Unemployment and the gendered economy in South Africa after Covid-19
As we approach the end of 2020, almost half of all people in South Africa who want paid work are unable to find it. This is almost unprecedented in the modern economic age, and is probably the single greatest existential threat to our society. Our unemployment statistics have ceased to shock us, but they should. And perhaps we should not be surprised at where we have arrived. The South African economy, like many capitalist states, has been built upon cheap labour through the systematic exclusion or discriminatory inclusion of all but white people, and on the unpaid labour of women. It is therefore predictable that while unemployment is widespread and persistent, it continues to cleave the country along the fault lines of race and gender, and this has been particularly starkly observed during the Covid-19 pandemic. In this piece I focus on the gendered nature of the South African economy, and how the pandemic has exacerbated the already precarious position of women. I argue that perhaps one answer to our conundrum lies in rethinking our conception of work and our policies for job creation.
The new normal
Between March and June 2020, 2.2 million people lost their jobs in South Africa (Statistics South Africa 2020), as emphasised by the precipitous drop in employed, pictured in Figure 1. As a country, we are rather inured to devastating unemployment statistics and so the significance of this number was eclipsed by a furious debate about the definition of un-employment. But this is a crisis of epic proportions. By my calculations, almost 47 per cent of the working age population is unemployed by the expanded definition. If we ignore the wrangling about who exactly counts as unemployed, and instead focus on that number, 2,200,000, the result is alarming: at the headline level, South Africa has lost [End Page 103] all of the jobs the economy has created since September 2010.
The fact that our unemployment rate has risen steadily over the last decade often obscures the fact that the economy has also created jobs–although nowhere near enough of them– and this decade of slow progress has been erased in three months, while all along the working-age population has been growing. So where are we as 2020 nears an end? Out of an economically active population of 39 million (that is, those between the ages of 15 and 64), 14.15 million are employed, while 4.3 million are unemployed and actively looking for work. 20.6 million people no longer participate in the labour market (we call them 'not economically active')–many because they have given up looking for work at all.
Click for larger view
View full resolution
South Africa has built its developmental ambitions on the promise of work for its citizens. This is precisely what the economy and the state have failed to deliver. The rate of job creation has trailed by more than half the rate of growth of the economically active population, leading to rising unemployment in the post-apartheid era. This, of course, is an economic disaster, but it is also a social disaster. There is little agreement about either the cause of our unemployment (some say it is structural, others that social and labour protection have driven wages above the equilibrium level), or the solutions to it. But much of the discussion about our unemployment problem takes the structure of the economy [End Page 104] as a given: our unemployment is created by a number of immutable economic forces which endure in the post-apartheid period. This view fails to acknowledge that the economy is also a social relation. Our persistent unemployment is a result of a social pact of sorts which is partly informed by how we think about work. We must start to acknowledge that the labour market, and indeed the broader economy, is the product of our choices, social, economic, and political. If this is indeed the case, why have we failed?
Thirty years into a democratic South Africa, one gets a strong sense of deja vu when examining our responses to our unemployment challenges. While the composition of our unemployment has changed, our overall unemployment rate remains in 2020 as it was in the mid-1990s (Hirsch 2005). If we review economic policy from Reconstruction and Development Progamme (RDP) in 1994 onwards, there is what now appears to be a misplaced optimism both that economic growth will create jobs, and that our economic and industrial policies will deliver economic growth. Furthermore, much of our economic policy in the post-apartheid era is founded on the assumption that the state is capable of delivering complex reforms that require intricate coordination across spheres of government and the social partners. Again this has been proven to be an unfounded assumption.
One result of this is that we have failed to address the pervasive gender inequality in the economy. There is significant inequality in income, access to employment and services, socio-economic status, education and household resources, both between women and men, and amongst women. As Dori Posel has argued, 'African women's access to resources has been limited both because of their gender and their race, and of all South Africans they face the highest risk of poverty' (2014:308). In the post-apartheid period, the gender gap in poverty rates has widened, and women continue to have higher levels of unemployment and lower earnings than men (Posel and Rogan 2009, Espi, Francis and Valodia 2019).
Historically, the South African labour market has been segmented both along gender and racial lines, with white men occupying the vast majority of high-paying jobs. The observed increase in female labour force participation in the post-apartheid era has benefited some women, particularly those in professional occupations, but, as Posel (2014) argues, the larger part of the increase in women's employment has been in low-earning and precarious work, including in the informal sector and in survivalist activities.
Many other women have simply been unable to find work at all (Casale and Posel 2002, Casale 2003, 2004). Furthermore, despite growing female labour [End Page 105] force participation in many countries around the world, women continue to perform most of the unpaid labour of housework and childcare (Folbre 1994). This is an issue which has been very well described in economics, but has had a limited impact on policy making, partly because economic policy has not shown as much interest in the unpaid economy, where, due to historical factors, women spend a significant proportion of their time (Folbre 1994, Himmelweit 2002, Budlender 2010).
