Credit Apartheid, Migrants, Mines and Money

Abstract

Migrant life has long required a careful balancing of responsibilities. Migrants travel to earn a wage in a capitalist economy while saving resources and honouring obligations that arise in a seemingly less-than-capitalist one. Various agents – rural patriarchs, traders, government authorities, appliance retailers – have used techniques to keep wages beyond migrants' control. Paradoxically, similar techniques have, on occasion, been eagerly embraced by migrants themselves, who know that these resources will need to be husbanded for the upkeep of home. This article explores these contradictions, showing that recent forms of debt build on expectations born of forms of credit that proliferated earlier, but differ in consolidating these forms of credit to produce an unimpeded flow of money into migrants' bank accounts and out of them again. It looks at the advantages and dangers of the recent expansion of credit to constituencies – like migrants – where it previously did not reach.

Key words:

• debt
• savings clubs
• moneylending
• hire purchase
• credit apartheid

The 2012 killing of platinum miners during a strike at Marikana was a cogent reminder that labour migration remains a salient part of South African life. It also revealed that the conditions of migrant life remain deeply problematic and in some ways have worsened or intensified. Among other factors that prompted the strike was an exponential proliferation of credit sources – including from microlenders practising unsecured lending – and hence a sharp increase in the indebtedness of migrants taking up offers of loans. That money is lent with such readiness is no accident: creditors can secure repayment with extraordinary ease. Miners' pay, automatically transferred into their bank accounts at month-end, is effortlessly removed from these again by those to whom they owe money. Two principle technologies of repayment prevail. One system – used by informal moneylenders or ‘loan sharks' – involves keeping borrowers' ATM cards and using these to withdraw funds, with interest of amounts of up to 50 per cent per month, from the cash machine at month-end. The other – used by banks, retailers and the new wave of microlending businesses that have sprung up since 1994 – is through direct debits, or by using and abusing the practice of garnishee or emoluments attachment orders. If a creditor is owed money and presents an employer with a garnishee order, the employer is obliged to enable that creditor to take a portion of the debtor's monthly pay before the employee receives it. The monthly earnings of miners, other migrant wage-workers, and workers and salary earners more generally, have been the principle targets of lenders who have been using these repayment technologies with ever-increasing frequency. The accumulation of multiple debts, paid back to a series of creditors in rapid succession as soon as payday arrives, means that many earners have little left to live on. While extending access to the formal credit facilities previously denied to black South Africans was an important aspect of the new financial inclusion which replaced credit apartheid, the Marikana episode suggests that its disadvantages outweigh its benefits.

The other factors underlying the episode should not, of course, be ignored. There is evidence, for example, that since the widespread dip in employment following the country's liberalisation in the 1990s,¹ each wage-earning migrant and/or miner is supporting a wider range of kinsmen than previously, with each pay packet required to go further (Stauffer 2010; Beinart 2014). This has been accompanied by widespread casualisation. In the mining industry, there has been a steady increase in the use of sub-contracted labour. Where, formerly, sub-contractors were employed for the short-term work of shaft sinking, while the majority of underground mineworkers were on permanent contracts with the mine, the core workforce of many mines now comprises large numbers of workers on lower wages and flexible contracts (Crush et al. 2001).² Mineworkers, previously seen to benefit from greater levels of security and unionisation than other areas of wage labour, are thus now facing conditions of precarity. Although one must certainly acknowledge that mineworkers at least ‘have jobs' (Beinart 2014), where many others are unemployed, it is partly the fact of their regular pay, ironically, that makes them vulnerable to the financial depredations described here.

The free market-style opportunism which led to the growth of informal and semi-formal microlending, alongside and inextricable from their formal financial counterparts such as banks, vehicle finance companies and retailers, has not gone unnoticed. After the Marikana killings, the Minister of Trade and Industry decried lenders' ‘outright preying on the vulnerabilities of low income and working people’ and undertook to implement more controls on those who make a living as lenders.³ Efforts had already been made, in the wake of South Africa's democratic transition, to regulate and control the ‘reckless lending’ of money at interest, notably through the National Credit Act (NCA), effective from 2007. Although some of these practices were always borderline illegal, and others were newly outlawed, most continue unhindered.

In seeking to explore how these borrowing, lending and repayment activities have taken root, despite attempts at regulation, it is necessary briefly to consider the phenomenon of financialisation – of which these practices seem to be a logical outcome – which is said to have taken hold in the South African economy in the late 1990s and early 2000s (Barchiesi 2011; Marais 2011:124–128, 132–139).⁴ Seen from creditors' point of view, the term describes a new ‘pattern of accumulation in which profit making occurs increasingly through financial channels rather than through trade and commodity production’; seen from that of borrowers, it means that they are ‘confronted daily with new financial products' (Krippner 2005:173–174), and that those previously unschooled in matters of saving, borrowing and repaying are enjoined to become ‘financially literate’ (Krippner 2005:173–174), often actively modelling their use of money along more formal lines in what has been called ‘financialisation from below’ (Krige Forthcoming).

These tendencies are evident in the way both smaller microlenders and larger corporate lenders – and both informal and formal creditors – have penetrated and capitalised on the many financial dealings and activities of wage earners in South Africa: from mobile money transactions and payday loans to business start-ups, micro-insurance, low-cost mortgages and the securitisation of migrant remittances (Hudson 2008; Roy 2010; Krige 2011; Schwittay 2011; Soederbergh 2013). While such initiatives have been lauded by some for being inclusive and for serving to democratise credit, for enabling the banking of the unbanked, and for opening up borrowing to those at the ‘bottom of the pyramid’ or BoP (Porteous with Hazelhurst 2004:89; Prahalad 2004), it has not gone unnoticed by others that ‘the effects of financialization … have been highly detrimental to significant numbers of people around the globe’ (Epstein 2005:5).

