collins 1994). 50 Generational transition in union employment But the centrality of union officials to the functioning of their organisations has been such that no account of South African trade unionism, past, present or future, can be complete without reference to their past and continuing role as employees, leaders, mobilisers, organisers, educators, negotiators, researchers, and administrators. In addition, an assessment of whether or not unions are becoming less democratic or more bureaucratic has to take account of trends in union employment. In this article, the term 'full-time officials' refers to all those unionists who are employed by trade unions on a full-time basis to perform a variety of administrative and co-ordinating duties. Although the literature generally makes a distinction between full-time officials (general secretaries, regional secretaries and organisers), on the one hand, and clerical (clerks and administrators) and support/ expert staff (researchers, legal officers, health and safety officers) on the other, the term is used here to encompass all these different categories of union personnel. Different unions use different methods by which to recruit full-time officials or staff. However, in the COSATU unions there is a broadly similar source of recruitment with officials almost invariably being drawn from one of three sources: the ranks of shop stewards or active rank-and- file members; or from the ranks of student or youth activists, some drawn directly into unions while others first join labour service organisations before applying for union jobs; or from the ranks of party or organisational political activists who see unionism as an important vehicle for their brand of politics. Most unions select their officials through an interviewing process, the format and depth of which differs from union to union. In years past, it was common for aspiring unionists to volunteer their services free of charge and then, when a vacancy appeared, to apply for paid employment. There were also in the past often more people willing to work for the unions than there were posts available. However, since the late-1980s many unions have lamented the fact that key officials have been leaving the movement. For example, the General Secretary's report to the 1989 National Congress of PPWAWU (19-20) observed: As stated above, we are still faced with a problem of high staff turnover ... This Congress must look for ways of minimising the high staff turnover. We must also discuss ways for PPWAWU to function effectively regardless of this turnover. The above shows that staff turnover in the unions is not simply a post- 1994 phenomenon. But, as the discussion in this paper suggests, the 51 Sakhela Buhlungu magnitude of the problem in the 1990s has grown to such an extent that it could change the character of the post-1973 unions. The rest of this paper examines the changing role of trade union officials and the organisational implications of staff turnover for COSATU unions. The central argument of this paper is that staff turnover, or the so-called brain drain, has a direct bearing on established traditions and patterns of organisation. Kelly and Heery's (1994) notion of'generational change' is a useful analytical concept which enables us to gain a better understanding of the changing composition of the cadre of full-time unionists. By this they mean changes in the composition of the 'officer' workforce as a result of changing social values. The attitudes of union officials are formed by the political and economic climate which prevails at a particular time in history. Kelly and Heery go on to argue that in Britain, officials who served in the 1950s and 1960s had their formative years in the 1920s when there was a general distrust of shop stewards among union officials. As a result, they tended to take a more bureaucratic approach to trade unionism, with little attention paid to building strong shop floor organisation. But the economic boom of the 1960s was accompanied by waves of worker militancy and the emergence of a new breed of officials who brought with them new values which sought to encourage strong workplace organisation in the unions. Following Kelly and Heery, this discussion has adopted the concept of generational change to explain the changes within this stratum of union leadership in COSATU. However, the term 'generational transition' is used here to denote not only the changes but also the continuities. This also helps to locate the South African changes in the context of a societal transition to democracy, which in turn is shaped by rapid processes of globalisation. In other words, what we see unfolding in COSATU (and the union movement in general) are not just changes but transitions. These transitions are about changes in the traditions and culture of the unions influenced by the profound re-organisation of society (particularly de- racialisation and class formation) and the restructuring of the global economic and political order. In this context, the term 'generation' does not simply refer to age. It is a sociological concept which refers to what Mannheim called a 'social generation', a social group which 'participates in the common destiny of that historical and social unit' (1952:303). In his study of youth and student politics in Cape Town in the 1980s, Bundy (1987:305) observed that in the process of grappling with 'a distinct set of 52 Generational transition in union employment social and historical problems', a social generation develops 'an awareness and common identity - a generational consciousness'. The concept of 'generational transition' suggests the existence of two generations of trade union officials, namely the old and the new, in the post-1973 unions. The consciousness of the old generation of officials was born out of struggles against, and confrontations with, a repressive apartheid state and intransigent employers. Under these circumstances, union officials assumed the role of liberation agents whose role demanded