The Covid-19 pandemic has exacerbated these problems. Between February and April, women accounted for two-thirds of all jobs lost (Casale and Posel 2020), and in recent months have shouldered an even higher share of unpaid work at home. For example, by June 2020, women were spending 3.3 hours per day more than men on unpaid childcare (Casale and Shepherd 2020:1). In the South Africa of 2020, women work far more unpaid hours, and where they are paid to work, they earn less than men even when doing the same jobs.
The new normal requires a pivot from our current approach–job creation as a by-product of economic development and economic growth–to an economic policy which has job creation as its primary goal, even if this means a reduced focus on other developmental objectives. It will also require policy makers to recognise that unpaid work (mostly care work) is work and needs to be funded appropriately. Is it not time that South Africa abandoned its ambitions to expand employment through economic growth, and instead target employment creation as the apex economic priority? This could be seen as anti-developmental, but it might be the least-bad option. We cannot continue to offer the same solutions to an employment problem which continues to get worse. The crisis has opened space for more radical policies (such as universal basic income) and perhaps it has created the space for more radical policies around work, for those who want it. There is some cause for optimism. There are some early discussions in government which advocate for paying for the work that needs to be done, and supporting employment creation in vulnerable sectors and funding social employment (The Presidency 2020).
One of the central obstacles to solving the pervasively gendered dynamic in our labour market is our understanding of what constitutes work in the first place. When we talk about work, what we generally mean is paid work. This ignores the vast amounts of unpaid work done particularly by women, around the world. Within economics, there is of course a vast literature which stresses the importance of considering unpaid work as a central part of the economy, [End Page 106] but this has not yet infiltrated mainstream economic policy making, at least in South Africa. This, the feminist economics approach as outlined by Elson (2000), highlights the need to understand gender discrimination both in the labour market, in the form of unpaid work in the household, and as a larger social relation. Using this approach allows us to understand how women's participation in the paid economy interacts with social reproduction, the labour market and the wider questions of the gendered economy. Elson argues that 'conventional conceptions of the way in which economies operate offer limited guidance for policies to promote women's empowerment and ways to combine gender justice with economic justice. This because they leave out much of the work that women do, especially the unpaid care work that women do for their families and communities' (Elson 2000:7). Furthermore, these conventional approaches do not examine the ways in which unpaid work relates to the market and the state (Folbre 1994). As a result, they treat labour as 'if it were like land–an input that exists without having to be produced, a "primary factor of production"' (Elson 2000:24).
One of the central features of the feminist economics approach is to interrogate the ways in which what we think of as neutral may in fact be gendered. The feminist economics perspective foregrounds the domestic care sector, and makes clear the links between the public and private sectors, the NGO sector, and the domestic sector. Elson emphasises that these sectors are linked both by market and non-market mechanisms. Susan Himmelweit argues that, of course, the economy does not depend solely on paid work, but is equally dependent on unpaid care activities carried out in the domestic sector. Work done in the domestic sector is vital to 'individual socialisation and to the production and maintenance of human capabilities on which economic life depends'(Himmelweit 2002:52) and we must find ways to incorporate it in our economic analysis. The unpaid care work which takes place in the domestic sector produces and maintains the labour force which is crucial to the reproduction of capital–this is social reproduction.
In an important contribution to the feminist economics literature, Humphries and Rubery (1984) identify two dominant methodologies for analysis of the relationships between the spheres of production and reproduction, which they term absolute autonomy, and the relative autonomy approaches. The former treats social reproduction as independent of the system of production–an approach taken by economic theory–while the relative autonomy approach views social reproduction as an integrated and adaptable part of the large system of production and therefore one that should be analysed within the same organising [End Page 107] theoretical framework. The dominant approach underlying economic policy has diverged from political economy approaches which rely on an historical materialist approach to building theory. Humphries and Rubery argue that we must proceed from the understanding that social reproduction develops, in all cases, in response to the productive system, but, importantly, that 'the form of this response must be understood historically. It is neither predetermined nor smoothly accommodating to the demands of the productive system, but depends on the dynamics of social reproduction' (1984:332). Renate Bridenthal wrote that 'the relationship between production and reproduction is a dialectic within a larger historical dialectic. That is, changes in the mode of production give rise to changes in the mode of reproduction' (1976:5). We have seen this and its inverse in South Africa, for example, in the ways in which our social protection system has in turn shaped our productive economy (Ardington, Case and Hosegood 2009).
The second obstacle to progress in addressing our unemployment is that our unemployment is so high–such a global anomaly–that our theoretical tools for understanding the labour market are at their limits when faced with an unemployment rate of nearly 50 per cent. One example of this surfaced during the debate about the adoption of a national minimum wage in South Africa. Those against it argued that high levels of social protection, and a wage level above the equilibrium, could destroy up to one million jobs. In the two years since the adoption of the minimum wage, this has not transpired, even in sectors such as agriculture and domestic work which are sensitive to wage increases.
