The electricity demand in South Africa has experienced vast changes both from a policy point of view and due to the country's great industrialisation and urbanisation. The literature dealing with the South African electricity case, to date, has not taken into account these changes and their effect to the symmetric behaviour of the residential and industrial electricity consumers. Hence, this paper aims at examining the asymmetric behaviour of the annual South African electricity demand (total, residential and non-residential) for the period 1960–2012. To do so, three different tests were used: the entropy test proposed by Racine and Maasoumi (2007. Journal of Econometrics 138, 2, 547; 2008. Econometric Reviews 28, 246), the conditional symmetry test proposed by Bai and Ng (2001. Journal of Econometrics 103, 225) and the Triples test proposed by Randles et al. (1980. Journal of the American Statistical Association 75, 168). The findings showed that there is weak evidence of asymmetry, given that the null hypothesis of symmetry can only be rejected at 10 per cent level of significance. Hence, econometric models examining the South African electricity demand during the same period are credible in their assumption of a linear data generating process.