One of South Africa’s biggest problems is that so many people lack employment. The main labour market survey in South Africa used by economists and policymakers is the Quarterly Labour Force Survey conducted by Statistics South Africa. Other surveys measuring employment include the General Household Survey, also by Statistics South Africa, and the National Income Dynamics Study, by the Southern Africa Labour and Development Research Unit, with funding from the South African Presidency. We compare employment estimates across these three surveys.

We show that, despite very similar or sometimes even identical sample designs and frequently identical questions about employment, the General Household Survey has produced higher estimates of total employment in South Africa than the Quarterly Labour Force Survey since around 2012, and this difference is even larger post-COVID-19. We investigate possible explanations and find that raw sample composition and weights do not contribute substantially to these differences.