The recruitment of a cheap and plentiful supply of semi-unskilled labour has always loomed large in the labour-intensive South African mining industry. At the heart of apartheid's migrant labour system lay mines’ monopoly on the supply of labour to the gold mines. The steep decline in gold production from the mid 1980s however meant that in South Africa's new democracy the booming platinum industry replaced gold as the state's flagship to racially transform the mining industry (Capps 2012). Despite some shifts in the platinum labour process in the early 1990s, a continuous and cheap supply of labour lay at the centre of this extraction. The Rustenburg platinum belt, for example, in March 2013 employed about 143,000 workers, a large proportion being semi-unskilled black workers (Mineral Resources 2013).
Dismantling the apartheid low-wage migrant labour system has been a state and trade union goal in democratic South Africa (Bezuidenhout and Buhlungu 2011; Manuel 2013), but how have platinum managements responded to this goal while simultaneously ensuring a plentiful supply of labour? Furthermore, how have such aims impacted on labour conditions and recruitment? Investigating the labour recruitment regime in Rustenburg provides some answers.
I have used the Rustenburg recruitment regime to assess how far the migrant labour system has been undone and with what effects. Central to all labour regimes is the need for capital to meet production requirements, and this means establishing control and securing labour's compliance whether by structural, negotiated, coercive or other means. The workplace regime is central, but recruitment systems are the critical point of departure in the South African mining sector (as elsewhere) since without a supply of cheap labour mines cannot realise the profits they envisage after paying wage and production costs. Recruitment regimes are premised on a range of socio-political forces which determine how and whether labour will make itself available. The nature of the industry, segmentation of the workforce, availability of labour, levels of unemployment, other employment options, state intervention through laws and regulations, and worker skill levels are all elements in shaping the recruitment regime. These factors begin the process of regulating power relations between labour and capital and include that recruitment systems are determined not by the mine enterprise alone but by factors in the larger society including local, national and global dynamics (Beckman and Gunilla 1999, 22).
The changing patterns of state formation in South Africa have affected platinum managerial practices including the need to devise new ways of ensuring an affordable flow of labour over which it has control. How labour recruitment takes place impacts on the entire labour regime and gives rise to many questions. Have recruitment patterns shifted and if so what has replaced them and with what implications? How have mines responded to the state's ‘transformative’ legislation? Has it been implemented, refashioned or subverted? Is labour more secure under this new order and how does it experience its new rights? Why do workers opt for certain methods of recruitment and do these mitigate or deepen their exploitation? What is the meaning of these transformational goals for communities in traditional sending areas such as Lesotho, Mozambique and rural South Africa? Have changing patterns created differentials amongst workers and do some belong in the new democracy more than others?
In formulating the concept of a recruitment regime, I draw on Burawoy's distinction between despotic and hegemonic labour regimes (Burawoy 1983). I also draw on his concept of ‘production politics', where he asserts that ‘the process of production is not confined to the labor process’ but ‘also includes political apparatuses which reproduce those relations of the labor process through the regulation of struggles’ (587, emphasis in original). This includes examining how both politics and apparatuses at the level of production differ from and relate to state politics and state apparatuses and that this means looking at the form and content of state intervention (596). Such labour regime distinctions are useful in historicising the Rustenburg platinum recruitment regime – through apartheid and post-apartheid periods – and in understanding how such regimes still find expression in the current period. This also allows for reflection on how recruitment relates to the politics of the factory floor, which in turn is impacted by recent state interventions in the labour sector. It also allows for an analysis of how mine managements have juggled overlapping recruitment regimes to ensure a steady supply of labour.
The recruitment regime is a usual lens through which to examine the fragmentation of the migrant labour system. Although I have not engaged in depth with an analysis of the labour process and work organisation underground, it is nevertheless at times important to reference underground work regimes to elucidate why certain forms of recruitment are preferred over others.1
In any discussion of labour supply in South Africa, unemployment looms large. About 30% of the country's labour force is unemployed, at 8.7 million (Stats SA 2013). Such figures place platinum mining and Rustenburg centre stage for job seekers. Rustenburg is the largest town in the North West Province and the fastest growing municipality in South Africa with the population rising from 387,096 in 2001 to 549,575 in 2012 (Stats SA 2012; Wikipedia n.d.). The town's growth is entirely due to the intensive mining of platinum group metals, largely driven by demand from the early 1990s from the environmentally conscious West for platinum in the manufacture of catalytic converters. Platinum mining in Rustenburg boomed until 2008 when the global recession hammered the auto industry in Europe, United States and China on which the industry is dependent. Platinum, unlike the fixed price of gold, is a global commodity dependent on the rand–dollar exchange rate. The industry is constrained by fixed input costs composed of 50% labour, and 5% to 10% power and water costs (Hartford 2012). The price of labour and the manipulation of its supply are the only areas of flexibility for mines (Hartford 2012). Three large companies dominate the industry: Anglo Platinum (Amplats) employing about 58,000 workers, while the workforce of Impala Platinum (Implats) is 46,500 and Lonmin about 28,000. Smaller players also exist such as Aquarius, Xtrata, Platmin and Royal Bafokeng Platinum. It is on the recruitment practices of a number of companies in Rustenburg that I base my observations, following intensive fieldwork in the area between December 2012–April 2013.2
The structure of the article is as follows. First, it outlines the features of the despotic migrant labour regime that evolved out of the mineral discoveries in the late nineteenth century, and was subsequently refined under apartheid. Second, it explores the shift to the new recruitment regimes that have emerged in the platinum sector since 1994, noting in particular the rise of outsourced contract labour alongside the development of a more permanent sub-stratum of mineworkers. Third, the article considers how this is internally differentiating the black mine labour force along lines of skill, security, pay and, ultimately, class position. The contradictory consequences for both labour organisation and the broader politics of production are the subject of the conclusion.
