Debt/credit,money and social relationships in the underground credit markets of Soweto,South Africa

Introduction

This article offers an ethnographic exploration of underground credit markets in the neighbourhoods of the urban township of Soweto in Johannesburg, South Africa. The workings of one small-scale neighbourhood lending business, as its owner conceives it, legitimizes it as a form of ‘entrepreneurship’, describing how he calculates interest, negotiates the terms of loan repayments, utilizes collateral, frames trust as a question of information, deals with risk and reputation, and relates to the borrowers and law. The reason why lenders borrow small amounts of money as loans is linked to the general economic precariousness among working class, poor and unemployed Sowetans. The moral evaluations of underground credit markets are considered and I argue that the vilification of the township moneylender (mashonisa) by trade unions, consumer activist groups and other civil society organizations (Madlala, 2005; Mohale & Moledi, 2005; National Consumer Forum, 2001a, 2001b; Nott, 2004), who describe him as unscrupulous, profit driven and individualistic, sits uneasily with many Sowetans’ view of the township lender. For them, ‘their’ mashonisa could be a necessary if expensive point of access to cash money in moments of financial crisis.

More generally, some Sowetans say that not being able to access money via a mashonisa may result in social death. As someone who ‘makes money from money’, offers loans of usually small amounts of cash to neighbours, family and friends at illegal interest rates, and operates beyond the purview of the law and the state, and yet has some local legitimacy, the moneylender is an important source of credit for working class and unemployed Sowetans, many of whom have yet to be incorporated into the formal financial and banking system (James, 2014; Siyongwana, 2004). However, the public condemnation of the township lender has in fact contributed to the ongoing financialization of the poor people’s money in South Africa. Since the 1990s, the state and civil society actors have encouraged poor and working class South Africans to access credit and deposit their savings through the formal banking system and thus to end a range of popular financial practices that have emerged during the colonial and apartheid era. Reluctant at first to give in to the political pressure placed on them by the ruling government to offer affordable banking services to this market segment, banks in South Africa have since the 2000s eagerly participated in the financialization of poor people’s money. This was made possible by crucial changes in legislation and lessons learned from a home-grown formal micro lending industry (James, 2014; Schraten, 2014), a worryingly close relationship between banks and the judicial system, and a much wider re-evaluation of poor people not as a risky market segment but as a source of considerable profits. As David Graeber (2011) and others have pointed out in the wake and aftermath of the 2008 financial crisis, it is the poor who typically repay their debts.

While a growing body of academic scholarship is critical of the global project to get the world’s ‘unbanked populations’ ‘banked’ (Elyachar, 2005; Peebles, 2014; Schwittay, 2011a, 2011b), the figure of the mashonisa has been deployed by powerful actors in South Africa to push for the criminalization of underground credit markets and the financialization of poor people’s monies. Nevertheless, Sowetans, are not as eager to do away with ‘their’ moneylender, and are not averse to embracing seemingly expensive credit relationships that are both personal and socially legitimate, even if somewhat shameful. This more positive evaluation of local moneylenders and accessing credit through such actors is in part a consequence of a widespread view that has been articulated in ethnographic studies of the region that show that money is not regarded to be antithetical to social relationships. These studies show that indebtedness, including the power differentials and hierarchies implied by the fact of being indebted, is a feature of all social relationships –what James Ferguson (2015) has aptly named ‘declarations of dependence’. Without wanting to develop a communitarian argument in which I romanticize the small moneylender in the face of the impersonality of the formal financial system, I suggest that the ethnography I present here echoes discussions about the role of debt and money in society (Graeber, 2011; Gregory, 2012; Hart, 1986, 2009, 2017a; Maurer, 2006; Peebles, 2010; Roitman, 2005).

Peebles, in his seminal review (2010), demonstrated that recent anthropological work insists on exploring the ‘credit/debt dyadic nexus’ rather than either debt or credit in isolation from one another. These works, Peebles shows, have revealed how the debt/credit nexus is productive of social ties, allegiances, enmities and hostilities. Anthropologists have been able to arrive at such conclusions, as they were reluctant to make normative pronouncements concerning whether credit is liberating and whether debt is debilitating (2010: 234). Instead they reaped the benefits of ‘not separating the economic effects of credit/debt from the moral debates over it’ (2010: 234), unlike economists who tend to favour material effects. The ethnographic record presents an incredibly wide array of moral pronouncements about credit as being beneficial and liberating and a source of productive power on the one hand, to debt seen as burdensome, imprisoning and a destructive weakness on the other. To further complicate matters, anthropologists have found research participants lauding lending as a form of saving (Shipton, 1995: 249). As Gregory has noted, there is much confusion about the actual difference between debt and credit. The only people who are not confused about the distinction are successful moneylenders ‘because their daily bread depends on it’ (2012: 383). For Gregory, credit exists as a potentiality that belongs to the future. Debt is credit realized, and then becomes history, ‘credit as the future in the form of potential debt and credit as history in the form of actual debt’ (2012: 384). Gregory further argues that the ‘positive moral evaluation of credit across cultures is grounded in the commonplace that traders in money everywhere must buy cheap and sell dear if they are to survive’ (2012: 395). This is why when we want to say moneylending is a good thing we use the word credit (as in microcredit), and when we want to say moneylending is a bad thing, we use the word debt (as in microdebt). ¹

In The Gift, Marcel Mauss (2002) discussed the dual if not paradoxical nature of exchange relations that produce debt and credit: they build hierarchy and dominance at the same time as they produce social ties and group solidarity. Mauss’s suggestion that there may in fact be something productive about debt has been taken up by some scholars, notably Roitman (2005), who has written about how moral evaluations of different forms of wealth create a system by which some people are included and others excluded. Debt here is productive in ways far beyond the economic. Similarly, I argue that the public and moral condemnation of the figure of the mashonisa assisted in the attempt to incorporate the cash monies of poorer and working class Sowetans into the financial system, ultimately with the aim of getting them ‘banked’ by turning them away from their township moneylender. The scholarly re-evaluation of debt, from theorizing debt as a burden and bondage that produces inequality, deepens power differentials and produces feelings of guilt and shame, has been accompanied by a re-evaluation of the distinction between gift exchange and commodity exchange in anthropological theorizing. As Peebles (2010: 228) puts it, ‘much recent work has started to look at the market itself as a place that creates credit/debit bonds between people’. Moreover, it is no longer so easy to retain an ideal–typical dichotomy between gifts and commodities as was famously defined by Gregory (1982) whereby gifts ‘establish[es] personal qualitative relationships between subjects transacting’ and commodities ‘establish[es] objective quantitative relationships between objects transacted’ (Peebles, 2010: 229). Mauss was concerned, as Hart (2007) has shown, not by the establishment of an opposition between gifts and commodities but in tracing the evolution of contracts in exchanges, paying close attention to the ‘non-contractual elements in the contract’ that jurisprudence tends to obscure. In the analysis of underground credit markets in Soweto I examine the social infrastructure (Simone, 2004) making possible contractual yet social agreements between lenders and borrowers in underground credit markets, as these are embedded (Polanyi, 2001[1944]) in shared social fields of action and meaning. In the case of Soweto, these social fields are not in any way organic, but have developed over three to four generations among a cosmopolitan population thrown together as a consequence of a brutal history of forced removals, social dislocation, underdevelopment and political subjugation (Krige, 2011).

