Social Citizenship, Decline of Waged Labour, and Worker Strategies – Barchiesi

Social Citizenship, the Decline of Waged Labour and Changing Worker Strategies

Franco Barchiesi

Sunday 5th September 2004 posted by dionysus

Employment and social citizenship in the constitutionalisation of wage labour

The social location of labour in post-apartheid South Africa has undergone dramatic shifts that underpin a reconfiguration of the material bases of the new democracy. This chapter will analyse changes in waged employment during the first seven years of democratic government (1994-2001), with particular regard to their relationships to workers’ access to social rights and citizenship and to evolving worker identities and responses. The aim is to establish a link between a quantitative decline of stable waged employment, on which virtually all statistical indicators concur, and what can be defined as a ‘peripheralisation’ of work in the context of life strategies, forms of income, and collective organisation. Unions’ perceptions and strategies in relation to such changes will be dealt with in greater detail in other chapters of this volume. However, the aspects here under examination are decisive in the study of democratisation and of its limits in a context of high expectations for social redress faced with a legacy of extreme socio-economic inequality. Conversely, the crucial role played by the South African labour movement in the country’s democratic transition has come under increasing strains as a result of re-insertion in a globally liberalised capitalism. This is a reflection of, and it actually reinforces, two concomitant trends: first, the emergence of new vulnerabilities in the labour market as a result of employment fragmentation and instability; second, the adverse consequences in terms of workers’ income of the erosion or enduring absence of social citizenship rights. Changes in employment and crisis in the access to social rights are therefore examined here as two converging trends in explaining the decline of wage labour in the ‘new’ South Africa The relationships between work and social citizenship have provided a central theme for sociological and political analysis of labour movements in twentieth century industrialised capitalism. The rise of mass production and the expansion of waged employment, often with the active support of the state, have indeed defined ‘social citizenship’ as a new generation of rights to expand and integrate previous liberal-democratic definitions (Marshall, 1950). This terrain, at the same time, acted as a vehicle for social compromise that opened the way for the ‘constitutionalisation’ of wage labour (Baldwin, 1990; Twine, 1994, pp. 103-04). Based on this, socialist parties and trade unions largely lost their ‘anti-systemic’ nature to become integral parts of the functioning of the institutions and apparatuses of state sovereignty. Such institutionalised compromises were able to make worker organisation and struggle functional to capitalist development and they largely took the form of ‘productivity pacts’. Here, increases in the social wage and negotiated predictability in workers’ income were exchanged for commitment on the part of organised labour to enhance production. In this way, welfare states and collective bargaining became crucial repositories of workers’ identities, while a relative stability of employment facilitated a widespread acceptance of factory discipline (Negri, 1996). The theme of ‘decommodification’ rose to prominence in these patterns of institutionalised collective identity, as a consequence of the expansion of the social wage. The concept came to define a generalised access to social rights and services independent from individual positions on the labour market and as a way to minimise market dependency. In terms of Esping-Andersen’s influential definition:

De-commodification occurs when a service is rendered as a matter of right, and when a person can maintain a livelihood without reliance on the market. The mere presence of social assistance or insurance may not necessarily bring about significant de-commodification if they do not substantially emancipate individuals from market dependence (..).. There is no doubt that de-commodification has been a hugely contested issue in welfare state development. For labour, it has always been a priority (..).. Decommodification strengthens the worker and weakens the absolute authority of employers. It is for exactly this reason that employers always opposed de-commodification (Esping-Andersen, 1990, p. 22).