Despotic migrant labour regime
The South African mining industry originating in the late 1800s can be described as a despotic regime where an autocratic regulation of mine labour was achieved through the colonial whip. Driving self-sufficient farmers into labouring on the mines was achieved through undisguised coercion as peasants resisted proletarianisation. Labour's compliance was enforced long before it began working underground as authorities squeezed rural agriculturalists into wage labour through taxation and land seizures (Bundy 1989; Wolpe 1972). In a short period, a new identity of worker/peasant emerged and the oscillation of migrant labour between mine and countryside became self-reproducing as families grew dependent on cash earnings. Unlike the proletarianisation of labour in countries like England, where workers had no other means of livelihood than through selling their labour power, mining managements in South Africa, in a delicate and fluctuating balance, ensured that the peasant economy was only partially destroyed so that families could reproduce themselves whilst simultaneously being wage dependent. This allowed for the payment of low wages in a labour-intensive industry. Further, annual contracts, influx control and pass laws enforced the return of workers to their rural homesteads (First 1977, 28, 150).
As the migrant system evolved, mine managements moved to more efficient and less competitive ways of ensuring plentiful supplies of labour. This was done through the Chamber of Mines recruitment agencies, which ultimately combined into Teba (The Employment Bureau of Africa). Teba recruited across southern Africa in Mozambique, Lesotho and Swaziland as well as in South African homelands where the recruitment journey ended in the allocation of a concrete bed in a mine hostel room often accommodating 22 people (Moodie 1994). Teba was in effect managements’ frontline institution for securing labour's compliance with the mine workplace regime. Attempting to strip people of their rural identities, undermining cultures of hierarchy and respect and subjecting them to sustained humiliation during the recruitment process were all part of this despotic coercion (Sithethi 2013).
Mines were still, however, plagued with fluctuations in numbers of returning migrants but the mid 1970s was a significant moment in its stabilisation (Crush 1992). In 1974 Malawian unskilled labour on the gold mines numbered some 109,731 but their contracts were terminated by the Malawian government responding to 73 deaths following a Wenela plane crash.3 To fill the vacuum, Wenela recruited Mozambican labour in greater numbers, peaking in 1975 at 115,309 only to be followed by a rapid decline to 37,000 in 1977 when Frelimo came to power and altered immigration requirements. Faced by an increase in internal labour unrest and a drop in its main Mozambican and Malawian sources of labour, the Chamber raised wages. It also intensified Basotho and South African recruitment to some 95,731 and 239,065 (a threefold increase) respectively by 1981 (First 1983, 55–56; Moodie 1994, 195–196).
The early 1980s saw a return of Mozambican and Malawian migrants as their economies experienced the impact of reduced remittances, but in an altered form. In order to stabilise its complement of labour, the Chamber now required that mineworkers be issued with Reengagement Certificates, which incentivised experienced labour to return to the mines by guaranteeing jobs and financial rewards for meeting a specified date. Simultaneously, novice recruitment was dramatically reduced. Thus, contracted migrancy progressively mutated into a permanent labour force by the mid 1980s. In 1989, however, following accusations that Malawians were spreading HIV/AIDS and must undergo forced testing on the mines, Malawian labour was permanently withdrawn (Crush 1997). In the breach both Mozambican and Basotho numbers grew, with Mozambicans numbering some 56,000 by 1995 (Cotterell 2012).
The early 1980s saw the rise of the National Union of Mineworkers (NUM) in gold and coal and in platinum in the early 1990s, which had substantial support from both non-South African and South African migrants (Allen 2003). Rustenburg, the centre of the platinum industry, like elsewhere in South African mining, became a trade union labour regime. The NUM played a large role in disciplining mine managements, establishing rigorous grievance and disciplinary procedures which ended a regime of arbitrary dismissals. The NUM engaged in negotiating standard paid leave accompanied by social wage benefits such as sickness and retirement provisions for its predominantly permanent workforce. The mines in their turn benefited from a stable, increasingly skilled and experienced workforce (Allen 2003, 361, 435, 438). The NUM was strongly opposed to the migrant labour system and in particular homed in on low wages, mine racism and on dismantling the degrading hostel system. Thus mines were forced to abandon arbitrary despotism as workers established a measure of independence through resistance to managerial domination. Managements now had to persuade workers that their welfare was aligned with the mine and that cooperation was in their best interests. The earlier brutally coercive despotic system was replaced with a hegemonic order where worker consent was required (Burawoy 1983, 590). Mine managements, forced to pay greater attention to their workforce, opened spaces for workers to manoeuvre. Mines still engaged in coercion but this was now circumscribed by rules and procedures. Post-apartheid workers’ independence from indiscriminate management control was reinforced by the passing of a tranche of laws and regulations favourable to labour through which the state aimed to regulate production politics. The new African National Congress (ANC) government, acknowledging the power of workers, formed a strategic alliance with labour to harness its ability to deliver the worker vote. In effect a corporatist regime was ushered in where the state and business were forced to cooperate with labour to win its consent. It should be noted however that under this hegemonic regime, unlike workers in advanced welfarist countries like those of Scandinavia and Western Europe, South African workers were still strongly reliant on the mine to fulfil their social and economic needs as the state provided little by way of social insurance. Mineworkers’ independence from the control of mine managements at the point of recruitment and in the workplace was thus circumscribed, and the balance between coercion and consent tipped towards compulsion reinforced by workers’ low skill levels and high unemployment rates.
