Worker Participation and Trade Unionism: Case Studies of Workplace Democracy in South Africa¶
Judy Maller (1994). Worker Participation and Trade Unionism: Case Studies of Workplace Democracy in South Africa. Economic and Industrial Democracy 15(2). SAGE (for Department of Economic History, Uppsala University).
| Type | journal article |
| Scope | Wider apartheid literature |
| Method | survey fieldwork |
| Period studied | 1970–1993 |
| Regions | South Africa, East Rand, Uitenhage, Eastern Cape |
| Themes | unions strikes, labour markets wages, transition post apartheid, manufacturing industrial policy, business and state |
| Citation key | maller1994 |
Argument¶
Maller argues that as South Africa democratizes and deracializes, meaningful workplace democracy requires combining oppositional, independent unionism with direct avenues for worker participation. Three COSATU-linked case studies (food, retail, motor cars) show three models: permanent opposition (Jabula), subordinate partnership in a non-union firm (Cashbuild), and a combined model (Volkswagen). The first is unsustainable under economic stress and the second leaves workers without collective power. The third shows that a strong shop-steward-based union can use management-initiated joint structures to extend its influence while keeping its autonomy. She concludes that unions must also attend to the production of wealth, not only its redistribution.
Evidence and method¶
Three workplace case studies drawn from the author's earlier research (Maller 1992), based on interviews with managers, shop stewards and workers, plus a survey of the metal, engineering and motor sector (Maller and Dwolatsky 1993), a COSATU shop steward survey (Pityana and Orkin 1992), official strike statistics (Department of Manpower via Bendix 1992) and ILO evidence on union penetration by sector.
Place in the debates¶
- Adversarial vs cooperative unionism / social movement unionism and industrial democracy (see Black trade unions: growth, wage effects, Wiehahn and strategy). Maller rejects both pure boycottist opposition and management-led participation without a union. She argues for a creative synthesis of collective bargaining and codecision-making, with union autonomy retained.
- Management participation schemes: genuine democracy or union by-passing (see Black trade unions: growth, wage effects, Wiehahn and strategy). She argues that most management-initiated schemes (quality circles, suggestion systems) share information but do not extend workers' decision-making, and often undermine unions. They are only meaningful if unions are placed at the centre.
Findings¶
-
Union density in South Africa had risen to a level comparable with industrial countries, at roughly 40 percent of the private sector workforce. (p. 2)
It currently stands at approximately 40 percent of the private sector workforce.
-
Shop stewards are broadly recognized by management and democratically accountable, which gives the labour movement a strong democratic culture. (p. 3)
the vast majority have formal recognition from management (97 percent).
-
Stay-aways shifted from protest to a strategic tool in bargaining with state and business, as in the 1991 VAT stay-away. (p. 3)
In November 1991, COSATU backed up its demands for democratizing the taxation system by leading a stay-away of 3.7 million workers against the unilateral imposition of value-added tax
-
The 1981 reform deracialized industrial relations but did not end the social-movement character of unionism because political enfranchisement lagged. (p. 3)
The restructuring of industrial relations in 1981 deracialized this system,° but it did not diminish the social movement aspect of trade unionism
-
Adversarial bargaining redistributed surplus but is unsustainable in conditions of declining competitiveness, pushing management toward cooperation. (p. 6)
Adversarial collective bargaining has succeeded in ensuring a more equitable distribution of surplus through organized labour's usage of its collective power.
-
Management participation schemes are widespread but mostly bypass unions and give workers little decision-making influence. (p. 7)
In this industry, 47 percent of companies have established shop-floor communication schemes which include quality circles, green areas and other channels for workers to contribute their ideas
-
At Cashbuild, the non-union internal representation system left workers fragmented and subordinate, with committees dominated by supervisors. (p. 8)
Most members of the management committees in fact hold formal supervisory positions, thereby reinforcing their domination of the decision-making process.
-
At Volkswagen, part-time shop stewards effectively work full-time on union business and hold considerable shop-floor control, including a de facto veto. (p. 11)
Shop stewards are consulted when production managers issue instructions to the workforce. In doing so, stewards have an effective right of veto.
-
The Volkswagen model remains limited: productivity has not risen substantially and conflict still causes downtime. (p. 13)
productivity levels have not increased substantially and a lot of production down-time can still be attributed to industrial relations conflict.
