Article Adversarial Participation and Antagonistic Cooperation?: workplace forums, employee participation and lean production Gilton Klerck Introduction Major developments in labour law usually occur in times of social and economic crisis. South Africa is no exception. The publication of the Labour Relations Bill on 10 February 1995 was without doubt a significant milestone on the path towards a post-apartheid South Africa. At the launch of the Bill, the Minister of Labour, Tito Mboweni, hailed it as a 'part of our broader programme for social and political transformation'. Unlike the bulk of the laws passed by the new government, the Labour Relations Act of 1995 (LRA) is not geared primarily towards the provision of 'things', but rather concerns itself with the transformation of power relations. As Kahn-Freund observed, labour law is mainly concerned with the regulation of social power - the capacity to effectively direct the behaviour of others (1977:3). In part, therefore, the new LRA must be evaluated in terms of its potential contribution towards the struggle for the structural transformation of South African society. This will require, amongst others, a clear grasp of the relationship between current changes and the sites of continuity. In a sense, South Africa is attempting to set in place a labour relations regime without parallel in the developing world: balancing global competitiveness with a stable and redistributive growth path. The institutional framework through which these goals are to be realised - a 'bargained' corporatism - will compel the labour movement to embrace the strategies and tactics of'strategic' unionism (Baskin 1996). The three central pillars on which this framework rests, operate on the macro (NEDLAC), meso (bargaining councils) and micro (workplace forums) levels. A sustained articulation between these different levels, underwritten by demand-driven Keynesian economic policies, TRANSFORMATION 40 (1999) ISSN 0258-7696 Gilton Klerck will provide the foundation for a virtuous cycle of growth (Webster and Gelb 1996). In so far as these policy prescriptions underlie the government's labour relations policies, they are distinctly out-of-step with the dictates of 'globalization'. The Minister of Labour at the time of the gazetting of the Act, was keenly aware of this fact: The world tends to see globalization in terms of down-sizing, you reduce democratisation of the workplace, you tighten things, and increasingly disallow unionisation and so on. We need to liberate productivity from that kind ofperspective... So, yes, there's a sense in which we' 11 be moving against the current. (Mboweni 1995:24) The ANC government aims to sustain a balance between political regulation and market forces that will serve the needs of both economic growth and social equality. However, relatively high levels of political mobilisation and vast socio-economic inequalities exacerbate the tensions between strengthening labour andmaking firms more competitive. The political changes that culminated in the democratic elections of 1994 were not accompanied by any significant shifts in the ownership and control of economic resources. South Africa continues to be characterised by extensive racial inequalities and vast disparities in wealth. Big business also remains largely outside the political and ideological hegemony of the Tripartite Alliance (African National Congress, South African Communist Party, and the Congress of South African Trade Unions). Hence South Africa's first democratically-elected government has to transform apartheid-capitalism fromaposition with little leverageover the'commanding heights' of the economy. The growing internationalisation of capital, ie the growth in its mobility as compared to that of labour, also means that capital flight is a real possibility in unfavourable circumstances. There is thus a powerful tension between the state's economic dependency and an increasing radicalisation of state policies. Corporatist arrangements therefore contain inherent limits on the degree of transformation possible: radical change is constantly undermined by the international mobility of capital and its unwillingness to engage in hostile or unfriendly policy-making forums. Unravelling the complex and contradictory unfolding ofthis new institutional framework obliges social scientists to look beyond the superficial and provisional and to uncover the conjunction of processes and conditions that constitute the terrain of labour relations in South Africa. The regulation of the employment relationship is neither a willful contingency nor a fateful necessity. To avoid both voluntarist and determinist explanations we need to recognise the duality of structure and the duality of agency. Social structures are constraining as well as enabling, while human agency can both transform and reproduce prevailing Workplace forums, employee participation and lean production institutional arrangements (Bhaskar 1989). The transformative capacity of human agency is always bounded by prevailing structures and conditions. Since social transitions are characterised by change as well as continuity, we need a theory that can adequately account for both purposive agency and structural causality. As Dunlop noted: the 'major characteristics of a national industrial relations system appear to be established at a relatively early stage in the development of a country... [and] appears to retain these characteristics despite subsequent evolution' (1958:307). The 'success' of Mitbestimmungm Germany, for example, is rooted in the fact that the labour movement, at an early stage of its development, regarded working through the established legislative machinery as a rational strategy. This ideological orientation of the German trade union movement generated' a closer attachment to the prevailing [capitalist] order of society' (Sorge, cited in Bean 1985:182). If changes in Soujh African labour relations are to move 'against the current' of the neo-liberal order significant socio-economic, political and ideological obstacles will have to be overcome. Not least of which are a deep- seated mistrust between management and trade unions, structural unemployment, and a lack of international competitiveness. The limits and possibilities of employee participation (EP) in South Africa are conditioned by the power relations underlying civil society and the state; a legacy of the compromises made during the negotiated dismantling of apartheid. The actual or potential impactofEP schemes cannot be adequately understood outside the inter-related set of choices and constraints generated by prevailing national and international conditions. In the immortalwords of Marx: people make their own history, but not under conditions of their choosing. The Socio-Economic and Political Context of the LRA The struggle around the future structure of labour relations in South Africa takes place within a specific global and local context. Internationally, the current conjuncture is one in which the labour movements of many countries are under attack, trade unionism is on the decline in its traditional strongholds, and management has seized the initiative in the restructuring of production and the transformation of labour relations. Much ink has been spilled in an effort to present new 'cooperative' strategies as a necessary part of capital and the state's response to the vagaries of globalization. The only choice sanctioned by the 'new world order', we are told, is to increase managerial control and emasculate the unions or attempt to harness the power of the unions by incorporating them in bargaining and participatory structures. Webster and Adler, for example, endorse 'a class compromise between capital and labour: Gilton Klerck a left version of social democracy' and insist that 'socialist solutions are unfeasible' (Webster 1998:59). The impact of the increasing globalization of production, together with the loss of traditional political identities, have led many unions in Europe to either a 'strategic disorientarion' or a 'syndicalist opportunism' (Hyman 1994:116). While the outlook for South Africa's labour movement is certainly more optimistic, its strategic options are severely curtailed by widespread poverty, low levels of economic growth, adversarial and low-trust labour relations, and neo-liberal economic policies. The sway and leverage of trade unions mean that employers in South Africa are unable to compete on the basis of labour costs with many newly industrialising countries. The pursuance of a low-cost strategy presupposes the suppression of the labour movement and leads to conflict with the unions or takes the form of 'productivity alliances' such as the enterprise unions in Japan. A relatively stable, productive and equitable labour relations regime in South Africa will demand incorporation and empowerment, rather than marginalisation and restraint, of trade unions. This, in turn, will require a move away from a low-skill, low-productivity, low-wage, low- investment in research and development, and a low-value-added economy. Such a drastic shift in the economy will, of course, be conditioned by South Africa's insertion into the international division of labour. While the labour movement has the power, influence and resources necessary to obstruct the cheap-labourroute to increased competitiveness, the question remains whether it can also constitute the driving force behind a high-wage/high-skill route. The modalities of such an endeavour will be decisively influenced by (amongst others) the scope for conflict and compromise within the institutions of labour relations. The currently fashionable emphasis on greater 'employee involvement' in the workplace must be contextualised within the framework of prevailing managerial strategies and governmental policies. The key to increased competitiveness, according to the proponents of 'new' labour relations, is production and employment flexibility. The latter, in turn, revolve around deregulation, the decentralisation of wage determination, dismantling institutions of shop-floor control such as job demarcations, the consolidation of a 'core' and 'peripheral' labour market, and the individualisation of the employmentrelationship. The 'flexible' or 'lean' firm is said to be characterised by an ability to adapt rapidly to changes in consumer demand. The alleged shift from economies of scale (mass production) to economies of scope (lean production) has necessitated a greater emphasis on employee cooperation, Workplace forums, employee participation and lean production multi-skilling, team work, and a delayering of the managerial hierarchy (see Womack et al 1990). The establishment of quality circles, just-in-time inventories, semi-autonomous work groups etc, are presented as a move away from alienated, deskilled employment towards new forms of participation and high trust relations. In fact, social integration, workplace harmony and equilibrium are central components in the account of a shift towards more flexible forms of employment. Since the employer constructs the terrain of employee representation, managerial strategies constrain the options and outcomes of participatory arrangements. The practical experience in various national contexts, Biagi argues, demonstrates that, 'in spite of institutional mechanisms, for most private sector employees the nature and the degree of their participation relates primarily to the human resources management policies of their employer' (1993:348). To an important extent, therefore, a prognosis of employee representation presupposes an analysis of employers (Hyman 1997:322). The contemporary fascination with lean production, human resource management and quality control has brought more individualistic forms of employee involvement to the fore. Employee share-ownership and profit-related pay schemes emphasise the individual's involvement (through his/her