Esping-Andersen’s operationalisation of decommodification is undoubtedly useful for the possibilities of comparison between diverse welfare systems it discloses. In this way, the concept of social citizenship can relativise the eurocentric bias inherited from its historical trajectory. However, this notion has also been the target of various criticisms that highlight the limitations of the criteria chosen to define decommodification. In particular, the determinants of decommodification cannot be located merely in the sphere of production, as in Esping-Andersen’s prioritising of pension, sickness and unemployment benefits. Rather, the provision of welfare services in the spheres of health, housing, education and municipal infrastructures constantly reorganises and rearticulates the borders between the spheres of production and reproduction and between waged and unwaged labour. More importantly, Esping-Andersen’s state-centric definition of citizenship carries serious limitations as a juridical construction of roles and entitlements that obscures the importance of social movements, popular mobilisation and power relations in shifting the nature and borders of such entitlements (Turner, 1986). Danilo Zolo (1994, pp. 29-31) notices from this point of view that the notion of ‘social rights’ does not depend so much (as in the case of civil and political rights) on a bureaucratically standardised implementation of strictly codified guarantees for the citizen. It rather predominantly relies on the interplay of social forces with different or conflicting expectations that try to re-orientate the allocation of resources generated on the market through policies that usually have a limited time horizon. In this sense ‘social rights’ become the universalist concept that legitimises material demands and conflicts over ‘social services’ that are mainly to be considered as ‘conditional opportunities’ produced by the relative strength of social subjects and movements. The dimension of power relations in explaining the ability of labour to define the terrain of ‘social rights’ as its own capacity to appropriate increasing shares of the social product has influenced more recent debates on global capitalist restructuring. Theories of globalisation often point at the negative impacts of neoliberal policies of liberalisation and state downsizing in two important respects. First, these processes increase the vulnerability of labour and tend to fragment its cohesiveness and stability, at the level of both forms of employment and identity (Moody, 1997). In this regard, while increased capital mobility jeopardises historically entrenched and unionised working-class constituencies, it also fragments the labour market in a panoply of ‘atypical’, flexible occupations that are vulnerable and unprotected. These also tend to be less organised, while the importance of non-manufacturing sectors increases. Second, these changes are increasingly experienced in the form of uprootedness, atomisation and powerlessness in working-class biographies where wage labour ceases to be a repository of collective identity, social rights, solidarity and emancipatory projects (Bauman, 1998a; Standing, 1999a). Ramesh Mishra (1999) argues that the globalisation of the market economy means that growth is no longer conducive to increasing employment, wages and job security. Conversely, the proliferation of vulnerable occupations, unemployment and working-class poverty leads to a growing diversification of social needs that increases the range of social benefits claimants while shrinking the fiscal capacity of the system. This provides arguments to advocates of further downsizing and leads to the welfare state’s retreat into means testing or coercive forms of ‘workfare’ (Peck, 2000). The decline in social security and new forms of ‘poverty trap’ are due in these scenarios to the fact that workers who enter low-wage jobs lose entitlements to welfare, while their wage is inadequate to provide decent living standards. The expansion of casual and atypical employment generally emphasises a growing gap between wage and household needs, especially in the case of low-skill employees (Heady, 1997; Rosenberg and Lapidus, 1999), an outcome facilitated by liberal deregulation of labour markets (Castel, 1995; Huber and Stephens, 2001). As a consequence, for an increasing proportion of waged employees, the wage is more and more inadequate to cover basic household expenditures. This inadequacy can be captured by a conceptual distinction between ‘wage’ and ‘income’. The former refers to a transaction as a monetary remuneration for productive performances. The latter includes also transfers of a non-commodity, and often non-monetary nature, such as social services and grants, employers’ contributions, and so on. It consequently encompasses those aspects of the household’s subsistence that are not subject to commodity exchange but depend on mechanisms of societal redistribution. Therefore, the difference between wage and income depends on the extent of decommodified components in workers’ livelihood. A corollary to this definition of ‘income’ refers to the ability of decommodified provisions to satisfy social needs. It follows that, apart from declining wage levels, working class income can be undermined by an increase in commodification in the form of individual payments for the financing of social services. This can be the consequence of a range of factors, from privatisation to ‘cost recovery’ policies adopted by the public sector. Recent analyses have however emerged to criticise views of globalisation that emphasise its dynamics of domination, which tend to represent labour as a helpless victim of impersonal economic forces while largely confining the possibilities of resistance to the defence of nation-state sovereignty. Hardt and Negri’s work, Empire (2000), advances in this regard an innovative and more convincing view. Their argument accepts that the nation-state is declining as the result of the globalisation of production regimes, financial flows and information, but they refuse to measure this decline at the level of mere numerical indicators of state intervention. They rather point out at a deeper disruption in the capacity of state sovereignty to articulate, represent and discipline social subjects. The social fragmentation produced by globalisation makes the constitutionalisation of strong social identities, as in the case of wage labour, increasingly unable to represent a growing complexity (or ‘multitude’) of social subjectivities, desires and demands. At the same time, the rise of this multitude (which as a concept is opposed both to ‘the people’ in representative democracy and to a unified wage labour in socialist imagery) is regarded not only as a by-product of an all-powerful globalised capital. Rather, the multitude is also a result of struggles against wage labour and factory discipline that, starting with the crises of the ‘golden age’ of welfare capitalism from the end of the 1960s, have counteracted the productivity-based view of citizenship proper of welfare states. The ensuing radical demands for decommodification of life could only be faced by a parallel, still largely incomplete, redefinition of the dynamics of sovereignty. These have become largely independent from the nation-state, to be embodied in a complex pattern of interaction (the ‘Empire’) of supra-national bodies, hegemonic powers, financial and information flows and global neoliberal policy-making, which tries to continuously reassert capitalist hegemony over an increasingly diversified multitude. In developing countries, neoliberal structural adjustment and public sector downsizing have also decisively affected the dynamics of social inclusion that post-colonial nation-states had tried to experiment with in their attempt at representing and modernising ‘the people’ in nation-building projects. Attempts at ‘decolonisation from above’ (Cooper, 1996) that combined the expansion of production for the market with state-driven forms of, more or less coercive, corporatism under the aegis of urbanised elites (Barchiesi, 2001) have been generally reversed by both nation-states’ crises and grassroots opposition. As Giuseppe Cocco (2001) shows in the case of Brazil, the crisis of social integration and control in countries with forms of ‘peripheral corporatism’ is deeper where earlier state-led developmentalism, protectionism and import-substitution industrialisation are displaced by a ‘post-Fordist’ dynamic of restructuring in a liberalising context. This facilitates the incorporation in positions of relative advantage of small high-tech, high-skill, ‘cognitive’ segments of the working class in the circuits of global competitiveness and in the commodity chains of large transnational corporations. At the same time, low-skill workers in uncompetitive, previously highly protected industries tend to slide towards permanent unemployment or most vulnerable jobs. As the following sections will show, this scenario largely applies to the South African case.