Rise of new recruitment regimes
The increase of union power and, a decade later, the rise of a new democratic state ushered in a range of workplace rules and procedures. These impacted both labour and mine managements and they responded in a number of different ways. The resultant recruitment system morphed into the complex coexistence of hegemonic and despotic regimes manifesting in the fragmentation of a formerly homogenous working class into different strata and substrata.
Recruitment geographies were transformed after 2003, evidenced by a steep decline in non-South African labour. About 15 years earlier, 60% of South African mine labour consisted of foreign nationals whilst 40% was South African. In the contemporary period, about 30% of mine labour is non-South African, and of the 70% South Africans its rural–urban origins are divided equally. In consequence, Teba mainly supplies South African labour, both migrant and local (Cotterell 2012).
The shift to a hegemonic labour regime was accompanied by new legislation in the post-apartheid era that promoted – and legitimated – the changed recruitment pattern.
First, in 2002 the Mineral and Petroleum Resources Development Act (MPRDA) was enacted, under which the Mining Charter and Social Labour Plans fall. The Mining Charter called for mines to promote employment and advance the economic welfare of South African mining communities. It also stipulates that while Social Labour Plans, which companies are obliged to submit in order to obtain mining licences, must give detailed information on the impact of mining on local communities, including numbers of jobs created. Second, the new Immigration Act (2002) has been interpreted by industry to mean a reduction in foreign labour and its replacement with local workers. Mines must obtain special exemptions to recruit non-South African workers and the Act regulates migration by charging employers a levy for every foreign migrant employed. The regulation promotes the recruitment of local labour; however, it also dictates non-discrimination towards non-South Africans entailing a gradual reduction as workers age and leave the industry. Thus the average age, for instance, of a Mozambican mineworker is 49 years (Ibid.).
As previously noted, by the mid 1990s Mozambicans were the largest non-South African migrant group on the mines, numbering some 56,000. But Mozambican numbers have dropped dramatically since 2003 to the current 28,000 – a figure which continues to decrease. Basotho, Swazi and Batswana labour is also in steep decline, now numbering 3000, 5000 and 2000 respectively (Cotterell 2012; Crush, Jeeves, and Yudelman 1991).
In 2005 the Chamber shed its recruitment wing to allow for the creation of a profit-making Black Economic Empowerment entity headed by former NUM president James Motlatsi. Teba continued to source labour for the mines, operating a database of 3 million workers, but, unlike its earlier operations, recruitment mainly led to permanent rather than contracted employment.
Teba recruits 240,000 migrants internally, with Eastern Cape workers at 35% of the total being the most numerous. However, in Rustenburg, migrants are fewer than in gold with locals numbering some 30% of the labour force (Cotterell 2012). This is in part due to some mines, particularly but not exclusively Implats, falling under the Bophuthatswana homeland in the apartheid era.4 Mines there were bound by a labour localisation agreement, and also because the Bafokeng tribal authority contested mining rights on its land resulting in an agreement with Implats to prioritise Tswana local labour (Capps 2010). In addition, owing to limited platinum mining until the early 1990s, Teba's recruitment role was historically less important. From the mid 1960s three other recruitment agencies were licensed to provide labour to mines not affiliated to the Chamber and these were mostly in platinum (ATAS was the largest) (First 1983, 23). Teba began to service the platinum belt in the late 1960s because platinum mines, unlike gold, had never combined to source labour and individual operations made their own arrangements. However, over time Teba became the main provider of labour, sourcing workers chiefly for Implats and Lonmin (and in Amplats in the last seven years) (Cotterell 2012).
Teba's recruitment role has nevertheless significantly declined. The coming of a hegemonic labour regime has meant that workers are no longer tied to a single mine or industrial sector but have the space to choose where and how they seek employment. Workers in democratic South Africa have used their freedom of movement guaranteed in the Constitution to directly seek work in booming Rustenburg. The repeal of influx control laws in 1986 and a subsequent crumbling of the Group Areas Act (repealed in 1991) had already partially lifted restrictions on African workers’ movements but it was only after 1995 that workers in large numbers took advantage of this. In addition, in the late eighties the gold industry began its steep decline. In 1980 it employed close on 800,000 workers, decreasing to just over 400,000 at the beginning of 2000 (Stats SA 2001). These experienced workers often seek work in Rustenburg, bypass Teba and go directly to the mine head. Some have contacts in Rustenburg and bribe their way into employment, as this permanent Amplats worker noted:
People give you information and say go to so and so. I paid R500 up front for a certificate to say I was qualified for the job plus R250 to be returned once I had a job … At the mine I showed my ID and certificate and they said this person is competent for the job. This happens often and the mines know about it so they always take you for training because if you get injured underground they will say this happened because you did not get proper training. It is not like before when Teba recruited and trained. (Worker A 2013)
Such recruitment has also lent itself to union corruption as some NUM officials fast-tracked their relatives into employment or received bribes for jobs (Chinguno 2013, 20). Companies had informally outsourced the recruitment of unskilled labour to the union, creating opportunities for more corruption (Hartford 2012).