Soweto: History, money and indebtedness

If the (former) African municipal township of the south western areas of Johannesburg – Soweto – resembled anything as a form under apartheid, it would be a temporary labour camp constructed by the politically powerful to host and control temporary visitors to the city, whose presence was tolerated solely as a consequence of their ability to sell their labour, and cheaply so. The legacy of Soweto’s origins as a labour camp hosting temporary African visitors to the ‘white declared city’ from apartheid mandated ‘ethnic groups’, situated in the rural areas, remain sketched on the economic landscape of contemporary Soweto, even as it has been amalgamated into the Johannesburg Metropolitan Region (Alexander et al., 2013; Marais, 2011). One aspect of this history is the uneven incorporation of working class and poorer Sowetans into the commercial banking system, a system to which they were denied access and which informed the development of a number of popular financial institutions and practices that scholars have typically described as ‘alternative’ or ‘informal’. The history of black South Africans’ relation to commercial finance remains to be written (see Porteous & Hazelhurst, 2004; Verhoef, 2009), but several factors prevented Sowetans from accessing credit during apartheid. Sowetan workers were typically paid wages in cash and in kind, but were often refused formal credit outside the workplace and the trading store. Other barriers to credit access were part of the effort, deployed particularly by the apartheid government, to obstruct the development of a black entrepreneurial and middle class in the cities. As the so-called Homeland system came into being, greater forms of credit became available to African businessmen operating in the Homelands at the expense of the development of urban townships, such as Soweto (Posel, 1991).

Subsequent to the 1976 Soweto Uprising and the 1980s ‘Era of Reforms’, new avenues for credit access did open up for Sowetans, although access to consumer and emergency credit took a back seat to the benefit of what the ruling elites considered to be more productive forms of credit. Retail stores, such as Ellerines and Edcon, had by then already pioneered retail credit among the urban black populations. But it was during the late 1980s that non-governmental organizations (NGOs) such as the Get Ahead Foundation and commercial lenders such as Group Credit Company, Louhen Financial Services and Credit Indemnity started offering consumer credit (Porteous & Hazelhurst, 2004: 80). The commercial lenders at the time ‘sensed market opportunity arising from the aspirations and needs of a growing, increasingly upwardly mobile urban population that was largely excluded from the traditional bank-based credit system’ (Porteous & Hazelhurst, 2004: 80), even as these were labelled as ‘high-risk’. However, these lenders operated outside the law, as the interest rates they levied on loans were prohibited under the Usury Act of 1968. In 1992, prior to the 1994 democratic elections, the then Minister of Finance signed into law an exemption to the 1968 Usury Act which effectively legalized the lending practices of these commercial micro-lenders. Under the exemption, lenders could charge usurious interest rates on shorter than 36-month loans that were less than R6,000. This was an important step in the commercialization of micro-lending and the success of some of the earlier micro-lenders ‘led to a frenzy of new lenders on the JSE Securities Exchange in 1997/8’ (Porteous & Hazelhurst, 2004: 81). It also led to the mushrooming of several Ponzi-type schemes such as Marburt and Krion Financial Services that promised investors that their investment monies were being used as capital for microloans or for bridging finance (Krige, 2012). The explosion of consumer credit in the market is evident in some of the industry analyses of the time: one report estimated that there were 3,500 formal micro-lenders operating by 1997, an increase of 192% over 1995; moreover, the estimated turnover of the industry over the same period had almost trebled from R3.6 billion to R10.2 billion and there were around 25,000 informal micro-lenders (Porteous & Hazelhurst, 2004: 82).

Since the abandonment of apartheid policies and institutions, drastic changes have taken place in the legislative and commercial environment when it comes to Sowetans’ access to consumer and productive forms of credit. As James (2014) has documented, the state and non-state financial services institutions have constructed a home-grown microcredit market that promised to offer to citizens access to mainly private (as opposed to donor-funded) consumer credit in a highly regulated environment. This followed a phase during the 1990s in which an unregulated microcredit market trapped many working class Sowetans in debt. The new and highly regulated microcredit market had been constructed by the state and financial regulators so as to bring down levels of consumer indebtedness, to cultivate higher savings ratios among the general population and to reform usury regulations. The state defended the construction of the said consumer market by demonizing and attempting to eliminate microcredit providers (including township moneylenders) that failed to register with regulators and continue with their ‘predatory lending’. Along the way, public authorities and consumer activist organizations developed a strong discourse to justify their regulatory moves. They focused their attention on how ‘vulnerable consumers’ needed protection from ‘predatory lenders’. The figure of the mashonisa exemplified one kind of predatory lender identified in the public discourse and its condemnation became an important aspect of the financialization of poor people’s monies (see Beresford, 1999; Hlatshwayo & Zulu, 2002; Joffe, 2002; Klein, 2000; Simanowitz, 1998).

For many unemployed and poor South Africans, accessing credit money from a regulated street-level credit provider was the first step to opening up a bank account and become ‘banked’. The other important avenue the state used to incentivize poor people to ‘get banked’ was through the extensive state-funded social security system (Vally, 2016). The emergent regulated microcredit market thus helped to incorporate into the formal sector monies that circulated among and in township communities – at very profitable margins for the financial institutions. By all accounts the flows of money inside township neighbourhoods across South Africa are considerable. The state had other reasons too for regulating underground credit markets and money flowing between groups of people rather than through bank accounts. Incorporating underground credit markets into the commercial financial sector, it was said, was meant to assist in building external (international) confidence in the South African financial sector. External confidence in this sector is of utmost importance given the increasing role that financial institutions have been playing in the South African economy, as the contributions to gross domestic product (GDP) from manufacturing and agriculture continue to decline (Ferguson, 2015; Marais, 2011). Stability within the financial system contributes to the way risk calculations are made by international credit rating agencies. Good ratings by international credit agencies allow the state to borrow money from global capital markets at lower cost. In this way the regulation of underground credit markets was linked to the financialization of South Africa’s economy and poor people’s monies. In the same way that effective formal financial institutions are crucial mechanisms for the incorporation of South Africa’s economy into the global economy, the effective regulation of underground credit was also central to the incorporation of the unbanked sections of the population into the commercial banking system. It was argued that incorporating the unbanked into the financial system would bolster the system by putting monies circulating outside the system to ‘work’. The mere size of the informal sector at the time – with 60% of the population said to belong to ‘the unbanked’ in 2000 – was reason enough for the financial services sector to want to incorporate ‘the unbanked’ (Porteous & Hazelhurst, 2004: 24), even as the government cajoled them into developing more affordable services. The state, through various legislative and regulatory reforms, argued that its aim was to ‘increase access’ to banking services and commercial credit while at the same time offering people ‘consumer protection’ coupled with ‘economic freedom’.