that they should continually devise new principles and modes of organising. In addition, this consciousness encompassed norms and values which guided the behaviour of individuals and the group, such as treating workers with respect, following democratic procedures and adopting an uncompromising stance towards employers and state institutions. Within some unions, informal codes of behaviour for officials dictated that officials should not take bribes, socialise with or be alone in the company of employer representatives or state officials. Changes in the position and role of full-time union officials The situation of full-time union officials today is a far cry from that which prevailed ten years ago. The first important point which comes out of the survey is that COSATU and its affiliates are significant employers in their own right. Although this point is an obvious one, its implications are not always so obvious. In general, there is a relationship between the size of a union's membership and the number of full-time officials it employs. The IPS has the smallest number of staff, with 18 officials, while NUMSA employs the biggest number: 208 officials/ However, staff policies in a number of unions are outdated and ad hoc and there are few unions with people dedicated to staffing and human development issues in the unions.5 Part of this is due to the fact that up until recently union work was seen by many as 'part of the struggle', rather than just another job. Jenny Grice of NUMSA argues that before 1994 union work was 'artificial' because there were few other job avenues for political activists who wanted to make a contribution in the struggle for democracy. So, union work was part of the struggle and unions did not have to improve conditions and salaries to attract and retain full-time staff. Although some officials did leave union employment, the turnover was comparatively lower than in recent years. Thus, it would appear that salary and conditions were not the main contributory factors at that stage as many officials tended 53 Sakhela Buhlungu to move into non-governmental organisations (NGOs) which offered more or less the same remuneration, full-time political organisation in other liberation organisations or went into exile. The following section of this paper reflects on the changing role of union officials and the nature of the emerging instability in union employment. Unions are big employers The results of the survey show that the growth of union membership in recent years has been accompanied by a rapid expansion in full-time official employment. This has had a number of implications for the functioning of the unions. First, most unions now spend at least half of their annual income on employment costs. With increasing numbers and upward pressures on salaries and other benefits, these costs are likely to continue rising. Many unions are increasingly finding that striking the right balance between staffing costs, on the one hand, and other administrative costs, on the other, is not an easy one. Table 1 gives a breakdown of the figures of officials by union and also shows staff-membership ratios for each union. Table 1: Full-time Officials by Union, 1996 Union Full-time Union total as % j Membership Staff/Member- Staff of Cosatu Total ! (1996) ship Ratio CAWU 59 3.5 [ 41 509 1:704 COSATU 52 3.09 | CWIU 77 4.58 ! 48 703 1:633 CWU 34 2.2 j 41 000 1:1 206 FAWU 108 6.4 | 145 124 :1 344 IPS 18 1.07 i 16 676 :926 NEHAWU 115 6.84 131 938 :1 147 NUM 187 11.1 357 198 :1 910 NUMSA 208 12.37 i 263 893 :1 269 POPCRU 38 2.26 j 43 520 :1 145 PPWAWU 62 3.68 i 53 900 :869 SAAPAWU 24 1.42 | 35 000 :1 458 SACCAWU 182 10.8 | 134 626 :740 SACTWU 120 7.1 I 160 000 :1 333 SADTU 60 3.5 i 114 962 :1 916 SAMWU 127 7.5 112 059 :882 SARHWU 52 3.0 40 421 1:777 SASBO 78 4.6 73 094 1:937 TGWU 4.7 I 67 288 1:841 TOTAL 1681 ; 1880 911 1:1119 54 Generational transition in union employment Second, large numbers of staff necessitate greater attention by unions to staff supervision, performance appraisal and administrative efficiency. This necessitates that staff management and development be designated as a function in its own right, rather than being a part-time task of the general secretary. It also means that unions have had to develop clear procedures on a variety of issues, such as recruitment, performance appraisal and disciplinary cases. Third, unions have moved away from flat salary structures by introducing grades and other forms of recognising responsibility and seniority, skill, service and workload. Changes to union officials' salaries and conditions resulted from attempts by unions to deal with a high turnover of staff, particularly in the early-1990s. In the 1970s and early-1980s, a number of unions had a very small resource base because they were relatively small and often did not have efficient methods of subscription collection such as stop orders. Even in cases where unions received outside financial support, this often went into paying for running costs such as meetings, travel, office rental and telephone charges. Thus many unions could not afford to pay salaries for some of their officials at all and their organisers often worked for many months as volunteers who were only provided with money for travel and lunch every day.* In many cases, unions which could not afford to employ large numbers of full-time officials also made use of the services of their shop stewards, for example, to do simple administration tasks or to sign up members on weekends. Those unions which could afford to pay regular salaries had flat salary structures, meaning that every official from the general secretary down earned the same salary. However, by the mid-1980s most unions had managed to achieve levels of financial stability which allowed them to pay their officials a regular monthly salary. Officials in financially stable unions could also expect to obtain an annual bonus and to take a month's leave every year. Today union salaries are graded and most unions have benefits such as housing allowances, car allowances and medical and provident funds. Table 2 gives salary ranges from the lowest to the highest in COSATU and its affiliates. Salaries and conditions for union staff have become critical issues today, and a union's ability to retain its staff has increasingly come to depend on that union's ability to offer a competitive package. This is far cry from the early years when officials joined the movement to serve and sacrifice rather than to make a living. 