financial stake) with the company. That is, new managerial strategies attempt to promote financial rather than participative-democratic forms of employee involvement. Many of the new managerial strategies involve a third 'channel' of representation-ie employer-initiated mechanisms of'direct' representation1 - alongside the first (collective bargaining) and second (joint consultation) channels. These management-inspired forms of EP extend certain privileges to the workforce rather than grant employees enforceable rights. However, a unilinear progression in workplace transformation towards greater EP finds no support in empirical research (Osterman 1994). Alongside the emphasis on team working and participation, conventional forms of authority relations overtly based on an assertion of managerial control, continue to exist (Duncan and Payne 1993, Moody 1997). Existing institutional arrangements, formal and informalrules of workbehaviour, andmanagerial strategies have abearing on both the form participation is likely to assume and whether trade unions reject, acquiesce or embrace these strategies. In many cases these new work arrangements could only be established through a more aggressive managerial style (Eaton and Voos 1992:191). A decisive contradiction for management is that a 'hostile approach always has to be an option under a lean production regime' (Steward 19 97:4). The ideology of'empowerment' isthusnuUifiedby the real imperative of lean production: getting employees to work harder. Gilton Klerck Industrial democracy is often said to involve a restructuring of managerial relations (power-centred) and a redesign of jobs (task-centred). Very rarely, however, is the relationship between the management of labour relations and control of the labour process explored in a sustained manner. The powers and liabilities of EP schemes must be conceptualised in terms of both the detailed control that workers exercise over their immediate work process and management's capacity to direct the workplace as a whole. The distinction between 'hard' production-orientatedand' soft' employee-orientated elements in the managerial process reflects the fact that EP may increase the scope of employee input into managerial decisions without altering the underlying structures of the workplace. Management-sponsored forms of participation usually mean that workers are expected to adopt the interests of the enterprise (quality, efficiency, etc) as their own. Managers have often been more interested in consultative participation to tap the knowledge of employees and increase their motivation to work than to extend their decision-making powers (Crcssey et al 1985:81, Wood 1988:20, Milkman 1997:140). As such, lean production and other forms of task participation do not necessarily involve a contraction in managerial authority, but rather its reconstitution in a different form. Hence the paradox: 'as workers were given more autonomy they were increasingly coming under tighter managerial control' (Geary 1994:648). No matter what labour relations strategy it pursues, management's basic objective remains a stable, predictable and cost-effective labour force. The provisions on workplace forums (WPFs) mirror much of the logic underlying lean production and flexible specialisation: securing employee commitment through participation; providing information and consultation arrangements; tapping the reservoir of knowledge about work processes possessed by employees; and an emphasis on the role of employee cooperation and harmonious labour relations in improving quality and efficiency. In the context of a managerial fascination with lean production it is no wonder that some commentators postulate the 'Japanisation' of works councils (Altmann and Dull 1990:124). That is, the ideology of lean production is likely to have a profound impact on management's view of the objectives and functions of second channel labour relations. An explicit focus on the impact of lean production on the shopfloor has highlighted the naive optimism in much of the literature on new managerial strategies (see Grenfell and Bendix 1994, Veldsman 1994, Horwitz and Franklin 1996, Anstey 1997). Several ethnographic studies have found that the working practices associated with new management strategies are stressful, Workplace forums, employee participation and lean production subordinate individual needs to those of the firm, lead to increased pressures on the work force, and have done little to empower workers or change then- attitudes towards employers (Briggs 1988, Dawson and Webb 1989, Fucini andFucini 1990,Babson 1993,Macshane 1993, Webb 1996). Kelly andKelly noted that the methodologically strongest studies of employee perceptions of new management strategies all revealed that there has been no change in the underlying 'them-us' attitudes of employees. They conclude that: there is little or no evidence to suggest that a variety of new industrial relations practices has altered workers' largely negative views of management in general... In reality, managements on both sides of the Atlantic have sought to guard their own authority, prerogatives and privileges against encroachment from below; they have claimed the lion's share of the gains; and they have been prepared to ditch schemes that no longer suited their purpose or were failing to deliver the goods. (1991:43- 44) The rhetoric of consensus and participation built up around lean production is thus readily exposed under close empirical scrutiny. Managerial initiatives which extend beyond mere consultation and involve delegative forms of representation remain remarkably rare. Management seems more concerned to convince employees of the need for, and the desirability of, change.' Conformity, not creativity, is the goal' (Moody 1997:89). Solidarity among workers is not only under threat from the individualism of new management strategies, but also from concerted efforts by the state and employers to reconstitute the relationship between management and workers (cooperation) and between workers in different firms (competition). These efforts reflect an attempt to create a fictitious identity of interests between employers and workers in the context of a competitive global economy. References in the Explanatory Memorandum accompanying the Labour Relations Bill (the Memo) to 'our' industries and 'our' economy encourage workers to think in terms of 'our' company. To be sure, low levels of economic growth clearly undermine the power of the labour movement. Job protection acquires increasing prominence, employee commitment and allegiance are orientated towards the survival of the firm, and broader processes of political exchange are weighted against the unions by adverse economic conditions. It does not follow, however, that a 'consensual' approach is the only viable option for the labour movement. The ultimate logic of such an approach is some form of'corporate patriotism'(Mahon 1991:305) at the workplace level where there is 'little room for powerful unions, anticapitalist consciousness or a stronger solidarity of employees with the community in general' (Alvesson Gilton Klerck 1991:364). This conception entails the replacement of the class struggle with the struggle for markets. Cressey et al refer to this as the 'lifeboat-democracy' argument: the common interest in survival forces both sides to work together thereby creating a unity of purpose (1985:156). Such conditions hardly provide fertile soil for the development of long-term cooperation and trust. Moreover, we shouldnever confuse mutual dependence with equal dependence. Employee Participation, Labour Legislation and Trade Unions In South Africa, black workers were only granted formal access to collective bargaining and trade union rights in 1979. Employers and the apartheid state devised a range of measures to forestall the development of autonomous unionism among black employees. Government and business leaders under apartheid went to great lengths to ensure that the wage-effort bargain of black employees was not established through 'free' collective bargaining. Legislative activity followed closely the various upsurges in labour militancy. Following the waves of industrial action in the early 1970s, for example, the National Party government passed the Black Labour Relations Act of 1973 in another attempt to contain the parameters of black employee interest representation. This Act expanded the rights of the existing system of enterprise-level works committees introduced by the Native Labour (Settlement of Disputes) Act of 1953, and introduced liaison and coordinating committees in an attempt to diffuse and institutionalise black labour militancy. By 1980, 2 745 liaison committees, 327 works committees and 5 coordinating committees had been established representing the interests of 799 369 workers (Anstey 1997:91). While the apartheid state may have portrayed these developments as an acceptance of its policies, in many instances these committees were in factused to build a base from which the independent unions of the 1970s was launched. This history of racial oppression and compromised forms of interest representation for black employees has resulted in a deep-seated mistrust of such forums and a corresponding commitment to collective bargaining through independent unions. The consciousness of black workers in South Africa is still in large part shaped by the collective memory of racial discrimination, state repression, economic disempowerment, autocratic employers, and appalling living conditions. The drafters of the LRA introduced a number of trade union controls over work place forums (WPFs) in (partial) recognition of this historical legacy. 'Workplace forums', the Memo warns, 'must not be conceived, and must never be permitted to be used, as alternatives for trade unionism' (1995:137). Workplace forums, employee participation and lean production WPFs are presented as a check on managerial prerogatives in exchange for unions' willingness to co-operate and refrain from adversarial tactics. Consultative institutions and processes - the so-called second channel of labour relations - are intended to supplement the processes of collective bargaining at workplace and industry levels. In the Memo, the argument for WPFs is presented as follows: South Africa's re-entry into international markets and the imperatives of a more open international economy demand that we produce value-added products and improve productivity levels. To achieve this a major process of workplace restructuring is required... Workplace restructuring has been most successful in those countries were participatory structures exist, for example, Japan, Germany and Sweden... Workplace forums are designed to perform functions that collective bargaining cannot easily achieve: the joint solution of problems and the resolution of conflicts over production. (1995:135) The strong voluntarist elements in the LRA makes it clear that the government's intention was not to impose a standardised form of EP by law.2 The role of the LRA is presented as providing a floor of rights above which various different schemes could be agreed to. The broad objectives of WPFs, according to the drafters of the LRA, are threefold: to promote the interests of all employees irrespective of union membership, to enhance efficiency in the workplace, and for employees to be consulted and to participate in joint decision-making with the employer. There has been an extensive debate over the use of concepts such as 'employeeparticipation' and 'industrialdemocracy'.Pateman( 1970)identified three different kinds of participation: (a) pseudo participation -management uses participation as a way of persuading workers to accept decisions that have already been made; (b) partial participation - two or more parties are able to influence the decision that are made but the final power rests with one of them; and (c)fall participation—a