-
Strikes peaked in 1987, with 1148 stoppages involving 591,421 employees. (p. 15)
1987 1148 591,421
-
Union penetration varied sharply by sector, high in leather, textiles, metals and mining and low in agriculture and domestic work. (p. 15)
Metal, motor, tyre 74 Mining 72
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Case study research at Jabula Foods (food processing, East Rand): interviews with managers, shop stewards and workers | interviews oral history | c. 1990-1992 (not stated precisely) | Quotations from managers, shop stewards and workers illustrate adversarial, low-trust shop-floor relations and permanent union opposition. |
| Cashbuild case study (building materials wholesaler): worker interviews and management statements | interviews oral history | c. 1990-1992 (not stated precisely) | Worker and management quotes show that non-union management committees give workers little real say and are dominated by supervisors. |
| Volkswagen South Africa (Uitenhage) case study: interviews, worker survey responses, and analysis of JUMEC meetings | author survey or fieldwork | c. 1990-1992 (not stated precisely) | Establishes how NUMSA shop stewards combine opposition and participation through joint committees and shop-floor communication. |
| Analysis of JUMEC (joint union/management executive committee) meetings at Volkswagen | company records or reports | not stated | Used to show the range of issues discussed in the joint forum (model changes, local content, community projects, union demands). |
| Maller and Dwolatsky (1993) survey of the South African metal, engineering and motor car sector | secondary literature compilation | c. 1992-1993 | Supplies the finding that 47 percent of firms have shop-floor communication schemes, mostly not negotiated with unions. |
| Pityana and Orkin (1992), Beyond the Factory Floor: A Survey of COSATU Shop Stewards | secondary literature compilation | c. 1991-1992 | Gives the number of COSATU shop stewards (25,000) and the survey finding that 97 percent have formal recognition from management. |
| Evidence submitted to the ILO Fact-Finding Commission on Freedom of Association (Table 1, union penetration by sector) | commission or official report | February 1991 | Table showing percentage of workforce unionised by sector, supporting claims about union strength and uneven penetration. |
| Department of Manpower strike statistics 1972-1990, quoted in Bendix (1992: 549) | official statistical series | 1972-1990 | Table 2 documents strike frequency and employees involved, supporting the claim that strike action peaked in 1987. |
| Bendix (1992) Industrial Relations in South Africa (registered employers' associations count) | secondary literature compilation | end of 1990 | Supplies the figure of 237 registered employers' associations to show fragmentation. |
| Labour Monitoring Group (1992) Survey of the November 4 and 5 Stayaway | secondary literature compilation | November 1991 | Source for the 3.7 million workers' stay-away against VAT. |
| Wiehahn Commission of Enquiry into Industrial Relations (1979-80) and Labour Relations Act of 1981 | commission or official report | 1979-1981 | Cited as the origin of the legal recognition of African workers and the 1981 deracialization of industrial relations. |
| Joffe, Maller and Webster (1993) on union density comparisons | secondary literature compilation | c. 1993 | Source for the comparison of South African unionization with Germany, Sweden, France, USA and Canada. |
New data¶
Three workplace case studies (Jabula Foods, Cashbuild, Volkswagen SA) built from the author's own interviews and fieldwork, published in greater detail in Maller (1992); a small survey of Volkswagen respondents; sample sizes not given.
Inference¶
Quantification: Descriptive tables.
Qualitative comparative case studies of three workplaces in different sectors (food, retail, motor), chosen purposively to epitomize different models of participation. Interview quotations carry the claims, supplemented by secondary survey figures and two descriptive tables (union density, strikes). Conclusions are normative and theory-driven, drawing on Burawoy's factory regimes.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Adversarial shop-floor relations at Jabula Foods put labour in permanent opposition and management in an authoritarian posture, producing a zero-sum game. (p. 245) | Jabula interviews with managers, shop stewards and workers (author fieldwork, Maller 1992). | Illustrative or anecdotal |
| Management-initiated participation schemes are widespread in manufacturing, mostly without union negotiation, and extend information-sharing rather than decision-making power. (p. 246) | Maller and Dwolatsky (1993) survey of metal, engineering and motor sector: 47 percent of firms. | Partial data |
| At non-union Cashbuild, internal committees give workers an inadequate, subordinate form of representation dominated by supervisors. (p. 247) | Worker interviews across Cashbuild stores and management statements. | Illustrative or anecdotal |
| At Volkswagen, NUMSA shop stewards exercise substantial shop-floor control and use joint committees such as JUMEC to extend influence while retaining union autonomy. (p. 250) | Interviews with managers and workers, a respondent survey (70% hold lunch-time meetings), and analysis of JUMEC meetings. | Partial data |
| The Volkswagen model remains limited: productivity has not risen substantially and conflict-related downtime persists. (p. 252) | Author's assessment from the case study; no productivity or downtime data presented. | Asserted |
| Union strength has grown substantially in South Africa (about 40 percent of the private-sector workforce, uneven by sector), with strike action peaking in 1987. (p. 254) | ILO Fact-Finding Commission evidence (Table 1); Department of Manpower strike statistics via Bendix (Table 2); Joffe et al. 1993. | Systematic data |
| Meaningful workplace democracy requires combining independent oppositional unionism with direct participation, placing unions at the centre of participation systems. (p. 253) | Synthesis drawn from the three case studies and theory; no comparative testing beyond the cases. | Asserted |
The author's caveats¶
The author states that the three cases are not statistically representative of new management strategies of participation or of union responses in South Africa. They were selected to illustrate and epitomize different trends.
Gaps in the evidence¶
The argument rests on three purposively selected firms and quotations, with little information on sample sizes, interview dates or selection of respondents, so generalisation to a 'model' of workplace democracy is untested. Firm-level data on productivity, down-time, wages and dismissals, and union and employer records (e.g. JUMEC minutes), would test the Volkswagen claims about participation's effects. Surveys of many firms and sectors with comparable measures of participation and union strength would allow wider testing; some of this existed at the time (e.g. the cited metal-sector survey) but is not exploited here.
Replicability: Data poorly documented.