55 Sakhela Buhlungu Table 2: Salaries in COSATU and its ,\ffiliates: Salary Range Officials Percentage Rl - 1 000 2 0.4 Rl 001 - 2 000 31 5.8 R2 001 - 3 000 167 31.3 R3 001 - 4 000 223 41.8 R4 001 - 5 000 57 10.7 R5 001 - 6 000 27 5.0 R6 001 - 7 000 10 1.9 R7 001 - 8 000 9 1.7 R8 001 - 9 000 3 0.6 R9 001 - 10 000 2 0.4 R13 001 -Upwards 2 0.4 TOTAL 533 100 It will be noted from the above table that most officials (79 per cent) are in the R1000 - R4000 salary bracket, with 73 per cent of them earning between R2001 and R4000. Looked at differently, the figures show that 90 per cent earn between R1000 and R5000. The average salary is R3564. In most unions there are salary grades for each occupational position (eg, organiser) and within each grade there may be different notches. Individuals are then placed on particular notches depending on a number of variables such as experience, educational qualifications and previous salary (for new staff). In this case, educational qualifications tend to carry more weight than long service. Thus the decision to place a particular staff member is negotiable. In a few large unions, such as the NUM, it is possible for highly qualified individuals to negotiate salary packages which do not conform to standard grades. However, this is not a widespread practice. The potential for tensions emerging within unions as a result of these differential forms of reward and recognition of contribution has become greater than ever before. In particular, the lack of clear criteria could lead to infighting and competition among officials for leadership positions. Finally, there is a lack of clarity about how union officials articulate and represent their collective interests. In a number of cases this has led to the emergence of fiefdoms or power bases among officials which then lobby for support among shop stewards, worker office-bearers, and branch or regional structures. This has often led to serious tensions in some unions. For example, when salary increases are due officials in most unions do not 56 Generational transition in union employment have a structure to represent their concerns. What tends to happen instead, is that some officials will 'caucus' certain office bearers or shop stewards for a favourable increase. Invariably, those officials who possess indispensable skills or who occupy key leadership positions have more bargaining power as they may threaten to leave if they do not receive what they consider to be a reasonable increase. Growing segmentation The research also revealed that the labour market for trade union officials is increasingly becoming segmented, particularly along the lines of income, occupational position, education, skill and gender. This is an inevitable consequence of the increase in size of unions and the emergence of occupational specialisation based on educational qualifications and differential remuneration systems for union officials. The income differentiation represented in Table 2 above is a clear indication of these new cleavages. Segmentation along the lines of gender is also becoming more prominent. Although gender inequalities have existed for a long time among union officials, the move towards differential remuneration and conditions will exacerbate segmentation on gender lines. The research showed that although there is a fair representation of women among union staff (41 per cent were women), most women occupy administrative and other junior positions while most men occupy more high profile, and therefore more powerful, leadership positions. Table 3 below shows the distribution of both men and women according to positions in the unions. Table 3: Gender and Positions Among Union Staff Gender Current Position Mile" "Femile Total in Fosition Women as V. of Total Local Administrator 2 32 34 94 Branch Administrator 1 24 25 96 Regional Administrator 3 48 51 94 H/O Administrator 5 53 58 91 Local Organiser 79 7 86 8 Branch Organiser 33 8 41 20 Regional Organiser 46 6 52 12 National Organiser 24 27 11 Branch Secretary 17 4 21 19 Regional Secretary 23 1 24 4 57 Sakhela Buhlungu General Secretary 10 0 10 0 Research Officer 6 3 9 33 Legal Officer (Regional) 14 3 17 18 Legal Officer (National) 2 3 5 60 Education Officer (Branch) 4 0 4 0 Education officer (Reg.) 12 0 12 0 Education Officer (Nat.) 11 3 14 21 Media Officer 5 0 5 0 Other 43 40 83 48 TOTAL 340 238 578 41 This finding confirms Nyman's (1990) conclusion that women often 'come up against a glass ceiling which prevents them from rising to positions of power in the trade unions' (1990:30). This 'glass ceiling' for women exists not only in relation to elected positions, but also applies to full-time positions. This means that not only do women lack influence and power in their jobs, they also earn less. The survey revealed that only 56 women trade unionists earned more than R5000, compared to 166 male unionists. Education, training and skills are becoming increasingly important for union officials as unions emphasise organisational efficiency and demand a high standard of performance from their officials. This is a significant change from the past when formal educational qualifications were less important, or not important at all, as criteria for recruitment, selection, and evaluation for advancement within the unions. Now, in particular in regard to the staffing of specialised departments and support positions, such as legal departments, accounts, research, etc, unions seek highly educated officials to deal with technical issues. From the side of union officials, as