process where each member of a decision-making body has equalpower to determine the outcome of decisions. The problem with such a demarcation of forms of representational participation is that it rides roughshod over the actual variety of possible types. Marked variations are possible in the extent or depth of EP; the range or scope of decisions subject to participation; the form that participation structures may assume; the person(s) that are party to participatory arrangements; the organisational levels on which participation occurs; and the purpose and outcomes of participatory actions. The forms which participatory initiatives may assume, for example, cover a broad spectrum from suggestion schemes, board-level representation, briefing Gilton Klerck groups to worker cooperatives. Recognising the different levels at which participation may occur allows for the possibility of an extension of participation at one level (eg workplace) and the restriction of it at another (eg corporate). As Geary's survey of EP schemes shows: 'management has at once become enabling and restraining' (1994:650). A continuum of possibilities - ranging fromno involvement to receiving information, joint consultationjoint decision- making and employee control - can provide for a more nuanced account of variations in the form and content of employee representational participation. The range or scope of issues subject to participation will depend on a number of inter-locking variables. These include the attitudes of the parties (especiallymanagement'scommitmenttoalong-termparticipatory endeavour); the extent to which participation is based on statutory compulsion or voluntary agreement; the degree of stability in product markets; the nature of ownership and organisational characteristics (size, degree of centralisation in decision- making, available resources, etc); and the levels of experience and skill among employees and their representatives to engage in participation (Blyton and Turnbull 1994:210). Considerable variation, even in countries where EP is compelled by statute and managerial commitment to employee involvement is high, suggests that we cannot ignore the many other variables which can influence the depth and scope of participation. Since the diversity in forms of EP cannot be accounted for in functional terms, explanations mustbe sought in the wider political, social and ideological context. Specifying the(complex and uneven) impactofthese 'environmental' factors is an area in which conventional industrial relations, with its affinity for systems approaches and functionalist explanations, is singularly unhelpful. In an attempt to explain cross-national variations in EP, Bean (1985:163-78) identified four sets of determinants: (a) specific cultural and ideological factors; (b) differences in collective bargaining structures; (c) the nature of power relations between the state, employers and trade unions; and (d) the particular historical circumstances which prevailed at the time when the decision to create participatory institutions was first made. It is important to note that most participatory institutions were introduced in the context of a crisis. Of course, a decision to implement EP does not guarantee its success. Instead of measuring it against some abstract standard, a more useful approach would be to investigate the conditions under which EP functions best. To this end, Bean identified a number of contingency conditions which have been shown to influence the outcomes of EP in various countries: the effect of trade unions; enterprise characteristics; the subjective orientations and attitudes 10 Workplace forums, employee participation and lean production towards participation held by workers and managers; and the influence of the external environment. Together these factors lead to a highly uneven development of EP schemes within and across different industries. The complex and contingent nature of participation in various settings highlights the need to study it along distinct dimensions of democratisation. Participatory structures and practices not only evolve in an uneven manner, but also seem to follow a distinctly cyclical pattern. Interest in EP has waxed and waned depending on factors such as managerial commitment, trade union organisation, and product market conditions. In an explanation of the 'cycles' of interest in EP, Ramsay (1985) stresses the importance of the different objectives of labour and management. Management's interest inEP occurs when its authority is being challenged and when it is offered the opportunity of securing the compliance of labour by encouraging the development of an 'enterprise consciousness'. Hence managerial interest in participation rises and declines in accordance with the extent of (a) the challenge posed by labour and (b) capital and the state's need to secure the cooperation of labour. Once these conditions abate, participatory structures generally fall into decay. This cyclical interest in EP, Ramsay argues, is evidence of its function only as a temporary means of accommodating apotentially disruptive labourmovement, rather than any underlying employer or state commitment to greater employee involvement in decision-making.3 The findings of several case studies suggest that EP schemes were only marginally successful in securing employee integration, cooperation and commitment. Applebaum and Batt report in their extensive study of workplace reforms that 'despite the reported gains in performance and the apparent acceleration of experiments with innovative practices, the overwhelming majority of US workplaces are traditionally managed' (cited in Milkman 1997:144). In fact, several studies have indicated that the single greatest improvement associated with co-determination is an increase in the flow of information (Bean 1985:179, Slomp 1995:296, Danford 1997:112). If these schemes were as successful as their proponents insist, it would be difficult to explain the general lack of depth to management's commitment to EP. The strength of Ramsay's (1985) analysis of managerial control is that it stresses ideological goals rather than directly profitable or productive ones. That is, the emphasis on participation creates the impression of a responsive, reformable enterprise structure, so legitimating the system. This approach implies a greater degree of scepticism aboutthe transformative capacities of EP. Despite 11 Gilton Klerck the reassuring discourse of EP, many of the changes of the past decade 'owe more to the attempts of management (and the state) to exert control over labour than they do to involve or empower employees' (BlytonandTurnbull 1994:207). EP schemes in practice almost always revolve around a' consensual unitarist' philosophy. It is simply assumed that common interests in the workplace are both desirable and achievable. Much of the literature on EP is silent on the contradictions between capital and labour. The fact that EP schemes are only ever partially successful is rooted in the basic contradictions underlying the employment relationship. Moreover, since workers' self-activity is rarely given institutional form, theories of EP under capitalism must of necessity remain incomplete and inconsistent. (As it derives from the nature of the object under investigation, it is more of an explanatory 'must' than a 'must' of logical necessity.) Participation is not, however, simply a matter of managerial subterfuge. Support for such schemes also emanate from workers and their representatives, usually on the basis that EP offers an opportunity to expand labour's influence and control. In order to understand the popularity of EP within any particular cycle, we need to explore the conditions that allow trade unions to challenge managerial authority in a manner that (a) compels management to look to participation in order to satisfy its own interests; and (b) assures workers that they have a good chance of using participation for their own different ends. Comparative research has found wide variation in the activities of and the influence exercised by works councils and trade union delegations, as well as in their relationships with management (Eaton 1990, Biagi 1993, Knudsen 1995, Slomp 1995). These differences extend across both national boundaries and regional and sectoral divisions within a particular country. Of particular importance in this regard is the significance of industrial branch (eg active councils in the auto industry and passive ones in construction) and workplace size (eg large firms have more extensive EP than small firms). The nature and dynamics of trade unionism, and hence of collective labour relations, also have a profound impact on the type of EP that is possible or feasible in a given context. EP tends to be most 'successful' in a context where employers and trade unions regulate the workers' participatory demands in a way that prevents any threat to either collective bargaining or cooperation in the workplace. Eaton's (1990) wide-ranging study of EP in the USA found that union control over participative programmes is less a function of bargaining power than of local union resources, national union policies towards EP, and perceptions of the threat the programme poses to the union. 12 Workplace forums, employee participation and lean production In Belgium, Germany and the Netherlands, 'integration, cooperation and union coordination or even discipline amount to a form of 'disciplined integration' between the works council and union representation in managerial decision-making' (Slomp 1995:305-6, see also Kirkwood and Mewes 1976, Sorge 1976). In the Nordic countries union representation enjoys priority over works councils, Sweden being perhaps the most extreme example. Since the extension of bargaining rights to all social and personnel issues in the Co- determination Act of 1976, Swedish unions have marginalised the works councils, leading to their demise. Trade union representation has become the only institution of EP in both centralised collective bargaining (distributive issues) and the workplace (production issues). In short, EP functions well without works councils. A study of works councils around the world during the 1970s by the International Labour Organisation noted that: there is often disenchantment with their functioning. There is a broad consensus in many countries that works councils have not lived up to the expectations that were placed in them when they werefirstinitiated. Many examples could be given from various parts of the world to show that the works council is not, perhaps, an ideal means of handling employer- employee relations at the enterprise level. (Cited in Cressey et al 1985:4) Drawing on comparative research to account for variations in the extent of EP in works councils, Slomp (1995:308) points out that extended co-determination in Denmark is regulated through collective agreements in the virtual absence of legal rules. In contrast, EP in France remains weakly developed despite the appropriate legislation. The existence of a legislative framework, in and of itself, is thus not a valid explanation for 'successful' co-determination. Moreover, labour legislation in the Nordic countries more or less followed existing practices, while the LRA aims at introducing a significant change in labour relations. The responses of employers and trade unions to the demands of co-determination will reflect the circumstances in which they find themselves. A fragmented labour movement, militant shop steward structures, adversarial labour relations and a lack of trust in the workplace decisively qualify the type of EP that is possible or feasible in South Africa. The current emphasis on EP in South Africa has to be understood in the context of (a) the attempts by management to increase its scope to manoeuvre on productivity issues and to secure its legitimation in a post-apartheid setting; and (b) the efforts of the labour