Skip to content

Mechanization and Labour Policies on South African Mines

Literature › Works

A. Spandau (1980). Mechanization and Labour Policies on South African Mines. South African Journal of Economics, vol. 48(2). Economic Society of South Africa.
Type journal article
Scope Economics and economic history of apartheid
Method mixed
Period studied 1960–1980 (projections to 1995)
Regions South Africa, Witbank coalfields, Sishen and Phalaborwa iron ore, Transvaal and Orange Free State gold fields, Zambian Copperbelt (comparison), Zimbabwe (comparison)
Themes mining migrant labour, labour markets wages, manufacturing industrial policy, unions strikes, education human capital, comparative theory
Citation key spandau1980

Argument

South African mines, coal first and gold later, are shifting from labour-intensive, unskilled, short-term migrant work toward capital-intensive production, despite low black wages. The author argues that mechanization is shaped by technical, economic and social factors, and that it also drives change in personnel policy. Pay increases, work-stabilization schemes (re-engagement certificates and bonuses) and in-mine training are the three pillars of the new policy. He explains these through internal labour market theory (Oi, Becker, Doeringer and Piore): fixed employment costs for recruitment and training make low turnover profitable. Neo-classical marginal-product wage theory is held not to apply. Coal is already largely mechanized; gold stoping is not, but a shift from unskilled to semi-skilled work is predicted to make mine incomes more equal.

Evidence and method

Chamber of Mines publications (February Report 1979 on recruitment and novices); Department of Statistics Quarterly Bulletin; King's coal-industry trend data and the Petrick Commission report; mining-company cost data supplied privately; discussions with mining engineers and the Chamber of Mines Human Resources Laboratory (estimates of the 1978 stoping workforce); comparative Zambian and German material; the author's own mimeographed research report (1979).

Place in the debates

Findings

  • Black wages in Zambia's Copperbelt rose much faster than equipment costs, driving large-scale substitution of capital for labour (used as a comparison). (p. 1)

    between 1947 and 1958, Black wages increased by 250 per cent, whilst equipment costs rose by only about 100 per cent.

  • Mechanization of iron ore mining displaced unskilled black workers in favour of skilled white maintenance staff, sharply lowering the black/white ratio. (p. 1)

    In the industry as a whole, the Black/White employment ratio dropped from 4,56:1 to 2,42:1.

  • Coal output and labour productivity grew very strongly in the 1970s. (p. 2)

    Output increased from 53,1 million metric sales tons (Mt) in 1970 to an estimated 114,0 Mt in 1980.

  • Hand loading of coal was almost eliminated between 1965 and 1978. (p. 2)

    hand loading of coal decreased from some 45 Mt (90 per cent of the total output) in 1965 to some 13 Mt (14 per cent of total output) in 1978.

  • A handgot colliery would employ roughly four times as many black workers as a mechanized one, which means a large hostel population to control. (p. 2)

    The handgot mine would thus employ 3 640 Black workers and in practical terms this would have required the establishment and running of a large

  • The 1973 strikes in the Witbank coalfield pushed collieries to mechanize earlier because large hostel workforces could not be controlled. (p. 2)

    In 1973, when a large number of strikes with occasional loss of life plagued some of the coal mines in the Witbank area, some collieries decided to phase out handgot mining in favour of mechanized mining

  • Stabilization schemes sharply reduced the need for new recruits through TEBA. (p. 4)

    recruitment figures reported by The Employment Bureau of Africa Limited (TEBA) dropped as follows: January 1977 to 90650, January 1978 to 54203 and January 1979 to 44 508.

  • The share of foreign novices on the mines fell steeply by the end of the 1970s, attributed to employers' fear of sudden withdrawal of foreign labour. (p. 4)

    Of the novices who entered the industry, some 80 per cent were recruited from foreign countries in the mid-1970s, but by the end of the decade this had dropped to 20 per cent.

  • Training is decentralized and short, and black operators remain barred from maintenance and repair work. (p. 5)

    Until the present time, Black operators have not been allowed to undertake maintenance and repair functions.

  • Gold stoping, the key underground operation, has seen no significant mechanization, unlike coal. (p. 5)

    In contrast to coal mining, no significant progress has as yet been made in the mechanization of stoping work on the gold mines.

  • About a quarter of the black gold-mine workforce is in stoping; a modest shift toward semi-skilled work had occurred by 1978. (p. 6)

    By 1978, a shift in skill demands from the unskilled to the semi-skilled level (mainly to B1-type of work on the Paterson scale) had already taken place

