In response to skyrocketing prices since the mid 1990s, the South African platinum mining industry has rapidly risen from comparatively humble beginnings to a position of increasing dominance at the heart of the post-apartheid mining economy. Between 1994 and 2009, platinum output grew by a staggering 67%, while production in the historically premier gold industry continued its long-term decline at a virtually identical rate of 63% during the same period (KIO 2010, p. 10). By 2010, over 24,000 more workers were employed in platinum mining than in gold, and platinum sales were generating higher returns than any other local mineral commodity (Chamber of Mines 2010, pp. 6–8). With global platinum demand predicted to keep rising (Genc 2008) and a domestic resource base estimated to be capable of meeting it for decades to come (Cawthorn 1999, 2010), industry commentators are now suggesting that the position of the South African platinum subsector is analogous to that of gold at its peak half a century ago (Stilwell and Minnitt 2006, p. 285). For all its significance, however, the story of platinum's meteoric rise has yet to be told from the perspective of political economy. This article takes a first step down that road by considering the critical role that the prevailing system of mineral property relations played in the industry's dominant accumulation strategy in the years leading to the current platinum boom. In keeping with Fine's (1992) contribution to a general theory of mining, it thus highlights the manner in which different, historically given configurations of landed property can either inhibit or promote the access of capital to the mineral resources embedded in the soil, and hence fundamentally condition the dynamics of productive investment and inter-firm competition in a specific mining sector. This analysis is developed in five parts.
The article first briefly introduces the South African platinum industry by locating its modern trajectory within the broader development of the apartheid-era Minerals Energy Complex. It then goes on to show how the unique geological and world-market conditions of platinum production combined under apartheid to generate a distinctive strategy of competitive accumulation, based on the control of South Africa's vast platinum reserves – by far the largest in the world. Part three investigates the uniquely favourable mineral rights regime that made such control possible, focusing on the mineral property forms of the former Bophuthatswana and Lebowa Bantustans (‘homelands’), where the bulk of these reserves were geopolitically located at this time. The fourth part establishes how the conditions of the reproduction of these strategic mineral property relations were secured during the course of South Africa's political transition (1990–1994), as the outgoing apartheid government rushed to protect the historic privileges of white monopoly capital and the Bantustans were (re)absorbed into the new, unitary dispensation. The article finally suggests that the industry's success in maintaining its proprietary control over the national platinum endowment would combine with an unprecedented surge in global platinum demand to launch it onto its new growth trajectory. Yet, this would also have the unintended consequence of turning it into the primary target of the new ANC government's minerals reform legislation, thus threatening the very conditions of its core accumulation strategy.
Platinum in the apartheid MEC
As is well known, the mining industry stands at the heart of the modern South African political economy. Fine and Rustomjee (1996) have captured its historically pivotal role in terms of what they call the Minerals Energy Complex (MEC). This concept refers both to the core set of heavy industries and institutions that have evolved in and around minerals extraction and processing, and to their interaction as a distinctive system of accumulation, whose linkages and dynamics have determined South Africa's unique industrialisation path. The critical actors in the MEC have been the six giant mining houses – or ‘group producers’ – that grew out of the ‘minerals revolution’ of the late nineteenth century, fused with Afrikaner finance capital in the 1960s, and diversified out of the minerals and energy core to spread an ever widening net of conglomerate control across the national economy through the 1970s and 1980s. Their meteoric rise was, however, no ‘miracle’ of the ‘free market’. The intense concentration and centralisation of industrial and money capital in the MEC rested on a combination of massive state support in key economic sectors, and the extreme exploitation and national oppression of the black majority: the first achieved through the creation of giant parastatals to service the energy and other needs of the mining industry, alongside favourable tariff and pricing policies; and the second through the systems of labour control, racial exclusion and geopolitical segregation that came to define ‘grand apartheid’.
The South African platinum mining industry may be conceived as a specific subsector of the MEC whose trajectory has been shaped by that which is specific to it, and by the changing dynamics and structures of the wider system of accumulation in which it is embedded. Although the commercial mining of platinum began in South Africa in the 1920s, and the odd burst of speculative activity aside, the industry itself remained small, marginal and even occasionally fallow for the first 40 years of its existence (Edwards and Silk 1987, pp. 9–24). However, platinum's fortunes dramatically improved from the 1960s as more favourable world market conditions combined with broader changes in the MEC to lay the productive foundations of the modern industry. These changes were of three closely related and mutually reinforcing orders.
First, the productive activities and holdings of the MEC began to diversify, both within and out of the mining and energy core (Fine and Rustomjee 1996, pp. 147–157). Diversification within the mining core saw platinum emerge as one of the ‘new’ raw and processed mineral products (others included vanadium, copper and ferrochrome) whose now more intensive development would gradually reduce (though by no means end) the economy's historic dependence on diamonds and gold (Fine and Rustomjee 1996, pp. 98–99). Second, the interpenetration of large-scale money capital into mining – the principal expression and form of the erosion of the inter-war disjuncture between Afrikaner political and English economic power – saw new platinum mining capitals emerge, and existing ones reorganised, in the context of accelerating mergers and acquisitions, often with direct state assistance (Fine and Rustomjee 1996, pp. 96–111). Finally, these twin processes of diversification and integration were facilitated by the evolution of a distinctive structure of corporate ownership in which numerous subsidiaries and groups were bound to the mining houses through interlocking directorships and extensive cross shareholdings (Davies et al. 1988, pp. 51–65, Fine and Rustomjee 1996, pp. 96–111).