movement, in alliance with the ANC government, to overcome the legacies of apartheid and expand its sphere of influence. So while managers are concerned with securing the consent and harnessing the cooperation of employees; trade unions are suspicious of forms 13 Gilton Klerck of participation which imply an integration of shop stewards into the management of the workplace or which establish competing lines of communication between employees and employers. Finding their prerogatives on the shop floor under increasing threat from militant shop stewards, employers are eager to shift the regulation of production relations out of the adversarial structures of collective bargaining. Hence the proliferation of quality circles, team working and the like. By involving employees with the aim of them gaining a better understanding of challenges facing the enterprise, employers hope that they will accept the 'inexorability' of managerial decisions. That is, trade union involvement in production issues requires 'an engagement with technical managerial ideas and concepts; an acceptance of certain managerially defined parameters of problems and issues' (Terry 1994:237, emphasis added). The provisions on WPFs in the LRA are geared towards carving out an arena of consensus and cooperation in which this engagement and acceptance can be insulated from the militancy of trade union structures and adversarial tactics. Workplace Forums, Co-Determination and Competitiveness There are two basic assumptions underlying the LRA's provisions on WPFs. First, if South Africa is to respond to the challenges of globalization, it must improve productivity levels. The latter, in turn, is best achieved by a more cooperative relationship between labour and management. Second, the issues that will make this increased productivity possible are unsuited to collective bargaining. While the proponents of this view recognise that conflict between management and workers cannot be completely eliminated, they emphasise that these conflictual relations mustatthe leastbe removed from the organisation of production. The Memo (1995:135-6) claims that thepurpose of WPFs is 'not to undermine collective bargaining but to supplement it'. They will achieve this by 'relieving collective bargaining of functions to which it is not well suited'. The LRA therefore envisages a' clear and strict institutional separation' between WPFs and collective bargaining. The rationale for this separation is 'to keep distributive bargaining and cooperative relations apart, so as to allow the latter an opportunity to develop'. Webster adopts the same position: The argument for a' second channel' rests on the assumption that there are limits to collective bargaining. Workplace forums are designed to perform functions that collective bargaining cannot easily achieve; for example, participation in organisational change and the restructuring of work. Put simply, collective bargaining tends to involve conflict over the economic surplus and only rarely cooperation over expansion. (1995) 14 Workplace forums, employee participation and lean production This line of reasoning contains several implicit assumptions and unargued causal links (see Lehulere 1995). First, there is the claim that issues of production and issues of distribution are fundamentally different. Second, the assumption that qualitative issues are mainly consensual and quantitative issues are predominantly conflictual. Third, the assertion that production issues can best be dealt with at the level of the individual workplace. Fourth, the claim that traditional collective bargaining structures are unsuited to regulate production-related issues. Fifth, the assumption that a' strict and clear separation' between so-called production and distributive issues canbe sustained through institutional means. Sixth, a causal relationship is postulated between participation, cooperation and increased productivity. Given their objective of reconciling the conflicting interests of capital and labour, WPFs exist and function well to the extent that employees are persuaded to contribute to efficiency in exchange for representation, and managers are persuaded to accept worker influence as a condition of cooperation. Rogers and Streeck — leading proponents of 'second channel' participation - had a profound influence on the sponsors of the LRA. The notion of 'industrial citizenship' is central to their arguments. Workers are granted quasi-constitutional rights to participate in decision-making at their place of employment that parallel therightsof citizens in political communities... Industrial citizenship of this kind can benefit democracy in society at large as well as within firms, and can improve national economic performance. (1994:104) Management must therefore become more democratic or just. But since management is primarily concerned with profits and control, 'industrial citizenship' implies the legitimation of these concerns. Given the immense disparities inpowerbetween employers and employees, the notion of industrial citizenship can only be a travesty of democracy. Labour lawyers have long recognised that: the law imported into the employment contract a set of implied terms reserving full authority of direction and control to the employer. Once the contract was defined as an employment contract, the master-servant model was brought into play... In this way, the continuing master-servant imagery lent a legal foundation to managerial prerogative... The prerogative contract gave the employer an open-ended, sovereign power. (Selznick 1969:136) We should acknowledge the limitations ofboth worker autonomy and repression within the employment relationship. When taken to their limits, neither strategy appears entirely viable, and social relations within the workplace will 15 Gilton Klerck therefore almost inevitably entail a fluctuating balance between the two. The choice is thus not between democracy versus efficiency or participation versus hierarchy, but rather the terms of their resolution. This means that precisely to the extent that unions succeed in representing the interests of their members by becoming involved in the process of management, they may simultaneously concede their ability to defend their members against management. The interior of a firm is not merely a series of exchange relations, but also consists of relations of domination and exploitation which facilitate the extraction and appropriation of surplus value. This is evident in the control-engage dilemma faced by employers: 'solving' problems of control exacerbates problems of cooperation and vice versa. The case studies conducted by Geary (1994) all revealed the difficulties of maintaining employee control together with engendering a spirit of cooperation and commitment. Despite their hype, participatory forms of management are neither inevitable nor necessarily more efficient. The question is not what is necessary, but rather what is possible in the context of a given institutional framework: 'rank-and- file disempowerment in task organization, rooted in a history of Taylorist practice, has become [sic] an obstacle to substantial restructuring of authority at the point of production' (Guarasci and Peck 1987:52). The 'success' of EP schemes therefore depends on their ability to legitimate what is in practice an asymmetrical reciprocity. This is often overlooked in the emphasis placed on the role of 'trust' in making the institutions of cooperation work: For firms to decentralise production decisions, managers must trust workers not to misuse their increased discretion. For workers to contribute to efficiency, they must trust management not to exclude them from the benefits of their effort. Generally trust is required to support cooperative exchange over longer periods, where outcomes and contingencies are not entirely predictable. (Rogers and Streeck 1994:106) Without mutual trust, participatory arrangements cannot function effectively. Management and trade union must therefore gain more faith in and commitment towards each other as a means to fostering cooperative relations. This argument begs the question: which comes first, trust or cooperation? As Marchington notes: 'team briefing has the greatest chance of success in those situation where it is least needed by management - that is, in situations where high trust and commitment already exist - and the least chance of success in those where it is needed most' (1992:68-9). With its history of adversarialism and low levels of trust, genuine cooperation between capital and labour in South Africa is bound to be of an ephemeral nature. 16 Workplace forums, employee participation and lean production There are definite limits and risks involved in the extent of autonomy that management can concede to any sector of the work force. However, employers have a general interest in all their employees displaying a certain degree of willingness to cooperate, and nowhere is this established in the absence of some degree of coercion. Management is both a technical function of coordination and a social function of control. For the voluntarists it acquires a predominantly technical role (Maree 1993, Webster 1998), while the determinists view it almost exclusively in social terms (Etkind and Harvey 1993, Lehulere 1995). Voluntarist approaches with their emphasis on positive- sum control, the creation of space for struggle, and labour's advance through participation, rest on assumptions of the vulnerability of managerial control. They also imply that EP has no inherent qualities that favour labour or capital: it is simply a tool that can be wielded by either party. Such abstracted, a priori reasoning will not get us very far. The assessment must be empirical whether participation has actually advanced large numbers of workers or, at least, offers better opportunities for such advance. Constraints on participation within the context of existing labour relations make it unlikely that (the idealistic notion of) a snowballing set of worker demands would inevitably follow. Sectionalism and polarisation with active managerial encouragement seem as likely. On the other hand, promoting the militant defence of their own sectional interests, condemns unions to forever remain outside the management process. Management's reluctance to accept accountability is therefore reinforced by the unions' aversion to responsibility (see Kelly 1988). The strengths of the voluntarist approach are therefore mirrored in the weaknesses of the determinist approach and vice versa. If co-determination is really more efficient or the assured route to incorporating labour, why do managers cling so stubbornly to their prerogatives? The truth of the matter is that, through the actions, management and trade unions clearly reveal their conflicting interests. The real limits within which EP can be sustained under capitalism are found, not in a determinist orientation that reduces management to a 'cypher' of market forces or a voluntarist conception in which anything is possible, but rather in 'the very pluralism of organisations, and the fact that alternative choices could be made' (Cressey et al 1985:171). Collective bargaining and workplace forums In his analysis of participation and industrial democracy, Clegg (1976) identified the structure of collective bargaining as a major influence on the extent and form of EP typical in a country. Where bargaining structures were most developed at workplace level (UK and the USA) much of the local need 17 Gilton Klerck for participation is satisfied through workplace union-management structures and procedures. Conversely, where collective bargaining is strongest at the industry or national level (Germany and Sweden) this leaves a vacuum of employee voice at the local level. For example: the capacity of Swedish workers to 'press wage claims was limited by centralized bargaining, while their ability to influence