The ambiguous promise of wage labour in post-apartheid South Africa

The crucial role played by organised labour in the anti-apartheid struggle has been identified by influential scholarly currents as one of the main elements of discontinuity between the South African democratisation, African decolonisation processes and, more generally, dynamics of transition to post-authoritarian political systems (Mamdani, 1996). South African writers on the democratic transition have emphasised this aspect as a progressive feature of the post-apartheid scenario in a liberalised macroeconomic environment. Adler and Webster (1999) contend that labour has operated in the South African transition through a combination of radicalism and pragmatism that significantly influenced the outcomes of the process. As evidence of this they discuss the establishment of NEDLAC to negotiate proposed socio-economic legislation, the adoption in 1995 of a Labour Relations Act (LRA) that enforces unions’ organisational rights and endorses the principle of collective bargaining, and the participation of the COSATU – together with the ANC and the SACP – in the ruling Alliance. It can be argued that view tends to overestimate the institutionalisation of labour as a decisive factor for social transformation in ways that neglect the impacts of changing production configurations, shifting forms of employment and new patterns of worker subjectivity. These factors are, on the other hand, of decisive importance in understanding the interaction of economic change, state policies and workers’ responses in South Africa’s labour history. The resumption of working class militancy in the 1970s, in a context of economic crisis and impending decline of the state-sponsored inward-looking industrialisation based on the ‘mineral-energy’ complex (Fine and Rustomjee, 1996), heightened a crucial contradiction of the apartheid regime. This referred to the state’s perceived need to gain the allegiance of a stabilised and relatively skilled African working class and to the impossibility of achieving that while an illegitimate unfree labour system presided over by a racist state form was in place. At the same time, the characteristics of the black working class composition (where migrants in contract employment played a decisive role), and their associated vulnerability, did not hamper its activism and politicisation (Friedman, 1987; Baskin, 1991). In the process, and regardless of its unstable and unrewarding nature, wage labour became in proletarian cultural imagery both the measure of subjective self-worth and the basis for legitimate claims to resources and rights in a future democratic South Africa. In other words, the experience of wage labour allowed a generation of working class activists to direct power and solidarity gained inside the workplace towards a broad vision of social emancipation. Trade unions’ involvement in popular struggles during the 1980s reflected these developments in particularly dramatic ways. Township struggles during that decade were developed around issues of social services, quality of life, political citizenship, and local government democratisation. These were conducted by a plurality of subjects (workers, students, women, churches, professionals, middle classes) with diverse, often contradictory political agendas that coalesced around national liberation themes that were subsequently hegemonised by ANC-aligned organisations. Such struggles required political choices from a union movement that up to that point had privileged workplace-based issues while maintaining a suspicious attitude towards national liberation politics. The involvement of labour organisations in that phase of the struggle was often the result of grassroots pressures that challenged the conservative preference of production-based conflicts on the part of the leadership (Baskin, 1982; Ruiters, 1995). This challenge was nurtured in shop stewards’ participation in extra-workplace practices of struggle, as in the case of boycotts of fees for municipal services, public transportation and municipal housing rents. The unions’ grassroots involvement in broader political opposition was therefore determined by a perceived need to bring the targets of the struggle beyond the terrain of the wage relation. This points at an important consequence. The struggle of the South African black proletariat expressed deep-seated desires for radical decommodification and liberation from wage labour that co-existed with the demand for social and political recognition deriving from being part of the industrialisation process. Trade unions’ participation in popular struggles for social services, boycotts and mobilisation against segregated municipal infrastructures during the 1980s (but also COSATU’s campaign against the introduction of the Value Added Tax in 1991) reflected an increasing unwillingness to confine demands for social emancipation to advances at the point of production. The control of workers’ militancy and demands became, therefore, a problem whose solution was required by the globalising agenda of a post-apartheid South African government that, in the absence and unfeasibility of an option of repressive containment of labour, put in place a system of tripartite institutionalised bargaining. In this context of uneasy coexistence of potentially contradictory imperatives (liberalisation and institutional recognition of a militant labour movement), analyses, like the ones noted above, focused predominantly on labour’s institutional role and exhibit a crucial weakness. They fail to recognise that policy-making and collective bargaining bodies, as in the case of NEDLAC, are not mere instruments whose successful use is largely a variable of the organised strength of role-players (in this case capital and labour). Rather, institutions of interest intermediation can act as conduits for the reproduction of ideologies and practices of what Foucault defined ‘governmentality’ where unions are called, in exchange for their own institutional recognition, to exercise restraint and moderation over grassroots constituencies (Götz, 2000). From this point of view, the introduction in 1996 of the GEAR strategy, discussed elsewhere in the volume, as a ‘non-negotiable’ strategy is part of a broader redefinition of the boundaries of contestation inside the priority of global competitiveness and business confidence. The removal of these policies from organs of societal bargaining and their confinement to technocratic decision-making tend to reduce the former’s potential as vehicles for labour’s influence. Therefore, corporatist-styled bodies like NEDLAC work in two concomitant directions. Not only do they reflect the strength of organised labour, but they also attempt to constitute organised labour by defining the systemic constraints and compatibilities of its demands and narrowing the range of available options (Mezzadra and Ricciardi, 1997). What seems to be at stake here is, once again, the control and disciplining of the radical demands enabled by the wage relation, and the attempt to impose this latter against a tradition of defection and resistance (Cooper, 1996; Moulier-Boutang, 1998). From this point of view, the kind of democratic constitutionalisation of wage labour spearheaded by the ANC resolved the contradictions experienced by the racist state by finally completing the transition from an unfree labour system towards the free contractual forms proper of a fully-fledged capitalist labour market. In this context, the requirements of global competition and the policy prescriptions of the neoliberal ‘Empire’ replaced state coercion as the source of regulation of the wage relation. This was supported by liberalisation policies aimed at defusing labour radicalism and the ‘excessive’ socio-economic redistributive expectations of the black proletariat (Bond, 1999). At the same time, however, the formal transition to a ‘free’ labour market, driven by the ANC and sustained by the institutionalisation of labour’s rights and influence, opened new contradictions. In fact, this shift was identified as necessary to provide the South African economy with the global competitiveness and the level of skills that had been hampered by apartheid. However, the economic liberalisation and the opening to the world market were accompanied by restructuring processes that carried adverse consequences in terms of rising unemployment and precariousness (Nattrass, 1998). These dynamics contradict the institutional influence of organised labour, threaten the representative capacity of the unions and of bargaining institutions, and emphasise widening areas of exclusion with little or no institutional presence. At the same time, the growing number of the excluded are further hurt by a rising commodification of services and the erosion even of very limited social rights previously enjoyed. Faced with this decline in wage labour’s capacity for cohesiveness and organisation in a general context of social citizenship crisis, new forms of identity and resistance are developed, for which wage labour is becoming more and more peripheral. The following sections will explore these dynamics, both in their structural and policy determinants, and in the redefinition of worker responses in a specific case of historically powerful union constituency.

