Skip to content

Changing Labour Market Conditions in South Africa: A Sectoral Analysis of the Period 1970-1997

Literature › Works

J.W. Fedderke; Martine Mariotti (2002). Changing Labour Market Conditions in South Africa: A Sectoral Analysis of the Period 1970-1997. South African Journal of Economics. Economic Society of South Africa.
Type journal article
Scope Economics and economic history of apartheid
Method mixed
Period studied 1970–1997
Regions South Africa
Themes labour markets wages, inequality poverty unemployment, manufacturing industrial policy, education human capital, methods data sources, mining migrant labour, transition post apartheid
Citation key fedderke2002

Argument

The authors review a panel of 48 SIC sectors for 1970-97 and argue that South Africa's poor employment creation is a long-standing structural problem dating from the 1970s, not a product of 1990s policy change. Real wages matter for employment, and the link between real labour productivity and the real wage is central: sectors where pay tracks productivity show stronger employment and sustainable wage growth. A pooled mean group labour demand estimate for 28 manufacturing sectors gives a long-run wage elasticity of about -0.5 to -0.55 and an output elasticity of about one. They conclude that supply-side labour market reform and human capital investment are needed. The paper presents itself as a preliminary, mostly descriptive study.

Evidence and method

TIPS (Trade and Industrial Policy Secretariat) sectoral panel of 48 SIC version 5 sectors, 1970-97, covering employment, real remuneration, value added and skills categories (highly skilled, skilled, unskilled) for manufacturing. Benchmark years 1970, 1980, 1990 and 1997. Population growth taken from Sadie (1993) and the World Bank. Econometric work uses a 28-sector manufacturing panel with T=28 and user cost of capital under 5% and 10% depreciation.

Place in the debates

Findings

  • Average sectoral employment growth was positive in 1970-90 but negative in the 1990s. (p. 8)

    the average growth rate in employment (again across all sectors, unadjusted for absolute levels of employment) from 1970-90, and 1990-97 was 2.16 per cent and -1.26 per cent per annum respectively.

  • Sectoral employment growth fell below population growth, so the problem predates the 1990s. (p. 8)

    It is notable that the rate of employment creation across sectors has been below average annual population growth

  • Number of sectors with negative employment growth rose sharply in the 1990s. (p. 10)

    Of the total of 48 sectors recorded, 31 had negative growth rates in employment over the course of the 1990's. By contrast, over 1970-90 only three sectors had negative employment growth rates.

  • Employment weakness is a long-standing structural feature, not due to recent policy. (p. 6)

    economic growth in South Africa has been poor at generating additional employment ever since the 1970's.

  • Skills intensity of manufacturing rose steadily over 1970-97. (p. 13)

    Even on average, by 1997 the ratio had reached 0.78, suggesting that skilled and unskilled labour may soon constitute an equal proportion of the labour force in manufacturing sectors.

  • Technological change has been capital-augmenting, and higher labour costs reduced labour competitiveness. (p. 14)

    What does emerge is that technological change has effectively been capital- rather than labour-augmenting over time, thus decreasing the capacity of SA industry to expand employment.

  • Every sector with falling employment shows a negative employment-real wage correlation. (p. 19)

    Every sector in South Africa that has showed negative growth rates in employment, also shows a negative correlation between employment and the real wage.

  • Productivity-wage correlation is positive and often strong across most sectors. (p. 22)

    Thus for 31 out of 46 sectors for which requisite data is available, the correlation conforms not only to economic theory, but is reasonably strong

  • Manufacturing labour demand has a negative wage elasticity of about -0.5 to -0.55. (p. 31)

    The implication is that the manufacturing sector is characterised by a negative wage elasticity in the region of -0.5 to -0.55.

  • Output elasticity near unity, so manufacturing growth is not jobless. (p. 32)

    It cannot, therefore, be claimed that South African manufacturing growth is jobless. Instead, demand effects on labour demand prove to be relatively strong.

  • Gold mining shows a strong negative employment-real wage association in levels. (p. 18)