working conditions was constrained by the [trade union federation's] acceptance of managerial prerogatives' (Mahon 1991:303). Works councils are typical of systems in which the trade union movement is characterised by a limited presence in the workplace. Conversely, this form of representation is uncommon in countries where the trade unions are more firmly rooted at the enterprise level and where there is greater emphasis on the collective bargaining function (Biagi 1993:325). WPFs are thus in many respects incompatible with the strong shopfloor orientation of labour relations in South Africa. Centralised collective bargaining shifts the point of gravity in labour relations from the workplace to the industry or national level. This has resulted in a 'neutralisation' or 'pacification' of the union in the workplace (Sisson 1987). Most enterprise agreements tend to follow the guidelines and policies negotiated in sectoral agreements. In the spirit of Tarifpartnerschaft, for example, German unions not only assume the obligation to refrain from industrial action, but also take on the commitment to hold their members to the peace obligation during the term of a collective agreement. For the unions this amounts to'apledge to prevent autonomous actionby the enterprise workforce' (Slomp 1995:309). This neutralisation of the workplace in countries such as Germany has facilitated the development of works councils as agents of cooperation rather than confrontation. Works council members (including trade unionists) view themselves not as adversaries of management but as co- managers (Kirkwood and Mewes 1976:296, Sorge 1976:278-9, Biagi 1993:325). In countries with more adversarial and decentralised systems of collective bargaining, conflict is not displaced to the industry level. Here it is management, not the relations between unions and management, which limits EP. In France, for instance, employers primarily seek a 'mutual understanding with the relatively weak councils, in this way bypassing the unions - and marginalizing the councils as a result' (Slomp 1995:309). Internationally the role of works councils and that of trade unions is often indistinguishable. The Commission on Industrial Relations found that most works councils in European countries were 'effectively collective bargaining institutions at plant level' (cited in Jackson 1991:153). The manner in which 18 Workplace forums, employee participation and lean production the role and functions of trade unionism are viewed is central to any discussion of 'second channel' forms of participation. The provisions of the LRA are based on a restricted view: a focus on market or exchange relations encourages an analysis of union functions that is restricted to the observable and quantifiable structure of wage and occupational outcomes within the enterprise or economy (Turnbull 1988). A corollary is that union interests are narrowly defined to include only those variables that impact on the act of exchanging labour for wage payments (Hyman 1975). The power, strategic capacity and organisational resources that the trade unions can bring to bear on their interactions with management have a crucial bearing on the nature and form of EP. Marchington (1992:136-43) distinguishes four different models of participatory structures in terms of their relationship to collective bargaining: (1) an alternative to collective bargaining; (2) marginal to collective bargaining; (3) competing with collective bargaining; and (4) an adjunct to collective bargaining. The latter is the model proposed by the LRA: collective bargaining and participatory structures are kept separate (with the latter handling issues not covered by the former), but these activities are viewed as being complementary. The logic of this approach is that strong workplace organisation will prevent consultative bodies from undermining negotiating bodies, the central role of shop stewards is protected through involvement in both channels, and management is committed to and perceives real benefits from involvement in participatory arrangements. Given the great variation in levels of unionisation within and across industrial sectors, significant variation in the levels of collective bargaining and the differences in managerial styles, we are likely to witness aspects of all four models in South Africa. The third approach, for example, is already taking place in centralised bargaining forums (von Holdt 1995:60). Unionists insist that EP programmes should be based on an extension of collective bargaining - an approach which the unions already understand (Ntshangase and Solomons 1993). Such an approach, it is argued, does not depend for its success on receiving cooperation from management, the union structure remains independent of managerial structures in the workplace, and the union retains control over the shopfloor component of the programme. The commitment of the parties to the Tripartite Alliance to centralised collective bargaining and its promotion by the LRA, have raised bargaining levels to the top of many unions' agendas. Adversarial bargaining at a centralised level may have a number of consequences for the initiation and development of WPFs. 'Enterprise opportunities forparticipation, information sharing and development may be sacrificed on the altars of centralized 19 Gilton Klerck bargaining and trade union controls' (Anstey 1997:115). Conventional labour relations wisdom suggests that it is the right to strike that separates collective bargaining from consultation as a process. In terms of the LRA, however, employees may lawfully strike on matters listed for consultation. (The problem is not so much which form of employee representation is authorised to call industrial action, but rather the distinction between lawful and unlawful strikes.) These provisions, Anstey argues, effectively ends any distinction between consultation and collective bargaining. 'In effect', he continues, South African workplace forums become an extension of collective bargaining to the enterprise rather than the complementary system envisaged by the drafters of the Act, which proposed mechanisms at the level of the enterprise to deal with matters for which traditional adversarial bargaining would be ill suited. Despite the stated intentions to bring about a clear institutional separation between collective bargaining and consultation, they have been confirmed in South African labour legislation as being one and the same. (1997:117) The argument that consultation procedures in the LRA simply reflect an extension of collective bargaining in another form suggests an unwillingness to entertain cooperation in the absence of the traditional tools of adversarial leverage (Anstey 1997:118). The provisions on WPFs, according to Anstey, will obstruct rather than facilitate a change in the 'confrontational attitudes' of the parties. In practice, the distinction between voluntary representation through trade unions (collective bargaining) and legally regulatedrepresentation through co-determination (consultation) has become increasingly blurred. Anstey's account reduces collective bargaining to an infrequent process of regulating market relations (wages, etc) to the exclusion of the day-to-day regulation of managerial relations (the way in which work is managed, etc). Even in the German system ofMitbestimmung—Anstey' s benchmark - we find a distinct blurring of the boundaries of different forms of interestrepresentation. The status of councils as bargainers has even been formalised in recent years with the trend to more flexible sectoral agreements. As Streeck notes: 'in large establishment, works councils have ... become de facto union bodies' (1992:153). Studies have also shown that what was relevant to the outcome of participation was not whether it 'took place under adversarial or co-operative conditions... [but rather] that some degree of power was ceded by, or usurped from, managers and transferred to workers' (Banks 1994:102). In short, Anstey is attempting to wish away the role of shop steward structures in workplace labour relations in South Africa. 20 Workplace forums, employee participation and lean production Productivity and employee participation The existence of participatory structures such as WPFs are presented as the key to meeting the challenges of international competitiveness. A causal relationship, is set up between the existence of 'participatory structures', increased productivity, and success in international markets. Given the increasing polarisation between the rich and the poor, the nature of economic 'success' which underlies many of these accounts must not go unchallenged. Despite the sheer volume of optimistic literature, some of its most telling critiques have come from researchers studying the impact of EP schemes in detail. This work revealed not only an enduring devotion to empiricism in the optimistic account, but also an obsequious commitment to several spurious and untested assumptions: Their presumption of worker motivation, for a start, relies on a remarkably long connecting chainfromincreased variety or responsibility to increased interest, and so to greater satisfaction, whence comes an attitude of enhanced loyalty to the employer and willingness to exert effort, and so finally greater productivity, lower absenteeism and labour turnover as behavioural output. Thechainhasnumerouspotential weaklinks. (Ramsay 1985:65) Through EP in managerial decision-making, it is argued, the efficiency of the firm is enhanced by an increase in the flow of information. The latter, in turn, will encourage informed decisions and tap the skills and insights of workers.4 Hence it is not participation per se that translates into enhanced economic performance, but rather the ability of management to obtain better knowledge of the production process and thereby avoid mistakes. 'Indeed', argues Webster ('Opportunity to Break Stalemate on the Shopfloor', Business Day March 10,1995), 'co-determinationprotectsmanagementfromnarrow,short- term responses to market signals, helping them to avoid costly mistakes arising from a lack of reflection'. Clearly the success of such endeavours depend on a whole range of factors such as the time and resources management is prepared to commit, the willingness and ability of employee representatives to volunteer more information than before, product market conditions, and levels of conflict in the workplace. Management's struggle to control the labour process has always revolved around attempts to gain access to workers' knowledgeoftheproductionprocess(Braverman 1974). Muchofthe managerial mechanisms associated with lean production — such as quality circles - are geared precisely for systematic access to the knowledge accumulated on the shopfloor. 