Teba is still the frontline migrant recruitment institution and for many its offices are still the gateway to mine employment. The reduction in migrants, however, means that sourcing labour is now only 10% of Teba's activities and it has diversified to offer a range of migrant services. Teba Bank (now U Bank) remits workers’ earnings to the tune of R1 billion a year; it operates a message service for migrants sending some 250,000 communications annually; it provides transport for the injured and deceased and a tracing service to track miners and their families (Cotterell 2012). Teba has also been forced to diversify as it faces competition from the new recruiters: labour contractors, or labour brokers as they are better known, who provide short-term labour to mines. By the mid 1990s, brokers had made significant inroads into the Rustenburg mines and today about a third of mine labour is contracted, numbering about 38,000 workers (105,069 permanent) (Mineral Resources 2013).
On the Rustenburg platinum fields, as in the gold mines (Kenny and Bezuidenhout 1999), contracted labour presents in different forms. Mines employ their own permanent unionised labour and also use the services of non-unionised, semi-unskilled temporary labour supplied by labour brokers. Contractors providing specialised services for a limited period also exist, such as shaft sinking and rope lacing, and supply their own labour for the duration of the contract. All mines and contractors make use of permanent labour as well as short-term unskilled labour supplied by labour contractors. At times the permanent employees of specialised contractors, such as at Murray & Roberts, are unionised and so earn the same or higher wages than permanent labour employed directly by the mines. However, mines also use the services of specialised contractors in less well-paid, vulnerable sectors such as cleaning and security and these workers alongside short-term labour supplied by labour brokers are often non-unionised and deeply exploited. It is the latter group, alongside mines’ permanent labour, that predominates on the platinum belt.
Such outsourcing and use of atypical (non-permanent) labour is associated with global neoliberalism, which has gripped the labour market since the 1990s and is characterised by peripheralising core productive activities. One of outsourcing's defining features has been that it bypasses organised labour and state regulations through the short-term hire of labour for both core and non-core activities, thus assisting mines to circumvent social wage obligations. Such labour is often supplied by profit-making, non-unionised external mediators or broking companies. Labour brokers have proliferated on the gold, coal and platinum mines since the early 1990s (Kenny and Bezuidenhout 1999; Kenny and Webster 1999).
Simultaneously with the rise of outsourced labour, the new ANC state was seeking to transform mine ownership and the colonial legacy of degradation and exploitation embodied in the migrant labour system. New legislation such as the Labour Relations Act (LRA), Mine Health & Safety Act, the Immigration Act and the MPRDA were all underpinned by this transformative spirit. In the labour broker recruitment system, however, some of the contradictions of government's post-apartheid laws have become apparent.
For mine managements, labour broking meshed with the partially dismantled migrant labour system associated with despotic labour regimes. The commonality has been low wages, low reproduction costs and no obligation to provide job or social wage security. Despite the new stringent unfair-dismissal and retrenchment provisions in the LRA, management identified a gap where the LRA regarded labour broking as a legitimate activity. The LRA failed to circumscribe the period for which workers could be ‘temporarily employed', thus profoundly undermining the notion of job security.5 In addition, the permissible three-month probation period allows for the employment of short-term labour without the protective provisions of the LRA (Theron 2011). Further, the Basic Conditions of Employment Act (BCEA), which governs all workers’ conditions of employment, was often ignored or manipulated to ensure workers were not covered. The BCEA for example does not apply to workers doing under 24 hours a month so brokers are careful to implement this limit, thus freeing them from observing laws governing overtime payment, paid public holidays or limits on hours of work.
Many retrenched gold workers and inexperienced young workers from across South Africa register with brokers. Low-skilled young workers in particular are recruited by the unregistered ‘bakkie brigade’ seen on the roadside near Rustenburg mines.6 They entered very short-term work, making union recruitment impossible and received wages as low as R60 a day. Brokered workers complained, and a larger more established contractor corroborated that many smaller brokers demand a R150 registration fee as well as taking half of the worker's payment (Esterhuizen 2012).
On the high end sit the larger labour contractors which provide a total service to the mine from sourcing through to supervision and payroll management. They typically hire out supervised unskilled team labour, but also provide an individual executive and artisan recruitment service. Like Teba, they keep a database of workers and engage in both active and passive recruitment. They record skill level and experience from numerous worker enquiries as well as actively engaging with local government, tribal authorities and community activists demanding the recruitment of local labour, in order to source workers from Rustenburg's 37 informal settlements and townships. Unlike Teba, however, they mainly recruit locally, although they prefer retrenched mineworkers as they come with skills and require little or no training. Many of these workers had previously been recruited by Teba for the gold mines but now return as retrenchees to register with brokers (Ibid.; Swart 2012). To get employed they present themselves as locals (sometimes buying/forging local identity documents), thus allowing mines to comply with the Mining Charter whilst simultaneously avoiding the levy on the recruitment of foreign labour. Labour contractors usually approach Teba to supply them with scarce skills offered by rural migrants and non-South Africans as it has the deepest reach into southern Africa and is able to rapidly effect border formalities. Thus, although Teba positions itself in contrast to the exploitative labour brokers, it too benefits from these insecure, short-term arrangements.