On top of these worldwide trends and effects, aspects of the national legal landscape arguably disadvantage borrowers still further. Overall, South Africa's legal framework has been noted as consistently advantaging creditors (Boraine and Roestoff 2002:4), and as providing little in the way of affordable bankruptcy or insolvency options for those, such as low-wage earners, who owe amounts of less than R50,000 (Schraten 2013). These factors make fairness for debtors difficult to pursue. Although recent attempts at regulation and reform by the state in partial collaboration with the non-governmental sector have been driven by a recognition that people need protection (even perhaps from their own profligacy), this has so far not managed to derail a deeply-rooted system of ‘external judicial control’ over wage-earners' finances (Haupt et al. 2008:51), often keeping them at arm's length from their pay packets and making even basic financial literacy less likely than it might have been.

This article provides an account of the origins of that system. In doing so, it seeks to transcend dichotomous views which either celebrate or condemn the recent expansion of credit. Instead it analyses them ethnographically and historically. Based on the authors' field research in Soweto, Rustenburg and Mpumalanga, and on literature which gives an account of earlier practices, it traces whether and how this newly proliferating set of borrowing and lending practices echoes earlier, apartheid- and segregation-era forms of saving, credit and debt (here glossed as credit apartheid). It shows how many of the main features of capitalism, including unbridled indebtedness, were already in evidence at around the turn of the 19th century. Various businessmen – from big capitalist corporations and companies to small-time traders and entrepreneurs – made profits from lending to a range of clients. Among those prospective borrowers sufficiently well-heeled to buy property, a system of mortgage finance was in place, together with the robust repossession of real estate from those unable to keep up with payments. Humbler people, including labour migrants, were having loans extended or advances granted to them by recruitment agents and then finding themselves having to work for a lifetime to pay these off. As time went by they started to rely on more apparently informal means of managing money, such as savings clubs and burial societies, later combining this with the purchase of movables on hire purchase (which were similarly subject to repossession when default occurred) and making use of bank accounts.

The increasing ease with which money can be (and is) borrowed in the present day thus builds on expectations born out of earlier forms of credit/debt. But contemporary borrowing and lending is also different in some ways. Varieties and sources of credit have expanded on the one hand, while, on the other, money is concentrated through the use of technological formats so as to facilitate its unimpeded flow through bank accounts. The article proposes that, in bringing economic inclusion, these processes have also re-inscribed older economic inequalities and created new economic vulnerabilities. There has been a progressive movement away from a more personalised to a more formal and financialised system – one less subject to influence, evasion or resistance on the part of the consumer. But certain informal techniques and strategies co-exist with these. As borrowers explore new forms of consumption they retain or newly develop a desire for continuity in socio-economic process. Alongside the move by the retail sector to formalising and streamlining credit rather than relying on repossession, and by the banking sector to encourage everyone to open a bank account, the informalisation of money-lending practices has been proliferating and intensifying.

Instead of tracking a single process of financialisation or formalisation, then, the article uses an approach that draws from recent writings in economic anthropology, which show how state-regulated and legal/formal economic arrangements interpenetrate with those less visible and less regulated (Hart 1973; Guyer 2004; Shipton 2007, 2009; Hart et al. 2010; Shipton 2011). It draws in particular on Jane Guyer's proposal that we reject binary assumptions, in which capitalism on the one hand is counterposed against local economic activities which resist it on the other. And it follows her suggestion that we explore how dynamic processes of formalisation are extended ‘piecemeal’ rather than uniformly and in a homogenising manner (Guyer 2004:157). Formalisation and informalisation thus interact with each other, producing a complex plurality.

The New Lending

While the ending of credit apartheid – like that of apartheid more broadly – progressed in gradual steps, certain developments in the post-1994 period were more dramatic (Porteous with Hazelhurst 2004:77). As black consumers began to take advantage of new credit opportunities, sectors of the white community started microlending businesses to cater to and exploit this new market. State policies, during the 1990s, enabled such developments by liberalising the economy and, more specifically, by lifting previous legislation governing ‘usury’ which had restricted the interest rate (James 2014:65). Countering market-driven impulses with regulatory ones in characteristic fashion, the government then sought, during the 2000s, to regulate the negative effects of this borrowing by passing new legislation to outlaw ‘reckless lending’ with its new National Credit Act.

The result of these developments was the emergence of three lending sectors: mainstream/formal financial; registered microlenders; and community moneylenders or ‘loan sharks'. Each, plugging gaps left by the other two, was supplying this new market in its own way, and came to define itself in contrast to the others – although borrowers, ignorant of the legislation, did not always distinguish clearly between the second and the third of these. First, and biggest, is the mainstream/formal financial sector, historically dominated by British-owned banks and rooted in the English-speaking capitalist sector (Verhoef 2009:157, 181). The ‘big four’ banks – ABSA, First National (FNB), Nedbank, Standard – took the lion's share. They offer credit cards, housing loans and vehicle finance. The sector also features store cards for clothing and food and hire-purchase loans for furniture and appliances. Some recent additions, originating in the second sector (below), have joined this mainstream corporate sector. They include African Bank and Capitec, both of which explicitly aim to cater to the low-income earning population. (The unsustainable character of the unsecured lending activities of the former, in particular, were exposed when in 2014 it sought a bailout from the government, causing Moody's to downgrade the credit ratings of the ‘big four’ banks in turn.)⁵ Perhaps more surprising, certain ‘home-grown’ yet equally corporatised players are claiming an increasing share of the BoP financial products market by targeting similar constituencies. Foremost among these has been the National Union of Mineworkers (NUM) and Chamber of Mines' joint corporate venture, the Ubank (see below). By leveraging proximity to, and (at times personalised) knowledge of, individual constituents, this move by the NUM marks a broader transformation in the relationship between union and workforce, redefining worker empowerment as financial opportunity, while turning members into clients/customers. Second is the new microlending sector, offering mostly smaller and short-term loans. It grew exponentially in the 1990s and was mostly run by Afrikaans-speaking former civil servants who invested their redundancy packages in these businesses. They did so after leaving state employ when the African National Congress (ANC) became the ruling party. Having at first been allowed to charge ‘uncapped’ interest rates, and to engage in practices – such as the use of borrowers' ATM cards by way of loan security – that were later prohibited, many of these subsequently registered as microlenders under the Act, which obliged them to charge monthly interest of no more than 44 per cent. Third are the mashonisas or neighbourhood moneylenders.⁶ Growing rapidly since the 1980s (Siyongwana 2004), this sector came to be defined by its difference to its newly-registered and hence legal counterpart: its protagonists, in contrast, were defined as loan sharks because they remained unregistered under the act and continued to charge monthly rates of interest as high as 50 per cent. The bigger mashonisas have continued to ensure repayment by using the ATM-card-confiscation technique: something that had been outlawed for use by those who had resolved to practise as registered – and more ‘formal’ – microlenders.