21 Gilton Klerck The claim that greater cooperation between management and labour will invariably increase productivity is simplistic to the extent of being misleading. That is: there is no simple relationship between the extent to which a change improves productivity, the extent to which it impinges upon an employee's work experience, and the extent to which the employee will require compensation for its acceptance... [The] success with which labour is managed depends less upon the relatively simple task ofmeeting employees' comparative wage aspirations, than on the endlessly demanding one of creating and maintaining the institutional forms that will maximise their willingness to work efficiently. (Brown and Nolan 1988:353) Kelley and Harrison, in an extensive overview of EP programmes, highlighted the severe methodological limitations within existing research. Their analyses reveal no statistically significant evidence that participation per se has any effect on levels of productivity, greater job security, greater power-sharing and more egalitarian systems of control over work processes (1992:248-9). In contrast to the pervasive emphasis on quantitative analysis, qualitative research has the 'ability to eschew statistical correlations between relatively remote variables in favour of the step-by step tracing of chains of causation from product market circumstances, through employer control systems, and on through trade union behaviour to wages outcomes' (Brown and Nolan 1988:349). Establishing a causal connection between EP and productivity, quality, etc, is impossible in the 'open' systems of the social world. It would require making allowances for all other associated changes in technology, work organisation, labour markets, etc. In short, EP programmes have different meanings, vary considerably across time and place, and never occur in isolation from other factors. Aggregating the variety of inputs into the production of goods or delivery of services, as well as the different products and services that are produced, has proved to be an intractable obstacle to an adequate conceptualisation of comparative productivity measures over time and space (Nolan and Marginson 1990:235). The introduction of EP may be more significant in removing obstacles to further productivity growth and enabling more efficient working practices than in providing a direct and unmediated contribution to the levels of output. A precise evaluation of the contribution of new management strategies to productivity, even in individual plants, has not been possible (Marsden and Thompson 1990:87). Unlike a contract of sale, an employment contract cannot stipulate 'the totality of relations' that arise between the parties (Turnbull 1988:110). This indeterminacy provides the formal conditions under which 22 Workplace forums, employee participation and lean production employers hire and utilise 'labour power' - the capacity to work. In order to realise the potential of labour power in a profitable manner, the employer must erect structures of control and consent to harness the cooperation, initiative and effort of the workforce. An important consequence of this open-ended character of the employment relationship is that the shaping of the labour market is itself a part of the process of cultivating labour productivity (Brown and Nolan 1988). Consensual production issues and conflictual distribution issues? Walton and McKersie (1965) drew a distinction between 'distributive' and 'integrative' bargaining in terms of the extent to which a particular bargaining topic lends itself to a mutually-beneficial agreement.5 While the former is geared towards compelling one party to give up something (eg wage negotiations), the latter is aimed at persuading both parties to take some action in order to realise a mutual gain (eg productivity agreements). The LRA is basedonthe assumption thatthere is a qualitative difference between integrative issues (changes in the organisation of work, health and safety, etc) and distributive issues (wages and conditions of employment). The arguments for second channel labour relations are predicated on this distinction. 'To avoid conflict interfering with the regulation of production, the role of labour in the production process is separated from its role in distribution of the product' (Streeckl994:88).Theformer(integrativeissues)aredealtwithinacooperative manner through WPFs and the latter (distributive issues) are subject to the power struggles of collective bargaining. This logic can just as easily be reversed. One could plausibly argue that quantitative demands have been central to collective bargaining because they do not challenge managerial prerogatives, their costs to the employer can be readily calculated, and they allow for a range of possible settlements. Unions are also more likely to prioritise issues on which concessions and agreement are most likely. By contrast, qualitative demands concern the actual conditions of work, the setting of effort levels and control of production/service delivery. Insofar as they challenge management's 'right to manage', such demands are likely to encourage bitter opposition from employers. This may lead to the dilution of qualitative demands in order to secure compromises in joint consultation. Many of these demands also have obvious cost implications. The extent to which different qualitative demands are capable of consensual resolution is contingent and should be demonstrated empirically. Experiences in South Africa and elsewhere clearly demonstrate that so- called integrative issues are just as rooted in, and imbued with, class conflict, 23 Gilton Klerck as the so-called distributive issues. The argument that production issues are essentially cooperative tends to rely on 'a highly optimistic account of the dynamics and outcomes of restructuring' (Terry 1994:244). Streeck, for example, is forced to concede that co-determination 'can be quite conflictual ... [and is] not necessarily co-operative' (1994:89). The attitudes that trade unions adopt towards different aspects of the employment relationship is an empirical question. A union may play a cooperative role in respect of certain aspects of its relations with management, while reserving the right to oppose others. The Biedenkopf Commission in Germany, for example, found that worker representatives on supervisory boards showed little inclination to want to influence the general business policies of their companies. They tended to defer to management representatives, particularly on matters such as investment and dividends. The Commission was apparently more impressed with the pacifying effects which co-determination has on organised labour than by the sense of participation which itimparted to individual workers (Bean 1985:178). A study examining the influences of trade unions in 150 cases of strategic decision-making found: (a) in those decisions that management define as strategic, union participation is low; (b) where unions do participate in the decision-making process, their influence in relation to management is low; and (c) unions almost always react to manageriaUy defined topics and very rarely initiate issues of their own (Wilson et al 1982:333). In general, therefore, unions only challenge managerial hegemony and power at the margins. The assumption that areas of cooperation and of conflict are susceptible to a single, overriding distinction is simplistic in the extreme. Even if management and the union could agree on certain objectives, it does not necessarily follow that they will also concur on how best to achieve these goals. It is therefore simply wrong to assume that production issues always lead to a 'win-win' situation. The efforts and capacities necessary to formulate policies on these matters can and has occurred at the expense of union democracy. A 'collaborative' engagement with employer policies of restructuring 'compromises the autonomy and capacity for resistance of workers' organizations and inevitably reduces their capacity to act as the militant representatives of their members' (Terry 1994:238). Moreover, where unions have succeeded in turning new production techniques into gains for their members, it has often been at the expense of 'peripheral' or 'marginal' workers. A central consequence of new management strategies has been a polarisation in the labour market between so-called core and peripheral employees. Far from being more realistic, the notion of 24 Workplace forums, employee participation and lean production a consensual strategy to increase the quality of working life now seems manifestly Utopian. Accordingly, trade union policy-makers are caught between the twin constraints of shaping an agenda which will attract, by reflecting the interests of, the changing constituencies which they seek to represent; yet which at the same time will be realistically attainable either by negotiations with cost-conscious employers or though political channels which are themselves conditioned by sensitivity to issues of national competitiveness. (Hyman 1994:126-7) A study of European unions reveals a correlation between the structures for handling production and distribution issues, and union approaches to production issues. 'Production policies require workplace-level acceptance and implementation, yet the formulation of a union response other than rejection or ad hoc bargaining demands a capacity for strategic thought which cannot emerge only from local structures' (Terry 1994:230). Proactive union policies on production issues tend to develop: (a) in the context of cooperative and consensual structures (eg works councils); and (b) where clear and strong links between local and national union structures exist. This articulation of local and centralised forms of interest representation is, however, contingent upon a chain of successive delegations which can break down at a number of points. In practice, the separation between issues subject to negotiation with trade unions and those that are subject to consultation with WPFs is seldom between so-called integrative and distributive issues. 'It is impossible in the operation of a modern workplace to separate technical problems of work organisation and management, from problems of the interests of the workers in the work organisation' (Streeck 1994:90). Hence it is not the exact line of separation between subjects that matters, but rather that they are in fact divided between negotiation and consultation. Shop steward structures and production issues Another important feature of WPFs is that there has been a conscious effort not to base them on, or in any way link them institutionally with, the traditional negotiating institutions in the workplace - the shop steward committees. The composition of a WPF will typically include both unionised and non-union employees, white-andblue-collarworkers, as well as relatively senior managers. The motivation for this institutional separation, according to the drafters of the LRA, resides in the fact that collective bargaining institutions are not suited to problem-solving because of their history of adversarialism. This approach has the potential, in certain sectors and industries, to divert the most active and militant institutions of the labour movement from central 25 Gilton Klerck aspects of industrial restructuring. Not surprisingly, many trade unionists have rejected this marginalisation of the shop stewards from production issues. The cooperationnecessary in order for WPFs to succeed may thus notbe forthcoming in the strongholds of the labour movement(eg steel and engineering industries). Case studies elsewhere have shown that the 'local unionhierarchyprovided the lines of accountability and democratic rank and file control which so disturbed the new management's vision of a trouble-free industrial relations' (Danford 1997:114). The relationship between trade union structures and WPFs will thus be decisively influenced by the prevailing balance between cooperation/ consent and conflict/coercion in the workplace. The need to overcome or at least minimise this institutional dualism emerges clearly in day-to-day shopfloor practices and in the trade union controls over WPFs contained in the LRA. Majority trade unions in South Africa possess the power to decide whether or not to trigger a WPF.6 Where workforce representation is 'a trade union affair, the councils function as an intermediary between management and trade unions ... where the unions are not able to forward such a claim, the works council performs that task' (Slomp 1995:304). The tendency is clearly that of predominance of one form over another, depending on their history and functions. The experiences of joint consultation committees in England suggest that 'shop stewards preferred negotiations to consultation, and would either boycott committees or change their character to make them indistinguishable from negotiating bodies' (Marchington 1992:129). The efficacy of works councils in expanding worker control does not compare favourably with workplace bargaining. In a comparative study of the USA and Germany, Herding found that co-determination has been least effective in providing the equivalent of US shopfloor control. This was particularly evident with regard to underdeveloped grievance procedures, and in humanising immediate job conditions relating to physical effort, heat, safety, and health conditions. In the German workplace itself, he concludes, 'management's discretion is hardly restricted' (cited in Bean 1985:175). For the trade unions, the central problems with these forums are that they create two sources of loyalty and decision-making in the workplace; possess weak or non- existent rights to bargain, mobilize and strike; and may undermine the traditional form of union organisation in the workplace (von Holdt 1995). These fears are not groundless and have several historical precedents.7 Worker involvement in production-related issues in South Africa - eg PG Bison and Volkswagen- is conducted through the shop steward structures (see 26 Workplace forums, employee participation and lean production Anstey 1990, Mailer 1992). EP schemes are generally more successful when supported by trade unions. Numerous studies indicate that participation in works councils is most effective in representing the interests of workers in those enterprises which are highly unionised (Bean 1985:173). Without strong union organisation in the workplace, 'works councils are effectively taken over by management, whatever the legal rights conferred on employees' (Terry 1994:246). A substantial body of research supports the contention that extensive EP is unlikely in a context where workers do not have independent sources of power to represent their interests (Eaton and Voos 1992, Marshall 1992, Streeck 1992). Works councils cannot function effectively without access to trade union resources. Unions, on the other hand, 'use the work council system as the institutional framework and the major source of support for their activities at the workplace andin the enterprise' (Streeck 1992:153). There is no doubtthat the unions could gain from the knowledge and experience of professional and highly skilled employees, especially inperiods of majorrestructuring (Altmann and Dull 1990:116-7). However, works council unionism is by no means assured of continuing success. It runs the risk of 'giving the impression that employees need not join unions, and engaging more with the immediate concerns of managers than of union members' (Terry 1994:247-8). Evidence of the resultant dissatisfaction - membership loss, the emergence of unofficial structures, and so on - is not hard to find. In South Africa the bulk of the layers that the LRA wants to bring into the forums on account of their skills has historically been closely tied to management. Representing the interests of these layers via a mechanism that is outside of the unions may have the effect of reinforcing management's position in the forums. A further problem with the institutional separation of shop steward structures and WPFs is that forum members nominated by non-union or staff association members will operate largely unchecked by the wishes of then- constituency. With little or no experience in democratic workplace structures and in the absence of clear lines of accountability and mandates, these employees are far more likely to pursue individual and/or sectional causes. Lehulere has convincingly argued that what is needed is a system that bases itself on the 'traditional institutions of the working class', and at the same time allows for a dynamic relationship with non-unionised workers. The advantages of this approach is that it provides an incentive for trade unions to cater for the needs of white-collar workers, expands the culture of democratic shopfloor structures and extends debates up the occupational hierarchy. In terms of such a system, he suggests that: 27 Gitton Klerck all workers will participate in elections for the forum, but only registered unions can put up candidates. Non-unionized workers will have to choose amongst the various union candidates. Unions should also be allowed to put up candidates who are not union members. (1995b:43) A weakness in the approach of the LRA is that it will ossify the relationships between the different strata of the working class as they have been inherited from the past. Factors such as skill, gender and age do not only impact on levels of unionisation, but also on levels of influence within the labour movement. Experiences in Germany, amongst others, highlight the frequent alliance of white-collar workers' representatives andmanagement as well as the increasing prominence of employee groups such as engineers which are traditionally difficult to organise (Altmann and Dull 1990:124-5, Jackson 1991:207). One of the important challenges facing the labour movement in the current period is to extend unionisation to these strata. By providing these strata with shopfloor representation outside the unions, the LRA will reinforce their indifference or opposition towards trade unionism. The mediating role of the trade union leadership can, paradoxically, be undermined when: (a) it grants too much or too little autonomy to the rank-and- file, and (b) it enjoys too much or too little of the confidence of the state and business. That is, 'labour leaders attempting to build more powerful organisations seem condemned to a precarious balancing act whose difficulty increases apace with the widening role of the state in economic affairs and the growing militancy of the rank and file' (Kirkwood andMewes 1976:298). The extent to which WPFs are a threat or a complement to the trade unions will depend on (amongst others) the nature of internal organisation and the degree of membership participation in the union concerned. The relationship between WPFs and trade union organisation is thus highly complex and will depend on a range of factors likely to vary between different workplaces. As such, it is matter for empirical investigation and not speculation. There is no iron law which condemns the forum to undermine the need for strong workplace unionism. In some workplaces this may well be the case, while in others the union may be strengthened by the presence of a forum. Managers may find WPFs to be either a new arena in which to battle with the union or a narrowly- defined and protected terrain of'cooperation'. The idea that a WPF 'will either undermine the union, or it will be so weak that it is not much use to anyone' (vonHoldt 1995:62, emphasis added), seems unduly restrictive. Conclusion Minor modifications in the structures of ownership and control in the workplace 28 Workplace forums, employee participation and lean production do not alter the broader power relations underlying a capitalist economy. Moreover, resolving a structural contradiction in one area of employment invariably reproduces it in another. Marx clearly identified a 'constantly recurring experience [whereby] capital, so soon as it finds itself subject to legal control at one point, compensates itself all the more recklessly at other points' (1967:490). Finding their prerogatives circumscribed by statutory regulation and trade union organisation, employers have shifted the burdens of economic uncertainty onto the shoulders of the most vulnerable sectors of the labour market. Witness the proliferation of temporary, part-time and fixed-term employment contracts. The all too common argument that trade unions in Germany and Japan were more'successful'during the adversities ofthe 1990s isbased on the assumption that unions should accommodate rather than contest managerial prerogatives and objectives. While legally-enforced organisational security for trade unions tends to encourage the leadership to embrace cooperative strategies (because the continuity of collective organisation is not endangered), it also tends to insulate the leadership from membership pressures. The German experience shows that 'the neutralization of direct member influence on union policy is a prerequisite for the unions to perform their functions of negotiating and dealing with employers... in a cooperative and authoritative way' (Jacobi et al 1992:234). 'Strategic' unionism, as a model for South Africa, may thus not be consistent with a strong union presence in the workplace. In the absence of strong organisation at the local level and in a context of adverse economic circumstances, 'company-level productivity coalitions can easily imply a competitive underbidding of eitherj ob protection or conditions of employment... Fragmentation into company unionism along Japanese lines represents the ultimate logic of this model' (Hyman 1994:134-5). Capitalist competition is first and foremost competition between and among workers. It boils down to competition among workers in the same enterprise, between workers of different enterprises, and between workers of different countries. There is no 'one best way' to manage labour: the struggle around the institutional forms of EP in South Africa will be carried out under different circumstances, by different subjects, and with different organisational resources. The idea of a collaborative or consensual path to greater competitiveness and social equality is not only unrealistic, but also completely divorced from the realities of workplace relations in contemporary South Africa. A more likely scenario is the uneasy coexistence of adversarialism and cooperation: ie adversarial participation and antagonistic cooperation. 29 Gilton Klerck Notes 1. A distinction is often made between employees participating directly in decision- making structures and participation that is achieved indirectly via the election of representatives. As Hyman notes, 'direct' forms of employee representation presume, usually tendentiously, that 'employees can articulate their own interests effectively without collective intermediation'. Even where forms of direct participation do not constitute a deliberate attempt to marginalise the unions, it may be the outcome as is the case with the move to team working in German manufacturing. He suggests that quality circles, team briefings and the like can be viewed as a form of 'Japanisation' informed by a primarily production logic (1997:327). 2. In terms of the LRA, four different options are possible in the establishment of WPFs:(l)abargainedWPF; (2) a statutory WPF with abargained constitution; (3) a statutory WPF constituted through the CCMA in terms of the provisions of Chapter V; and (4) a trade union-based WPF. The essential contrast is therefore between total self-determination by the employer and majority union and external imposition by the CCMA. The option that is adopted will be influenced by the extent of trade union representation and the degree of compulsion necessary in situations where the parties fail to reach agreement on the form and functions of the WPF. 3. The 'cycles of control' argument has been criticised for (a) implying that periods of upsurge are followed by periods of decay with little or no overall change in depth and scope of participation; (b) failing to explain an increase in the level of interest in participation during the 1980s when many of the pressures for it were absent; and (c) assuming that a common set of circumstances has given rise to interest in EP at different times and places, and that there is an all-embracing theory which applies in the same way in all workplaces (Blyton and Turnbull 1994:218-9). While wishing to retain the notion that interest in participation ebbs and flows, the concept of waves (as opposed to cycles) of participation aims to overcome these shortcomings: No static taxonomy can do full justice to the flux and development which characterise the history of participation... Our preferred image of participation 'waves', hopefully avoids any such determinism, leaving the whole subject open to empirical enquiry. Such waves ... may correspond to the organisation's own internal development, or to the wider material pressures and ideological opportunities available in the surrounding society during a particular period, or, as is often the case, to an interaction of the two (Ackers et al 1992:281). 