Callebert usefully critiques a ‘dual economy’ approach to South Africa. This model, posited by scholars and politicians, casts a poor ‘second’ economy as separate and excluded from a capital-intensive, globally connected ‘first’ economy. Versions of this diagnosis imply particular solutions: either bringing the second economy into the first, or bringing people from one sector into the other by lowering wages.

As Callebert shows, matters are more complicated. The employed are hardly secure. The interests of the formally employed and the ‘underclass’ are not as sharply opposed as may be imagined. Families contain people in different socio-economic positions, and the employed have to support large numbers of kin. Moreover, people and families combine strategies to make ends meet, straddling the supposed formal/informal divide. And formal jobs offer resources and opportunities for informal business. Callebert contends that all this complexity – evident in South Africa's past and present – was masked by an exceptionalist, teleological narrative of proletarianization. Foregrounding the interdependences within a single economy reveals the limitations of lowering wages to employ more people (since wages support other kin). And doing so enables proper historicization of the connections between livelihood strategies.

Callebert's point is well made. His critique of calls to expand the ranks of the working poor is convincing. Here, I engage not so much with his overall point as with his underlying conceptual framework. The article interrogates the formal/informal nexus. As it shows, the formal and the informal are tightly intertwined, in South Africa and elsewhere. So much so that, as Callebert notes, the very distinction can quickly become problematic.

Yet the formal/informal distinction appears here in rather abstract, even universalist, terms – despite Callebert's sensitivity to the historical contingencies of people's livelihoods. The reason is that he limits the analysis to resource distribution, and to ways of making ends meet. ‘The economy’ itself is presented as a self-evident entity. ‘Formality’ and ‘informality’ then appear as ‘sectors’, however entangled.

Waged employment has always intersected with other areas of life in more complex ways than can be captured by material distribution. It is this complexity that shapes what ‘formal employment’ has meant in South Africa. By stepping back, Callebert's argument about interconnections can be extended. Rather than ask ‘How does formal employment intersect with the informal sector?’ we might begin with a more open question: ‘What does waged employment do amidst unemployment?’ How, for example, do workforces shape not only income flows, but also notions of moral regulation, kinship, space, and time? We might take a lead here from seminal South African scholarship, which did in fact emphasize the messy diversity and interconnectedness of ‘delayed proletarianization’, but from a broad, ‘social’ perspective (see Beinart Reference Beinart2001 for an overview).

Take moral regulation. Under apartheid, labour migrancy was fraught with moral questions. Eastern Cape workers on the mines were famously subject to pressures from rural kin. Rituals were intended to ensure that migrants stayed on the straight and narrow, remitted, and came home (McAllister Reference McAllister, Spiegel and McAllister1991). Such pressures were carried into workplaces, as workers kept an eye on one another. And, crucially for my point here, the mines were themselves sites of moral regulation. Rules (imiteto) fused management requirements, a recognized and hierarchical ‘moral economy’, and ‘informal customs among workers’ (Moodie 1994: 84).

On apartheid-era mines, such ‘law’ appears to have been confined to clearly contained workforces (at least in Moodie's account). But elsewhere – on the farms, for example – matters were less circumscribed, as people moved in and out of residential labour compounds (see Waldman Reference Waldman1996). All the more so in contemporary South Africa, with its skyrocketing unemployment. On the farm on the Zimbabwean border where I conducted resident fieldwork, residents include not only employees, but also dependants of different kinds, and businesspeople such as shebeen Footnote ¹ and spaza Footnote ² owners (some of whom are also workers or dependants). As Callebert notes, this is a world in which formal and informal livelihoods intersect in intricate ways. But there is more to it. All of these people are subject to the authority of senior workers. Led by the foreman, such elders constitute a court (khoro) that sits in judgement on grievances in the compound. In cases of adultery or violence, for example, picking supervisors hear testimonies and strive for consensus. Meanwhile, the farm's security guards direct mob justice when necessary. Agricultural estates in South Africa's margins may represent an extreme case, but they illustrate how the intersections between ‘formal’ and ‘informal’ go well beyond the economic. The very notion of a formal workplace is complex and multivalent.

Everyday regulation is perhaps a dimension of the formal/informal nexus that is too far from Callebert's concerns. But the same complexity pertains when it comes to social reproduction. ‘Households and families combine both formal and informal sources of income in their livelihoods’ (Callebert this issue: 125). But a prior question is how workforces shape the constitution of households and families in the first place. On the border farms, the everyday law and order just described is part of a wider situation in which permanently employed men call the shots (see Bolt Reference Bolt2013). Regulation is on their terms. As in other labour compounds in South Africa (see Elder 2003), women have to take on roles in relation to these men – often in domestic arrangements, from short-term relationships to more enduring farm marriages. It is not enough simply to recognize that people with wages sustain people without. The conditions are crucial.

What all of this suggests is that the wider meanings of ‘formality’ have spatial and temporal entailments. Informality in apartheid-era South Africa depended precisely on the strategic use of the spaces of formal employment (see Preston-Whyte and Rogerson Reference Preston-Whyte and Rogerson1991) – labour compounds, the houses of domestic workers’ employers. Appreciating how wages affect wider populations – and how people's lives ‘transcend dual economies’ – must mean taking a spatial perspective (see Bolt Reference Bolt2012). Crucial dimensions of the formal/informal nexus emerge when we focus on workforces’ capacities to control the space around them: who lives where, who is able to move where, and who is able to access what resources as a result. Equally, workforces have their own rhythms and timeframes, which in turn determine the lives of countless people around them. Wage cycles shape strategies for storing and borrowing money (Bolt forthcoming 2014). At a larger scale, formally employed workers on the border farms mediate the transience around them. As they do so, they hold out the possibility of incorporation into something rooted and settled. But they may also reinforce the vulnerability around them. For some women, temporary relationships – and the pregnancies that follow – lead to ever-greater reliance on what may remain a highly unstable form of inclusion.

None of this obviates Callebert's conclusions, all of which are extremely useful. Instead, what I have tried to do here is extend his argument, in the spirit in which it is made. Callebert calls for a more holistic view of the economy, ‘transcending dual economies’, with a focus on the formal/informal distinction. I suggest that ‘the economy’ itself needs transcending. Formal employment – and especially key forms of compounded labour and domestic work in South Africa – have shaped the worlds around them in ways that cannot be glossed simply as economic.