The established and emergent platinum capitals were intimately bound up in this pattern of corporate consolidation, being little more than subsidiaries of one or more of the great mining houses, often held alongside a range of other mineral, industrial and even agricultural concerns. By the mid 1970s, the typical institutional form of a platinum capital was that of a semi-autonomous company locked in a wider, and distinctly labyrinthine, structure of conglomerate control. Three such producers dominated the platinum industry from this period onwards: Rustenburg Platinum Holdings (RPH), Impala Platinum Holdings (Implats) and Western Platinum Mines (Westplats), which, by the mid 1990s, accounted for 51.6%, 39.1% and 9.3% of national platinum production, respectively (Vermaak 1995, p. 69). RPH and Implats were in turn the subsidiaries of Johannesburg Consolidated Investments (JCI) and Gencor, themselves held by the giant Anglo American and Sanlam corporations, while for its part, Westplats was majority controlled by Lonmin, the mining arm of the London-listed Lonrho. The intensely monopolistic character of the apartheid platinum industry was not, however, simply the result of its location within the MEC. It also reflected a particularly powerful tendency towards the centralisation of capital that arose from the interrelationship between the accumulation dynamics specific to the platinum industry and the distinctive system of landed property that mediated its access to South Africa's platinum resources. We shall now establish how the geological and world-market conditions of platinum production shaped that accumulation dynamic, and then go on to investigate the historically specific system of mineral property that the industry confronted.
The geological and world-market conditions of platinum production
The starting point in the analysis of the accumulation dynamic that has historically characterised the platinum industry is the geological properties of the resource that it has sought to exploit. In any mining sector, such properties both set absolute and relative physical limits on the course of industrial development and competition, and constitute the natural basis of the surplus profits which, in the presence of landed property, may be transformed into royalty revenue and other forms of ground rent (Bina 1989, Fine 1992, Massarrat 1980). As the platinum mining industry was from its earliest days global in scope, the peculiar qualities of South Africa's mineral endowment must, moreover, be set in their world context. The concern here, then, is with international and national differences in the quality of platiniferous ore and reef as the geological drivers of uneven industrial development and differential production costs in the South African platinum industry.
Platinum is one of the family of six chemically similar elements that make up the platinum group metals (PGMs). The others are palladium, rhodium, iridium, osmium and ruthenium. Although PGM deposits are known throughout the world, commercially viable lodes are rare and intensely spatially concentrated. South Africa is estimated to hold 87% of the world's PGM reserves and, in 2009, accounted for 76% of world platinum production and 33% of palladium (Chamber of Mines 2009, p. 31). Russia was the second largest PGM producer, while the remainder was made up by mines in North America and elsewhere, which extract PGMs as a by-product of other minerals, such as nickel.
South Africa's unique PGM deposits are located in a vast geological formation, the Bushveld Igneous Complex (BIC), which also contains the world's largest chromium and vanadium reserves. Shaped like an enormous bowl, the BIC spans an area of 66,000 km2 with a western limb rising in what are now the Gauteng, North West and Limpopo Provinces and northern and eastern limbs breaking out in Limpopo and Mpumalanga (see Figure 1).
The BIC system encompasses three narrow but extensive PGM Reefs that vary in quality and location. The first and most significant is the Merensky Reef, which is found in both the western and eastern Bushveld and has been at the heart of South Africa's platinum mining industry since discovery in the 1920s. Second is the Upper Group 2 (UG2) Chromite Horizon, which runs at variable depths below the Merensky; and finally the palladium-rich Platreef, exclusively located in the BIC's northern limb near Mokopane (formerly Potgietersrus).
As this brief sketch suggests, there are significant natural variations both between the PGM deposits in South Africa and elsewhere, and within the BIC itself. Unevenness within the BIC includes differences in ore quality, PGM composition, reef structure and depth. Together these natural variations dictate different mining and extraction techniques; different applications of labour and capital; and hence the differential production costs which form the basis of surplus profits in the South African industry. This in turn has generated powerful competitive pressures to win control of superior portions of the BIC.
The prospect of surplus profits is not, however, the only incentive to monopolise superior natural conditions of production. Three distinctive factors have also combined to shape the productive organisation of the South African platinum subsector, giving it an extreme monopolistic character. The first, as noted above, is the natural rarity and intense geological concentration of the world's PGM reserves in the BIC. Second, the extraction and beneficiation of platinum requires considerable applications of capital, technological capacity and expertise.1 Yet, paradoxically, the high price of platinum commodities has historically resulted in a narrow and highly volatile range of global demand – the third factor.
The platinums are unusual metals in that their commercial applications (i.e. use values) are limited and the end-users extremely sensitive to price (Edwards and Silk 1987, p. 47). This is mainly because PGMs are utilised in sectors which produce consumer durables with relatively high income elasticities of demand. In the closing years of apartheid, the jewellery, chemical, electronics, dental and autocatalyst industries had emerged as the major PGM consumers – and the demand of each was profoundly linked to the overall health of the global economy and its regions of greatest purchasing power (Vermaak 1995, pp. 70–90). Moreover, much of this consumption is vulnerable to non-PGM substitution, efficiency drives or recycling particularly when prices become too high. The tension between the (limited) use and (high) exchange values of the PGMs has resulted in a notoriously volatile commodity market historically punctuated by sharp booms and slumps, and hence an industry characterised by extremes of productive expansion and contraction. Faced with PGM reserves whose scale could easily flood the world market, the South African producers pursued a strategy of regulating the rate of global supply from their inception (Edwards and Silk 1987, p. 11).