4. The Memo (1995:136) argues that 'representative consultation' contributes to economic performance in 'a number of ways: (a) the flow of communication between management and the workforce is improved; (b) the quality of decisions is improved because proposals are carefully scrutinised, flaws are discovered early and the range of alternatives explored is enlarged; 30 Workplace forums, employee participation and lean production (c) the implementation of decisions is facilitated where the decision is the result of informed input from the workforce; and (d) the top level of the organisation is provided with feedback on its middle management'. 5. The difference between a 'zero-sum' and a 'positive-sum' outcome in game theory is based on a similar distinction. 6. Only six WPFs have been successfully set up since the promulgation of the new LRA in 1995 {The Star, April 19, 1999). 7. In the United States during the 1920s and 1930s, when trade unions were weak and still struggling to establish themselves, many employers introduced 'participatory' structures in the hope that employees would accept them as an alternative to trade unions. More contemporary examples are also easy to find. One of the reasons for the weakness of co-determination in the German coal and steel industries seems to be the split in labour representation on the supervisory board between the union and works council (see Jackson 1991:213-4). Joint consultation committees in certain industries iri the UK also helped to reinforce the view that unionisation has little to offer the staff (Marchington 1992:136). References Ackers, P, M Marchington, D Wilkinson and J Goodman (1992) "The use of cycles? explaining employee involvement in the 1990s', Industrial Relations Journal (23)4. Altmanns, N and K Dull (1990) 'Rationalization and participation: implementation of new technologies and problems of the works councils in the FRG', Economic and Industrial Democracy 11. Alvesson, M (1991) 'Corporate culture and corporatism at the company level: a case study', Economic and Industrial Democracy 12. Anstey, M(1990) Worker Participation: South African options and experiences. Cape Town: Juta. (1997) Employee Participation and Workplace Forums. Cape Town: Juta. Babson, S (1993) 'Lean or mean: the MIT model and lean production at Mazda', Labour Studies Journal 18. Banks, A (1994) 'Participating in management: union organising on a new terrain', South African Labour Bulletin 18(5). Baskin, J (1996) 'Unions at the crossroads: can they make the transition?', South African Labour Bulletin 20(1). Bean, R (1985) Comparative Industrial Relations: an introduction to cross-national perspectives. London: Croom Helm. Bendix, W (1995) 'Workplace corums: shadows of a shady past or beacons for reform?', Industrial Relations Journal of South Africa 15(2&3). 31 Gilton Klerck Biagi, M (1993) 'Forms of employee representational participation', inRBlanpain and C Engels(eds) Comparative Labour Law and Industrial Relations in Industrialized Market Economies. Deventer: Kluwer Publishers. Bhaskar, R (1989) The Possibility of Naturalism: a philosophical critique of the contemporary human sciences. Hemel Hempstead: Harvester Wheatsheaf. Blyton, P and P Turnbull (1994) The Dynamics of Employee Relations. London: Macmillan. Braverman, H (1974) Labor and Monopoly Capital: the degradation of work in the twentieth century. New York: Monthly Review Press. Briggs, P (1988) 'The Japanese at work: illusions of the ideal', Industrial Relations Journal 19(1). Brown, W and P Nolan (1988) 'Wages and labour productivity: the contribution of industrial relations research to an understanding of pay determination', British Journal of Industrial Relations 26(3). Clegg, H (1976) Trade Unionism under Collective Bargaining. Oxford: Blackwell. Cressey, P, J Eldridge and Maclnnes (1985) 'Just Managing': authority and democracy in industry. Milton Keynes: Open University Press. Danford, A (1997) 'The "New Industrial Relations" and class struggle in the 1990s', Capital and Class 61. Dawson, P and J Webb (1989) 'New production arrangements: the total flexible cage?', Work, Employment and Society 3. Duncan, D and B Payne (1993) 'Just in time South Africa: Japanese production systems in the South African motor industry', Capital and Class 50. Dunlop, J (1958) Industrial Relations Systems. New York: Henry Holt. Eaton, A (1990) "The extent and determinants of local union control of participative programs', Industrial and Labor Relations Review 43(5). Eaton, A and P Voos (1992) 'Unions and contemporary innovations in work organization, compensation and employee participation', in L Mishel and P Voos (eds) Unions and Economic Competitiveness. Armonk: M E Sharpe. Etkind, R and S Harvey (1993) 'The workers cease fire', South African Labour Bulletin 17(5). Fucini, J and S Fucini (1990) Workingfor the Japanese: inside Mazda's American auto plant. New York: Free Press. Geary, J (1994) 'Task participation: employees' participation enabled or constrained?', in K Sisson (ed) Personnel Management. Oxford: Blackwell. Grenfell, D and W Bendix (1994) 'Design and implementation of lean manufacturing system in two Eastern Cape companies', South African Journal of Industrial Relations 14(2&3). 32 Workplace forums, employee participation and lean production Guarasci, R and G Peck (1987) 'Beyond the syndicalism of workplace democracy', The Insurgent Sociologist 14(2). Horwitz, F and E Franklin (1996) 'Flexible work practices come to SA\ People Dynamics (October). Hyman, R (1975) Industrial Relations: a Marxist introduction. London: Macmillan. (1994) 'Changing trade union identities and strategies' in R Hyman and A Ferner (eds) New Frontiers in European Industrial Relations. Oxford: Blackwell. -(1997) 'The future of employee representation', British Journal of Industrial Relations 35(3). Jackson, M (1991) An Introduction to Industrial Relations. London: Routledge. Jacobi, O, B Keller and Muller-Jentsch (1992) 'Germany: codetermining the future', in A Ferner and R Hyman (eds) Industrial Relations in the New Europe. Oxford: Blackwell. Kahn-Freund, O {\91iyLabour and the Law. London: Stevens. Kelley, M and B Harrison (1992) 'Unions, technology, and labor-management cooperation', in L Mishel and P Voos (eds) Unions and Economic Competitiveness. Armonk: M E Sharpe. Kelly, J (1988) Trade Unions and Socialist Politics. London: Verso. Kelly, J and C Kelly (1991) '"Them and Us": social psychology and "the new industrial relations'" British Journal of Industrial Relations 29(1). Kirkwood, T and H Mewes (1976) 'The limits of trade union power in the capitalist order: the case of West German labour's quest for co-determination', British Journal of Industrial Relations 14(3). Kjellberg, A (1992)' Sweden: can the model survive?', in A Ferner and R Hyman (eds) Industrial Relations in the New Europe. Oxford: Blackwell. Knudsen, H (1995) Employee Participation in Europe. London: Sage. Lehulere, O (1995) 'Workplace forums and the struggle for workers' Control', unpublished paper. (1995b) 'Workplace forums: co-determination and workers' struggle1, South African Labour Bulletin 19(2). Macshane, D (1993) 'Lean production or mean production?: Japanese auto plants in the US', South African Labour Bulletin 17(3). Mahon, R (1991) 'From solidaristic wages to solidaristic work: a post-fordist historic compromise for Sweden?', Economic and Industrial Democracy 12. Mailer, J (1992) Conflict and Co-operation: case studies in worker participation. Johannesburg: Ravan Press. Marchington, M (1992) Managing the Team: a guide to successful employee involvement. Oxford: Blackwell. 33 Gilton Klerck Maree, J (1993) 'Trade unions and corporatism in South Africa', Transformation 21. Marsden, D and M Thompson (1990) 'Flexibility agreements and their significance in the increase in productivity in British manufacturing since 1980', Work, Employment and Society 4(1). Marshall, R (1992) 'Work organization, unions, and economic performance', in L Mishel and P Voos (eds) Unions and Economic Competitiveness. Armonk: M E Sharpe. Marx, K (1967) Capital. New York: International Publishers. Mboweni, T (1995) 'Interview with Karl von Holdt', South African Labour Bulletin 19(1). Milkman, R (1997) Farewellto the Factory: auto workers in the late twentieth century. Los Angeles: University of California Press. Moody, T (1997) Workers in a Lean World: unions in the international economy. London: Verso. Nolan, P and P Marginson (1990) 'Skating on thin ice? David Metcalf on trade unions and productivity', British Journal of Industrial Relations 28(2). Ntshangase, W and A Solomons (1993) 'Adversarial participation: a union response to participatory management', South African Labour Bulletin 17(4). Osterman, P (1994) 'How common is workplace transformation and who adopts it?', Industrial and Labor Relations Review 47(2). Pateman, C (1970) Participation and Democratic Theory. Cambridge: Cambridge University Press. Ramsay, H (1985) 'What is participation for? A critical evaluation of "labour process" analyses of job reform', in D Knights, H Willmott and D Collinson (eds) Job Redesign: critical perspectives on the labour process. Aldershot: Gower. Rogers, J and W Streeck (1994) 'Workplace, representation overseas: the works council story' in R Freeman (ed) Working under Different Rules. New York: Russel Sage. Selznick, P (1969) Law, Society and Industrial Justice. New York: Wiley. Sisson, K (1987) The Management of Collective Bargaining: an international comparison. Oxford: Basil Blackwell. Slomp, H (1995) 'National variations in worker participation', in A Harzing and J Ruysseveldt (eds) International Human Resource Management. London: Sage. Sorge, A (1976) "The evolution of industrial democracy in the countries of the European Community', British Journal of Industrial Relations 14(3). Streeck, W (1992) Social Institutions and Economic Performance: studies of industrial relations in advanced capitalist economies. London: Sage. 34 Workplace forums, employee participation and lean production (1994) 'Co-determination and worker participation: what does it mean for unions?', South African Labour Bulletin 18(5). Steward, P (1997) 'Striking smarter and harder at Vauxhall: the new industrial relations of lean production?', Capital and Class 61. Terry, M (1994) 'Workplace unionism: redefining structures and objectives', in R Hyman and A Ferner (eds) New Frontiers in European Industrial Relations. Oxford: Blackwell. Turnbull, P (1988) 'The economic theory of trade union behaviour: a.critique', British Journal of Industrial Relations 26(1). Veldsman, T (1994) 'Skills flexibility as an integral part of world-class manufacturing systems', Industrial Relations Journal of South Africa 14(2&3). Von Holdt, K (1995) 'Workplace forums: undermining unions?', South African Labour Bulletin 19(6). Walton, R and R McKersie (1965) A Behavioural Theory of Labor Negotiations. New York: McGraw-HilK Webb, J (1996) 'Vocabularies of motive and the "New" management', Work, Employment and Society 10(2). Webster, E (1998) 'The politics of economic reform: trade unions and democratization in South Africa', Journal of Contemporary African Studies 16( 1). Webster, E and S Gelb (1996) 'Jobs and equity: the social democratic challenge', South African Labour Bulletin 20(3). Wilson, D, D Jones and DRoos(1982) 'The limits of trade union power in organisational decision making', British Journal of Industrial Relations 20(2). Womack, J, R Butler, D Cray, D Hickson and G Mallory (1990) The Machine that Changed the World. New York: Rawson. Wood, S (1988) 'Between Fordism and flexibility? The US car industry', in R Hyman and W Streeck (eds) New Technology and Industrial Relations. Oxford: Basil Blackwell. 35