Despite the concerted efforts by public authorities and the banking sector to reform legislation and to create incentives for working class and poorer Sowetans to hand over the management of whatever cash they have to the formal financial system, Sowetans have rather doggedly refused to get rid of the various practices they have developed over time to deal with individual, household and community financial management. This is true, not only for the underground township moneylender, but also for other financial institutions in Soweto’s ‘popular economies’ in the mid-2000s (Krige, 2011). A good example is the case of rotating savings and credit associations, which have a long history among African communities in urban and peri-urban South Africa (Bähre, 2007; Kuper & Kaplan, 1944; Lukhele, 1990; Verhoef, 2001), and are also common among urban communities in Africa (Ardener, 2010; Chipeta & Mkandawire, 1992) and Asia (Geertz, 1962). Since the 1990s attempts have been made to incorporate such stokvels into the financial system and it was expected that these neighbourhood-level associations, which rely on personal face-to-face between people and which are embedded in neighbourhood and township social networks, would cease to exist. It was assumed that Africans would en masse opt for the modern and impersonal institutions of commercial finance. While Sowetans certainly embrace the opportunities offered by banks and insurance companies (Bähre, 2012; Collins, 2005) and today make use of all available financial and banking services, this has not resulted in the demise of the stokvel, the colloquial name for rotating savings and credit associations. In fact, there has been an upsurge in the popularity of these mutual financial associations to the extent that banking institutions now develop tailor-made financial services for such groups, with the ultimate aim of persuading such associations to stop stacking their collective cash savings under their mattresses.

Concurring with Peebles’s (2014: 600) argument that ‘it is time to stop exclusively pitying the unbanked and dropping our jaws at their collective “ignorance” for not banking’, perhaps we should ‘start respecting the logical conclusions they draw about why they do or do not turn to banks’. As Peebles (2014: 602) recounts, ethnographic work has shown how people in Africa and elsewhere ‘store economic value in eminently rational ways outside the banking system’, including elements of resistance by insisting that ‘structures other than their own need reform’. In concluding his argument, Peebles (2014: 609) emphasizes that ‘unbanking, in many instances, may therefore be more fruitfully seen as a form of active resistance against certain types of subject formation and the corresponding attempt to redraw community boundaries, or to destroy a prevailing form of financial mutuality’. Yet unbanking is also increasingly difficult in an urban context like Soweto where financial and telecommunications technologies are converging and integrating, such as in the disbursement of social security payments (Vally, 2016).

Growing inequalities within the general population, along with a decline in opportunities for wage work, amongst Africans in South Africa were pointed out by Ferguson and others (Barchiesi, 2011; Ferguson, 2015; Hunter, 2010). Since the 1970s South Africa has changed from a ‘labour-shortage economy to [a situation of] massive labour surplus’ (Ferguson, 2015: 4). While fewer people are working, more money is certainly flowing into Soweto’s neighbourhoods than ever before. This is in part the result of greater commercial property development, especially shopping malls, but also because of infrastructural investment by local and national government. In Soweto, as elsewhere in South Africa, the redistributive aspects of the government’s social security programme have become a major source of income for many households (Hunter, 2010). ² In this context of greater flows of money to and within Soweto, amidst the decline of wage earning opportunities, it has become more important to be able to make a claim on other people’s income than to hold down a job (Ferguson, 2015; Neves & Du Toit, 2012). In a capitalist urban economy where people are reliant on market exchange and money to access food, and where access to housing is also increasingly dependent on money in order to cultivate patron–client relationships with local and city-level politicians and officials, we see the emergence of two tendencies: the increased intensification and dependence on both money and social relationships. In all likelihood, wealthy Sowetans have recourse to the impersonal and bureaucratic nature of commercial financial institutions in order to evade or lessen claims that dependents including relatives can make on their income (colloquially known as ‘black tax’), whereas poorer Sowetans are pushed into intense personalized relationships and flows of money.

The relationship between money and social relationships has long been a concern for academic scholarship. In the fields of economic anthropology and the sociology of money, recent work has placed this question at the heart of the theories about capitalism, money and market exchange (Guyer, 2004; Hart, 2017a; Maurer, 2006; Zelizer, 1994). In their introduction to Money and the Morality of Exchange (1989), Parry and Bloch argue that most scholars from Western traditions share the view that money acts as a powerful agent of profound social and cultural transformations, with the power to revolutionize society and culture. For Simmel, they claim, money was important in the development of the cognitive world we now live in, as it helped to promote rational calculation in social life. For Marx, on the other hand, money’s power was subordinate to the phenomenon of production for exchange, but he nonetheless associated money with inevitable individualism and the destruction of solidary communities. Parry and Bloch (1989) are critical of this tradition represented by Simmel and Marx that attributes to money the growth of individualism, the destruction of bonds of solidarity between persons and households, and changes in the way humans who use money think. Bloch (1989) and Hart (2009) use similar examples about money, intimacy and sex to show that in Madagascar and Ghana respectively, money is not associated with impersonal transactions and is not awarded the power to transform persons. Parry and Bloch (1989: 8–9) argue that the impersonality of money is something peculiar to Western intellectual traditions and culture, because there ‘money signifies a sphere of economic relationships which are impersonal and amoral, and which constitute a separate domain’. They argue that in societies where the ‘economy’ is not seen as a separate domain, but where it is embedded in the society’s relationships, money and monetary exchanges are unlikely to be expressed as antithetical to social ties formed through kinship and friendship. This point is borne out by the existing ethnographic archive on southern African societies, which shows that in various contexts money has been used to initiate and strengthen social relationships through processes of conversion (Ferguson, 1985; Kiernan, 1988) and claim making (Bank, 1997; Khunou, 2006), as much as money may also be viewed to be detrimental and destructive to relationships (Krige, 2014). This is so despite the historical presence and contemporary significance in southern African societies of traditional forms of Christianity with its doctrinal opposition to money and material acquisition, and despite a long history of wage labour. Yet even in societies where there has been a concerted political attempt to separate the economy from society as documented in The Great Transformation (Polanyi, 2001[1944]), where the sphere of impersonal market exchange is ideologically opposed to the sphere of intimate and personal relationships (Carrier, 1990), money is not always seen as the antithesis of sociality (Peebles, 2010; Zelizer, 1994).

The case of Seb, a small-scale neighbourhood moneylender

One day, during the course of my field research in 2004, I was visiting a friend, Loli, at his house in one of the predominantly Zulu-speaking neighbourhoods of Soweto. There, I was introduced to Seb who turned out to be, amongst other things, a small-scale neighbourhood moneylender. I had met Loli several months before at a club where he was working as a bartender. That the club was within walking distance of the backroom I was renting for the 24 months of my stay in Soweto was very convenient. Over a period of more than a year I became better acquainted with Seb and he was very interested in the research I was undertaking about his ‘line of work’. I had several formal interviews with him, spent time hanging out with him and his friends, and managed to persuade him to start keeping a loan book in which to record his loans and repayments.

At the time of my field research on underground credit markets in Soweto, Seb was 35 years old and held no formal employment. Like many other young Sowetan men in his position, he was unlikely to be ever employed in the formal economy. He had a criminal record, did not speak much English and had not completed secondary school. During the day, the tin shack in the front of his mother’s house where he slept doubled as a tuck shop (spaza) from where he was selling bread, eggs, sugar, carbonated cool drinks and beer when he had stock and when his refrigerator was working. He did not have a father or relative who could support his business endeavours by extending his capital. He could not write an adequate business plan and approach a commercial bank or a state-funding agency for a loan. Yet he had an entrepreneurial drive, and, as I learnt later, engaged in petty theft and dealing in stolen goods. Together with the money he earned from selling consumables, his major source of income was from dealing in illicit goods and his moneylending business no doubt assisted him in justifying his income, however meagre that turned out to be.