As elsewhere in the minerals sector, the South African producers attempted to control output, costs and competition through intensive horizontal and vertical integration (Fine 1992, pp. 15–25). There were a number of elements in this strategy that appeared at different levels of the value chain. First, starting downstream and moving up, the major producers heavily invested in ‘marketing research’. This typically translated as the technological development of new platinum use values and aggressive promotional drives to establish and expand existing markets. The South African industry's attempts to ‘create and grow demand’ were historically coordinated by the producer organisations Platinum Guild International (PGI) and, later, the International Platinum Association (IPA) (Vermaak 1995, p. 95). However, the PGI and IPA were not quota-fixing cartels. In the early 1990s, around 85% of platinum was sold directly to established clients (such as the motor industry) through long-term contracts that guaranteed supply at specific future dates (Vermaak 1995, p. 95). This hedging strategy enabled the majors to plan productive expansion in line with future demand. But they still needed to strike the balance between meeting their supply obligations without flooding the market, and so the control, if not total elimination, of rival producers and new entrants remained a pressing issue throughout this period (Vermaak 1995, p. 197).
Second, the complex and highly specialist metallurgical processes required to extract, concentrate and refine the PGMs from mined ore forced massive scale economies in mineral processing. This compelled the South African producers to develop and integrate beneficiation capacity, which often overlapped with private research and sales activity. The classic industry example is Johnson and Matthey, the UK-listed refiner and marketing agent that had been controlled by Johannesburg Consolidated Investments (JCI, itself a subsidiary of the Anglo American Corporation) since 1946. Impala likewise developed its own beneficiation plant at an early stage (c. 1970), some of which was later ‘contracted out’ to smaller domestic producers. Aspects of these mineral-processing operations were located in South Africa while other facilities were based overseas. Whichever the case, the horizontal and vertical integration of mining and beneficiation capacity itself represented a major form of monopolistic concentration in the platinum industry that simultaneously generated mid-range barriers to entry, and shaped the character of productive expansion, across the apartheid platinum sector.
Finally, and most significantly, the platinum producers achieved an unusual degree of effective control over mineralised land, during the apartheid era. This enabled them to unilaterally deploy or sterilise their reserves in response to boom and slump, and so attempt to stabilise, or at least ride out, the sector's characteristically extreme commodity price fluctuations. As Fine (1992, p. 16) notes, competitive control over current or future mineral reserves does not usually act as the major source of cartelisation in the mining industry. Rather, mining capitals tend to establish market dominance through regulatory mechanisms in the chain of supply. However, two unique factors allowed the South African producers to regulate platinum supply through their collective control over the minerals resource base. The first, as seen above, is that the world's known reserves are overwhelmingly concentrated in South Africa. This means that control over the BIC effectively translated as control over global platinum supply. The second is that a distinctive mineral property system emerged under apartheid that made such control possible. It is to this that we now turn.
The apartheid mineral property system in the Bantustans
Fine (2006, p. 139) observes that, ‘For Marx, the particular form and content taken by landed property has a more-or-less favourable impact upon the accumulation of capital both as potential promoter of, and barrier to, access of capital to the land’. The particular form and content taken by landed property in the South African platinum mining industry was determined, in the first instance, by the generic mineral rights regime of the apartheid state, and, in the second, by the specific types of minerals ownership that had emerged in the Bophuthatswana and Lebowa Bantustans, where the major platinum reserves of the BIC were at this time geopolitically located. The result was a unique configuration of mineral property that would prove particularly favourable to the accumulation strategy of the platinum capitals. We will first consider the universal features of the apartheid mineral rights regime and then the specific forms that it assumed in these two Bantustans.
The jural framework regulating minerals ownership and exploitation under apartheid turned on three key distinctions. The first was a legal distinction between mineral and surface rights. The historic development of South Africa's minerals legislation was extremely uneven, with each new legislative development a response to the changing imperatives of either the state or the mining industry (Badenhorst et al. 1994, pp. 1994–1995, Cawood and Minnitt 1998, pp. 369–372, Dale 1997, pp. 15–16). Nevertheless, it was grounded in the basic common law principle that ‘the owner of the surface of the land is the owner of the whole land and the minerals underneath it’ (Smith and MacDonald 1998, p. 621). Over time, however, an ad hoc combination of administrative practice and legislation led to the statutory recognition of the severance of mineral rights from surface rights (Smith and MacDonald 1998, p. 621). The extent to which the separation of mineral and surface rights took effect varied between the major territories of South Africa (LRC 2002, p. 4), but its overall effect was to constitute each as a separate form of landed property defined and regulated by a specialised subset of property law.2 A South African ‘mineral right’ was thus constituted as a tradable title that variously combined rights to possess, sell, use and lease specified mineral resources (Badenhorst et al. 1994, pp. 289–292, Department of Minerals and Energy [DME] 1998, pp. 10–11).
Second was a distinction between public and private categories of mineral rights ownership, which evolved in tandem with that between surface and mineral rights. Although ‘Western-legal’ conceptions of individualised private property had been ideologically dominant from the earliest days of conquest, South Africa's fledgling colonial administrations were not adverse to expropriating mineral resources from private landowners when it suited their strategic purposes (Dale 1997, pp. 15–16, Cawood and Minnitt 1998, pp. 370–371). These initial forms of public minerals ownership were generally short lived, but, with the political unification of South Africa the early twentieth century, the state had fully emerged (and would continue to grow) as a major landowner in its own right. At this point, the principle was established that any mineral rights attached to this ‘state land’ were a national resource under public control, and hence that they would remain in government hands, even if the surface rights were later sold on. The basis was thus laid for the development of a ‘dual’ minerals ownership system in which mineral rights could be held both by private entities (individuals, corporations) and by the state, which would receive any royalties arising from their exploitation (Cawood and Minnitt 1998, p. 373, DME 1998, p. 10).