    Gold & uranium ore mining -0.81 -0.4

Evidence ledger

Sources used

Source Kind Period How used
TIPS sectoral panel dataset: 48 SIC version 5 sectors, employment, remuneration, value added/real GDP, 1970-97 official statistical series 1970-1997 Underlies all descriptive tables on employment, real remuneration and labour productivity and the panel regression.
TIPS data set documentation (public availability statement) official statistical series 1970-1997 Indicates the full data set and its documentation can be obtained from TIPS.
Manufacturing skills composition data (highly skilled, skilled, unskilled employees) for 24 manufacturing sectors official statistical series 1970-1997 Used to compute the 'skills ratio' and show rising skills intensity in manufacturing (Table 3).
Panel of 28 three-digit SIC version 5 manufacturing sectors, annual 1970-97 (employment, real wage, user cost of capital, output) official statistical series 1970-1997 Used for the pooled mean group estimation of the manufacturing labour demand function.
Sadie (1993) demographic data, A projection of the South African Population 1991-2011, Research Report 196 secondary literature compilation 1970-97 (population growth estimate) Supplies the 2.38% average annual population growth against which sectoral employment growth is compared.
World Bank Development Indicators (economically active population growth) international trade or finance data 1970-97 Gives 2.56% annual growth of the economically active population, strengthening the claim that employment creation lagged labour supply.
Fedderke, Shin and Vaze (2000), Trade & Labour Usage, ERSA Working Paper 15 secondary literature compilation 1970s-1990s Prior econometric results are relied on to dismiss trade liberalisation as the explanation and to claim technological change was capital-augmenting.
Fedderke et al. (2001) sectoral capital market study, Development Southern Africa 18(4) secondary literature compilation 1970-1997 Cited to show capital markets display a structural break in the 1990s similar to the labour market, and for data documentation.
Bowles and Heintz (1996) wage elasticity estimates secondary literature compilation not stated Benchmark wage elasticity for the whole economy against which the manufacturing elasticity is compared.

Inference

Quantification: Econometric (causal design).

Descriptive exploratory analysis of sectoral panel data (rank tables, growth rates, rank and simple correlations at benchmark years 1970, 1980, 1990, 1997), followed by a dynamic heterogeneous panel (Pooled Mean Group, ARDL error-correction) estimate of a log-linear labour demand function for 28 manufacturing sectors, 1970-97. The causal interpretation rests on theory (cost minimisation) and correlations, not on a natural experiment.

Key claims

Claim Basis Strength
Poor employment creation is a long-standing structural problem dating from at least the 1970s, not a result of recent policy change. (p. 2-3) Sectoral employment growth rates and ranks from the TIPS panel (Tables 1-2), average growth across sectors 1970-97. Partial data
1990 marks a structural break in employment: 31 of 48 sectors had negative employment growth in the 1990s versus only three in 1970-90. (p. 10) Average annual sectoral growth rates 1970-90 and 1990-97 from Table 2; a correlation of 0.49 between the two periods' growth rates. No formal break test is reported. Partial data
Employment creation across sectors has been below population growth, and large traditional employers have shrunk while small newer sectors grew (correlation of -0.27 between 1970 employment level and growth rate). (p. 9) Sector growth rates compared with Sadie (1993) population growth (2.38%) and World Bank economically active population growth (2.56%); cross-sector correlation. Partial data
Skills intensity in manufacturing has risen steadily; the average skills ratio reached 0.78 by 1997. (p. 12-13) Manufacturing skills-category employment counts for 24 sectors, 1970-97 (Table 3). Systematic data
A negative employment-real wage relationship is present: every sector with negative employment growth also shows a negative employment-real wage correlation. (p. 19) Sectoral correlations between employment and real remuneration per worker, in levels and growth rates (Table 5). Simple correlations on a short time series, with no controls. Partial data
Sectors where real wages track labour productivity more closely show higher employment growth and more sustainable real wage gains. (p. 23-24) Sectoral correlations of productivity (Y/L) with real wage (W/P) in Table 7, grouped by correlation strength and compared with employment and wage growth (Fig. 1). Grouping is descriptive. Partial data
Manufacturing labour demand has a long-run wage elasticity of about -0.5 to -0.55, an output elasticity of about unity, and a negative effect from the user cost of capital, so growth is not jobless. (p. 31-32) PMG estimation on a 28-sector by 28-year manufacturing panel, with 5% and 10% depreciation assumptions (Table 9); Hausman test supports pooling of long-run parameters. Systematic data

The author's caveats

The authors note that data are for a few benchmark years in the rank tables, not yearly. Manufacturing is disaggregated far more finely than services and mining, which disadvantages manufacturing in cross-sector comparisons, and no finer public data exist for non-manufacturing sectors. They restrict the panel to manufacturing because data reliability and definitional consistency are weaker elsewhere. They also call the paper preliminary and exploratory, and flag that population growth estimates are controversial.

Gaps in the evidence

The argument that wages rather than policy or demand drove employment falls rests mainly on simple sectoral correlations and one manufacturing panel, with no identification of wage-setting institutions or controls for the capital-augmenting technological change it invokes. The 1990 structural break is inferred from eyeballing growth rates rather than a formal break test. Firm-level or household labour force data, bargaining-council wage settlements and skill-specific wage series would test the mechanisms; some of these (October Household Surveys, firm surveys) existed by the late 1990s. Non-manufacturing sectors, which dominate employment, get no econometric treatment.

Replicability: Data described but not public.