Possible sources of credit have thus proliferated since the end of apartheid. A consumer is able to borrow from many banks, use many credit cards, buy a car with finance and furniture on hire purchase, and hold store cards from an array of retailers, as well as having access to microloans both legal/formal and illegal/informal. Lenders such as Ubank offer short-term loans which promise to offset high fees with free airtime and discount vouchers for retail outlets and bus companies.⁷ Often, consumers, having been able to take out loans with very few checks being made on their income or credit-worthiness, borrow from one source in order to repay another: a practice that has not noticeably reduced despite the passage of the Act (James 2014:85, 157). At the same time the means by which these debts are collected has shrunk drastically. Advanced IT systems, rooted in apartheid's preoccupation with surveillance, were put in place, aimed at making it possible to use a single hi-tech system for identification, cashing in social grants, buying and banking (Breckenridge 2005:272–273, 2010) and even opening up the commercial use of such data (Breckenridge 2005:281). Within the mining industry specifically, the system of migrant labour recruitment through TEBA (formerly The Employment Bureau of Africa, established in the 1920s to recruit labour for the mines), coupled with the policing of domestic space in the hostel compound, has relied on a hi-tech infrastructure for identification and surveillance which fuses employment ID with financial ID. Identification and surveillance of employees' financial lives is facilitated, and the NUM's own Ubank is ideally placed to move in on this terrain. Such technologies overall have played their part in transforming what were previously distinct ‘pockets' of finance into one single flow. Sources of credit, then, have become more diverse, but modes of collection have narrowed and become streamlined.

Recent changes thus affect how migrants use their money; how widely they are required to redistribute it; how much of it they can save; and – especially – whether their outgoings exceed their income.

Credit Apartheid – A Historical Perspective

From soon after the onset of proletarianisation, a variety of agents and forces attempted to cajole black cultivators to become wage-workers in the mine economy. Success in doing so was not guaranteed, but where this was accomplished, the regular payment of wages opened up a stream of income for small-scale traders. At the same time, South Africa's patterns of racialised land ownership and exclusion were gradually being put in place, eventually yielding a situation in which blacks were moved into reserve areas (later ‘homelands'). The resulting relations of debt and credit differ markedly from the situation well-documented in many South East and South Asian countries, where relationships combining paternalistic dependence and exploitation developed between tenants and their landlords – who doubled as their creditors, and where landlords often cultivated that dependence, milking their borrower/tenants dry over several generations (Mosse 2004; Martin 2010; Murray Li 2010; Shah 2010). South Africa had a more definitive land dispossession, yielding less personalised debt relations. More often than landowners, it was traders and recruitment agents, and later the owners of companies and large corporations, who were the major lenders; but alongside these, especially since the early 1980s, local community moneylenders have become a prominent feature of life, alongside the recently formalised microlenders.

Following colonial conquest, black people in South Africa – especially those who were or would later become members of the emerging middle class – made efforts to procure their own land in the late 19th and early 20th centuries. Some were granted title in recognition of their former occupancy (Murray 1992:37–44), others formed syndicates and bought land via missionary and other intermediaries (Trapido 1978:28; Cobley 1990:157; James 2007:53). Given the extent and influence of 19th-century mercantilist capitalism in the region (Trapido 1978), opportunities for land speculation were considerable, and some of these black owners either lost their land after having borrowed against it (Beinart 1986:265; Beinart and Delius 1986:24) or willingly sold it to pursue education and other modern investments (Murray 1992).

The subsequent passage of several draconian laws in 1913 and 1936 – although they took decades to fully enact – secured most farmland for white owners. The eventual commercial success of some of these farmers depended not only on those laws but also on other measures, including putting pressure on black occupants to become rent- or labour-tenants who, while continuing to cultivate, also worked for farm owners (Trapido 1978:30–31; Beinart and Delius 1986:33–34; James 2007:6, 39, 131–139). Other measures were the provision of state subsidies; and the establishment of state marketing boards (Beinart and Delius 1986:29–42; Morrell 1986). Credit was given on favourable terms to white farmers and was withdrawn from black ones (Beinart and Delius 1986:29–30; Morrell 1986:379–380, for an earlier period see Ross 1986:68–70). For blacks who remained on the farms attempting to cultivate, increasing restrictions on their time and decreasing access to the market – and to credit – eventually caused many to leave for the reserves, far from urban centres of commerce. The town-ward migration of many then began in earnest.

Cultivator households still based on the land, with some household members travelling to town to work as migrants, were restricted to borrowing from traders, or buying on credit from them (Van Onselen 1996:253; Krige 2011:137; Whelan 2011:88–89, 93–94). Access to such traders was differentiated depending on whether borrowers resided in white farm areas (Van Onselen 1996) or in the ‘native reserves' (later homelands) (Beinart 1979; Whelan 2011). The racial/ethnic profiles of these merchants, in a setting where many occupations were restricted and hence defined in such terms, also differed from one place to another. Merchants belonged to the established white settler constituency in some places (Beinart 1979; 1986:266–267; Roth 2004:62; Whelan 2011), but other merchants, from ethnic minorities – Gujarati-speaking Muslims from the Punjab and Jewish refugees from Russia and its borderlands – sold (and lent) to blacks in areas where their counterparts from more majority settler backgrounds did not venture (Cobley 1990:43; Krige 2011:137). Such traders were not, however, necessarily exploitative; indeed they were ‘often sympathetic’ to their black customers, themselves being ‘emigrants from societies dominated by peasant economies' (Van Onselen 1996:186). Some traders' enterprises remained as small ‘native’ trading stores (and mine shops), probably combining such sympathy with profit in a fine balance, while others expanded and grew into large retail companies (Kaplan 1986:327). These retailers later took to selling their goods on hire purchase (Phillips 1938:40–41; Kaplan 1986:167); and profit eventually outstripped more personal considerations.