Finally, this public/private duality closely mapped the contours of the racial order, with its generic distinctions between the property rights of whites and blacks. Private minerals ownership was the exclusive domain of whites, who enjoyed not only the full rights of mineral property law and minimal state regulation, but a range of additional measures designed to maximise the benefits of minerals exploitation. According to the Legal Resources Centre, white mineral proprietors, surface owners and tenants thus shared the exclusive right to prospect on their land, and a first option to take out a prospecting permit and subsequent mining lease. These leases guaranteed white mineral owners a 25% share of profits or royalty payable to the state, while white surface owners or state lessees were entitled to at least a 32% share in mines on their land (Legal Resources Centre 2002, p. 3). In contrast, black minerals ownership was subject to the same restrictions as all other forms of African landed property under apartheid. Black mineral rights were thus overwhelmingly constituted as group rights owned nominally or absolutely by state trusts, and hence embodied a distinctive public dimension in contrast to the fully commoditised rights of whites.
The key distinctions in the apartheid minerals framework – between surface and mineral property rights, and private/white and public/black ownership – combined to generate a highly complex and fragmented structure of mineral property in South Africa. Nevertheless, it was one in which two dominant patterns of resource access by the major platinum producers would emerge.
First and foremost was the access to the mineral resources owned and controlled by the Bantustan states. A striking feature of the platinum mining industry under late apartheid was that the bulk of its mineral reserves fell within the borders of two of the nominally independent homeland territories into which the black majority had been racially segregated and ethnically divided. Bophuthatswana encompassed a substantial swathe of the western Bushveld while the lion's share of both the northern and eastern limbs of the BIC rose within Lebowa. The access of platinum capital to the greater part of its mineral resources was thus mediated by the dominant forms of minerals ownership that had emerged in these Bantustans. For present purposes, we may briefly identify three such categories of Bantustan mineral property that would play a significant part in the platinum story.
State trust property
The first form of black minerals ownership was common to all Bantustans and may be termed ‘state trust property’. This refers to the vast areas of mineralised and other land that were legally owned and administered by the South African Development Trust (SADT). The registered property of the SADT included both the land ‘reserved’ for African occupation in terms of the 1913 Natives Land Act and the additional ‘trust’ or ‘quota’ land ‘released’ through the 1936 Natives Trust and Land Act for addition to the reserves after 1936 (Platzky and Walker 1985, pp. 89–93). In 1971, all SADT land falling within the jurisdiction of the new Bantustans was legally transferred to their emergent governments and simply renamed ‘state land’.3
The Bantustan territories were overwhelmingly comprised of state land, which was formally conceived as a ‘national resource’ and therefore had an overt public character. This public status in part reflected the fact that large areas of state land were occupied by the homeland population. While these areas fell under the local administrative jurisdiction of tribal authorities, all major decisions concerning surface and mineral rights in state land were the preserve of central (i.e. Bantustan) government.
The Bantustan administrations had the sole right to negotiate surface leases with mining capital and appropriate the surface rents in respect of state land. These revenues were normally paid into state-land trust funds and earmarked for the acquisition of more state land (usually before transfer in 1971) or for ‘economic development’ (South African Institute of Race Relations 1964, p. 165). The situation with mineral rights was considerably more complex. Land ‘released’ in terms of the 1936 Act could come with minerals attached, or with specified mineral rights separated by the previous proprietor, often on sale to the state. These separated rights would remain the property of private (white) owners and could potentially be sold or utilised independently of the state-land administration. Conversely, the state could itself separate mineral rights for sale to third parties; or it could lease mineral rights in state land to mining capital and realise royalty revenue in the much the same way as surface rents. These mineral revenues were generally set at an average, and staggeringly low, rate of 10% of mine profits (ibid.).
The rights of the occupants residing on state land were minimal in relation to mining activity. As the legal owner of surface and/or mineral rights, the Bantustan administration was charged with a fiduciary duty to consider the interests of any people living in the vicinity of proposed prospecting activity or mine development. This could extend as far as setting conditions, ‘by operation of law’, to benefit the local population (Legal Resources Centre 2002, p. 7). However, what these ‘benefits’ meant de facto varied greatly, and mining leases were invariably negotiated without the active participation or majority consent of those affected. Forced removals to make way for mines on state land were common (Levin, Solomon and Weiner 1997).
Tribal trust property
The second type of black minerals ownership tended to predominate in Bophuthatswana and may be termed ‘tribal trust property’. This refers to land historically purchased by Africans through a variety of routes, and subsequently registered to a state official ‘in trust’ for a recognised ‘chief and his tribe’, in terms of the distinctive property laws that emerged in the colonial Transvaal during the late nineteenth and early twentieth centuries.4 Although far less common than state land, such tribal trust land was acquired on a particularly significant scale in what were then the Rustenburg and Pilanesberg Districts of the western Transvaal, where it would form the basis of the local tribal authority areas (‘reserves’) designated by the Land Acts. Crucially, these areas encompassed some of the richest platinum resources on the western limb of the BIC, and would later be incorporated into the Bophuthatswana ‘national’ territory when that Bantustan was granted its fictive independence from South Africa in 1977.
Although the mineral rights had often been severed from tribal trust land, in some cases they remained attached. This potentially endowed the tribal authorities in question with far greater autonomy in respect of mining activity than their counterparts on state land. However, under apartheid, the interests of mining capital were safeguarded by the trusteeship element of this property arrangement. In his capacity as the trustee of tribal land, the Minister of Bantu Affairs mediated every aspect of the relationship between such tribal authorities and the mining corporations, from the negotiation of the initial prospecting agreement to the monitoring and control of royalty payments. Although the tribal authority was supposed to play the leading role in the contracting the process, with the minister merely ‘authorising’ its decisions, the minister could unilaterally modify and consummate the requisite mineral and mining leases, in turn raising the question of which entity – the tribal authority or the trustee state – formally represented the corporate interests of the ‘tribe’ at the ownership level. Moreover, while the minister was duty-bound to ensure that ‘the proposed contracts will be beneficial to the tribes concerned’ (South African Institute of Race Relations, 1964, p. 165), neither the extent nor the content of this fiduciary responsibility had ever been formally defined in South African law (Bennett and Powell 2001, p. 619). On ‘independence’ in 1977, the powers (and ambiguities) of state trusteeship were transferred wholesale from the South African government to the Office of the President of Bophuthatswana (Jeppe 1980, pp. 60–61).