Seb started lending money at interest after realizing that, even though he was formally unemployed, he often had some disposable cash. He lived a frugal life, did not have to pay rent, and did not need much money to get by. Some friends of his realized he was ‘sitting on money’ and they started borrowing small amounts of money from him, R10 or R20. ³ Such loans carried no interest as they were transactions between friends. Unsurprisingly, some friends did not settle their outstanding loans. They ‘played’ elaborate games in order to avoid him. Yet more friends and neighbours started arriving at his shack with similar requests for loans, ‘crying’ about the fact that they had no money, some even using the old political slogan of ‘hasina mali’ (‘we have no money’) used by civic organizations during rent and municipal boycotts from the 1940s onwards. Frustrated because of unpaid loans to friends, Seb nonetheless realized the ready demand for credit among his neighbours and the opportunity to ‘make his money grow’. This led me to query how Seb dealt with the difficulties arising from combining friendship and income generation.

Lending money to friends, Seb complained, was an arduous and problematic affair. In his own words, friends ceaselessly ‘took advantage of’ and ‘abused’ his friendship: ‘They cry a lot when they come here’, he told me. Seb said that these difficulties aside, being a moneylender was a straightforward if not easy business: you do not require substantial amounts of capital to start with and once you are able to extend and retrieve loans effectively, all you have to do is ‘sit and spread your money’. You do not require a certificate of some sort in order to register or any form of (expensive) training, nor did it require a lot of work –‘the money grows by itself’, he said. Advertising was not necessary, as word spread through the neighbourhood. Besides, old clients regularly came around to introduce a new client, hoping to stay in Seb’s good books by recruiting new borrowers. ⁴

Borrowers took out small cash loans, ranging between R20 and R500. This was partly because of the constant shortage of capital Seb himself experienced, thereby making it impossible for him to extend more than a few, small loans at any given time. But it was also because he had no proper recordkeeping system and could only remember the details of a few loans. The recurring delays in loan repayment by some debtors also put a strain on his already limited capital funds. It was risky for him to lend large amounts of money – he knew from experience that larger loans were more difficult to recover. It was less risky for him to spread his capital over several smaller loans than investing it in fewer larger ones. Moreover, borrowers owing large amounts, with sometimes considerable interest added on top, were more likely to seek protection from the state or an outside party when struggling to repay a loan; a situation he preferred to avoid.

In the public discourse, the high interest rate typically charged by unregulated moneylenders on loans is used as a key argument in the vilification of the township moneylender. As Keith Hart (2017b) noted in his reflection on learning the business of moneylending in urban Ghana by doing it himself, the headline interest rates in underground credit markets that are mentioned in public do appear astronomical. But the key factor in the business of moneylending is the default rate and a moneylender tends not to reveal this rate ‘since he depends on the mystique that he does not allow default’ (2017b: 218).

Seb had a standard and fairly simple way of calculating interest (or ‘penalty’ as he sometimes called it). Take for example a small loan of R100. For such a loan, Seb charged no interest when the full amount was returned to him within seven days. After the seventh day, however, there was an immediate ‘penalty’ of R50 which Seb added on top of the original loan amount of R100 borrowed. This 50% ‘penalty’ on the original sum effectively covered only a seven-day period and was only activated after the seventh day. After the 14th day, or two weeks, the penalty increased from R50 to R75. The amount payable to Seb was then R175. Should another week pass, the penalty would increase by a further R50 to R125, so that the repayable amount after four weeks would total R225. This translated into 125% interest on a loan over four weeks. After four weeks, no further interest or ‘penalties’ were added. This meant that Seb did not charge clients more than R225 repayment on a loan of R100. Such practice raises the question of why a limit or cap existed.

Seb offered me a number of reasons for this limit. First, he had learnt from experience that when the penalty increases indefinitely, the amounts inevitably became too high for borrowers to be able to realistically repay him. Second, since he operated in an environment where his clients were neighbours and friends, such a ceiling left room for further negotiations on the terms of payment. This flexible approach is of great importance for any neighbourhood lender in the vicinity in order to cultivate a reputation for fairness. Despite the popular image of the moneylender as an unscrupulous and profit-driven operative, it is important for a neighbourhood lender not to be marked in his immediate community as an exploiter or as anti-social. Such a reputation would not just reduce his potential clientele in a neighbourhood where other lenders also operate, but could result in existing clients refusing repayment as a group. Moreover, as the practice is embedded in neighbourhoods where values and practices are regulated by social perceptions, reputations and the possibilities of witchcraft accusations if not community action, lenders do not want to make enemies. Third, it is also the case that when amounts became too big, debtors are more likely to have recourse to the police in their efforts to annul or question the loan arrangement. Seb recounted to me an incident when he explained this to me.

A few months prior to our meeting, one of Seb’s clients who had taken out a rather large loan refused to repay Seb according to their contractual agreement. The client wanted to pay back just the original loan amount and not the penalty amount. Needless to say, Seb refused and insisted on the agreed upon loan and penalty repayment. Unwilling to oblige, the client went instead to the nearby Jabulani police station where he reported the case and provided the police with Seb’s address and his complaint. The police officers arrived at Seb’s home and questioned him about the loan. After learning about the nature of the loan contract they ordered the debtor to repay Seb the original amount plus interest. In my experience it was only in exceptional instances that repayments transcended the arrangements made between lenders and borrowers. In just a minority of the cases that I followed were repayments settled through police mediation. In some instances, the matter was taken to a local ward councillor or even a local priest. Importantly, whether these cases were concluded through the mediation of local authority, they rarely – if ever – were resolved in relation to the specifications of the then Usury Act. When I questioned lenders and borrowers and ward councillors about the legal prohibitions against usury, they were generally completely unaware of the various stipulations of the Usury Act (see Lukhele, 1990). In most cases repayments were settled without outside mediation. In instances where clients took longer than the agreed-upon time to repay a loan, Seb sometimes granted additional time for repayment. As such, the terms of repayment seemed always to be negotiable. When such negotiations failed, the next step for Seb was to visit the debtor’s house and confiscate (or ‘repossess’ in his words) consumer items belonging to the debtor approximating the value of the loan, such as a refrigerator, television set or DVD player. Seb usually retained such items for three days, allowing time for the client to make additional plans to come up with the money to settle the loan, before selling these items to ‘write off’ the loan. Seb denied that he utilized the threat of violence to force clients to make repayments. But it would not be farfetched to imagine that at times it was his sole (or last) resort.

Lenders: Entrepreneurialism, legitimacy and information

It is important to note that here I do not differentiate between full-time moneylenders and those who occasionally lend money. The term mashonisa is generally used in popular parlance to refer to any person who lends out money informally, whether that person does it in conjunction with any other business or not. Seb conceived of his illegal and usurious lending practice as a ‘business’ and an entrepreneurial activity. Although the interest rates he charged were certainly high and usurious in terms of the then Usury Act (and thus illegal), his practice was not regarded by his neighbours and clients as exploitative. Borrowers and even some police officers and ward councillors regarded his loan arrangements as above-board and socially legitimate if not entirely acceptable. Both lenders and borrowers at times made use of police officers at Sowetan police stations to formalize their (illegal) loan contracts. Such practices were indicative of the historical role the police play in urban township communities as local brokers and adjudicators rather than as enforcers of state law. As a consequence, the police officers’ participation in the underground credit market added some social legitimacy, if not legal sanction, to these practices.