Suppliers of labour experience a shortage of rock drill operators (RDOs), who are chiefly sourced from Lesotho, Mozambique and the Eastern Cape. RDOs occupy a shadowy space between skilled and unskilled work where their accuracy and fearless strength combine in a job that is central to the production process. Shortages have become so acute that companies poach from each other. Myths abound that migrants have a natural talent for this work.7 Some Eastern Cape operators boast of coming from a long family tradition of powerful drillers, however a Basotho driller simply commented, ‘Employers like migrants as they know how desperate they are. Most South Africans shun this kind of work. They have options which we don't have … ’ (Worker G 2013). Although 60% of Impala's workforce come from the North West Province, it still recruits 15% from the Eastern Cape and Lesotho because they are willing to do permanent RDO work. The average age of RDOs at Impala is 50 years and they will retire in 12 years’ time, ‘so this is a huge problem. When we call for locals to train as RDOs very few apply,’ commented Chris Prinsloo, senior human resources (HR) manager (Prinsloo 2012). Mines cannot rely on Teba to provide sufficient drillers and have been forced to implement their own training. Impala and Lonmin embarked on training local Tswana RDOs between the ages of 20 and 25, however at least 50% dropped out before completion.8 This category is thus set to persist for southern African migrant labour, despite the decline of non-South Africans.
Mines use brokers for various reasons but an important one is to circumvent unions. As a labour broker put it, ‘Mines don't want to go through all employment procedures, especially when they want to just subcontract sections of a mine for a short time’ (Swart 2012). Mines harness the services of brokers through the signing of a ‘recruitment’ or service-level agreement stipulating wages and conditions including length of employment, which is not available to workers. Some workers on longer projects join unions but this is mainly for protection against unfair dismissal and not to bargain wages and other benefits. A broker commented:
We have no union members and we don't bargain wages as our service-level agreements with companies have their own rules and regulations and the pay is stipulated in the contract. Some do join unions but there is no bargaining. (Esterhuizen 2012)
Brokers know that discouraging union membership is key to sustaining low wages so that they can offer competitive rates to mines whilst securing their own profits.
The contract at the point of recruitment thus once again, as with the migrant labour system pre-1980, is the whip that allows the employer to forego responsibilities for the long-term reproduction of its labour force. Workers under the migrant labour system worked an annual contract which ended with the Christmas break without any negotiation of the terms under which they would be employed. Brokered workers also labour under an unbargained, undisclosed service-level contract which deeply impacts on their job security and rates of pay. A difference lies however in the earlier migrant's ability to sustain himself and his family as a semi-proletarianised peasant. In contrast, the modern-day outsourced mineworker has little or no access to workable land, and/or the means to work it, that can generate social reproduction. Both however labour under despotic regimes driven by recruiters and the mine overseer or shift boss.
Permanent semi-unskilled labour, including many Eastern Cape migrants employed directly by the mines, however is well organised into unions. Such labour enjoys the benefits of the NUM's earlier struggles which provided job security, improved wages and other benefits. This labour aristocracy has allowed mines to present a picture of adherence to labour legislation and the negotiated contract while in fact a large proportion of their labour is hidden, unorganised or weakly organised brokered workers. Through brokered arrangements mines have divided permanent and contract workers from each other and deeply fragmented the labour force (Bezuidenhout and Buhlungu 2011). When permanent rock drillers on the platinum mines in 2012 revolted against the NUM, which, they believed, had been co-opted by management and demanded higher wages, they did not include contracted labour in their demands. Permanents view contract labour with pity but also with fear as they recognise that much brokered labour had once been permanent and after retrenchment had signed up with brokers. They also fear that if unions negotiate for permanent and contracted labour under one umbrella, this will lower their wages (Worker B 2013). Thus, unionised permanent labour is cast in a structural position that undermines its resolve to combine and fight for decent wages for all mineworkers.
Wide pay differences result between contracted and permanent labour doing the same work, at times alongside each other. One Amplats permanent RDO earned R8804 per month basic, plus an RDO allowance of R1000 which, if taken with a living out allowance (LOA) of R1737 and overtime of R1000, amounted to R11,641.9 Another permanent Implats RDO earned a basic of R6540 per month, and LOA of R1850, and an extra shift of R536 amounting to R8390. By contrast, one contracted RDO earned a basic of R3060 plus a bonus of R1080, amounting to R4140 (with net pay of R1688 after deductions including a garnishee court order deduction to pay creditors at the point of salary payout of R969). Another contracted RDO earned a flat R3920. At Implats the basic rate for a winch scraper was R4488 per month plus a bonus of R3657 and LOA of R1296, which amounted to R9441. The basic rate for a contracted winch scraper was R3230.