well, education, training and the acquisition of skills are becoming important for a variety of reasons. The most important of these are job security, upgrading skills to deal with complex issues and improving qualifications so as to take advantage of advancement opportunities within the union. The survey made two key findings with regard to education, training and skills among union officials. Firstly, levels of formal education are high in the unions, with the overall average qualification being matric. Put differently, only 24 per cent of union officials do not have matric and around 75 per cent have matric and above as their formal education qualification. Also revealing is the figure of full-time officials with tertiary qualifications. The total of officials with technikon/university certificates, 58 Generational transition in union employment university degrees and post-graduate qualifications is 197. This contradicts Baskin's (1996a) conclusion that few officials with tertiary education remain in the union movement. If the number of officials who are currently studying is anything to go by, then the number of officials with tertiary education is going to increase rather than decline. Table 4 shows a breakdown of formal education levels among union officials. Table 4: Formal Education Levels Among Union Officials: Formal Education Level No. of Officials Percent (*/•) None 1 0.2 Up to std 2 2 0.3 Std 3 - 5 9 1.5 Std 6 - 7 18 3.1 Std 8 - 9 113 19.3 Matric (Std 10) 235 40 Technikon Certificate 84 14.3 University Certificate 34 5.8 University Degree 44 7.5 Post Graduate Qualification 35 6.0 Other 9 1.5 TOTAL 584 100 A second finding related to education is the fact that almost half (46 per cent) of union officials who responded were doing private studies of one sort or another at various institutions. Most (103) were doing special certificate courses (eg, at Damelin), while 61 were doing a university degree/certificate, 42 a technikon diploma/ certificate and 26 matric. These courses and degrees were not all relevant to the needs of the unions. Instead, it appears that some officials are improving their qualifications to be more marketable outside the unions. Where officials who leave the unions go in employment terms warrants a separate investigation. However, existing evidence suggests that the majority of ex- unionists have moved into politics (national and provincial), local government, the civil service, parastatals, the private sector (especially as personnel and human resource officers), union investment companies and in some cases NGOs (see for example, Buhlungu 1994b; Macun and Buhlungu 1996; von Holdt 1996). The growing importance of formal education qualifications is putting pressures on unions to create career paths for officials who improve their 59 Sakheia Buhlungu education and to reward qualifications accordingly, thus widening the gap between the bottom and the top of the salary scale. The dilemma for unions, however, is that union work does not offer many opportunities for upward mobility and thus the opportunities for the creation of career paths are limited. In many cases, promotion happens through election rather than appointment based on skills and qualifications. For example, an able and educationally qualified local organiser has only two positions above him/ her (regional secretary and national organiser) before he/she becomes general secretary. Even then, he/she can only obtain the positions of regional secretary and general secretary through election by a regional or national congress. Thus, a local organiser may have to wait 5 to 15 years before he/she gets to the top rung of union employment. Even then, he/she may lose that position after a few years. The net effect of this state of affairs is that many officials tend to look elsewhere where they believe there are greater opportunities for upward mobility and where their qualifications will be better rewarded in financial terms. Unions, therefore, are increasingly becoming waiting stations for unionists with a variety of ambitions. In this context, COSATU's alliance with the African National Congress (ANC) and the South African Communist Party (SACP)7 and the emerging sector of union investment companies provide many union officials with a new career path for upward mobility which did not exist before. Additionally, the stigma which many unionists attached to working for management and government has virtually disappeared. In a context where many unions have business arms (investment companies), those who 'cross the floor' no longer necessarily face the prospect of being ostracised by their comrades. In a nutshell, the high turnover of full-time officials may be explained in terms of'push' and 'pull' factors. Chief among the 'push' factors are the relatively low salaries and benefits, long hours of work, limited opportunities for upward mobility (career paths), inadequate recognition of educational qualifications and the fact that for some, union work does not carry the same status and profile of a middle or senior position in government or the private sector. Factors which 'pull' unionists into government and the private sector include generous salaries and benefits,8 enhanced status and better prospects for upward mobility. It is not, however, all union officials who have these options. Only those with relatively high educational skills or qualifications, senior union positions or who are active ANC or SACP members are realistically placed to take advantage of them. 60 Generational transition in union employment The changing nature of union work A trend which has accompanied some of the changes noted in this article has been the gradual restructuring of union work. This has been driven by the imperatives of organisational efficiency and a need to address complex issues in a changing national and international environment. In the past, the work of officials was largely unsupervised and there was little attention to performance appraisal. What little supervision there