The benefits of this arrangement for mining capital are illustrated by a celebrated case in the platinum-rich Rustenburg area between Gencor's Impala Platinum and the Bafokeng Tribal Authority, over the control of the mineralised land historically registered ‘in trust’ to the latter.5 Impala had operated in the Bafokeng area since the mid 1960s in terms of a highly favourable minerals lease brokered with the tribal authority under the oversight of the (South African) Minister of Bantu Affairs. Such was the mine's success that, by the mid 1980s, Impala was seeking to extend its operation into a new and unexploited portion of the Bafokeng reserve. However, at this point, the Bafokeng chieftaincy began to assert its interests as a form of landed property by questioning the level of past royalty payouts and, more significantly, offering the rights to the new mining area to a rival venture. With its planned expansion blocked, Impala appealed to Lucas Mangope, the autocratic president of Bophuthatswana, to intervene in his (inherited) capacity as state trustee of the tribal land in Bafokeng. The result was an entirely new set of agreements, unilaterally negotiated by Mangope ‘on behalf’ of the Bafokeng tribal authority, which not only opened the new mining area to Impala on a low royalty rate, but granted it perpetual rights in its existing lease area. Despite repeated legal challenges from the Bafokeng chieftaincy, these agreements would remain in place for the remainder of the homeland period, thus underlining how the interests of platinum capital could be served by the state trusteeship element of this property form.
Mineral trust property
The final type of black minerals ownership considered here was unique to Lebowa, and may be termed ‘mineral trust property’. In 1987, the mineral rights in the former SADT land held by the Lebowa government were separated and transferred wholesale to a new structure – the Lebowa Minerals Trust (LMT) – in terms of the Lebowa Minerals Trust Act (No. 9 of 1987). The LMT was explicitly defined as a corporate body possessing mineral property in a similar manner to a private rights holder, as opposed to mineral rights held by the state (DME 2000, p. 2). This endowed the LMT with absolute authority to grant mineral rights to third parties to prospect and mine, and to receive the entire revenue.
The LMT was a particularly transparent instrument of appropriation for the (notoriously corrupt) Bantustan elite (Delius 1996, pp. 173, 211). In 2000, the LMT was reported to control 1500 title deeds to mineral rich farms, of which 1432 were registered in the trust's name. The total value of these assets were estimated to be in the region of R280–R300 million, while the LMT's royalty income was close to R20 million a year (Kindra and Beresford 2000). The LMT was also a means of ensuring that the ‘barrier’ of landed property was distinctly porous on the eastern Bushveld. By 1994, Anglo American's Rustenburg Platinum Holdings (RPH) had gained exclusive control of the mining rights to 26 farms through a series of agreements with the Lebowa government, which resulted in the formation of several joint-venture companies (ibid.). As mining rights were perpetual rights in law, this effectively enabled RPH to monopolise the entire eastern limb of the BIC.
Despite their variations, it is apparent that these different categories of mineral property collectively presented platinum mining capital with little more than a partial barrier to accumulation in the Bantustans. Homeland mineral leases were easily acquired, royalty rates low and the Bantustan regimes always on hand to ensure that the way was eased for white mining capital to access and control the giant reserves falling within their fictively autonomous domains. This, however, was not only the pattern of resource access in the platinum industry. Portions of the BIC also fell within ‘white’ South Africa, and here a second pattern of resource access emerged based on the direct ownership of mineral rights by the platinum capitals themselves. This may be considered more briefly.
Under apartheid, the majority of the rights to the smaller and less economical platinum-ore bodies in ‘white’ South Africa were held by petty speculators, farmers and other landowners (Vermaak 1995, pp. 197–198). However, the rights to the more significant reserves outside of the Bantustans were the property of the platinum corporations. As South Africa's sole producer in the two decades after the Second World War, it was Anglo American's Rustenburg Platinum Holdings (RPH) that had been able to gain direct control of the most significant platinum resources: first, by pursuing an aggressive strategy of buying up mineral rights (with or without surface rights) when prices were low, and hence options cheap; and second, by ‘inheriting’ the mineral property of the smaller companies it had absorbed through the serial mergers and acquisitions that characterised the ‘grand apartheid’ phase of the MEC (Edwards and Silk 1987, p. 27). In 1972, for example, RPH gained mineral rights covering 460,000 hectares of the BIC through its acquisition of the Rand Mines–Anglo American–General Mining consortium (Edwards and Silk 1987, p. 27) and by 1995 controlled an estimated 80% of known platinum resources across the BIC.6 By incorporating and combining the otherwise separate functions of resource ownership and production, RPH was thus able to abolish the barrier of independent landed property in relation to those portions of the BIC it directly controlled (Neocosmos 1986, pp. 30–32).
Although the major platinum producers were locked in a fiercely competitive struggle, the net effect of these two patterns of resource access was to generate a common interest in the reproduction of the apartheid minerals property regime. The combination of Bantustan state control and direct minerals ownership significantly weakened, if not totally eliminated, the potential obstacle of landed property in the platinum mining industry. South African platinum capital thus had far greater freedom to exploit (and sterilise) its vast domestic reserves in response to changing world market conditions than might otherwise have been possible under a more stringent minerals regime. Moreover, by virtue of their effective control over the major limbs of the BIC, the South African producers were able to erect a powerful barrier to entry by potential competitors. The historically given form of the apartheid mineral property system thus represented the essential jural condition of the entire industry's accumulation strategy. It would therefore be imperative that the conditions of the reproduction of these property relations were secured as the political transition from apartheid got underway. It is to this process that we now turn.