Incorrect job descriptions and grading amongst contracted workers also resulted in large differentials. At one company, a contracted load haul dumper (LHD) was earning R15,600 whilst a contracted colleague doing the same work earned R4750 because he was graded as a transporter rather than an LHD (Worker C 2013). Variations in shifts also meant contract labour was unable to predict monthly earnings. A belt attendant contract worker paid on a shift basis rather than on a basic rate earned R1367 for 12 shifts and another R3283 for 19 shifts. Contracted labour nearly always earned significantly lower rates, sometimes by as much as R3000–R4000, than permanents in the same job category. Sanitation assistants are all contracted and earn below all other categories underground – a R1853 basic plus a R360 underground allowance (Worker D 2013). All mineworkers do overtime, often up to 12-hour days in order to meet targets, but whereas permanents received production bonuses for meeting targets, contracted labour was frequently not remunerated on the basis of targets and either not paid or inaccurately remunerated for overtime done (Stehring 2013; Worker E 2013). All permanent staff received benefits (medical aid, pension or provident, LOA). A few contracted labour earned benefits, although lower and fewer than permanents and most received none.10
Another benefit of low unionisation in the contracted sector for mines is the ease with which labour can be discarded when the mine restructures by simply not renewing the service agreement. In the 2014 Amplats’ threat to retrench 14,000 workers, negotiations with unions reduced this to 3000. This was in part achieved by terminating contracted labour and was supported by all unions including the Association of Mineworkers and Construction Union (AMCU), which has recently replaced the NUM as the dominant black majority union on the Rustenburg platinum fields (Stehring 2013; Business Day 2013; IOL Business Report 2013).11
The service agreement must include the broker's obligation to observe well-intended safety legislation but it has resulted in unintended consequences for contract labour. Under the MPRDA a mine inspector can close a mine if it endangers workers’ safety. This is problematic for mine owners who lose production time and it occurs fairly frequently in Rustenburg mines. Thus, brokers know that breaches in safety regulations will rebound on them and result in the cancellation of the contract. In consequence they hide injuries and workers are instructed not to make compensation claims. A worker explained how he injured his leg in an underground accident, making it impossible to continue working underground. He was instructed by the company to claim the accident happened on the surface but later received a company text message saying, ‘Don't worry you don't need to change the story. We will fill the WCL [worker's compensation] forms.’ He has heard nothing further and believes he will never receive compensation (Worker D 2013). All jobs in the platinum industry are insecure, but contracted workers live in a state of precarity within precarity – and health and safety laws contribute to this.
Managements also bypass safety laws through the use of brokered labour because their unorganised status means they consent to excavate the most dangerous areas, or continue working in areas no longer mined by permanent labour. Permanent workers have noted how it is common for brokered labour to have insufficient equipment, making their work more unsafe and arduous: ‘We often lend them chain block lifts to raise heavy rock which they have to lift with their hands’ (Mabutyana 2013). Under the MPRDA, the mine is responsible for ensuring that contractors adhere to the law. But this is problematic as mines become both referee and player and so oversight is uneven. In addition, mines do not adhere to all legislation, as their neglect of the Mining Charter demonstrates (Mzukisi 2012).
Mines have also benefited from the flexible brokered recruitment because the definition of local and migrant has been blurred. Some workers who call themselves locals are in fact retrenched migrants from the gold fields. Industry claims significant local labour, in some cases overwhelmingly more than migrants, but all platinum mines use brokers and, as an Implats manager commented, ‘Contractors use many more migrants’ (Prinsloo 2012). Local workers frequently state that employers prefer migrant labour and there is some truth to this. Migrants are driven from their rural homes by poverty, underdevelopment and unemployment while locals are more likely to resist dangerous, demeaning work. However, many more locals are willing to work underground than formerly, as alternative jobs have become scarce and local labour demonstrates a dependence on formal employment, showing a reluctance to engage in trading, an area dominated by foreign nationals in Rustenburg's informal settlements.12 After the Marikana massacre, however, Lonmin is considering reducing its dependence on migrant labour and replicating the Implats model where 60% of the workforce is local. As Natascha Viljoen, an executive vice president of Lonmin Plc, explained, ‘Our vision is to create more sustainable communities where people can raise their rural families near the mines … ’ (Viljoen 2013). Besides this altruistic sentiment, the use of local labour will end the six-week production downtime on Lonmin mines when migrants return home at Christmas. It will allow for uninterrupted production, and will place less pressure on the company to arrange regular worker access to families in sending areas as well as the imperative to provide both rural and local housing and services.
Implats defines ‘local’ as anyone living within 60 kilometres of the workplace, so that commuting to work is possible (Prinsloo 2012). There are indeed many such people working on the mines but companies often refer to migrants from the North West Province as ‘local’ even if they come from Vryburg, 339 kilometres, or Taung, 410 kilometres away. These are clearly migrants but not from traditional sending areas such as the Eastern Cape or Lesotho, and employers thus circumvent the obligation to hire labour from the greater Rustenburg area when it suits them.
Skills and shifting class position
The move to a hegemonic labour regime from the mid 1980s opened up important spaces for the permanent sub-strata of the mine working class. Some amongst this group seized the opportunity to access new skills and to earn a higher income. In the process they separated from their historical roots in the despotic migrant labour system and their class position began to shift. This shift was accelerated by an ANC government which soon promulgated laws to fast-track the acquisition of skills for previously disadvantaged black people.