was came through the constitutional structures of the unions. Although these structures monitored the administrative aspects of officials' work, the greater part of this monitoring was political. Thus, union work was largely self-supervised and the union tended to rely on the political commitment and goodwill of the official to perform his/her work satisfactorily. This research has noted changes to this situation. Unions now put more emphasis on staff supervision; new procedures have been introduced which require officials to account not only for the general aspects of their work but about the details as well. In addition, union work is no longer just about membership recruitment, organising, political education, and negotiating wages and conditions annually. It now requires that, in addition, officials should do at least basic research, run workshops, use computers, read company information and prepare detailed proposals on technical issues in and outside the workplace. Thus a skilled official is one who is not simply good at handling the traditional issues, but one who is also able to tackle new ones. A revolving door Over the last few years, much has been said and written about the so-called leadership drain from the union movement, particularly the loss of some key full-time union officials (Buhlungu 1994b). Kgalema Motlanthe, ex- general secretary of the National Union of Mineworkers (NUM), has pointed out that the departure of large numbers of union officials should be seen in the context of the new democratic dispensation which presents 'new vistas of opportunities' (Buhlungu 1997). The reasons for the exodus have been summarised elsewhere as a combination of the following: • 'greener pastures' in the form of attractive packages and other benefits in NGOs, government and the private sector; • family pressures and responsibilities; • 'poaching' by management; • loss of vision and direction (Buhlungu 1994). 61 Sakhela Buhlungu However, little effort has been made to quantify this loss. This research has revealed that there is restlessness and instability among union officials. The survey attempted to establish the length of service of all union officials currently working in the unions. Table 5 looks at this question. Table 5: Length of Service: Year Employed by Union No. Percentage 1973 -1977 2 0.4 1978-1982 8 1.4 1983 - 1987 59 10.8 1988 - 1992 166 30.3 1993 - 1996 313 57.1 TOTAL 548 100 A number of conclusions can be drawn from these figures. First, the majority of officials (57 per cent) had not completed four years in the unions at the time they were surveyed. This figure is made up of 8.6 per cent who were employed in 1993,15.1 per cent who joined the unions in 1994, 16.4 per cent who joined in 1995 and 17.2 per cent who were employed in 1996. Second, looked at from a different angle, the table shows that only 24 per cent of officials from the 1970s and 1980s remain in the unions. The 43 longest-serving officials (those with more than 10 years service or who were already there by 1986) were in PPWAWU (2), CWIU (7), NUM (4), SASBO (4), SARHWU (2), CAWU (1), SAAPAWU (1), POPCRU (1), FAWU (1), SACCAWU (4), and NUMSA (16). Although some recent trends in the development of South African trade unions seem to confirm some of the assumptions made by the Webbs, Michels and later generations of sociologists, such as growth in size, increasing centralisation of decision making and a widening gap between leaders and the rank-and file, the above figures contrast with an assumption which is crucial for these theorists, namely, the emergence of a long- serving layer of full-time officials who make trade unionism a life-time career. The high turnover of staff in COSATU unions has prevented the emergence of a permanent bureaucracy characterised by longevity of service among full-time officials. This makes South African unions different from other trade union movements in countries such as Britain, Australia and the United States. However, it is premature to argue that South African unions are unique in this regard. The high turnover of staff may be a passing phenomenon Generational transition in union employment engendered by the specificity of South Africa in the current conjuncture. Indeed, the fluidity occasioned by the turnover may provide the new generation of full-time officials with opportunities to introduce new traditions and practices which will lay the foundations for the emergence of a permanent oligarchy when avenues for mobility outside the unions have been exhausted. In some of the bigger unions such as the NUM this is already happening. New skilled officials are able to negotiate individual salary rates, there have been upward pressures on salaries paid to retain officials and some general secretaries and presidents have been provided with luxury cars as an incentive for them to stay in the unions. Thus, those officials who do not succeed in securing the conditions they desire will continue to exit, but a few are succeeding in securing conditions which may soon match those in government and the private sector, thus making the incumbents of certain key positions keen to stay on. Third, the high turnover has meant that the staffing situation remains fluid. This 'revolving door' syndrome is so serious that although the survey was conducted at the end of 1996 and early-1997, several officials have left their unions since they were surveyed. This phenomenon shows no signs of letting up despite predictions by some that it was a phase that would soon pass. Although it is reasonable to predict that things will stabilise at some point, the research shows that the end has not been reached yet. The research also found a changing attitude to union work among officials. Many officials now see union work as a form of employment like any other. This means that many no longer see it as a commitment to a struggle or a set of political beliefs. They tend to judge it primarily on what it has to offer them in material terms. In addition, many of those who join