Showing less evidence of sympathy and more of profit, there were elements in settler society that directly profited by linking lending with labour recruitment. In Pondoland before about 1913, rural cultivators were induced into work contracts or tempted to leave employment in one sector in favour of another, by local traders, doubling up as semiformal recruitment agents, who gave ‘cattle advances' against migrants' future earnings (Beinart 1979). In other settings – such as Bechuanaland (now Botswana), a British Protectorate at the time – labour agents recruiting for the South African mines ‘induced’ black cultivators to enter into contracts by paying them wages in advance, thus indebting them (Schapera 1947:108). Both parties abused this system, with agents often extending such large advances that the borrower ‘remained in debt even after having worked for several months', and with borrowers often accepting advances from several agents at the same time, with no intention of honouring their debts to any of them (Schapera 1947:109).

To counteract practices which – if left unregulated – might have led to unsustainable levels of debt on one hand and to the collapse of agents' enterprises on the other, a series of measures were put in place by the colonial authorities. Cattle advances, enabling rural patriarchs to control the wages of young men, were seen by officials as exploitative, and eventually abolished (Beinart 1979). But such prohibitions did not result in migrants' getting free access to their earnings. Instead, a system developed of deferring part or all of miners' pay rather than giving it to them at the work site (Schapera 1947:106–107; First 1983): the authorities feared that cash received immediately would be too readily spent or diverted from ‘legitimate’ uses – including the payment of various colonial government taxes and levies – or might encourage migrants to neglect or desert their families. An intermediary system of paternalistic control, which later developed in the context of life in the mining hostels or compounds, was that through which hostel managers attempted to monitor and manage the ‘responsible’ remittance of migrant miners' meagre wages to families ‘back home’.

A combination of free-wheeling enterprise on the one hand, and its regulation by paternalistic authorities on the other, thus laid the basis for a system in which wage earnings, and eventually the bank accounts used to transfer these, have been viewed as legitimately controlled and/or regulated from the outside. (The reliance on measures through which migrant earnings were subject to external or social control rather than being individually accessed by workers themselves, proved to be long-lived: garnishee orders might be seen as its most recent variant, see Haupt & Coetzee 2008). Such attempts to regulate both the excesses of capitalism and the profligacy of consumers, however, could alternatively be seen as representing just another version of something that migrants and miners have long practised on their own initiative. Making efforts to save their wages from immediate consumption, they have ring-fenced them, investing them to prevent them from being frittered away (initially in cattle, and later through fixed deposits, ‘lay-bys' and furniture instalments) or turning them into a socialised asset by putting them into savings clubs.

Banking and Saving: Work Cycles, Life Cycles and Marital Arrangements

Migrants in need of systems for saving and transmitting money, long before ‘banking the unbanked’ initiatives were rolled out, have long been aware of and made use of them. Probably starting soon after labour migration itself, they have invested money in savings clubs. Founding and becoming members of what were known locally as stokvels, itimiti or megodišano – South African variants of a worldwide phenomenon – workers have grouped together to put aside some of their wages for specific purposes. Among Sepedi- and Ndebele-speakers, for example, groupings of men, mostly from the same village, formed rotating-credit clubs on the mines or, less formally, collected money to help return migrants' bodies home for burial (Delius 1996; Lekgoathi 2006, 2014). Starting in a later period after female migrancy began in earnest, women's clubs enabled members to pool their savings and gave each a turn to receive a payout (James 1999:60; Verhoef 2009; Mager & Mulaudzi 2012:308): the enforced savings system was explicitly used by women as a means of disciplining themselves not to use money in other ways. More recently, some clubs have been used by men and women as a means to buy ‘big ticket items' such as appliances, while others lend money at interest in an attempt to make the clubs' money ‘grow’ (Bähre 2007). Alongside other arrangements, and themselves increasingly acquiring a formal and more evidently financialised aspect (Krige Forthcoming), they have been one among many means of managing money.

The use of bank accounts, although commencing more recently, has also served as a means of judiciously storing and managing money. For men now retired who worked on the mines or in industry, selective use of savings or transmission facilities became common, as fieldwork in Mpumalanga revealed. One former miner spoke of being given a pay-slip at month-end which he would submit to the ‘time office’, exchanging it for cash. In some cases, he might deposit his pay in a building society account in order to put it aside, later drawing this out to buy telegraph orders from the Post Office which were sent home and which his wife could redeem locally (James 2014:108). In others, trusted drivers of mini-bus taxis were relied upon to repatriate earnings. Bank accounts were often used for saving, to hinder rather than enable an easy flow of money, or were opened and closed again as workers moved from one job to another or from rural to urban areas and back again. Sometimes the opening of a new account and the closing of an old one accompanied the switch from one spouse to another, or reflected the domestic distrust that went along with relationships that, although conflicted, were more long-lasting. Bank accounts were single- rather than multiple-use, and often served to block, rather than facilitate, the channeling of wealth.

The use of bank accounts was combined with other arrangements, intricate and requiring considerable powers of recall to manipulate and manage: households used a complex portfolio of arrangements to manage their finances (Collins 2008; Collins et al. 2010). People with commitments to kin or spouses sometimes strategised to make their money inaccessible to these dependents, by putting it in fixed deposits, arranging with their employers to help them commit to enforced savings practices (Krige 2011:137), putting money aside with a retailer in the ‘lay-by’ system (making a deposit on an item in the expectation of paying the rest of the price within a set time period or forfeiting the deposit (Roth 2004:72), or by paying instalments for furniture on hire purchase).

Migrant labourers' use of banks thus reflected marital, spatial and temporal discontinuities. But as time went by, increasing numbers of employers during the 1990s rendered these strategies more difficult to pursue by requiring that wages or salaries be paid into bank accounts (Porteous with Hazelhurst 2004:77, 81). At around the same time, the government, attempting to enable regularity of payment, similarly encouraged pension and grant recipients to open accounts (Breckenridge 2005), then requiring all recipients to re-register in the 2000s (Vally 2013). Being banked increasingly began to enable the unimpeded flow of money into the account at month-end, and out of it again. Wages paid directly into employees' bank accounts enable employees to borrow without collateral or ‘use their expected wages as a collateral substitute’ (Roth 2004:78). What were wealth stores have gradually become wealth conduits, enabling creditors to reclaim their advances. Among these creditors are the new microlenders who, as outlined earlier, began their operations with the lifting of the restrictions on usury.

Unexpected new players have also been drawn into the lending landscape: most strikingly the NUM, which has turned its focus to a new programme of financial support and services for members. One of the most recent in the series of schemes to keep migrants' wages from immediate spending, and a continuation of the kind of attitude that earlier found expression in ‘deferred pay’ (Schapera 1947:106–107), the NUM's TEBA bank was established in 1976 as a savings initiative for mineworkers. Now transformed into the NUM's own form of BoP financial innovation, it has been reconfigured as the Ubank. Established by the NUM's fully financialised investment arm in partnership with the Chamber of Mines, the Ubank has become a major source of credit for mineworkers, aiming to offer a legitimate alternative to the loan sharks (see below) by providing an array of financial products from payday loans to small enterprise start-up loans to mortgages. ‘Our vision is to be the workers' bank of South Africa,’ Ubank's then chief executive, Mark Williams, announced in 2010, as the bank embarked on an aggressive strategy of growth to augment its existing 500,000 clients by consolidating its dominance in mining communities and beyond.⁸ Initially dubbed the ‘workers' bank’ – an ethical counterweight to corporate banks cashing in on lending to the poor – Ubank's ‘moral highground’⁹ seems to have become rather more precarious, as rank-and-file NUM membership have proved to be more the generators of profit than its recipients. Broadly speaking, this reflects a shift from a register of collective action and worker solidarity to the promotion of individual self-actualisation and economic empowerment in line with the national corporate-sponsored discourse of patriotic capitalism. The images of empowerment projected in Ubank advertising portray an aesthetic of new South African citizenship, embodying the core virtues/values of upward mobility, youth and entrepreneurialism.

At odds with this aesthetic is the reality of life underground and in the shacks of the squatter camps where many miners live. For the poorest-paid mineworkers, and sub-contractors, low job security and low wages make repayment on an unsecured payday loan of up to a significant portion of the value of their net pay¹⁰ extremely difficult. Rock driller Mbuzi Mokwane claimed:

My payday is the most miserable day for me … on the day you're given a pay slip that says R4,000 [£300], you start to calculate your outgoings and you can go crazy. There is daylight robbery in the mines.¹¹

Heightened expectations spurred on by the post-apartheid promise of economic empowerment since democracy have put greater pressure on pay packets across the board (Daniels 2004:842; Posel 2010; James 2012), including for migrants themselves (Reddy 2014:258). But testimonies of migrant miners at the bottom of the pay-scale and under conditions of increasing insecurity show that lending and indebtedness in this sector can often be less about pursuing mobility and the fulfilment of aspirations than about getting by and supporting increasing numbers of dependents at a distance. These demands mean that migrant mineworkers have proved a lucrative target for formal lenders seeking to ‘bank the unbanked’.

Furniture and Hire Purchase

Investing in items of furniture and appliances – fixed assets substituting as the ‘next biggest thing’ where other forms of property were disallowed, and enabling the decoration of council-owned township houses in black townships – was a key aspect of the black South African experience, especially from the 1950s onwards. Pioneered in urban areas by town-dwellers seeking respectability (Krige 2011:138, 172), it was soon adopted by migrants as well. Where the latter had initially taken a conservative approach to investment in property and the purchase of material goods, often consolidating the wealth of the homestead by buying cattle (Ferguson 1992), those migrating more recently have tended to favour the purchase-on-credit of household furnishings. A bride's parents often provided an item of furniture as part of her trousseau, later investing in further items, paying each off in turn, with the eventual aim of equipping all rooms in the house. Although interest rates were and remain high (Schreiner et al. 1997), typically more than doubling the price, many householders have kept up their repayments in a prudent fashion. Furniture purchase thus contributed to a ritualised life-course as well as involving aspirations to sophistication and modernity. It exposed householders to gradually increasing expenditure, and expanding credit access, over time. Providing a means of ‘saving’ money by making it unavailable for other things, it could also however, when unregulated, lead to unsustainable levels of debt.

This credit system mixed unlike styles. Meticulous record-keeping and mailing out of invoices in brown envelopes reminded purchasers of what they still owed, but trust was also important, since the social and geographical distance between retailers and their customers in villages made it necessary to rely on intermediary/agents and lent the system a personalised dimension. Considerable profits could be made from targeting low-paid black migrants, but relying on these buyers, with their low earnings and – in some cases – reluctance to keep up repayments, also made business risky for smaller retailers. The high costs of hire purchase compensated for defaults and helped pay for repossession operations. Success in making money in this sector (despite these problems) attracted much competition (Tlali 1979:26, 30, 116), which made business owners increasingly determined to increase their profit margins over those of their rivals while simultaneously reducing what they paid to employees and agents.

Customers who might have fallen on hard times, when handed reminders of payment due, often threw them away or hid them under the bed, only to endure the shame of having items repossessed. The alternative – showing that ‘trust’ was often honoured more in the breach than the observance – was that some clients entered into ‘scams' with agents or paid them bribes to depart (Tlali 1979:82–83, Cohen 2004:42–46). From the customers' perspective, these forms of collusion were one way to escape the high interest rates charged. But such efforts did not necessarily make matters easier for customers. One such agent in a Mpumalanga village, despite having been fired by the retailer for crooked practice, continued to travel around to prospective customers, taking advantage of villagers' ignorance of his dismissal and continuing to pocket the deposits they were paying in expectation that their furniture would be delivered. The agent later left the area to go into hiding in order to escape their wrath once villagers discovered his trickery (James 2014:105).

By the early 2000s, despite banks' efforts to abolish credit apartheid by extending financial formalisation to all sectors of society, this furniture hire-purchase system remained in place to some degree. Increasingly since the 1990s, however, migrant consumers have been getting into hock to clothing retailers via store-cards and to microlenders instead of, or as well as, to furniture retailers. Those buying appliances often now do so using microloans rather than procuring credit from the furniture stores themselves on hire purchase. Indeed, many of these retailers – such as members of the JD group – have branched out into financial services and microlending as an equally or more profitable aspect of the business. Lending has become increasingly financialised, and customers' being banked means less risk to the lenders.

Borrowing: Informal Moneylenders

Informal moneylending has long been practised in South African township areas and rural villages alike. Not all accounts of this practice depict the lenders (mashonisas) as violent loan sharks, however (Roth 2004; Siyongwana 2004; Krige 2011). The structural factors glossed as credit apartheid have limited householders' options to borrow from the formal sector at reasonable rates, inclining them instead towards illegal moneylenders. Such loans are often cheaper than those available from formal lenders who add extra charges to cover their administrative costs and to counteract the risks of non-repayment (Roth 2004:52; Gregory 2012:394). Detlev Krige's research shows that neighbourhood lenders often have a personal connection to borrowers. This community embeddedness helps to ‘cap’ the interest rate, since loans are intended to be repaid at month-end but lenders often extend the loan without calculating an escalation of the interest rate. Doing so would make repayment increasingly difficult for borrowers, thus giving the lender a reputation for unfairness, increasing the chances that violence be used against him, and prompting complaints to the authorities (Krige 2011:154–158). Thus, ‘the termination of future contracts' by community members acts to regulate such moneylending (Roth 2004:99), and calculation of profit is regulated by community-mindedness (Krige 2011:154–158). The desire of the mashonisa to stay in business controls the terms under which he seeks repayment.

A generalised moneylending, less attached to specific individuals, became pervasive in urban townships and small-town settings in the former homelands from the 1980s onwards (see Roth 2004; Siyongwana 2004; Krige 2011:136–181). Borrowers and lenders cannot – except at the extreme ends of the continuum – be easily separated. Some start as one and later become the other; some are both but at different times and in different registers. Such lenders, operating beyond the system and aware of the illegality of their activities, nonetheless aim at greater economic formality themselves. Ironically, policy-makers' attempts to ‘bank the unbanked’ were here at odds with state regulation of illegal moneylending. Rebecca Matladi, a teacher of financial literacy, was often approached by small-scale lenders for advice on how to bank their own proceeds and securely store the proceeds of their enterprise without drawing attention to its illicit character (James 2014:113).

Smaller lenders often lend only small amounts and, being embedded in local neighbourhoods, must adjust their collection arrangements to fit in with local norms. They often lend amounts of less than R300, charge about 15 per cent interest monthly, are relatively flexible in the calculation of interest over time, lend to self-employed or low-paid people, and have no formal system of collateral. Such lenders do not compete with larger ones, their products are distinct and aimed at a different market (James 2012:25–26). But as moneylending expanded in size and remit, some lenders, modelling their actions on those of retailers in the formal sector, began asking the neighbours of prospective clients to stand surety, or started inventorising clients' assets with a view to repossessing these in cases of default (James 2012:113). From the mid-1990s onwards when credit apartheid was, in theory, coming to an end (Siyongwana 2004), illegal moneylending with interest began to acquire its more financialised techniques, aping or consolidating those being used by the new microlenders which were to be outlawed at the end of that decade. Lenders ask borrowers for their ATM cards as loan security, later returning these to their owners after withdrawing the money owed to them on payday. Whereas banks and regulated lenders require a pay slip before agreeing to offer credit, informal lenders do the equivalent after the event, by taking the borrower's card and withdrawing the money owed to them directly from the bank. (At month-end in some rural areas, people wanting to use ATM machines often find that these are being monopolised by mashonisas, who use a succession of cards from a variety of customers to withdraw what is owed to them.) Borrowers, shorter of money than previously, then often borrow again, once again yielding up their ATM cards (James 2012:10, 2014:113). When they try to escape by cancelling their ATM cards at the bank and applying for new ones, lenders, aware that it is impossible to get a new ATM card without an ID book, retaliate by asking to keep their ID books as well. The result can be a debt-bondage cycle embodied by the term sekôlôtô (from the Afrikaans skuld, debt). Although, being in need, borrowers willingly participate in these arrangements, they gradually come to resent the fact that they are ‘working for mashonisa’.

Many borrowers do not distinguish the registered, and more formal, microlenders (in the second sector discussed earlier) from the unregistered and illegal informal moneylenders or loan sharks (in the third one): they use the same word – mashonisa – to describe both. While small-scale moneylending with a more neighbourly feel certainly remains in place, its more streamlined version – with all the unsustainability that implies – seems to be on the increase, having escalated from the mid-1990s onwards (Siyongwana 2004; Barchiesi 2011:200, 210). The proliferation of such lenders, added to that of other possible credit sources, seems by many accounts to have contributed to the levels of indebtedness which caused such concern in Marikana.

Housing, Property and the Migrant Conundrum

Much decried by critics, the denial of property rights to black people in South Africa was an intrinsic element of credit apartheid. When migrants invested wages in buying furniture, this represented – among other things – a tacit recognition that titled property had been put beyond their reach. Instead of settling at their place of work, the prevailing structure of accommodation provided to lower-ranking migrants was one in which a resident was ‘expected to live as a “bachelor”’ within the confines of the hostel’ or compound, until he went on leave and returned ‘“home” to visit his family’ (Ramphele 1993:20). While a strong commitment to supporting rural homesteads played its part in this system of oscillating migration (Reddy 2014:256–257), the inability to own accommodation at the workplace was thought to play its part as well.

Under apartheid, the government had restricted the provision of family housing at the mines to no more than three per cent of the black workforce. The mining industry, sensitive to criticism that it had been complicit in perpetuating such a system and eager to erase the legacy of compounded labour, invested resources in upgrading hostel accommodation for migrant miners, restyling them ‘single accommodation villages' (Anglo Platinum 2008). But, while hostels were improved, less progress was made in the provision of family housing to low-wage mineworkers; something that had been promised as far back as 1944 by the mining industry. The industry protested the government's restriction on family housing, and expressed a desire to ‘experiment’ with settled labour at the mines, but failed to provide permanent family accommodation to even this small fraction of workers. In its submission to the Truth and Reconciliation Commission (TRC) hearing on business, Anglo American notes this failure, referring to it as a ‘missed opportunity to oppose apartheid and hasten its demise’ (Anglo American 1997:12). Yet in the two decades following the end of apartheid, progress on the transformation to family housing has remained slow (Fassin 2007:186).

Mining companies now make available, to permanent employees, a living-out allowance of around R1,800. Given the lack of rented accommodation, and the unaffordability of other options,¹² miners mostly rent shacks from slumlords in the squatter camps at the mines. This is especially the case in and around Rustenburg and Marikana where the platinum boom between 2004 and 2008 led to an inflation in the property market of over 100 per cent and a substantial shortage of low-cost housing. Pointing out that ‘many workers now have two families here and back home’, one migrant miner bemoaned ‘the past’ when the mine would supply a concrete ‘bed’ and meals. ‘Workers could send home to the rural areas most of their pay … [Now] on our small wages we have to pay for our own beds and meals.’¹³ Since the late 1990s the numbers of miners opting to live outside the paternalistic confines of the hostels has multiplied, even though the ‘living-out allowance’ provided is too little to support a family, let alone multiple dependents. The choice was simple, says Prosper Masinga, a shop steward at the platinum mines:

80 per cent I think live outside the hostels and many are in the informal settlements … if you live outside the hostels … you have your space. You have freedom.

To counteract this defection of employees to the informal settlements, some mining companies have sought a solution not in the extension of company-owned family accommodation – which is available, for the most part, only to manager-level employees – but rather in the market; in this case the burgeoning market for low-cost housing construction and BoP mortgage facilities. Anglo Platinum, for example, has expressed its commitment to addressing the deep structural reliance on migrant labour, by launching a low-cost new home ownership scheme to support lower ranking employees on lower wages in order to enable them to buy a family house (Benchmarks Foundation 2012). However, critics point out that many miners prefer to invest their earnings in building houses back home, alongside education for their families (Reddy 2014:260). The company-provided houses are also, in the case of the platinum mines, often of sub-standard quality.¹⁴ Celebrated among BoP advocates as vehicles for democratising the housing market, providing financial security, agency and education to those who previously had no access to credit, such schemes often fail to account for the widespread experience of foreclosure in the so-called ‘subprime’ mortgage sector around the world. In the context of rising job insecurity in the mining industry and the precarity of migrant life, employer-provided mortgages are likely to increase the vulnerability of migrant workers under the banner of delivering financial security and social cohesion. If schemes such as Anglo Platinum's were to work, mining companies would recoup more in interest paid on mortgages over a longer period of time than it cost them to build the housing, not only externalising the costs of providing and building housing for migrant employees onto workers themselves, but profiting from it. As the Benchmarks Foundation puts it, ‘does this represent an investment by Anglo Platinum in the community, or is it a community investment in Anglo Platinum?’ (2012:54). All-in-all, housing has become a fulcrum of the new dimension of economic precarity among the migrant workforce, making it simultaneously the target of attempts at financial education and BoP investment, and the rallying point for a new social struggle as the emblem of economic exclusion in post-apartheid South Africa.

The experience of migrancy by miners such as the striking rock drillers of Marikana thus differs in important respects from its earlier forebears. It is not only that miners' range of dependents has increased and their access to credit has widened while their means for repaying it – under the insecure conditions of contract labour and ongoing retrenchments – have shrunk; and that a far wider range of consumables are now considered essential. It is also that the mining companies no longer provide board and lodging to the increasingly substantial portion of their workforce that is subcontracted: the role of compounds has declined and miners are left to find their own accommodation and food using scarce resources in a setting of soaring prices. There is a sharp disparity between the promises of upward mobility through financial inclusion, and increasing precarity within the migrant workforce, as new forms of economic vulnerability combine with much older legacies of disenfranchisement in the mine economy, unsettling the stark dichotomy between job and no job (Cooper 1987:181).

Conclusion

The ideology that underwrites BoP finance (Prahalad 2004) claims an apparently universal ideal of economic and social mobility. But we have highlighted here how the project of ‘banking the unbanked’ takes on very specific local forms, as it grafts onto historical social, political and economic relations, drawing in and relying on the energies and resources of local actors and agencies as much as big corporate players such as banks and furniture retailers. These local responses embody the fact that migrant life has long required a careful balancing of responsibilities. Often thought of as caught between two geographical worlds and two divergent systems (although these of course are inextricably intertwined), migrants travel to earn a wage in a capitalist economy while finding ways to husband their resources and to honour obligations and reciprocities which arise in a seemingly less-than-capitalist one. Techniques used to keep wages aside and beyond migrants' control have not only been used by various agents – rural patriarchs, traders, government authorities, appliance retailers – but also at times eagerly embraced by migrants themselves, in the knowledge that these resources will need to be guarded for the upkeep of home, for future retirement, or for the costs associated with death and burial. However, if credit has a positive aspect in that it allows a migrant to ‘borrow speculative resources from his/her own future and transform them into concrete resources to be used in the present’ (Peebles 2010:226), the motives of those who extend such loans cannot necessarily be assumed to be benevolent. In South Africa, where the ‘advantage to creditor’ principle has long been in evidence (Boraine and Roestoff 2002:4), there are signs that an increasingly financialised capitalism has irrevocably transformed the landscape within which migrants make their calculations about family obligations and the life that is to come. There are hints, however – in the collapse of African Bank as this article was going to press and the downgrading by Moody's of the ‘big four’ banks as a result¹⁵ – that such financialisation will eventually have repercussions, in turn, for those creditors.

Acknowledgements

Part of this research was funded by a grant from the Economic and Social Research Council of the UK (award RES-062-23-1290). Opinions expressed are our own. Thanks to the School of Social Sciences, University of the Witwatersrand, South Africa, where Deborah James was a Mellon Distinguished Scholar and was working on aspects of the museum exhibition Ngezinyawo – Migrant Journeys held at the Wits Art Museum (WAM) from 10 April to 20 July 2014, for which a version of this article was written and while it was being completed. Deborah James gratefully acknowledges the help of Isak Niehaus, Sputla Thobela and Rosemary Ngunyula, and all those who spoke to her in the field. Dinah Rajak is grateful for advice and insights from James Nichol and Margaret Renn; Catherine Dolan, Jamie Cross, Alice Street and Paul Gilbert at the Centre for New Economies of Development. Both authors gratefully acknowledge those with whom they held discussions in their research settings.

Note on Contributors

Deborah James is Professor of Anthropology at LSE and a Mellon Distinguished Scholar at the School of Social Sciences, University of the Witwatersrand. She works on South Africa, in Mpumalanga and Limpopo provinces and their urban hinterland, the Witwatersrand, and has a special interest in popular economies.

Dinah Rajak is a Senior Lecturer in Anthropology and Development at the University of Sussex, UK and is a co-founder of the Centre for New Economies of Development. Her research concerns the anthropology of global capitalism, development and new moral economies, with a focus on the extractive industries and South Africa.

Notes

1. David Smith ‘Anglo American sheds 15,000 jobs as profits are hit by falling metals prices' The Guardian 31 July 2009. See also Haroon Bhorat and Morne Oosthuizen: ‘the growth in the number of jobs has been far outstripped by the expanding labour force. While employment for some groups has grown rapidly relative to the increase in the size of the labour force, resulting in relatively high labour absorption rates (specifically, non-Africans and White females in particular), none of the groups investigated experienced actual employment growth even in the region of the target growth rates. Thus, unemployment rates have continued to rise for these groups. The fact that the economy has been unable to provide sufficient employment opportunities for the South African population means that even greater demand for job-creating growth will be placed on the economy in the medium-term’ (2005:154).

2. In 2011 almost a third of Lonmin's workforce – that is 9,131 full-time contractors of a total workforce of 33,046 – was sub-contractors (Benchmarks Foundation 2012:72). While Anglo Platinum have drastically reduced their use of such labour from 14,014 in 2009 to 5,513 in 2012, this may well reflect (and therefore veil) mass retrenchments at the Anglo Platinum mines in the wake of the financial crisis in 2009/10 (Benchmarks Foundation 2012:45).

3. Lynley Donnelly ‘NCR raises alarm at unsecured lending in Marikana’ Mail & Guardian 11 October 2012.

4. In the absence of investment in manufacturing and production, financialisation has been seen as accounting for South Africa's ‘jobless growth’ during the same period (Marais 2011:124–128, 132–139).

5. Tim Cohen of the Financial Mail pointed out that ‘we now know … that this level of microlending is unsustainable’, although the former CEO, Leon Kirkinis, defended the industry, saying that ‘the unsecured lending market had grown from about R30bn in 2006 to over R159bn by 2012', Tim Cohen ‘Editor's note: Un-Abil to unwind’ Financial Mail 14 August 2014. Soon after African Bank was placed under curatorship, Moody's downgraded South Africa's ‘big four’ banks by one notch to Baa1. The financial strength rating of Capitec, the largest unsecured lender left standing after African Bank's collapse, was also slashed by two notches. Agency Staff ‘Moody's downgrades Standard Bank, Absa, FNB and Nedbank’ Business Day 19 August 2014. In seeming disregard, one of African Bank's executives, Tami Sokutu, estimated to have made more than R50-million in share options alone during more than a decade at the bank, indicated that he cared little about those blacklisted because of their inability to repay, or investors whose savings in the sector overall would now be affected. Thekiso Anthony Lefifi ‘“F*** the poor!” is the message from a top executive at African Bank’ Times Live 17 August 2014.

6. The Zulu word mashonisa relates to the verb stems -shona (to sink, become poor, die) and -shonisa (to impoverish, cause to become poor) (Dent & Nyembezi 1969:481). It may be translated as ‘one who impoverishes' or who ‘takes and continues to take indefinitely’ (Krige 2011:144). In popular parlance the plural is mashonisas (Krige 2011:151; Siyongwana 2004:851).

7. Maya Fisher-French ‘Teba gives birth to Ubank’ Mail & Guardian 7 October 2010.

8. Ann Crotty ‘Ubank branch shut at Lonmin's Marikana hostel’ Business Report 16 October 2012 <http://www.iol.co.za/business> (accessed 13 March 2013).

9. Ibid.

10. Lisa Steyn ‘Marikana miners in debt sinkhole’ Mail & Guardian 7 September 2012: ‘In accordance with the national credit regulations, a maximum of R1 257.50 in interest and fees can be charged on a short-term loan of R1 000 – that is more than 25% a month, or 300% if annualised’.

11. David Smith ‘South Africa's mines: gulf between employees above and below ground’ The Guardian 26 September 2012.

12. David van Wyk ‘After eight debate’ SAfm 8 April 2014.

13. Solly Mapaila ‘A Marikana story that isn't being told’ Umsebenzi Online <http://www.sacp.org.za/main.php?ID=3736#one> (accessed 7 September 2012); see also Reddy (2014:257–258).

14. David van Wyk ‘After eight debate’ SAfm 8 April 2014.

15. See note 5 above.