By the mid 1980s, mineworkers were moving into skilled positions some years before the repeal in 1988 of the last vestiges of job reservation under the Mines & Works Act (1912), which had entrenched white miners’ rights to a range of occupations (Backer 2008, 33). The combination of higher wages for now unionised unskilled workers, which made expensive skilled whites increasingly untenable, and an artisans skills’ shortage made this a necessity for mine managements. The Wiehahn Commission Report (Wiehahn 1982) significantly recommended the introduction of an Unfair Labour Practice (ULP) clause into labour legislation to pacify the white right-wing Mynwerker Unie, which could argue that a unilateral change in job description and pay constituted a ULP (De Villiers 1990). The clause however ultimately favoured the emerging non-racial unions which used it to establish rights for black workers. Again, following the 1988 job reservation repeal, Mynwerker attempted to restrict skills training to white miners by arguing that it could only be conducted in English or Afrikaans, knowing that black mineworkers were not proficient in these languages. The Chamber lobbied the government, enabling the lifting of this restriction in 1993 when white privilege finally crumbled, undermining a central tenet of the migrant labour system. Mozambicans as team leaders, who had long performed the ‘charging up’ (the process of preparing explosives in drilled holes in final preparation for the blast), were the first to enter training as blasters (rock-breaking qualification), followed by Basotho workers (De Villiers 1990, 2013). The white artisan unions responded strategically by opening their doors to all colours and merging to become the United Association of South Africa (Uasa). As a result, the racial profile of Rustenburg's approximately 30,000 skilled mineworkers has changed dramatically and with it recruitment patterns. Uasa now has a 75% black membership and, post-apartheid, the launch of the Mining Qualifications Authority (MQA) under the Skills Development Act (1998) has fast-tracked black mineworkers into managerial, technical/engineering and artisanal training. Uasa's Franz Stehring noted, ‘Previously whites were recruited from school and went to a mining school, or to Wits University … but now it is more a case of people climbing the hierarchy and training, and moving into more skilled positions’ (Stehring 2013).
The current skilled racial mix has meant recruitment diversification. Historically, skilled miners approached HR departments directly having heard of positions through a newspaper or word of mouth. The Employment Equity Act has now formalised recruitment through panels which include union representatives as observers. But, as Richard de Villiers, former HR director at Village Main Reef, observes: ‘Some HR departments are lazy and just have the shop stewards bring candidates forward, hence corruption – as shop stewards bring family, friends and get sexual favours and money’ (De Villiers 2013).
Some newly trained black miners who have climbed the hierarchy and entered the skilled category were originally recruited by Teba as unskilled workers and remain migrants. Older mineworkers, however, have not entered skilled categories. This is because of persistent low literacy levels and the failure of mine-based Adult Basic Education and Training to equip workers to Grade 12, which is now an entry requirement (Stehring 2013). In recent years, Teba and labour contractors have turned to recruiting skilled workers in managerial, artisanal and miners’ categories. This embraces skilled older and younger whites as well as young black migrants and better-educated Setswana locals who have taken advantage of the MQA or mine-based training (Cotterell 2012). It was widely observed by workers that poor education in traditional sending areas such as Pondoland has meant a growing contingent of local Rustenburg skilled recruits where better education prevails.
The collapse of the Zimbabwean economy has also meant an increase in well-trained artisanal migrants to the area. In 2006 the mines bled artisans to the better-remunerated Gautrain project, resulting in unions negotiating a ‘skills retention allowance’ and in companies negotiating a three-year contract with the Zimbabwean government to provide 600 artisans. Many of these migrants still work in Rustenburg (Stehring 2013).
At least 80% of skilled workers have permanent positions at mines or with specialised contractors to the mines. Their skilled permanent status endows them with dramatically higher salaries than that of semi-unskilled workers, in the bracket of R36,000–R48,000 or more in monthly income including a mine house and access to home loans (Ibid.). A semi-skilled labourer formerly residing in a hostel or informal settlement can thus be catapulted into an altered class position overnight. Sithethi describes this early process of assimilation into the lower ranks of the middle class:
Management who wanted to show they were progressive especially in a time of reform, tried to persuade me to leave the hostel saying you are a black umlungu [white] now … So some went to white suburbs including me in the early 1990s. White neighbours moved out so we encouraged the process of making black neighbourhoods … Church services, braais [barbecues] on the street – slowly turning suburbs, like Noord, Geelhout into townships. We called them townships … (Sithethi 2013)
This class shift is symbolically represented by two-thirds of former unskilled workers from about 2009 moving out of NUM. They believed they were inadequately serviced and they moved into Uasa. In 2007 Uasa produced a Strategic Development Plan as a guide to changing its culture and way of operating to attract such workers. It sought to provide for the ‘special needs’ of skilled workers through better pension plans, housing loans, retention bonuses, death benefits, advanced legal services and experience in the negotiation of individual contracts for managerial positions (Backer 2008, 179–180). Uasa has thus assisted in the process of individuation replacing the former collective struggles of mineworkers which characterised the NUM.
The remaining 20% of skilled non-permanent workers value the portability of their skills and contract out to the highest bidder. This group is characterised by job insecurity, and will be the first to be retrenched (as in the Amplats cuts where about 4000 Uasa members voluntarily took severance packages) (Stehring 2013), but this is experienced as a benefit rather than a threat. Early retirement provides the opportunity, especially for migrants, to return to sending areas, make home improvements and live off retrenchment packages. They await an upturn in the economy with the knowledge that their skills will be in demand and then approach Teba for re-recruitment.13
Conclusion
Recruitment systems are important in any labour regime as sufficient, appropriately skilled and priced labour is central for capital to meet production requirements and ensure a profitable enterprise. It is also the point at which mines first assert their control over the workforce and begin the process of regulating power relations. In the new democracy, the undoing of apartheid's migrant labour system has been a stated goal and evidence shows deep fractures in the migrant recruitment regime. However, a steep decline in non-South African labour and the fracturing of the ‘kraal to compound’ migrant system has not significantly disrupted mines’ procurement of plentiful labour. Several factors have converged to ensure the continuance of a cheap labour regime through the meshing of hegemonic and neo-despotic labour regimes.