The apartheid mineral property system in the political transition
As South Africa stumbled towards a negotiated settlement in the early 1990s, the apartheid state rushed to protect the historic privileges of white corporate capital in a democratic dispensation. The long-held fear of white industry and agriculture was that an ANC government would adopt the radical programme outlined in the 1955 Freedom Charter. With the mass struggle at its height, and in the absence of coherent economic policy statements to the contrary, the National Party took the ANC at its written word and pursued a twin-track strategy. On the one hand, the outgoing apartheid government successfully negotiated a universal ‘property clause’ in the constitutional settlement to safeguard (white) private property from redistributive pressures in the new dispensation (Ntsebeza 2007). On the other hand, it hastily legislated a range of measures designed to protect and promote white capitalist interests in discrete economic sectors. Key here was the Minerals Act No. 50 of 1991. As with the Abolition of Racially Based Land Measures Act (the ‘Land Act’) passed the same year, the 1991 Minerals Act was a key component in the National Party's last-ditch outpouring of state handouts and prophylactic legislation on white capital's behalf (Bernstein 1996, pp. 24–25, Francis and Williams 1993). Its principal aim was to narrow the scope for the statist control and redistribution of mineral property in the new dispensation, and its main instrument the further privatisation of mineral rights through the selective deregulation of the minerals property system (Badenhorst 1991, p. 113, Badenhorst et al. 1994, p. 289, Dale 1997, p. 16). Accordingly, the legislation provided for the disposal of state-owned mineral rights in ‘white’ South Africa, so ‘releasing’ them for market acquisition by the mining houses; ‘encouraged’ the market transfer of private rights from ‘small’ holders (like farmers) to the same; enhanced the protection of any mineral leases acquired in terms of previous legislation; and radically reduced overall state regulation of the industry (DME 1998, p. 12, Smith and MacDonald 1998, pp. 513–515). In the words of future platinum ‘empowerment’ entrepreneur Patrice Motsepe, the Act was thus ‘a preposterous attempt to maintain the status quo of privileges to those in the mining industry who benefited in the past, and to constrain any restructuring attempts by the new government’ (1993, paraphrased by Smith and MacDonald 1998, p. 513).
How, then, did this impact on the platinum industry's strategic imperative of securing its proprietorial control of the BIC? Certainly there is little doubt that, as Cawood and Minnitt (1998, p. 371) put it, the 1991 Minerals Act represented ‘the biggest step towards a system of exclusive private mineral rights ownership in the history of South Africa’ (Cawood and Minnitt 1998, p. 371). Yet, as they go on to note, the outgoing apartheid government did not in the end exercise the politically charged option of redistributing state-owned mineral rights to the white conglomerates in the dying days of the old republic. Rather, a view was taken that there were sufficient provisions in the new legislation to enable the mining corporations to maintain exclusive control of this category of mineral property through their existing agreements, which, in effect, could be renewed indefinitely (Cawood and Minnitt 1998, pp. 371, 373). At the same time, those capitals already in direct possession of mineral resources in ‘white’ South Africa, like Rustenburg Platinum Mines, would see their extant private rights consolidated and strengthened. But, given the geopolitical location of the bulk of its resource base, of greatest significance to the platinum industry as a whole would be the essential continuity of the dominant forms of Bantustan mineral property in the new, non-racial dispensation itself.
After the first democratic elections of 1994, immediate moves were made to (re-)absorb the Bantustans into the unitary post-apartheid state through the dissolution of their old structures of central government and the generalisation of key national laws and institutions. Yet, the actual development of new policy and legislation was markedly uneven. This tended to force the new ANC government to work within existing jural and regulatory frameworks in what were now renamed the ‘communal areas’. Critical here is the manner in which the slow development of new minerals and land reform legislation would converge to keep the old mineral property relations of the former Bophuthatswana and Lebowa territories fundamentally intact.
In its 1994 election manifesto, The Reconstruction and Development Programme, the ANC had pledged to develop a new minerals policy that would radically transform the apartheid-era mineral rights system by vesting the ownership of all mineral resources in the democratic state (ANC 1994, p. 99, Cawood and Minnitt 1998, pp. 374–375, Dale 1997, pp. 17–19). The antithesis of the 1991 Minerals Act, this policy of resource nationalisation thus represented a full-frontal assault on the proprietorial interests of the white mining houses in general and the platinum mining industry in particular. However, the ANC was reluctant to take on South Africa's most powerful economic sector on assuming power, and so the minerals policy development process would not meaningfully get underway until the end of the 1990s.7 In the interim, the 1991 Minerals Act was generalised throughout the former homeland areas under the administrative control of the still largely ‘unreconstructed’ Department of Mineral and Energy Affairs (DMEA), which had developed strong formal and informal institutional linkages with mining capital since its formation in 1980 (Fine and Rustomjee 1996, pp. 97–98).