While entrenched unemployment is unsurprising from the perspective of heterodox economists and other social scientists, it is seen by more mainstream economists and economic policy makers as an aberration–a departure from equilibrium–rather than an equilibrium itself. SouthAfrica's economic policies are founded on the assumption that if we correct our economic 'fundamentals', growth will accelerate and unemployment will fall. But in the long arc of South Africa's economic history, our unemployment comes as no surprise (Feinstein 2005, Freund 2019).
The third obstacle is that we do not have, at our finger tips, data which would allow us to craft gender-responsive responses to crises. Rather predictably, South Africa's post-Covid economic recovery policy is founded on massive infrastructure investment by the public and private sectors. But, as has been persuasively argued (Elson 1997, Himmelweit 2002, de Henau, Himmelweit and Perrons 2017, International Labour Organisation 2018), infrastructure investment is a gender-blind economic policy that does little to address the [End Page 108] underlying causes of women's oppression in the economy and in society. In the United Kingdom, this argument is forcefully and regularly made by the Women's Budget Group (WBG), which argues, among other things, that funding unpaid care work is a far better use of resources than amorphous infrastructure investment. In the years after apartheid, South Africa had something similar, in the form of the Women's Budget Initiative (WBI), which ran from 1996 to 2000, which argued persuasively for budget allocations that would address the systemic gendered nature of the economy. The WBI, and the women's budget which it prepared, has been replaced by the notion of gender-responsive budgeting, but this has failed to gain traction in the mainstream of South African policy making. One of the most useful and powerful results of data on the gendered impact of budget decisions is that it allows us to motivate for spending that addresses our gendered problems. In the UK, the WBG uses data on the impact of spending (multipliers and the like) to argue persuasively that investment in the care sector can have better broader economic impacts than spending on infrastructure (see Himmelweit 2002, and de Henau et al 2017, for examples of this approach). Others have taken a wider view, arguing that the solution lies not only in economic policy, but in a broader feminist movement (Fraser, Bhattacharya and Arruzza 2018). If we had similar data for the South African economy, we would be able to motivate persuasively for spending on the care sector and on unpaid work.
Notion of the global
One of the most striking things about the gendered impacts of the Covid-19 pandemic is how they have played out in such a similar fashion around the world. Of course the magnitude of the impact has differed, but around the world, women have been put in an even more precarious position by the pandemic (with some therefore calling it a 'she-cession'). The staggered impact of the virus itself, starting in the global North in January 2020 and spreading around the world, means that activists and policymakers have had some foresight into the repercussions. This allowed, for example, a group of academics and activists to motivate for a social policy intervention which strongly targeted women–the caregivers top-up grant in April.
The crisis also highlighted the ongoing question of whether this common experience can become the foundation for a global movement towards a fairer economic system. For one example, Fraser et al (2018) offer 11 theses as the basis of a new, anti-capitalist, internationalist feminist movement. Inspired by the international feminists strikes of 2017-2018, they offer a number of [End Page 109] solutions to the hegemony of global capitalism. One of their theses is that feminist internationalism is the answer to the gendered ills of capitalism. It is too early to understand the longer term impact of the Covid-19 pandemic on such international ambitions, but the crisis appears to have reinforced the nation state, and it is not yet clear what implications this has for the building of international movements. Regardless, their argument points to the interconnectedness of feminism with the ecological struggle when they ask: 'Will we continue to pursue "equal opportunity domination" while the planet burns? Or will we reimagine gender justice in an anti-capitalist form, which leads beyond the present carnage to a new society?' (Fraser et al 2018:114).
One of the things that the crisis has illustrated is the ways in which our economic problem is a gender problem. That women account for a full two-thirds of the jobs lost in South Africa at the height of the pandemic is indicative of the extent of the pervasive gender skew in our economy. What is infuriating is that we do not lack either the theoretical tools to understand why this is the case (in large part thanks to social reproduction theory and the feminist economics approach), nor solutions both from our past (the WBI) and from other countries (the WBG) to how we can address the problem. While the gendered problems in the South African economy seem insurmountable, there are some things that could be accomplished fairly easily. The first would be to pick up on the work of the Women's Budget Initiative to develop an accurate and detailed understanding of the impact of our budgetary decisions. This would allow us better to motivate for spending which is advantageous to women–funding unpaid care work and supporting the care sector, for example. There is a growing appreciation of the importance of the care sector as a constituent part of the economy, and a more detailed understanding about the gendered impacts of various spending choices would be beneficial. This is given impetus in South Africa as we contemplate not additional spending, but spending cuts. If we are not to further entrench gender discrimination in the economy we need to understand the gendered impacts of spending cuts as a matter of urgency. The second is that we urgently need to rethink our economic growth strategy. There are many alternatives, but given our unemployment, an economic strategy for employment is the most pressing. We can no longer target the 'investment climate' in the hope that it creates jobs. We need an explicit focus on job creation in both the productive and reproductive spheres of the economy. [End Page 110]
David Francis is the Deputy Director at the Southern Centre for Inequality Studies, and a PhD candidate in Economics at the University of the Witwatersrand.