The racial employment gap in South Africa

The aim of this paper was to estimate whether the probability of an employment gap between White and African, and Coloured and Asian labour force participants had changed between 1995 and 2002. For each of the four years the expected racial employment likelihood hierarchy emerged, with Whites more likely to be employed, followed by Asians, Coloureds and Africans, The initial findings confirmed official statistics and independent studies in that unemployment rates (both official/narrow and broad increased in the period. Using a probit employment function comprising of consistent variables for the five years of data used and through decomposing the observed employment likelihood figures into explained and unexplained terms, it was found that the upper limit estimate of White-African employment discrimination had increased absolutely from 12.0 per cent to 17.4 per cent between 1995 and 2002. The equivalent estimates for White-Coloured and White-Asian employment discrimination increased from 4.7 to 9.5 per cent and from 0.9 per cent to 5.5 per cent respectively. Whilst for all racial groups the observed likelihood of being employed declined in the period, the probability gap between White and non-White groups increased, signalling a worsening employment likelihood of African, Coloured and Asian economically active individuals relative to Whites. When the observed racial employment probability gaps were decomposed further, it was found that Whites were benefiting more from both unexplained advantages in 2002 than 1995, increasing from 9.0 per cent to 14.9 per cent, and from explained employment likelihood characteristics, increasing from 15.8 per cent to 18.0 per cent relative to the average (pooled) economically active individual