Evidence ledger

Sources used

Source Kind Period How used
Department of Statistics, Quarterly Bulletin of Statistics, Vol. 13, June 1979 official statistical series 1975 and 1978 (coal 1960-1978) Provides White and Black employment in iron ore mining (Table 1) and the falling Black/White ratio.
Hansard (House of Assembly Debates), Questions and Replies, 22 February 1977, Qu. 7 parliamentary debates 1977 Documents the Minister's statement that an enquiry into capital-intensity would be made, supporting the claim of no settled policy.
King, Peter, 'Long Term Trends in South African Coal Mining', Coal, Gold and Base Minerals of Southern Africa, Vol. 27 No. 3, March 1979 secondary literature compilation 1965-1980 (estimates) Source of coal output, labour productivity per man-shift, and hand-loading decline figures.
R.P. 63-1975, Report of the Commission of Inquiry into the Coal Resources of South Africa (Petrick Report) commission or official report 1975 Reviews coal mining practices and the potential productivity of different mining methods.
Mine cost data supplied by Dr A. W. S. Schumann (General Mining and Finance Corporation) company records or reports 1977/78 Comparison of capital and running costs between mechanized and handgot mines, and estimated employment (Table 2).
Chamber of Mines, 'Labour Force Becoming More Stable', February Report 1979 chamber of mines or recruiting records 1977-1979 TEBA recruitment figures falling from 90650 to 44508, as evidence of the success of stabilization schemes.
Chamber of Mines, 'Profile of a Novice Mineworker', February Report 1979 chamber of mines or recruiting records mid-1970s to 1979 Shares of foreign novices (80% to 20%) and average length of induction training (9-17 days).
Discussions with staff of the Human Resources Laboratory of the South African Chamber of Mines, mid-1978 interviews oral history 1978 Basis for the estimated allocation of Black workers to stoping tasks (Tables 3 and 4).
Discussions with mining house engineers and coal machinery importers interviews oral history c.1979 Informs the account of coal mining equipment adaptation and work sequences.
Anglo American Corporation, Chairman's Statement 1979 (H. Oppenheimer) company records or reports 1979 Quoted as an example of mining house views on irreversible capital-intensive trends.
Letter from Mineworkers' Union Welfare Officer to H. Griffiths, 28 August 1979 union or movement records 1979 Evidence of continuing union support for job reservation against Black mineworkers.
Spandau, Labour Research Report No. 22 (mimeo), University of the Witwatersrand, September 1979 secondary literature compilation 1978-1995 projection Supplies the author's own model projecting a shift from unskilled to semi-skilled gold-mine employment by 1995.
Baldwin, R.E., 'Wage Policy in a ... Economy - The Case of Northern Rhodesia', Race, 1962; Turner Report (ILO, 1969) secondary literature compilation 1947-1959 / mid-1960s Provides the Zambian Copperbelt wage and mechanization comparison used as the foil for South Africa.
Trade and technical journal articles (Joughin 1976, 1978; Salamon 1976; Mining Survey 1973; SA Mining and Engineering Journal 1978; Gallie 1977) newspapers periodicals 1973-1978 Technical background on stoping, mechanization prospects and work reservation.

Inference

Quantification: Descriptive tables.

Descriptive and analytical argument that combines a Zambian comparison with a few official employment figures, a hypothetical costing of two coal mine types, and a Becker/Oi/Doeringer-Piore theory of internal labour markets. It is applied illustratively to coal and gold mining, with a projection from the author's own model published elsewhere.

Key claims

Claim Basis Strength
In iron ore mining, mechanization cut the Black/White employment ratio from 4,56:1 in 1975 to 2,42:1 in 1978, as skilled White maintenance workers replaced unskilled Blacks. (p. 2) Department of Statistics Quarterly Bulletin figures in Table 1; the explanation is asserted. Partial data
Mechanized coal mining has been more competitive than handgot mining since about 1958 despite low Black wages, and uses far less labour. (p. 2) A single hypothetical mine costing from a mining executive's data (Table 2), plus King's series on hand-loading. Illustrative or anecdotal
The 1973 Witbank strikes and labour control problems pushed collieries to mechanize, so the labour unrest itself reduced employment. (p. 3) Unreferenced narrative account; no cited source. Asserted
Mines stabilized their core workforce through pay, stabilization bonuses and re-engagement certificates, shown by falling TEBA recruitment and a falling share of foreign novices. (p. 4) Chamber of Mines February Report 1979 figures; the causal link is inferred. Partial data
Mechanization has forced changes in coal mine personnel policy: group incentive pay, stabilization and machine-operator training done mine by mine. (p. 4) Interviews with engineers, trade literature and Doeringer-Piore theory; one mining house's bonus figure. Illustrative or anecdotal
Gold mine stoping has not been significantly mechanized: about 142,000 Black workers are in stoping and 63.4% are in unskilled work, with only a small shift to semi-skilled work by 1978. (p. 6) Author's estimates from discussions with Chamber of Mines Human Resources Laboratory staff; no census data exist. Partial data
Gold mechanization between 1978 and 1995 will shift employees from unskilled to semi-skilled and make mine income distribution more equal. (p. 7) The author's own model in a mimeographed report; the details are not given here. Asserted

The author's caveats

The author notes that census data on the occupational composition of the underground workforce are unavailable, so the stoping figures are estimates. He also notes that equipment modifications are too minor to be formally recorded, and that no government findings on capital intensity have been released. Gold mechanization timing is described as the subject of widely differing expert opinion.

Gaps in the evidence

The argument needs systematic time-series data on mine employment by race and occupation, wages, and the capital-labour ratio, instead of one hypothetical costing and oral estimates. Chamber of Mines and individual mining house records on wages, turnover, training costs and mechanization decisions would test the internal labour market claims, and many of these exist in archives. Mine-level data on the 1973 unrest and on the timing of mechanization would be needed to support the claimed causal link.

Replicability: Data poorly documented.