Wage labour decline and social citizenship crisis

Recent South African research has focused on the rise of ‘atypical’ or ‘non-standard’ (temporary, casual, contract, part-time) forms of employment as a trend that mirrors the opening up of the economy to international competition (Standing, Sender and Weeks, 1996; Crankshaw and Macun, 1997; Kenny and Webster, 1999). These works show that, while ‘atypical’ employment has strong continuities with past labour market structures, the expansion of the permanently waged condition is not part of the social basis of the first years of democracy in South Africa. The rate of unemployment, measured according to the ‘expanded definition’ used by Statistics South Africa has grown from 31.5 per cent in 1994 to 37.0 per cent in 2001 (39.2 per cent to 42.9 per cent in the case of the African economically active population). However, a recent survey for the Department of Labour estimates total unemployment at approximately 45 per cent of the overall economically active population and at more than 50 per cent of the African one (Torres, 2002). In 2000, full-time occupations employed only 42.6 per cent of the economically active population in Gauteng, the country’s industrial heartland, decreasing to 35.9 per cent in the mostly rural Northern Cape (Statistics South Africa, 2001; NEDLAC, 2000). Various factors contribute to this outcome. According to Bond (1999), post-apartheid South Africa has been unable to overcome the legacy of the ‘mineral-energy complex’ with its dependence on ‘upstream’ raw material extraction and beneficiation and on consumer durable sectors that catered for a limited, ‘white’ demand downstream. The development of a significant intermediate goods sector capable of absorbing an expanding African proletariat is still hampered as a result. Therefore, while cheap energy and minerals facilitate high capital intensity, the limited size of the consumer market boosts a bloated financial sector as an alternative to productive investment and as a source of shareholders’ pressures for short-term returns based on cutting labour costs. This encourages casualisation, outsourcing and layoffs. As a result, workers are exposed to cost-cutting pressures produced by a renewed international competition (see Table 7.1). Estimates from the 1996 National Census and NEDLAC’s ‘Infrastructure Delivery Report’ (2000) suggest that casual and contract employment accounts for around 18-19 per cent of the economically active population, reaching 22 per cent in Gauteng.

Table 7.1 Forms of employment by population group

Population Group Unemployed (%) Casual and Part-Time (%) Full-Time Employed (%) Total (%) % of Total Population African 47.3 20.1 32.7 100.0 76.4 Coloured 8.6 20.0 71.4 100.0 8.9 Indian 21.8 16.4 61.8 100.0 2.6 White 8.9 5.8 85.3 100.0 10.9

Source: NEDLAC (2000); South African Institute of Race Relations (2000).

Research has documented how ‘atypical’ employment is especially growing in labour-intensive sectors (mining, retail, construction, clothing, footwear) and in small and medium business enterprises, where most new jobs are now created in such forms (Kenny and Bezuidenhout, 1999; Mosoetsa, 2000). At the same time, the rise of ‘homework’ means that an increasingly vulnerable workforce, particularly female, is inserted in unprotected, unstable, and exploitative occupations (Theron, 1996). ILO researchers have shown that where government’s export promotion policies have been successful – as in the automotive industry – they have facilitated downsizing, retrenchments and increasing capital intensity in a labour market that is already considered highly flexible (Hayter, Reinecke and Torres, 1999). Finally, dynamics of casualisation reflect new forms of polarisation of the labour market. During the period under examination, employment in South Africa has undergone a process of ‘feminisation’ coupled to rising female unemployment and to the rise of female occupations in vulnerable informal self-employed activities where the number of women nearly equalled that of men by 1999. These outcomes seem to be related to a ‘push’ dynamics, where women enter the labour market – in unfavourable and unprotected conditions – largely to replace income lost by male members of the household (Casale and Posel, 2001). The seriousness of this situation is deepened by the fact that, with company restructuring and increasing capital intensity as crucial determinants, the only segment of the labour force for which a significant increase in demand has been noticed is the technical-professional. At the same time, restructuring has swollen the ranks of retrenched, low-skill, potentially long-term unemployed workers (Bhorat and Hodge, 1999). For the unions this scenario implies an increasing fragmentation whereby de-unionisation of large, ‘Fordist’ working class concentrations has to be counterbalanced by renewed organising efforts in a dispersed geography of production. While this stretches the unions’ resources, it also makes it more difficult to create, reinforce, and represent comprehensive workers’ demands. On the contrary, unions are often confronted with basic issues of survival, recognition, and little respect for minimal worker rights, which leave comparatively little time and resources for information and solidarity-building on broader issues. The data mentioned above point at the fact that relatively stable forms of waged employment are by now part of the experiential world of a decreasing share of the South African working class, probably the minority of the economically active population. The loss of centrality of waged employment in forms of income is reinforced by the high level of inequality in the access to benefits for permanent and ‘atypical’ workers, and therefore by highly skewed patterns of decommodification (Table 7.2). Much recent research has associated atypical employment to higher commodification. In particular, casualisation is generally indicated as a major cause for disparities in employer-subsidised medical and provident schemes. Studies in sectors with a high density of atypical employment, such as retail (Kenny and Webster, 1999), have shown significant wage and benefit gaps between permanent and atypical workers. In this sector, hourly wages of full-time permanent workers (90% unionised) can be 50 per cent higher than those of casual employees (37% unionised), while a R1800 monthly wage for a normal 45 hrs/week for permanent workers contrasts to the R650 for a 24 hrs/week for casual workers. The following table illustrates benefit differentials for permanent and fixed-term workers according to an influential survey.