Establishing whether lenders did in fact make a considerable income from lending out small amounts of money at interest was very difficult for me, and it depends much on the rate of repayment of loans. Lenders were certainly not going to boast about their income from such moneylending to me, in all likelihood fearing an outcry from the community and the audience of my research. I did not see Seb getting rich from extending small loans to neighbours. It may have helped him in purchasing a car one day, if he saved his profits judiciously, but his income from moneylending was too precarious and insignificant for him to ever grow rich from it. Moreover, I cannot draw many correlations based on the number and variety of lenders I interviewed between income levels, social class, ethnicity, education levels and informal neighbourhood credit providers. But these were also not the primary questions I wanted to ask, nor the questions my research methodology were designed to answer. I limited my research focus to small-scale neighbourhood lenders, excluding larger moneylenders associated with the taxi industry and with the organized criminal underworld, informally known in Soweto as ‘The Mafia’. Thus, one variable I used in order to differentiate between various lenders was the scale. At the one end of the scale were the lenders with sizeable amounts of capital who were able to lend large amounts of money to many borrowers. On the other were the small-scale operators like Seb who made small loans to a few neighbours. The larger the scale of a lender’s operations, and the size of his loan book, the likelier he was to also be involved in other capital-intensive businesses, whether legal or illegal. These lenders typically also run shops, taverns (shebeens), own a few taxis and may be involved in drug trafficking. On the one hand, the profits from these other enterprises provide them with the capital against which they are able to extend loans at very profitable interest rates. On the other hand, the illegal, but seemingly accepted practice of informal moneylending provides them with the opportunity to effectively ‘wash’ ill-gotten monies into a socially accepted if not legal enterprise. This also involved investing in social relationships of all kinds.

At the other end of this scale were the small-scale neighbourhood lenders who combined their lending with other income-generating activities, such as selling contraband goods, running a spaza shop and the like. They tend to run their various economic activities out of sight of the state and regulatory authorities, paying no taxes and doing little in terms of bookkeeping. I met such small-scale lenders who were officially unemployed and who were also engaged in criminal enterprises they would explain away as ‘working from home’. But not everyone I met who lent money and was known as a mashonisa was engaged in the criminal economy, or used moneylending as a way to launder illicit income. And not all lenders were unemployed or self-employed. I encountered small-scale lenders who were employed in the formal sector and used a portion of their salaries as capital against which borrowers could draw interest-bearing loans. Such workers, who lend money to neighbours after hours and on weekends, typically earn enough to support their family and have few expenses in terms of housing or transport. In interviews, lenders who conceptualized their lending as a form of investment, explained that they invest surplus income in the informal credit ‘market’, where they hope to get some handsome returns. This form of investment, albeit risky, makes sense given the high banking and transaction costs involved with investing or saving monies with formal financial institutions. An important category of these small-scale, salary-earning informal lenders were teachers, nurses, policemen and other government employees. Those with few dependent relatives or minimal outstanding debts or bond payments they have to serve were able to use their salary savings as a capital against which borrowers could borrow. However, given slightly different economic circumstances of salaried residents or their households, the same category of state employees equally constituted an important category of borrowers. If they had recently bought a car or a privately mortgaged ‘bond house’ in one of the newer middle class neighbourhoods and were struggling to service their mortgage, they turned to formal and informal lenders for credit. One lender I interviewed, who resided and operated in a neighbourhood where many police officers and nurses resided, told me that the largest pool of his clients were in fact police officers and nurses who were not able to keep up with the repayments on their ‘bond’ houses. These borrowers resorted to borrowing expensive credit in the underground credit market so as to pay off house loans in the formal sector. They were of course also trying to avoid being blacklisted by the powerful (and then largely unregulated) credit bureaus while risking the consequences of not repaying loans in the underground credit market.

Another way to demonstrate the importance of social relations and information in the social aspect of informal credit operations and to point to how credit relations are not separate from social life, is to examine the way in which lenders such as Seb deal with risk and secure collateral. As a rule, informal neighbourhood lenders have no recourse to the law. This means that they have to obtain other forms of security that promise a return that equals the value of the outstanding loan amounts. The clients of the small-time lenders (the borrowers) typically reside in the same neighbourhood as their lenders. This is a central dynamic, as one of the few forms of security against loans available to lenders is possession of as much information as possible about borrowers. This information, they hope, would prove sufficient in terms of collecting loans efficiently or in terms of repossessing moveable assets of defaulting borrowers. It became evident to me that in order for a lender to be able to repossess consumer items from a defaulting debtor, personal information on the debtor was required: their names, addresses and often also their type and place of employment. Lenders as a rule did not provide loans to strangers or to individuals from whom they could not obtain such information. Formal financial institutions obtain information from borrowers so as to calculate their risk, or pursue legal action, repossession or blacklisting. Lenders such as Seb had to obtain information for similar reasons. Lenders talked about this information as an important part of constituting the ‘trust’ between a lender and borrower. Seb, for instance, defined ‘trusting a borrower’ as knowing a borrower’s name and address in addition to having the ‘right feeling about a person’. ‘Introductions’ are also an important part of ‘establishing trust’: as a new client you are more likely to be provided with credit if you are introduced by an existing client who can vouch for you and ‘make an introduction’.

Lenders did sometimes need security or collateral other than information from borrowers. One practice which was often mentioned in media reports at the time, and which was outlawed in 1999, is the practice whereby lenders confiscated the identity documents or ATM bank cards of borrowers as a form of security until the loan had been repaid. This was widely practised not only by informal township lenders but also by the newly emerging, formal micro-lending industry, before the practice was criminalized by legislative reform (James, 2014). The confiscation or ‘safekeeping’ of ATM cards by lenders was widely practised during the 1990s.

As more and more workers’ salaries and wages were paid out electronically through the banking system and no longer in cash, lenders turned to confiscating the banking cards of defaulting borrowers. By the time I interviewed several small-scale neighbourhood lenders, many lenders and borrowers were aware of the legal prohibition against this practice and lenders told me this was no longer practised. It is, however, more likely that large-scale lenders and micro-lenders confiscate bankcards as neighbourhood-level lenders could rely on other forms of security. The further the lender is removed from operating in a face-to-face community, or the lesser degree to which they are embedded in the fields of social relations in which both lenders and borrowers participate, the more likely it is that the lender will confiscate identity documents and bank cards.

I established another unexpected way in which Seb minimized his clients’ exposure to higher levels of indebtedness. He would, as a matter of course, inquire from his borrowers whether they were also indebted to other lenders, whether at work or with formal credit providers. Furthermore, he would occasionally meet up with four other informal lenders who operate in his neighbourhood; these lenders informally discuss a number of the clients who have active loans with more than one of them. Seb indicated that they are all well aware which people are indebted to which lender(s). The lenders would at times agree amongst each other not to advance any more loans to a specific client who was perhaps too indebted to several of them. In this manner they act as informal regulatory institution by pooling information concerning clients and minimizing their exposure to high-risk borrowers.