The despotic migrant labour system was followed by the emergence of a post-1970s hegemonic labour regime where workers experienced a more labour-friendly environment and mine productivity benefited. Semi-unskilled retrenchees from gold (non-South African and South African) entered the platinum labour force sometimes through the back door; local mine employment has grown and young inexperienced workers have flooded the area. The emergence of skilled black labour on the platinum belt has also benefited both parties.
Mines, in accommodating the once militant NUM from the 1990s onwards, bought the consent of their permanent workforce to labour under arduous conditions in return for negotiated contracts. Yet, as Chinguno shows in this special issue, this consent, with the cooperation of NUM, was challenged over time by workers and fuelled a rock drillers’ revolt in 2012 against the NUM. This in turn led the workers to embrace a new union, AMCU, which they believed better represented their interests. Indications are that an accommodation with AMCU will also be negotiated but this remains to be seen.
Paralleling the move to a more consensual relationship with permanent labour has been the emergence of a neoliberal labour market characterised by flexible patterns and accompanied by the subversion of post-apartheid transformative laws, which has further enabled the mines to fulfil their labour needs. In a return to despotic relations, managements have evaded their responsibilities to a large sector of labour through the employment of non-unionised contracted workers. However, mines have been able to project an appearance of adhering to the spirit of transformative laws through their engagement with their unionised permanent workforce. Mines have further colluded in this mirage by obfuscating the definition of local versus migrant, with the assistance of labour brokers, which has allowed them to access short-term unskilled labour and experienced migrants which new laws oblige them to reduce. Mines no longer need to rely on a centralised Teba to deliver labour but still benefit from its migrant networks, particularly with regard to RDOs.
Over the past 15 years of the platinum boom, despite ostensibly high levels of unionisation and a partial crumbling of the migrant labour system, the power relations in the mine recruitment regime have continued to favour mines. Managements have been able to meet production targets by buying the consent of their permanent workforce through negotiated contracts and through the use of contracted labour disciplined by the spectre of unemployment. They have also manipulated transformative laws which the Department of Mineral Resources’ weak inspectorate has failed to monitor.
For workers, the fracturing of the migrant recruitment regime has however had contradictory repercussions. In the move to more hegemonic relations, workers have benefited from the post-apartheid open labour market emptied of extreme humiliation and fear. Reduced foreign migration has also eliminated substantial job competition for South Africans. The opportunity to train and move into higher grades has also favoured particularly South Africans including migrants, many of whom have retained links with rural homesteads where their increased remittances have a positive impact. Many young skilled workers with permanent appointments see in mining a career path whilst contracted miners have a safety net in the portability of their skills. However, for the many older permanent uneducated and unskilled workers, the deprivations of the migrant labour system will never be undone – no career path lies ahead. Further, many older formerly organised workers are now recruited by union-bashing labour brokers, as are young inexperienced workers, and their labour continues to disproportionately benefit the employer. In the absence of a state welfare net they are forced to consent to job insecurity and to hard, dangerous work.
Non-South African migrants have in some measure benefited from a hegemonic labour regime but have mainly been disadvantaged. Semi-unskilled foreign migrants benefit from the platinum's chaotic hiring regime and at times successfully manipulate the contracted system to conceal their non-South African identity. Yet, like unskilled South Africans, many were previously unionised but now suffer at the hands of labour brokers and mines bent on rapid capital accumulation which similarly lay at the centre of the migrant labour system of old. The new focus on local recruitment threatens jobs for both South African and non-South African migrants whilst the continued employment of migrants gives rise to local resentments. Redundancies in the gold industry, accompanied by reduced recruitment, have devastated parts of the Transkei as former workers experience a decline in status, while their communities encounter deeper poverty through the loss of remittances (Ngonini 2007). The reduction in foreign migrancy has also heavily impacted on the economies of countries like Mozambique and Lesotho (Crush 1997; Gaspar 2006). Lesotho's prime minister, for example, has called for the reintroduction of recruitment to South African mines, referring in particular to the loss of deferred remittances to its economy (Business Day 2012).
The politics of production plays out in the jostling for employment on the terrain of the former migrant labour system. Managements in different ways are manipulating the system; different unions are contesting for power; and former homogenous worker interests are fracturing as their class positions shift and they move into new strata and substrata. Former migrants, some in NUM's leadership, are climbing the skills ladder and altering their class allegiance. Meanwhile, organised workers have challenged the NUM's hold whilst older non-literate migrants continue to identify with the masculine tradition of rock drilling but have no possibility of class advancement. Contracted workers who constitute new sub-strata of the working class also have little opportunity to alter their class position, while class solidarity between South Africans and non-South Africans has been fractured through the production politics of state laws.
Changing opportunities and recruitment patterns have created large differentials amongst black workers, with many being unable to claim full citizenship in democratic South Africa.
The production politics of the mines and the state have cleaved the solidarity of labour, with the consequence that all workers experience precarity. Non-South African and contracted labour is exposed at the extreme whilst permanents fear retrenchment and entry into the world of precarious brokered labour and unemployment. High competition for jobs has also meant that labour has consented to poor wages and conditions.
The despotic coercive tactics that characterised South African mining during large parts of the twentieth century are now redundant. The overt coercion of apartheid mines in sourcing cheap labour has mutated into a less obvious form which nevertheless still mainly benefits the mines, which have won labour's consent both within the workplace and through the state's politics of production which co-exists with the rigours of the labour market.