Table 7.2 Entitlement to Enterprise Benefits According to Employment Status (1996)

Benefit Regular workers (%) Temporary workers (%) Paid vacation 96.1 16.4 Paid sick leave 96.5 15.1 Medical aid 67.8 1.4 Medical facilities on site 53.0 33.8 Subsidised housing/ housing allowance 12.9 0.5 Childcare services 2.2 0.5 Incentive bonuses 37.6 6.8 Profit share bonus 12.5 1.4 Severance pay 77.5 8.7 Transport allowance 29.3 5.9 Occupational health service 45.0 25.9 Provident fund 85.7 8.7 Paid maternity leave 61.6 2.3 Pension 62.1 6.4 Hostel benefits 5.2 0.5

Source: ILO, ‘South African Labour Flexibility Survey’, quoted in Standing (1999a, p. 226)

Such differentials indicate that ‘atypical’ employees are less reliant on services funded through employers’ contributions, which increases commodification of social security and medical expenses. On the other hand, the inadequacy of wages to address households’ subsistence needs is not only linked to casualisation. A recent ‘Poverty and Inequality Report’ commissioned by the Government indicates (May, 2000, p. 39) that wages still constitute the most important source of income for the majority (40%) of poor households, followed by state transfers (26%), and remittances (17%). The same report argues (May, 2000, p. 80) moreover that 45 per cent of self-employed workers earn less than the poverty line (76 per cent of these are African), which seems to confirm that employment shifts from large companies towards small, medium, and micro-enterprise implies a high probability of substantial decline in living standards. Moreover, the figures for self-employed workers whose income is below the poverty line are not much lower than those unemployed under the poverty line (including those that receive some form of grant and social security transfer), set at 55.4 per cent by the 1996 National Census. This indicates that unemployment in itself is only partially accountable for working class poverty. This impression is confirmed by further data. According to the union-aligned research institute NALEDI (1999a, p. 40), formal wages, rather than casual labour or state transfers, still provide the most important source of income to the poorest 40 per cent of the population (23% and 44% of households in the first and second quintile respectively), while remittances constitute a slightly more important source for the poorest 20 per cent. At the same time, 47 per cent of ‘poor households’ and 35 per cent of ‘very poor households’ have at least one economically active member in employment (in 26 per cent and 19 per cent of cases respectively all the economically active members are employed). These figures are higher or close to those for ‘poor’ and ‘very poor’ households with no economically active members (25% and 28% respectively) or where all economically active members are unemployed (28% and 37% respectively). These figures underline the existence of large areas of working class poverty in the South African society, which seems to be only partially explained by a decline in living standards due to the rise of flexible and ‘atypical’ forms of employment. More generally, they indicate an enduring, structural inability of waged employment to satisfy basic necessities and uplift households from the poverty line. This gap between what was previously defined as actual ‘wage’ and ‘income’ needs is deepened by pressures on households’ expenditures due to the commodification of public services. To this contribute ‘cost recovery’ policies implemented by municipalities that are encouraged to become more ‘self-reliant’, with consequent increases in water and electricity user fees (Bond, 2000b). These trends reinforce the impression of a gradual ‘decoupling’ or collapse of the nexus between wage and income. On the other hand, the loss of income related to the decline of waged employment and the fragmentation of the labour market is not compensated by a system of social security that remains largely residual, targeted to the poorest of the poor, and predominantly means-tested. All these factors actually concur in accentuating the already high levels of commodification of working class lives. They also witness a crisis of social citizenship that mirrors and reinforces the collapse of wage labour. Current welfare and social security expenditures are largely aimed only at areas of extreme exclusion. Nearly 70 per cent of all social grants paid by the Department of Welfare (now Department of Social Development) in 1999 went to old-age pensions (NEDLAC, 2000) allocated to extremely poor recipients with no other sources of household income. According to the 1996 Census, only eight per cent of the population (men aged 65 upwards and women aged 60 upwards) – subject to means test – were entitled to such pensions, whose maximum monthly value in 2001 was R548 (USD 52) per capita. Taking also into account the ‘child support grant’ administered by the Department of Social Development (R150 per month in 2001, for caregivers with children below 7 years and a household monthly income not exceeding R800 in urban areas and R1100 in rural areas), South African citizens aged between seven and 59 (women) or 64 (men) are totally deprived of any form of social security coverage. Other aspects of the South African social security system depend on separate government bureaucracies or are fully private in nature. The Unemployment Insurance Fund (UIF) is a contributory scheme to cover unemployment, short-term sickness, maternity and death, administered by the Department of Labour in terms of the Unemployment Insurance Act of 1966. It defines as beneficiaries only workers in regular employment and excludes independent contractors, self-employed and job seekers (Olivier 2000, xx). A Ministerial Task Team appointed in 1998 noticed that less than 10 per cent of South Africa’s unemployed were benefiting from the UIF. The inadequacy of the UIF can be traced to the fact that it is predominantly a temporary economic relief from need (which seems to inform the rationale of including maternity, sickness and death benefits), rather than a real income replacement to address unemployment. Second, the vast areas of exclusion from the UIF, especially unemployed job-seekers, do not take into account the structural inability of the South African labour market to absorb new entrants. They therefore codify unemployment or under-employment in terms of temporary accidents of life, rather than enduring, long-term forms of exclusion. Finally, the system of retirement benefits in South Africa is fully privatised. The country does not have a national public pension system, and company or industry-based schemes provide the totality of retirement coverage. The system has been greatly shaped by past unions’ struggles for social citizenship rights. During the 1970s the long-standing exclusion of low-skill (mainly black) workers from pension funds was abolished, while the 1980s witnessed a diversification of schemes into pension and provident funds. This was the result of strikes, particularly in 1981, demanding the possibility for workers to receive cash payments when changing employer (Friedman, 1987, p. 256-63). As a result, since then hourly-paid workers in many industries have opted out of ‘pension funds’ to join ‘provident funds’. Both schemes are funded by employers’ and employees’ contributions (usually 7.5 per cent of the normal wage for workers) and belonging to one of them is compulsory for eligible employees. While the former operate in the form of monthly payments after retirement, provident funds – a South African peculiarity – allow for lump sum payments at the termination