Skip to content

Employment and Unemployment in South Africa

Literature › Works

Charles Simkins (2004). Employment and Unemployment in South Africa. Journal of Contemporary African Studies, 22(2). Taylor & Francis (Routledge).
Type journal article
Scope Post-1994 South African economy
Method mixed
Period studied 1857–2013
Regions South Africa
Themes inequality poverty unemployment, labour markets wages, methods data sources, transition post apartheid, bantustans removals influx, demography urbanisation, unions strikes
Citation key simkins2004

Argument

Simkins reconciles South African labour-market data sources for 1990-2002 and shows that the Survey of Employment and Earnings (SEE) has increasingly undercounted formal jobs, while censuses and household surveys undercounted informal work. Statistical uncertainty is large, but a substantial labour surplus persists and will remain in 2013 under plausible growth and employment-elasticity scenarios. Probit analyses show that Africans, women and the less educated are slowest to be absorbed into work. He locates the problem in the political economy of post-1994 labour policy, the double-class compromise, trade unions, and the limits of social grants and public works. A brief opening history frames labour shortages turning into surpluses, and the apartheid-era wage cliff and influx control displacing unemployment from cities.

Evidence and method

Survey of Employment and Earnings 1990-2001, population censuses 1996 and 2001 (SuperCross tabulations), October Household Surveys 1997-99, Labour Force Surveys 2000-02, Survey of Average Monthly Earnings, ASSA/Spectrum population projections, the Wiehahn Report, and secondary literature (Peires, Wilson, Hobart Houghton, Desmond, Surplus People Project).

Findings

  • Apartheid-era influx control and rural relocation displaced unemployment from the cities until the system collapsed. (p. 2)

    The equilibrium in the middle of the apartheid era was one with a wage cliff and unemployment displaced to urban areas.

  • Africans were massively redistributed from white farms to tribal areas. (p. 2)

    In 1951 there were 128 Africans in tribal areas for every 100 on commercial farms. By 1996 there were 751 Africans in tribal areas for every 100 on commercial farms.

  • The SEE shows a fall in formal employment, but the censuses show a rise, contradicting it. (p. 6)

    Formal employment as estimated from the censuses then rises from 8 367 913 in 1996 to 8 786 797 in 2001, an increase of 418 884, or 5 per cent.

  • SEE coverage fell between the 1996 and 2001 censuses, indicating a measurement problem. (p. 6)

    This implies a substantial drop in SEE coverage between 1996 and 2001, indicating an emerging measurement problem with the SEE over that period.

  • Unemployment in September 2002 was very high on both official and expanded definitions. (p. 8)

    Unemployment rate 30.5 per cent 41.8 per cent

  • Most official unemployed are first-time job seekers, and absorption into work is slow. (p. 8)

    Sixty per cent have never worked before, so the bulk of the unemployed are searching for their first job.

  • Long unemployment spells are common even among those who have previously worked. (p. 10)

    41 per cent of the unemployed on the official definition who have worked before stated that they had been looking for work for more than three years

  • Employment rose only modestly between 1997 and 2002, and the estimate is imprecise. (p. 7)

    The net rise is put at 584 thousand (5.4 per cent) between 1997 and 2002.

  • Under median growth and elasticity assumptions, unemployment falls only slightly by 2013, leaving a large surplus. (p. 13)

    shows a reduction of the expanded unemployment rate from 43.0 to 39.3 per cent over 11 years and a reduction of the official unemployment rate from 32.7 to 28.6 per cent.

  • Real wages of unskilled formal workers fell while average earnings rose, suggesting rising pay inequality and a skills shift. (p. 6)

    the real wage of unskilled workers in the formal sector dropped between 1997 and 2002.

  • Under early and middle apartheid, minority unions secured rents by excluding African workers from bargaining. (p. 17)

    minority unions could negotiate with employers and have agreements extended to non-unionised parts of the relevant industries without having to deal with representatives of African workers.