unions today see union work as a stepping stone, a training ground to gain practical experience which can then be used to secure employment on better terms elsewhere. The implications of these trends is that the solution to the situation is more than just improving salaries and conditions. It is also about the unions articulating and projecting a clear set of political objectives which can form the basis of new bonds of political solidarity and value system. These two approaches, improving salaries and conditions and forging new bonds of political solidarity, will go a long way towards attracting people to the trade unions and stabilising the situation. Once that happens, officials would remain in the unions not because they are in relatively well-paid positions or because they have no alternatives, but because they want to stay. 63 Sikhela Buhlungu Lack of vision and direction What seems to have held the union movement together in the apartheid era and which made union work meaningful and rewarding for many officials was the fact that there was a much clearer sense of purpose, political objectives and value system than today where there appears to be a lack of political direction accompanied by confusion regarding the future of the workers' struggle. Old value systems and bonds of solidarity have virtually disintegrated. The context for this is that in the broader society, notions of sacrifice and the collective ethic of the struggle days have been replaced by individualism and a quest for personal wealth accumulation. These factors now influence the thinking and practice of union officials. Collins noted that: working for the Union was an act of struggle. The reward for one's work was not monetary, but the contribution that one was making to the liberation of society. It would appear that for some, since the struggle is now over, so too, is this value system. Many 'comrades' are now demanding that their services be valued in terms of entirely different criteria. (1994:23) These trends have, since the collapse of socialism, been accompanied by a decline in idealism and Utopian notions of a radical transformation of society. Notions of socialism which some unionists still espouse are now little more than rhetoric, an article of faith that has little or no relation to union practice. In practice, many unionists have abandoned any radical pretences with the result that unions are no longer seen as agents for radical change. A generational transition The notion of a generational transition is helpful for explaining these developments. It refers to the fact that union officials who participated in the birth and growth of the trade union movement are gradually moving out and being replaced by a new generation which may not have seen or participated in the struggles that accompanied the birth of the movement. The demise of apartheid and the de-racialisation of society have created a new context for union officials. More importantly, this has resulted in the emergence of a new consciousness among union officials as a social group. This change in context together with the emergence of a new consciousness and the 'brain drain' noted in this paper allow us to identify two generations of union officials - an old and a new. The defining features of each are outlined below. 64 Generational transition in union employment The old generation • embraced a tradition which regarded union work as a sacrifice and an act of struggle; • emerged at a time when union work was least attractive in terms of conditions and pay; • operated in unions at a time when union democracy was emphasised, particularly in relation to worker control. Leadership accountability in the form of mandates and report-backs was also strongly emphasised; • did not rely on negotiating skills alone but employed also the sheer power of the workers to achieve gains; • had their work and work schedules shaped by the concerns and militancy of the rank and file; • sought not to discourage worker militancy but to use it to achieve gains and victories; • directed their efforts towards the vision of a future where the working class would attain total victory. Although most of the officials who joined the union movement in the early stages fall within this 'old' category, it is not defined by long service alone. It is also defined by their experience of these traditions and values. There are officials who joined the unions in later years but who were committed to the politics which these traditions represented. The new generation • emerged when the tradition of sacrifice was on the decline and union work was changing from being struggle-oriented to being just another job but, with greatly improved salary and working conditions; • emerged when the traditions of worker control, leadership accountability and strong shop stewards were on the decline. Writing on what she termed a ' new type of officiaF, Keet (1992:3 7) argued that they resented, and did not fully understand, the notion of worker control. She went on to quote the view of some' long-standing union organisers' who suggested 'that some of the new "careerist officials" actively counter emerging shop steward leaders in a variety of ways'; • rely more on negotiating skills and less on the strength of the work force. There is also a tendency to rely on the advice of lawyers and other so- called experts rather than to resort to a political approach to resolve problems or disputes. These 'new' leaders take a service organisation approach to workers' problems; 65 Sakhela Buhlungu • operate with less clarity on the overall objectives and goals of the union movement in particular, and the working class in general. This has tended to leave many new generation officials without an overall perspective and vision of the working-class struggle. Again this generation is not defined simply by when they started working for the unions. There are many long-serving officials whose thinking and approach to union work has been transformed in recent years and who now hold the view