of each relation of employment. In some industries, such as chemical, paper and metal-engineering, provident funds are structured in national schemes with the participation of union representatives in their boards of trustees. The growth of the provident fund industry stems from the unions’ realisation of the structural nature of unemployment and employment insecurity, which requires a flexible use of limited financial means beyond the provision of an income at retirement age. At the same time, COSATU maintains an orientation that favours the establishment of a comprehensive social security system articulated into an effective unemployment coverage and retirement benefits to be paid entirely at the end of the worker’s career (Congress of South African Trade Unions, 1999). This reflects the inherently contradictory nature of provident funds, which have come to provide (more importantly than the UIF itself) the main means through which workers are protected against a wide range of very different occurrences, including short-term unemployment risks. Such contradictions are serious challenges to the cohesiveness of trade unions and to leadership-grassroots relationships. In fact, accumulated contributions that can be paid in the form of a lump sum of even tens of thousands of Rands could be used not only to maintain the family but also to start self-employed small businesses in alternative to low-wage jobs. This prospect encourages workers to exit formal employment or accept voluntary retrenchment packages as an alternative to wages that often do not even allow the repayment of family debts or meaningful savings. In many instances workers claim the payment of benefits in this form against the advice of the unions, which determines conflicts and tensions. As a result, the distinction between unemployment insurance and retirement benefits tends de facto to disappear. At the same time, this also questions a notion of ‘employment’ as a phase of life in which the worker builds rights and entitlements to be enjoyed at retirement. Rather than being a repository of social rights, work is here reduced to the mere access to (direct or indirect) cash payments, which is up to the individual discipline of the worker as a consumer to administer. Scarce and overstretched resources such as provident funds have to cover a wide range of risks and uncertainties without any support on the part of fiscal redistributive mechanisms. The result is that, in the absence of a adequate forms of social security to face non-employment, the individual employee is called to make choices based on calculations of opportunities between keeping increasingly unrewarding and unstable jobs and accepting a monetary compensation for the expulsion from the labour market. These examples reinforce conclusions already emphasised in the earlier discussion of labour market and employment changes. In particular, they confirm a convergence of two mutually reinforcing trends – commodification and changing status of formal employment – that result in a dynamics of collapse of wage labour as a central factor for the access to rights, income and protection for an increasing proportion of the South African working class. The consequences of this process are far-reaching and the challenges they raise for organised labour are massive. As discussed earlier on, wage labour has been a powerful actor in South Africa’s democratisation because it has linked organisational ability to structure broader social struggles with a discourse of redistribution, citizenship and basic social rights (COSATU, 1996b). Having access to a wage represented, from this point of view, the moral, symbolic, and political foundation of a claim to a largely decommodified income. All this framework of expectations, which has historically nourished strong patterns of solidarity and militancy, is shattered by the social scenario of the democratic transition. Not only has the institutionalisation of organised labour been unable to respond to those expectations on the terrain of policy-making, but, more importantly, dynamics of institutionalisation of organised labour have not even managed to contain, or at least cushion, the most adverse impacts of liberalisation in terms of downgrading of conditions of work and increasing commodification of everyday life. Instead, the real possibility has emerged for the wage form to turn from a symbolic and political factor of integration in the struggle against the racist state into a factor of increasing exclusion and disintegration of society under the new democratic dispensation. The following section will outline some possible developments in terms of responses to these changes.

Responses to the collapse of wage labour: the revenge of the multitude?

The argument developed in previous sections moves from the realisation that adverse changes suffered by the South African working class in the post-apartheid transition have to be understood in terms of long-lasting processes of formation of workers’ subjectivities and desires, in relation to attempts at control and discipline on the part of the state and capital. From this point of view, the successful imposition of market discipline on the part of the ANC regime in the context of the ‘Empire’ is not a product of the imperatives of globalisation. It is rather the result of strategic choices that responded to the development of worker struggles that had led to the collapse of apartheid. It can be questioned, however, to what extent the successful disarticulation of wage labour operated in this way has actually impacted on forms of collective worker identity and solidarity. An investigation of this kind would be extremely ambitious and wide in scope. I will try to approach it by looking at the case of the metal industry in the East Rand region, one of the most unionised and historically militant worker communities in the country. Throughout the 1990s this area immediately east of Johannesburg has been affected by vast processes of industrial restructuring that led to a significant decline of large-scale manufacturing and shrank the contribution of the sector to the region’s income. This process of economic decay and deindustrialisation (Barchiesi and Kenny, 2002) has in particular been accelerated after the introduction of GEAR in 1996. As a result, approximately 80,000 manufacturing jobs were lost during the 1990s, leaving a total manufacturing employment of 150,000 in 1999. The largest metalworkers’ union in the country and one of the largest COSATU unions, the 200,000-strong NUMSA, has in the East Rand one of its historical strongholds. However, between 1996 and 2000 NUMSA has lost 26 per cent of its members in the area. Downsizing, retrenchments and the fragmentation of the manufacturing texture in smaller, un-unionised workplaces as a result of restructuring in an increasingly competitive environment are indicated by union organisers as responsible for this outcome. While these new workplaces are more dispersed and they tend to ‘stretch’ the union’s organisational capabilities, they also recruit large numbers of casual and temporary workers, notoriously difficult to unionise. Interviews conducted in three metal-engineering plants reveal that workers’ experiences are increasingly shaped by a link between employment uncertainty and the loss of rights and power. Casualisation is a decisive factor questioning established and deeply ingrained life strategies. This is accompanied by the feeling that not only the union is inadequate in this regard, but that being a union member constitutes a specific target for management’s unfettered authority:

They retrench today and they hire tomorrow. Today they retrench 20, tomorrow they hire five on contract. Since 1982 I have been retrenched and called back four times. NUMSA has tried many times to talk with them, but at the end of the day they are still retrenching people. (..). Workers are no longer coming to union meetings, maybe they are afraid of being retrenched. Now they have started firing also white supervisors, but otherwise NUMSA members are still those most likely to be retrenched (Interview, 12 September 1999).

Restructuring trends and the associated job losses are generally received by workers with a sense of uncertainty that articulates a perceived threat that contrasts recent experiences of relative employment stability. While this stability had facilitated in the past forms of collective consciousness, organisation and militancy, now powerlessness accompanies a growing sense of inadequacy of traditional militancy on workplace issues:

A: Management is reducing employees numbers, they say workers must be expandable, by which they mean you must be able to do many jobs. Employees lost are never replaced. This is one of those tricks the company uses to its employees. Maybe the next time you’ll come back only not to find me. You never know, these days anything is possible. Today I’m here, tomorrow I’m not, after 16 years…

Q: Well, they can’t fire people just like that…

A: They can’t, but they have many means you can lose your job: frustrate you, make you run around, make you feel lost, many things (Interview, 16 July 1999).

On the other hand, the existence of a ‘government of the black people’, as one worker puts it, is not enabling a re-codification in racial or nationalist terms of patterns of solidarity and expectations of social promotions. Rather, the fact that a government representing the majority of the population is in power inside a formal alliance with the union movement, while enforcing spending constraints and disciplining socio-economic expectations, contributes to workers’ disorientation and lack of direction. In fact, these developments sanction the continuity between workplace-based and community demands that had provided a fundamental contribution to the definition of black working class identities in the East Rand:

This government, we don’t trust it any more. They say they are going to make changes, but workers don’t know nothing about the changes. It’s only us who are feeling who are going to be suffering. We say we are going back. We voted for our government, and we thought things would be better, but our government is making us suffer (Interview, 12 October 1999).

These opinions are generally linked to a deeply entrenched pessimism about the union’s ability to provide comprehensive responses to these trends. In fact, while the union continues to play a role in representing workers’ grievances and in expectations to redress power imbalances in the workplace, often linked to the persistence of racially despotic management and favouritism, these expectations tend to disappear when faced with prospects of retrenchment or casualised employment. Reasons advanced for this are the legal impediments created by the LRA on unions’ action against retrenchments and, more significantly, the fact that ‘the union does not help those who don’t pay subs’, such as retrenched workers and casuals. Disillusionment with the government and sense of alienation and powerlessness with regard to union organisation can also become, however, determinants of renewed radicalism and militancy. These processes can be usefully read in the framework of Hardt and Negri’s ‘politics of the multitude’ in a crucial regard. They are, in fact, the product not so much of labour market ‘fragmentation’ facilitated by neoliberal politics. Indeed, the experience of fragmentation has always been a constant component in the lives of a South African black proletariat whose class formation has historically coincided with the experience of migrant labour and coercive short-term contract employment. However, in the case of current worker responses to social fragmentation, what is decisive is the recomposition of social identities and struggles that directly respond to the crisis faced by labour organisations. Desai (2002) provides a path-breaking account of these dynamics of redefinition of oppositional identities in his investigation of social movements against the commodification of services. This is reflected in the resumption of community activism in recent years in many of the country’s urban centres. At the heart of this process are local practices of resistance around demands for free basic services (for example water and electricity) or against the coercive enforcement of payment for use. Prominent among these practices are mobilisations against evictions and disconnections of rate defaulters. In Soweto, the largest township in the country, these measures are believed to affect 200,000 residents, or 20 per cent of the population, every year. As Desai (2002) argues, these movements are based on three main characteristics. First, they are not represented by, and do not identify themselves in, the tradition of labour movement organisations. Rather, the majority of unemployed, pensioners and welfare recipients in their ranks reflects a declining centrality of wage labour in individual and communities’ experiential field. This is mirrored by the lack of access to institutionalised and corporatist forms of interest intermediation. Second, these movements privilege a terrain of direct action in the form of reappropriation of basic needs. Their ‘anti-privatisation’ and ‘anti-eviction’ campaigns (as in the case of the Soweto Electricity Crisis Committee) express desires of decommodification of services, which are not delegated to organisations in charge of defining comprehensive societal alternatives. Third, they make a remarkable use of grassroots cognitive skills (as in the case of illegal reconnections of electricity) that replace the traditional functions of cadreship and apparatuses. The fact that they are organisationally un-representable and that they privilege an immediate terrain of reappropriation make these movements similar to what Hardt and Negri (2000, pp. 156-7) discuss as the identity of the ‘poor’ as resistant subject:

The poor is destitute, excluded, repressed, exploited – and yet living! It is the common denominator of life, the foundation of the multitude (..).. The poor is in a certain respect and eternal postmodern figure: the figure of a transversal, omnipresent, different, mobile subject (..).. Finally today, in the biopolitical regime of production and in the processes of postmodernization, the poor is a subjugated, exploited figure, but nonetheless a figure of production.