Borrowers: Emergency credit and the precariousness of everyday life

While neighbourhood moneylenders may have been employed or unemployed, their clients tended to be drawn from the employed and those who had access to some income such as a social security grant. In most cases, lenders wanted prospective borrowers to be able to ‘prove’ their employment or sources of income. While neighbourhood lenders did not always enquire about why their borrowers needed credit, they knew well not to always believe the reasons clients gave them for requiring a loan. Some of the lenders I interviewed kept records of every loan contract but did not allow me to peruse their records for a more accurate picture of the size of their loan accounts. Unsurprisingly, they were reluctant to give me access to them, as it may have constituted proof that they were engaging in informal moneylending. As my friendship with Seb developed, he agreed to keep a record of all his loans for me and to ask borrowers their reasons for taking out loans. He had not kept records before my request, mainly because he rarely had more than seven or eight active loans at any given time. During subsequent interviews, Seb told me that his clients aired freely their reasons for asking for a loan, usually as part of their efforts to convince Seb of the legitimacy of their need. But Seb, as he was ready to tell me, could sometimes see a debtor at a tavern drinking away some of the money he had borrowed from him the day before. He was very pragmatic about it: ‘It is up to them if they tell stories and lies – me, I just want my money’.

A look at Seb’s loan accounts revealed the following about a number of his clients. One client was a Mozambican man from Maputo who had been staying for more than a decade in a tin shack in the back yard of a house down the street. At the time this 50-year-old man was married to a South African woman, was working as a ceramic tiler in Johannesburg and was introduced to Seb by the owner of the house where he was renting the shack (who also used to be a client of Seb). The Mozambican man had borrowed money between five and ten times and had, according to Seb, a good payment record. His most recent loan at the time consisted of R100 he borrowed in order to meet his rent payment for his shack. He duly returned R225 at the end of the agreed-upon month.

Another client was a 45-year-old man who stayed around the corner from Seb in his mother’s house. This man borrowed R45 in order to pay for his transport costs to and from work. The contract was that the client would pay no penalty if he repaid the loan within one week. After every seven days an automatic penalty of R15 would be added to the original loan amount. In the notebook I had given Seb he had written down some of the reasons offered by borrowers for why they wanted to borrow money from him. These included monies required in order to pay for school fees, to repair a cellular phone, to settle a debt with someone else and to pay for a traffic fine. All the clients who had active loans with him at the time resided in his immediate neighbourhood and he knew their addresses, names and ages from memory. He had a good idea of where most of them were employed, if they were. Seb clearly had a remarkable knowledge of the lives and histories of his clients and the people in his neighbourhood in general. However, having accurate knowledge about his clients was not the only way for Seb to minimize the risk taken in lending out money.

As mentioned earlier, using local authority figures, such as police officers or priests, is one of the ways Seb used to enforce a loan contract. Even if that contract was legally flawed and in contravention of the Usury Act, he had in the past utilized the police as a source of local power to underwrite such agreements. For example, Seb related to me the case of a client who in 2003 came to borrow an unusually large amount of R1,000 from him. Seb was at that time in a position to provide such a large loan. Seb agreed to the loan on condition that the two of them ‘make a contract’ and get a police officer to witness the transaction and loan contract. The borrower was agreeable and so they went to the nearby police station. After informing an officer of their request, the officer made them sign an affidavit that spelled out the terms of the loan contract. It also included their names, identity numbers, and places of work and residence. According to Seb, the police officer stamped the contract with an official police stamp and gave each one of them a copy. I was curious to know whether they had to pay the officer a commission or bribe him in order to give the contract approval, but Seb maintained that the police officer did it for free because he saw it as part of his duties as a police officer. The police officer did not consider the procedure out of the ordinary and other instances recorded during interviews indicate that this is a popular practice. The debtor honoured the contract and paid Seb back the outstanding R1,500 after two weeks. In Seb’s eyes this arrangement worked well. He maintained that he would ‘make a contract with the police’ again in the future, should a client require such a large loan. But he also recounted incidents where borrowers took refuge with local police officers, especially those who defaulted on loans that had become excessively large. He suggested that he initiated the previously mentioned cap on interest because borrowers would get upset when their loans grew infinitely large. Hence his statement to me, ‘[that] is why I stop with R225 – there is no cop in that money’, suggesting that the benefits of acting without police sanction perhaps outweighs the disadvantages of capping the interest.

The borrowers I interviewed were among the main Sowetans who obtained loans from small-scale neighbourhood lenders in order to access emergency credit. In addition to regular borrowing from friends and family, these respondents typically borrowed from a neighbourhood lender in order to meet unexpected emergency costs. These emergency costs were invariably linked to the conditions under which many working class Sowetans live: in an environment characterized by high levels of unemployment, and so forth. There were a few examples where residents borrowed money from neighbourhood lenders for productive purposes, such as capital for starting a business like purchasing an asset such as a computer, but the main loans were for ‘emergency’ credit.

I mentioned some of the reasons Seb’s clients came to him for credit. Other common reasons included the need for emergency cash to pay for unexpected situations such as unforeseen hospitalization costs, the burial of a family member, bail money for a friend or family member, and the like. I heard accounts of borrowers who had to borrow money from an informal lender in order to pay off mortgage bonds and loans in the formal sector when there was the threat of getting blacklisted by one of the national credit bureaus. ⁵ Borrowers I interviewed were not ignorant about the workings of formal financial services. They were able to articulate the advantages and disadvantages of taking a loan from an informal moneylender as opposed to obtaining one from a commercial bank or regulated micro-lender. They emphasized the fact that the sort of credit that neighbourhood lenders are providing was mainly for emergency credit. They also stressed the advantages of neighbourhood credit providers: it was quick to access a loan; it required no complicated paperwork; the terms were easy to understand and they could make the loan arrangement with someone in their neighbourhood who speaks their language. Moreover, they were quick to complain about their experiences with formal banking institutions where they had to wait for long periods in queues, were unable to explain themselves confidently in English or Afrikaans, and where encounters with these institutions left them feeling embarrassed to have to fill in forms they did not fully understand. Having noted the above, it raises the issue of whether these are reasons enough for residents to access expensive credit. It also calls into question the matter of whether neighbourhood moneylenders exist merely because people struggle to access the formal credit market.

There is no doubt that the opportunities for accessing credit in post-apartheid have increased for Sowetans. They no longer have to rely on expensive credit in the underground credit market. But at the same time we have witnessed a rise in poverty levels, higher levels of urban unemployment among the unskilled and rising living costs (Marais, 2011; Mosoetsa, 2011). These conditions have pushed a section of the urban working poor into a precarious situation in which they constantly have to rely on expensive emergency credit that they can often access solely through neighbourhood lenders, especially when their claims on kin and friends discussed by Ferguson (2015) have been unsuccessful. For many of them, their neighbourhood lender is their ‘lender of last resort’. In this context, both lenders and borrowers emphasized the personal and social dimensions of credit relations.

Informal credit relations: A domain separate from social life?

Every Monday Seb meets with a number of other local businessmen who all belong to the same rotating savings and credit club. That club has an account at a reputable commercial bank and members deposit their savings at a local branch every week. At some point the members may decide to withdraw their deposits from the club’s bank account to purchase vehicles, renovate their homes, use their savings as capital for a moneylending enterprise, or to purchase a large quantity of stolen liquor that has become available on the black market after ‘falling off a truck’. A few men who run legal and illegal businesses come together on a weekly basis to pool parts of their savings that they then deposit at a reputable commercial bank, only to withdraw it at some point in order to invest it in further legal and illegal economic activities. I encountered numerous case studies such as this one that seem to blur the neat categories of formal and informal, licit and illicit. ⁶

The predominant distinction drawn between formal and informal is in some way reliant on boundaries constructed by the state and by regulatory authorities. Borrowers easily move between these as they navigate credit and debt relations in Soweto. Some lenders borrow money in the underground credit market in order to settle debts in the formal financial sector. Seb is at once a moneylender that provides illegal credit to neighbours while at the same time a member of a local savings club, an institution respected and lauded by the state and commercial financial institutions. The boundaries that economic textbooks draw between the categories of credit and savings and investment are in practice very hard to separate and to distinctly isolate in space and time. It is also then very difficult to offer moral evaluations of credit as good and productive and debt as bad and debilitating. An extreme example of this comes from Andrew Lukhele’s book on stokvels in South Africa. Lukhele (1990: 20) writes about a savings club which ‘forces’ its members to borrow from their own savings at usurious rates as a mechanism of enforced savings, so that members’ year-end ‘performance bonus’ (1990: 20) is related to their ‘level of indebtedness and repayment’. ⁷ We find then that lenders such as Seb regularly talk about their lending practices as a form of business, in the same way as some savings club members would talk about the ‘business of saving’. Put differently: certain financial mutual saving societies use the pools of money they collectively save as a capital fund against which members and non-members can borrow (at high rates). Such agreements are typically reached among members at the start of the year and the capital and profit from the interest obtained through members’ loans is then divided among members at the end of the year (see Krige, 2011).

This brings me to a central aspect of this chapter: the disjuncture that emerged between, on the one hand, everyday understandings and practices of credit relationships in Soweto and, on the other, the public discourse on moneylenders that paints them as unregulated, exploitative and profit-hungry proto-capitalists. Trade unions and consumer activist groups have sought to fight for greater working class access to formal credit by attacking the figure of the moneylender. A remarkable consensus that moneylenders were to be done away with, not only regulated, emerged in the early 2000s. The production of narratives in which the most vulnerable members of society, particularly African women pensioners, were portrayed as the victims of moneylenders was central to the emergence of this consensus. An example came from the ANC Member of Parliament M. M. Sotyo in Parliament on 17 February 2003, when he said: ‘We say to those unlicensed moneylenders who steal the pensions of elderly by borrowing this poor senior citizen’s money unlawfully, we are watching you and you will be arrested very soon’ (Sotyo, 2003).

With this background in mind, it was notable that a majority of lenders and borrowers I spoke to were much less moralistic and judgemental regarding the neighbourhood moneylender than the discourse emanating from the mainstream media, the commercial banks, trade unions and consumer interest groups. While elements of the public discourse on the exploitation and victimization of borrowers were evident in the voices of some well-educated and politically informed Sowetans I interviewed, many informants expressed a decidedly more pragmatic view. They did not locate the danger and power attached to moneylending in the lender as an individual, but in the practice of lending. They did not regard the lender as evil, but detested the practice of lending and the reasons for the existence of the practice. They felt no need to condemn the person or the figure of the lender the way the public discourse does. As Bähre (2007) has argued, South Africans often do not consider other humans to be either ‘good’ and ‘evil’ but rather to have the capacity to act in both evil and good ways. Such views were no doubt informed by the fact that lenders viewed their own lending practices as a form of business and that, while it may not be legal even if it is socially legitimate, they were able to enlist police officers and ward councillors in underwriting loan contracts. Many lenders and borrowers shared the sentiment that informal neighbourhood lenders were providing an important albeit expensive service. Many of those I interviewed articulated a discourse of individual responsibility and social contract. In such talk the onus in informal credit transactions was seen to rest squarely with the borrowers. After all, it was said, they are the ones who initiate the relationship in the first place. Here there was little room for a blame-the-system discourse; rather, the victims of the public discourse were castigated by my research respondents for not being able to repay loans that they had willingly agreed to and in fact initiated. Thus lenders repeatedly conveyed to me the view that they never force clients to come to them to borrow money; borrowers come of their own accord and therefore the moral weight of the transaction rested with the borrower. The borrower is, after all, the one who initiates the relationship and who is obligated to return the loan plus interest, as they seem to agree. There is always, as neighbourhood lenders were quick to point out, a verbal contract between the two parties (which in South Africa is legally binding) and borrowers are always aware of the nature and agreement of any loan. Lenders and borrowers alike rarely make reference to a structural context ‘forcing’ borrowers to initiate relationships of indebtedness in the first place.

In a similar vein, the relationship between a moneylender and a borrower is not articulated as necessarily exploitative, despite the power the lender has over the borrower given his access to cash. It is not phrased or talked about in the same moralistic language that characterizes the views that have developed historically in (the then) industrializing Western Europe, notably around the historical practice and image of the Jewish Shylock (Ferguson, 2008). All banks, Niall Ferguson (2008) reminds us, started out as moneylenders. At times, some of my informants explicitly expressed the notion that it is not altogether ‘bad’ to be indebted to a neighbourhood moneylender. I am certainly not advocating for greater indebtedness to moneylenders, whether a small-scale neighbourhood lender or a global banking corporation. Yet poorer and working class Sowetans have been pushed into such intensely personalized relationships and money flows that they talk about the inability to access cash, even if through a loan from a mashonisa, as a kind of social death. Some informants suggested that remaining indebted with your neighbourhood moneylender, and not settling your debts, is a way to ensure the continued existence of the relationship. Indebtedness keeps the lines of communication and possible future credit lines open, which is no doubt of great importance for those who live under precarious economic conditions. That is why one informant remarked to me that paying off an entire loan amounts to ending that specific relationship. This counterintuitive practice, of remaining indebted, and viewing this as an important resource, was echoed in what some Sowetans said about the wealth transfers that accompany bride wealth transfers as part of the marriage process. Some said that a husband is in fact never supposed to fully settle the outstanding amount of the bride wealth transfer (lobola) to his in-laws. Monetary debt makes concrete the social obligations that sustain the relationship that connect the families. While younger women were not in support of this view, seeing it as yet another argument husbands use not to settle debts to their in-laws, the point is that credit relations transcend narrow economic dimensions and form part of a wider system of social and personal exchange relationships that create bonds of dependence and hierarchies and obligations between humans. This may be one reason why some borrowers told me they never want to terminate their relationship with ‘their local lender’ so when they do pay off a loan they take out another, even when they don’t need the emergency cash. Others explain this ‘vicious cycle of indebtedness’ to one lender as a form of addiction.

The idea that indebtedness can be seen as a kind of ‘reciprocity’ or productive of social bonds rather than as exploitative is not new to anthropology (Leach, 1982). In her work on women moneylenders in rural Senegal, Donna Perry (2002) writes that borrowers do not view local moneylending activities as exploitative, but rather as a form of reciprocity. ⁸ Similarly, some of my informants’ views on the desirability of being indebted to a local moneylender echo this desire for continued relations between people, even when mediated through indebtedness, given the social proximity between lenders and borrowers.

Concluding remarks

In this article I have presented data about the organization and meaning of credit relations in the underground credit market that continues to flourish in Sowetan neighbourhoods despite the notable consensus between public authorities, civil society and financial institutions that informal township lenders should be done away with. I demonstrated how small-scale neighbourhood lenders are embedded in the very neighbourhoods in which they live and work. Because borrowers and lenders live in the same social context, the social ties between lenders, borrowers and the wider community effectively regulate lenders. Lenders have to ensure that they do not develop an anti-social reputation. They have to keep open the option of negotiating contracts with borrowers. They treat information about borrowers as their main source of collateral, with such information obtained through interviews and gossip as well as the sharing of information between lenders operating in the same neighbourhood. The regulation imposed on lenders by the social context in which they live and work put a limit on the interest they can charge on loans and it puts limits on their ability to use violence to recuperate loans. Lenders legitimize their business by adopting discourses of ‘entrepreneurialism’ and ‘business’ that echo government discourses. They further legitimize their lending businesses by extending credit to borrowers such as police and nurses who work for the government, while using police officers and ward councillors paid by the local state to make loan contracts ‘official’ despite the fact that these agreements are illegal. They frame their lending enterprise as a form of investment and invest their earnings in reputable community savings clubs.

I further demonstrated how borrowers are required to prove their income to lenders and how important this relationship is for many borrowers. Borrowers in Soweto access expensive credit in the underground credit markets mainly for emergency costs arising from the precariousness of everyday life. While borrowers expressed frustration and discomfort with the process of accessing credit at commercial banks and regulated microcredit providers, they were generally quite aware of the risks and costs associated with accessing credit through a neighbourhood lender. This does not mean that cost–benefit analyses could explain why borrowers utilized expensive credit in the underground markets. Borrowers may borrow money from the neighbourhood lender at high cost in order to make a payment to a rotating savings and credit club. Like lenders, borrowers emphasized the personal and social nature of the relationship between them and their lender(s). This provided them with the ability to negotiate the terms of the loan contract and appeal to the lender who typically had multiple and, at times, dense social connections to the borrower.

In my analysis of the organization and meaning of relations of credit and debt in the underground credit markets of Soweto I showed how difficult it is to make productive use of the informal and formal distinction. I indicated how lenders are hardly ever only lenders but also businessmen, neighbours, members of a local savings club, and so on. I showed how both lenders and borrowers adopt quite a pragmatic attitude to the business of moneylending, emphasizing the contractual dimension involved when adults initiate loan contracts. This sits uneasily with the public consensus that has sought to condemn the figure of the mashonisa. The public discourse stems from a tradition that views debt as a form of bondage and that views money as having the power to destroy social relationships. Sowetans, however, especially those who are facing the consequences of a decline of wage work and who are pushed as a result into personalized relationships of dependence in order to survive, do not demonize the mashonisa, money and debt.

I agree with Soederberg (2013: 536) when she argues that the contemporary and prevalent narratives of debt since the onset of the austerity crisis in Europe and elsewhere tend to ‘ignore important questions of power, but also to reduce the ecological and social aspects of indebtedness to the exigencies of markets’. For her this also means examining how ‘debt relations facilitate capitalist social order(s)’ and politicizing debt so that we can expose the ‘technical and economistic cloak of the official discourse of debt’ (2013: 537) that makes all kinds of political projects possible. A starting point she argues is not to render the economy as ‘a harmonious, naturally evolving space distinct from society’ (Soederberg, 2013: 538). In this article I attempted to understand the dynamics of ukumashonisa – the business of lending money to others at interest – from the vantage point of those who participate in this practice as lenders and borrowers. I focused, from one end of the spectrum, on informal moneylending: those small scale and unregistered lenders who operate within township neighbourhoods and who are not linked with large-scale transport or financial enterprises. Unlike the mashonisas Bond (2013) wrote about and who operate as microcredit providers to mine workers around Marikana minefields, the lenders I write about live and reside in the same neighbourhood as their borrowers. The consequence of this is that lenders and borrowers participate in the same social field and their economic and social worlds are not set apart. This is not true for the lenders Bond discusses, and he points to how the mine workers are now not only exploited at the point of production but also in ‘super-exploitative debt relations, in which financial desperation is compounded by legal abuse’ (2013: 583). While Bond’s attempt to understand and connect debt on macro- and micro-economic levels does not do enough justice to the actual dynamics of credit markets in Marikina, his macro perspective points to an important change in how capital has viewed South African workers: whereas townships like Soweto were created as temporary labour camps, they are no longer framed by capital as spaces from where to extract only labour but as spaces from where to extract poor people’s money. The financialization of poor people’s money in the example I offer here was achieved in part by the vilification of the township moneylender. I argue that a closer examination of the actual practices and dynamics of microcredit markets is warranted. Whereas Ferguson (2015) points our attention to the importance of ‘redistributive relations’ in livelihoods, including practices that are ‘bound up with relations of dependence that are unequal, hierarchical, and often exploitative and abusive’ (2015: 25), the material presented here shows the importance of contractual relations around the debt/credit nexus in situations where those with little income and with limited capacity to make redistributive claims on kin and friends, access cash credit in the underground credit markets of Soweto. The public vilification of the mashonisa may unintentionally contribute to severing the overlapping ties the precarious ‘unbanked’ have to actors in the underground credit markets, furthering their dependence on to the commercial banking sector.

Marcel Mauss’s The Gift (2002) was also a prehistory of our modern kind of legal and economic contract: from total prestations between groups in which material goods are just one item amongst a whole range of non-economic transfers (including services and favours), to gift exchange between persons as representative of groups, to modern market exchange between individuals. He was interested, following Emile Durkheim, in theorizing the ‘non-contractual element in the contract’ (Hart, 2007: 3) that makes possible the three obligations contained in gift giving: (1) to give, (2) to receive, (3) to reciprocate; he shows the importance of social obligations entailed in social relationships, contracts and exchanges. While there is growing agreement in the literature that it is difficult to retain a hard and fast distinction between gifts and commodities, it may be useful to view gift exchange as the exchange of inalienable objects between interdependent persons and commodity exchange as the exchange of alienable objects between dependent individuals. Aspects of the underground credit market I have described above certainly suggest that at times the exchange of cash money between lenders and borrowers takes on the appearance of gift exchange rather than commodity exchange. How alienable is money when one’s livelihood depends on it?

In Peebles’s discussion about the centrality of hoarding in capitalism and exchange, he more broadly suggests that ‘we may be justified in seeing the spread of formal banking as a history of conquest … and the entrenched unbanked as a global resistance movement’. More important is Peebles’ contention that unbanking can be seen as a desire not to give up one’s right to hoarding part of one’s wealth and to retain some power over it. Banking then is about deciding to ‘unite a hoard with a larger one [and this] is very much a declaration of dependency – an agreement to take a secondary position in a new hierarchy of potentially promising financial mutuality’ (Peebles, 2014: 609). What I described above suggests that many poor and unemployed and working class Sowetans are not in agreement with the public authorities and civil society organizations that have made it their mission to get Sowetans to completely eschew neighbourhood moneylenders. Instead, they retain their connections with such lenders even as they get ‘banked’ and access formal banking services.