that some of the democratic practices and traditions of the unions are outmoded. It should be obvious, then, that in reality this categorisation and the trends it represents is not as clear-cut as portrayed here. There are continuities from the past and the generational transition involves much contestation between the two generations, and between full-time officials and union members. There are also still overlaps in the consciousness of the two generations. It is not a given that the new generation of officials will succeed in imposing new value systems and politics on the unions. At the same time, unions cannot ignore the demands and concerns of the new generation of officials. As the number of new officials increases inside the unions, the traditions, value systems and ethics of union work will change. This means that the way officials see their work, the way they relate to workers and shop stewards, their demands for improved salaries and conditions, their perception of disciplinary actions against them and so on, will be different from the previous era. Generational transition is also not necessarily a negative development but it is important to take it into account when devising strategies for the future of the movement. Finally, this transition is an unfolding process and it is. therefore, difficult to state with any certainty where it is going. It is, however, possible to identify some of the dangers it holds for the unions in COSATU. The remaining sections of this paper discuss these dangers. The danger of bureaucratisation (the 'managerial model' of unionism) The research also threw up issues related to the danger of bureaucratisation in the unions. This is not an inevitability, but in their quest for organisational efficiency and the professionalisation of union work, unions could find themselves unwittingly planting the seeds of what has been termed 'a managerial model' of unionism (Bramble 1995) The generational transition could exacerbate this trend. According to Bramble, the key features of this 66 Generational transition in union employment model of unionism are 'its emphasis on trade union officials playing a servicing role towards members, with the latter seen in the role of customers of union products rather than active industrial participants' (1995:406). This is not to suggest that organisational efficiency and professionalism are not important. It just means that these efforts should take place in the context of a broader political strategy to take the unions into the future. Organisational implications This paper has noted a number of changes in union employment, particularly in unions affiliated to COSATU. While all the factors identified have organisational implications for these unions, it is the issues related to turnover and generational transition which are the most far-reaching. First, as the discussion above has noted, the high turnover of full-time staff is linked to a generational transition and has resulted in an erosion of some of the practices and traditions which have been the hallmark of the post-1973 trade unions. This could bring about a new orientation to the values and politics of the unions. Although the trajectory of union development in South Africa has a uniqueness of its own, this gradual transformation of the trade union movement here is not dissimilar from what has been observed in other movements in other parts of the world. In his study of American unionism after the Second World War, Richard Lester (1958:21) noted a process of settling down or 'maturation' in these unions. He argued that these unions tended to follow a 'natural evolution of organisational life' characterised by stages of development beginning with a radical and democratic stage. However, as unions matured, they tended to shed or outgrow their youthful characteristics and become both less radical and less democratic. South African unions could be entering into a similar stage, and staff turnover (coupled with a high turnover of worker leadership) together with the generational transition could accelerate this process. Second, the erosion of the tradition of worker control could result in competition for power and material benefits which higher leadership positions carry. In some cases such competition could lead to in-fighting and a debilitating fragmentation. Recent developments and splits in, for example, the Food and Allied Workers Union (FAWU) could be replicated in other unions. The FAWU case, where close to 40 per cent of the membership broke away to form a new union, resulted from allegations and counter-allegations of corruption among officials. It is illustrative of a new 67 Sakhcla Buhlungu culture of squabbling and competitiveness within the labour movement and within which union officials are playing a central role. Third, generational transition and the consciousness of the new generation could result in a significant shift of power away from workers and constitutional structures towards full-time officials. Such a shift would be manifested in diminishing leadership accountability to the rank-and-file and goal displacement in these organisations. This means that union officials would increasingly dominate trade union affairs while the democratic culture of the post-1973 unions would decline. It could also mean that the COSATU unions could find themselves edging closer to Michcls' conceptualisation of oligarchy, where a small but powerful coterie of full-time leaders come to dominate the rank and file. Finally, and in spite of the above-mentioned implications, worker mobilisation and contestation could act as a countervailing force to minimise, but not eliminate, the possibility of these tendencies. Hyman (1971:25) has argued that the presence of anti-democratic tendencies in trade unions does not inevitably result in oligarchy. He has observed that 'the same social conflicts which in the first instance generated unionism persist as counter- tendencies to the specific processes of integration, oligarchy and incorporation'. For the post-1973 unions, the processes identified in this paper could be the harbinger of a new phase where notions of democracy and worker control are subject to contestation and struggle between union officials and rank-and-file members. Notes 1. The author would like to thank Eddie Webster for his useful comments on an earlier draft of this article. 