The case of new community movements in South Africa symptomatic precisely of this linkage between the condition of poor as exploited, un-representable and institutionally voiceless on one hand, and its being at the centre of a changing geography of production of another. Irredeemably expelled from a collapsed wage economy, activists in these movements use communicative and cognitive skills to arrange a life in the networks of informality, crime or, occasionally and precariously, at the margins of service activities. But these skills can also become important weapons of resistance. This renewed social movement politics stands in stark alternative to traditions of labour organising that are, on the other hand, generally suspicious of these processes (see interview with Appolis and Sikwebu, in this volume). Desai (2002) provides from this point of view a useful comparison between strikes that during 1999-2000 took place at the Volskswagen factory in Uitenhage (Eastern Cape) and at the Engen oil refinery in Wentworth (Durban). In the first case, a strike over wage and social security issues escalated into a full-scale confrontation between workers and their union, NUMSA, which refused to support the strike, and this led to the retrenchment of ‘insurgent’ shop stewards (Rachleff, 2001). In the second case, a dispute over working conditions and safety erupted between Engen and casual workers from the surrounding township that are usually employed in maintenance works during the annual ‘shutdown’. The Volkswagen strike was concluded with a defeat for the workers, following a split in the union, a decisive repressive intervention of COSATU national leadership, and NUMSA’s endorsement of a total capitulation in the ensuing agreement, signed amidst government warnings and threats of ‘disinvestment’. Conversely, the Engen strike made some, however limited, inroads in worker demands and, especially, succeeded in preventing the victimisation of strikers. While the hostility of COSATU unions to these strikes (which in both cases involved breakaway unions) was a common element, their divergent outcome is convincingly explained by Desai as due to remarkably different workers’ strategies. In the first case, in fact, the workers’ demands remained confined to a workplace terrain, making little or no effort to mobilise the surrounding communities. In the second case, residents of Wentworth were sensitised and radicalised by the action of worker and community groups that emphasised the link between the downgrading of working conditions and the general process of environmental degradation in the area due to the cost-cutting activities of the refinery. As a result, community mobilisation and material support for the strike was a decisive factor both in bringing Engen to a standstill and in ensuring continuity to the workers’ action. This example can be taken as illustrative of a trend that, although embryonic, is bringing to the surface the growing distance between traditional forms of working class organisation and identity and rising patterns of social subjectivity and struggle. The decline of waged labour as a source of structuration and representation of oppositional subjects is central in this process. From this point of view ‘fragmentation’ is not necessarily negative, as long as it implies a multiplication of sites of contestation. What all this decisively interrogates, however, is the possibility of defining new networks among these sites. If these trends are confirmed in the future, this would probably lead to the challenge of replacing views of opposition as ‘ontologically’ centred on production and wage labour. A strategic response to the collapse of wage labour will then have to be thought as an alternative to mere individual alienation and despair. The conditions for entrenching and generalising such a response are probably difficult to evaluate at this stage. In any case, the questions raised by these movements are already particularly urgent for organised labour in its current crisis.


This chapter has used a specific angle to look at the widely recognised significance of organised labour in the transition to post-apartheid South Africa. Wage labour has provided in this case a vehicle for collective identities that were at the heart of black workers’ citizenship demands and vision. In this role, wage labour has played an inherently contradictory and multifaceted role. On one hand, it provided the basis for production-based organisations. On the other hand, workers’ social demands have tended to exceed the boundaries of wage bargaining to encompass broader perspectives of social transformation and rights. To address this contradiction, the democratic state has defined a terrain of institutionalisation and constitutionalisation of labour as crucial to compromises between capital, labour, and the state that could facilitate the establishment of a citizenship project. At the same time, this project has been encroached, and ultimately severely questioned, by the vicissitudes of a different priority for the ANC in power: the establishment of a fully-fledged capitalist labour market – with its own juridical forms – as a prerequisite for insertion in globalised capitalism and for disciplining the working class around its imperatives. This process has led to a decline in the stability of wage labour (and often of wage labour, tout court) both in ‘macro’ quantitative terms and as a significant component of everyday experiences. The ensuing dynamics of social fragmentation and exclusion indicate, however, that the role of ‘globalisation’ in this regard should not be decontextualised. In fact the ANC government has been oscillating between formal corporatism and practical pro-market choices as part primarily of a range of instruments to control the demands and expectations of labour developed throughout a decades-long trajectory. The rising fragmentation of the working class has, therefore, helped the new regime in controlling workers’ expectations and moderate their demands, often at the level of mere survivalism. But this is also leading to the abdication of erstwhile intentions of institutionalising social dynamics. This ultimate contradiction in the ANC’s strategies is opening new spaces of political demands and actions by social subjects that are traditionally unorganised and unrepresented. While a reduced capability of the ANC-ruled state seems to undermine its role of containment with respect to oppositional social movements, their development strikes at the core of the government’s policy choices and of their direst consequences for the everyday life of communities. It is probably too early to evaluate the potential of these movements, but their appearance indicates nonetheless a turning point for practices of struggle that have shaped twentieth-century South Africa.

Franco Barchiesi


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: