This paper explores the relationship between education and levels of poverty and inequality in South Africa, the country with the world's worst inequality. The World Bank argues that one of the most important contributing factors to this inequality is poor schooling in South Africa. The dominant strand of educational research and policy thought in South Africa draws on Human Capital Theory and is promoted by a group of economists who have become influential in educational research and policy, ‘The Stellenbosch School'. In this paper we argue against two strong claims made by the Stellenbosch School and the World Bank.

The first is that educational inequality is being reproduced by teacher unionization; weak teacher content knowledge and pedagogical skill; wasted learning time; and weak institutional functionality. These are, according to the Stellenbosch School, the four binding constraints which prevent improvement in education, and therefore prevent improvement in the economy. However, in numerous research reports and publications, the Stellenbosch School present large sets of data which clearly show a bi-modal distribution of school quality that mirrors the harsh economic reality of South Africa. Their own data demonstrate that on almost all measures – teacher knowledge, school management, pre-school preparation, class size, curriculum coverage –challenges related to educating in poverty minimize whatever effects interventions have on performance in poor schools. In other words, although the work of the Stellenbosch School demonstrates that poverty is the single most dominant binding constraint on improving educational outcomes, it places the solution in individuals utilizing resources at hand effectively, with improved management the most important factor. We argue, by contrast, that without disrupting social and economic conditions of extreme poverty that reproduce economic relations, it is unlikely that we can substantively improve learning.

The second is that because high levels of educational achievement are positively correlated with employment, it is possible to work backwards from economic marginalisation and a lack of labour market participation by intervening in the education system. Here the flaw lies in the assumption that supply will create its own demand in the labour market. This is a simplification of the complex ways in which a range of factors in the South African context intersect and ignores the social determinants of both education and labour market outcomes. We demonstrate that the interface between education and the economy via the labour market creates relationships between credentials and work that are far less rational and more complex than what HCT assumes. We argue that the small size of—and lack of dynamism in—the labour market and economy seem to be shaping the nature of the education and training system in perverse ways, undermining the possibility that it will produce educational outcomes that lead to greater levels of equality. This is because the highly unequal South African labour market creates a negative feedback loop in how education interacts with labour markets and undermines the possibility of strengthening the production of technical and vocational skills. The possibility for changing the interaction between education and the labour market by changing aspects of the education and training system seem very remote because the dynamics which are driving it don't emanate from the education system.