2. This article does not examine the position and changing role of the other layers of union leadership, particularly shop stewards, local, branch and national office bearers in the unions. Although the role they play is often similar to or dovetails with that of full-time officials, the latter constitute a distinct layer of leadership which is marked off from others by the fact that they are full-time employees of the union. In addition, traditions of'worker control' within the unions have always drawn a clear distinction between this stratum and union members (rank-and-file, shop stewards and worker office-bearers). 3. This report is based on responses by these officials. However, more questionnaires were received in the course of 1997 and are being incorporated to enlarge the database. 68 Generational transition in union employment 4. The number of officials in NUMS A has declined significantly since the early- 1990s. At the end of November 1993, the union employed 341 full-time officials (Buhlungu 1994a). 5. The exception in this regard is the NUM and TGWU, both of which have employed human resources managers. 6. Maree (1992:188) cites the example of Thizi Khumalo, a dismissed textile worker in Natal, who subsequently worked as a volunteer organiser for the National Union of Textile Workers (NUTW) for six months. There were many other cases of dismissed worker or youth activists who started off as volunteers. 7. The use of a union's alliance with a political party as a stepping stone to other career options is not unique to South Africa. In countries such as Australia and Britain many unionists have used their alliances with the Labour Parties as an avenue for upward mobility. 8. In some cases ex-unionists in government, business and the civil service earn four to five times their union salaries. For example, a general secretary of a large union who earned about RS000 with allowances per month recently left the union to join one of the provincial legislatures under the banner of the ANC. He was subsequently appointed a member of the executive committee (MEC) and now earns at least R20 000 per month plus other benefits such as car and other allowances. References Baskin, J. 1991. Striking Back: a history ofCOSATU. Johannesburg: Ravan. 1996a. "The social partnership challenge: union trends and industrial relations trends'. In: J Baskin (ed), Against the Current: labour and economic policy in South Africa. Johannesburg: Ravan. 1996b. 'Unions at the crossroads: can they make the transition?', South African Labour Bulletin 20(1). Bramble, T. 1995. 'Deterring democracy? Australia's new generation of trade union officials', Journal of Industrial Relations 37(3). Buhlungu, S. 1994a. 'COSATU and the elections', South African Labour Bulletin 18(2). 1994b. "The big brain drain: union officials in the 1990s', South African Labour Bulletin 18(3). 1997. 'Working for the union: a profile of union officials in COSATU', Labour Studies Report 8, Sociology of Work Unit, University of the Witwatersrand, Johannesburg. Sakhela Buhlungu Bundy, C. 1987. 'Street sociology and pavement politics: aspects of youth and student resistance in Cape Town, 1985*, Journal of Southern African Studies 13(3). Collins, D. 1994. 'Stopping the gravy train', South African Labour Bulletin 18(3). Filita, T. 1997. 'COSATU: marching forward', South African Labour Bulletin 21(1). Heery, E and J Kelly. 1988. 'Do female representatives make adifference? Women full-time officials and trade union work', Work, Employment & Society 2(4). Heery, E and J Kelly. 1990. 'Full-time officers and the shop steward network: patterns of co-operation and interdependence'. In: P Fosh and E Heery, Trade Unions and their Members: studies in union democracy and organisation. Basingstoke: Macmillan. Hyman, R. 1971. Marxism and the Sociology of Trade Unionism. London: Pluto. Keet, D. 1992. 'Shop stewards and worker control', South African Labour Bulletin 16(5). Kelly, J and E Heery. 1994. Working for the Trade Union: British trade union officers. Cambridge: Cambridge University Press. Lester, R. 1958. As Unions Mature. Princeton, N.J.: Princeton University Press. MacShane, D, M Plaut, and D Ward. 1984. Power!: black workers, their unions and the struggle for freedom in South Africa. Nottingham: Spokesman. Macun, I and S Buhlungu. 1996. 'When the rain comes, it rains for everybody: interview with Gwede Mantashe', South African Labour Bulletin 20(1). Macun, I and A Frost. 1994. 'Living like there's no tomorrow: trade union growth in South Africa, 1979-1991', Social Dynamics 20(2). Mannheim, K. 1952. Essays on the Sociology of Knowledge. London: Routledge and Kegan Paul. Maree, J. 1982. 'Democracy and oligarchy in trade unions', Social Dynamics 8. 1992. 'Developing trade union power and democracy: the rebirth of African trade unions in South Africa in the 1970s'. Unpublished manuscript. 70 Generational transition in union employment Marie, B. 1992. 'COSATU faces crisis: "Quick Fix" methods and organisational contradictions', South African Labour Bulletin 16(5). Michels, R. 1959. Political Parties. New York: Free Press. Nyman, R. 1996. 'The glass ceiling: the representation of women in COSATU', South African Labour Bulletin 20(5). PPWAWU. 1989. Annual Report to the 3rd National Congress. Von Holdt, K. 1996. 'Worker councillors in Witbank: new ways to fight for the living wage', South African Labour Bulletin 20(1). Webb, S and B Webb. 1920. Industrial Democracy. London: Longman, Green and Co. 1926. The History of Trade Unionism. London: Longman, Green and Co. Webster, E. 1985. Cast in a Racial Mould: labour process and trade unionism in the foundries. Johannesburg: Ravan. 71