In contrast, the development of land reform policy, under the Department of Land Affairs (DLA), was ostensibly more rapid. A commitment to the three basic elements of the post-apartheid land reform programme – land redistribution, legally secure land tenure, and the restitution of land – had already been enshrined in the constitution and indicated an important area of exception to the universal property clause. However, the necessary legislation to put these principles into effect was developed at markedly different speeds – a reflection of the distinct political pressures to which each area was subject. Of these three elements, land tenure reform was the most significant in respect of the dominant categories of Bantustan mineral property. The prospect of tenure reform initially held great promise for former homeland populations whose land and mineral rights had been subject to the institutions and practices of state trusteeship. Early policy statements from the DLA signalled that the titles previously held by the erstwhile Bantustan states and the SADT would be devolved to the (long-term) occupants of state and tribal land, to be held and administered by a variety of forms ranging from individual, through group to tribal (now ‘community’) ownership. Title transfer thus had the potential to dissolve large swathes of landed property under the control of the central state and, in its place, create new and wholly independent types of black minerals ownership. This in turn raised the possibility that the terms on which mining capital accessed the mineral resources in these areas could change. However, the development of this tenure reform policy was hamstrung by the increasingly fraught politics around the place of the chieftaincy (now ‘traditional leadership’) in the democratic dispensation (Cousins 2008, Ntsebeza 2005). This would force the new government to institute – and then keep in place – a series of ‘interim measures’ whose overall effect was to reproduce the extant mineral property arrangements in these areas.
When the new South African Constitution (Act 200 of 1993) came into operation on 27 April 1994, the ownership of all land and minerals held by the former homeland governments reverted to the (new) national government (DME 2000). From 1995, all tribal and state land (including ex-SADT land) was thus unified under the trusteeship of the Minister of Land Affairs, who repeatedly stated his intention to treat the occupants of that property as putative owners, pending tenure reform. Yet, this would also mean that all of the old mineral and mining leases brokered by the apartheid and homeland authorities would remain in force, until there was substantive change in the mineral property regime. Moreover, for reasons that require further investigation, the Lebowa Minerals Trust was exempted from the new pattern of state trusteeship and instead placed under the administration of the (new) Minister of Minerals and Energy in 1996, pending specific legislation for its abolition at a future date (DME 2000). Critically, then, the platinum industry's juridical control over the mineral resources of the former Bophuthatswana and Lebowa territories would remain untouched, thus safeguarding the essential condition of its accumulation strategy. At the same time, however, there was the possibility that the actual development of these mineral rights through new platinum mining activity might be subject to more onerous regulation in the new dispensation. In his (inherited) capacity of trustee in the new dispensation, the Minister of Land Affairs now had a clearly defined fiduciary duty to ‘protect and uphold the rights’ of all those living on state and tribal land.8 This formally committed the DLA to ensure that there was proper consultation with rural ‘communities’ in respect of new mine developments and to enter into tripartite negotiations with mining capital that would guarantee them agreed benefits, including surface rent (DME 2000, Legal Resources Centre 2002, p. 6). The Department of Minerals and Energy (DME) was also obliged to consult the DLA and provincial government before granting prospecting and mining authorisations in respect of such land. However, the notoriously poor institutional capacity and coordination of regional DLA and DME offices meant that these good intentions were rarely put into effect. Over the course of the mid to late 1990s, the DME unilaterally approved a number of prospecting and mining deals in the former homeland areas that offered few if any ‘benefits’, which were generally brokered by chiefs and other agents claiming to represent such ‘communities’, often without their knowledge, let alone majority consent. Thus, it was not only the case that the platinum corporations retained their exclusive control over the national platinum endowment through the essential continuity of the dominant categories of Bantustan mineral property. The politically reconstituted practices of state trusteeship would, by default if not design, continue to present only the most marginal barrier to the exploitation of these resources whenever it suited the corporations to bring them on stream.
‘Globalised restructuring’ and the platinum boom
If these were the ways in which the platinum interest secured its strategic property relations during and immediately after the negotiated transition, it would also have much to gain from the ANC's deepening commitment to a neoliberal macroeconomic policy position at this time. Critical here were the terms of South Africa's re-engagement with the world economy. Though ineffective in many ways, the sanctions imposed on South Africa by the ‘international community’ in the final years of apartheid had succeeded in cutting off substantial sections of national capital from the world economy and significantly reducing the rate of foreign direct investment. However, the privileging of corporate interests in the ANC's macroeconomic policy now meant that the beleaguered South African conglomerates could take full advantage of the opportunities presented by the brave new world of capitalist globalisation. From the perspective of the platinum corporations, the immediate effects were twofold.
First, the phased reduction of apartheid-era capital and exchange controls enabled the conglomerates to relocate to the world's leading financial centres, thus increasing their capacity to tap global equity markets, export capital and assets, and so discipline their home state. Second, they embarked on a vigorous programme of domestic restructuring (‘unbundling’) in which the diverse holdings and subsidiaries of the group producers were broken up, sold off where they were weak, or amalgamated into powerful new companies where deemed internationally competitive. This in turn enabled the divesting corporations to secure political cover and influence at home by offloading their ‘non-performing’ assets to aspirant ‘black empowerment’ entrepreneurs drawn from the ranks of the ruling party, while retaining the most profitable ventures for themselves. The conglomerate's ‘globalised restructuring’ strategy thus simultaneously reproduced and modified the determining influence of the MEC in the post-apartheid political economy by integrating their core domestic operations with their fast-expanding international interests, while increasing the domestic power of the burgeoning financial sector (Fine 2008, also Freund and Padayachee 1998, Makgetla 2004, Marais 2001 pp. 110–116, Saul 2001, pp. 195–213).
The first and most significant example of ‘globalised restructuring’ in the platinum subsector was the formation of the Anglo Platinum Corporation (Amplats) out of the break-up of Johannesburg Consolidated Investments (JCI). In May 1995, the giant Anglo American Corporation (AAC) ‘unbundled’ its JCI holdings into three separate companies: JCI Ltd with interests in gold, ferrochrome and base metals; Johnnies Industrial Corporation (Johnnic) containing its nominal industrial holdings; and Anglo Platinum, which amalgamated the industry leader Rustenburg Platinum Holdings (RPH) with JCI and AAC's other platinum assets into a single, ‘focused’ entity majority owned by AAC (Coakley 1998, p. 3).9 Accounting for 40% of global output and with a blue-chip London listing, Anglo Platinum was thus indisputably positioned as the world's pre-eminent platinum powerhouse.10
With the advent of the ‘new’ South Africa, then, the platinum industry had not only managed to retain exclusive control of its giant resource base, but produced a new and streamlined breed of mining company retooled to compete on the global stage. The combination of these factors would in turn enable the platinum interest to take advantage of a historically unprecedented platinum price boom that began in the mid 1990s. Driven by a powerful surge in autocatalyst demand and the phenomenal growth of the ‘white metal’ jewellery market, the world platinum price began a steep climb in early years of the new dispensation that, as the price chart in Figure 2 graphically illustrates, would show no signs of abating until the great financial crash of 2008.11 The effect of this ‘platinum explosion’ was to propel the South African industry on a steep growth trajectory distinguished by three principal features.
First, the major platinum producers each announced, and then upwardly revised, ambitious programmes to expand mining and refining capacity. In 2001, the combined investment plans of the platinum subsector stood at R30 billion over seven years (Graulich 2001b) and accounted for 34% of all committed capital projects (2000–2004) in the national mining industry (Chamber of Mines 2001). Crucially, this growth was based on a combination of extensive and intensive development; that is, the opening of new mines in new areas, as well as the deepening of existing operations. Here the mining corporations were able to reap the gains of their continued jural stranglehold over the national platinum endowment. On the one hand, they could selectively bring on-stream the best resources from their vast portfolios of unexploited rights, and, on the other hand, continue to prevent potential competitors from gaining entry to the BIC. The rewards were enormous. In 2001, Anglo Platinum became the first South African company to report a profit of US$1 billion (R6918 billion) earned solely from its domestic operations (Graulich 2001a), while its annual profit rate increased by 87% from 1996 (Anglo Platinum 2001). Impala's headline earnings likewise doubled between 2000 and 2001, with sales revenue rocketing by 71% to hit US$1573 million (R1197 billion) that year (Impala 2001).
Second, platinum's overall strategic ‘weight’ in the MEC significantly increased relative to the historically premier gold industry, which had been locked in a spiralling decline since the early 1980s (Freund 1991, Marais 2001, p. 103). The nature of the diverging relationship between South Africa's platinum and gold industries is clearly illustrated by the graph sets in Figures 3 and 4, which compare sales, production and employment trends in the period under consideration.12 The weighted comparisons in particular show that the PGMs began their rise relative to gold's decline in the early to mid 1980s (depending on the indicator used) and that this medium-term trend was vastly accelerated by the boom in PGM prices, starting in the mid 1990s. That the total sales value of the PGMs would eclipse gold by 2000 is particularly striking and, while there was no guarantee that this position would be maintained given the commodity's historic volatility, it was apparent that platinum was fast emerging as the new dispensation's star primary export and leading mineral stock.
Finally, however, the political consequences of platinum's meteoric rise would prove deeply contradictory. As was briefly noted in the previous section, the ANC had committed itself to radically overhauling the apartheid-era mineral rights system through the nationalisation of South Africa's mineral resources in its 1994 election manifesto. Although the development of this policy had been effectively stalled by the power of the mining industry in the early years of the new dispensation, it would increasingly gather momentum towards the end of the ANC's first term. Thus, in October 1998, the Department of Minerals Energy published its Minerals and Mining Policy for South Africa White Paper (DME 1998). This not only reaffirmed the ANC's intent to transfer all privately held mineral rights to the state through new legislation, but elaborated that one of the main rationales for doing so was to ‘prevent hoarding or sterilisation of mineral rights’ which acts as ‘a barrier to entry against potential competitors’ (S1.3.2[iii] and S1.3.3.1[ii]a). Yet, this was the precise moment that the platinum subsector was emerging as the most dynamic element of the post-apartheid MEC. Moreover, as a Business Day editorial would put, it was the only one ‘in which the mining houses are sitting idly on vast payable ore deposits’ (Business Day 2001). The platinum industry's very success in securing its control over the national platinum endowment thus had the unintended consequence of turning it into the primary target of the ANC's stated policy of nationalising and redistributing all categories of mineral property. It would therefore now be confronted with the prospect of nothing less than the dissolution of the essential conditions of its core accumulation strategy.
Conclusion
As this article has shown, the unique geological concentration of platinum resources in South Africa had historically combined with the extreme volatility of the world platinum market to generate an accumulation strategy based on monopoly control of the BIC. The essential element of that strategy was a unique mineral property system that enabled the major platinum producers to lay exclusive claim to the great platinum reserves in the homeland areas, while hoovering up private rights in ‘white’ South Africa. Despite the redistributive pressures building up in the transition from apartheid, a variety of political factors had enabled the platinum corporations to simultaneously reproduce and strengthen their strategic mineral property relations in the new dispensation. Yet, whilst this placed the industry in the optimal position to take advantage of the new global platinum boom, it also made it the primary target of the ANC's new minerals reform legislation, which was released for public consultation in December 2000. The platinum corporations had, in short, become victims of their own success. Their subsequent struggles over the final form and benefits of the new minerals dispensation (c. 2000–2002) would thus open a dramatic new chapter in the platinum story. This is beyond the scope of the present article, but will be considered in a future contribution to this journal.
Note on contributor
Gavin Capps is a Postdoctoral Research Fellow in the National Research Foundation Research Chair in Land Reform and Democracy in South Africa at the Centre for African Studies, University of Cape Town. His current research is on the new processes of rural social differentiation and struggle that are being set in motion by the expansion of platinum mining activity in the former homeland areas of the North West and Limpopo Provinces, South Africa.