Skip to content

Crossing Boundaries: Mine Migrancy in a Democratic South Africa

Literature › Works

Crush, Jonathan (ed.); James, Wilmot (ed.) (1995). Crossing Boundaries: Mine Migrancy in a Democratic South Africa. IDASA / International Development Research Centre (Canada).
Type book
Scope Economics and economic history of apartheid
Method mixed
Period studied 1902–1994
Regions Witwatersrand and Orange Free State goldfields, Lesotho, Mozambique, Malawi, Transkei and Ciskei, southern Africa (SADCC region)
Themes mining migrant labour, labour markets wages, unions strikes, transition post apartheid, origins segregation, demography urbanisation, inequality poverty unemployment, business and state, macro fiscal finance debt
Citation key crush1995a

Argument

This edited volume stems from a June 1994 conference. It examines how the gold-mine migrant labour system, built on the segregationist settlement, has persisted and changed. It argues that migrancy was never an economic necessity: it was kept in place by a political bargain among mining capital, white labour and farmers, and by the state. Since the 1970s, internalisation, career mining or 'inflexible migrancy', computerised control, the NUM, and downscaling after 1987 have changed the form but not the content of the system. The costs of retrenchment fall on rural South Africa and on Lesotho and Mozambique. The editors call for 'normalisation' of miners' status and for regional policy, set against a backdrop of crisis in gold mining.

Evidence and method

Conference papers drawing on Chamber of Mines and Teba statistics, company records, government commissions, interviews and focus groups with miners and wives, an NUM survey, HSRC hostel research, a household survey in Mandela Park, case studies of co-operatives, a Lesotho national accounts series (1980-91), and epidemiological and autopsy data. Two chapters are earlier journal articles (Labour, Capital and Society 1992).

Place in the debates

Findings

  • Entrenchment of migrancy was political, not dictated by mining economics. (p. 31)

    The entrenchment of the system should not be seen as the inevitable product of economic necessity in the mining industry.

  • The 1911 Union laws settled mine labour policy through a white-supremacist political compromise. (p. 25)

    Significantly, the economics of mining did not directly dictate this outcome. It was more the result of a political bargain laboriously constructed in 1906-7.

  • The 'unwritten law' limiting married quarters meant virtually no stabilisation in 1950-75. (p. 30)

    Between 1950 and 1975, no gold mine even came close to the 3 percent figure.

  • By the early 1970s the mines were heavily dependent on foreign migrants. (p. 30)

    By the early 1970s, almost 80 percent of mineworkers were migrants from Malawi, Mozambique, Angola, Botswana, Lesotho and Swaziland.

  • The mid-1970s shock to foreign supply came from Malawian and Mozambican decisions. (p. 37)

    Malawian president Hastings Banda's recall of 120 000 Malawian miners from South Africa in 1974, and a drop in the Mozambican complement from 97 000 to 35 000

  • By the 1980s foreign labour was discretionary: the mines could have replaced it with unemployed South Africans. (p. 38)

    The mining industry could probably have replaced all unskilled and semi-skilled foreign labour with South Africans over the course of the 1980s with only minimal disruption to production and profit.

  • Downscaling after 1986 hit South African miners harder than foreign miners. (p. 40)

    In the last six years the gold-mining industry has lost nearly 180 000 jobs to down-scaling and retrenchments (Seidman, 1993).

  • Career mining or 'inflexible migrancy' tied miners to one mine, with very high return rates. (p. 44)

    Return rates on most mines had reached over 90 percent by the end of the decade.

  • The gold crisis was mainly a revenue problem, and gold's fiscal contribution collapsed. (p. 186)

    The proportion of tax revenue collected from the gold-mining industry dropped from 26,4 percent in 1981 to 2,0 percent in 1992.

  • The 1987 NUM strike was defeated at heavy cost to workers. (p. 189)

    The NUM was forced to accept the original pay offer at the end of the strike which cost 43 000 miners their jobs and in which R105 million in wages was forfeited

  • Coplan and Thoahlane argue retrenchment was partly a strategy against the NUM and for cutting labour costs. (p. 164)

    Large-scale retrenchment, we contend, is part of a strategy to undermine the NUM, reduce militancy, avoid the consequences of collective bargaining, reduce labour costs

  • Lesotho statistical evidence suggests migration does not raise domestic output. (p. 178)

    However, the overall implication of the evidence is that the GNP multiplier is less than one (and autonomous GDP multiplier negative)

  • Mozambique's role in mine employment has fallen sharply since independence. (p. 148)

    Today they make up only 13,6 percent of the total labour-force and 28,4 percent of the foreign labour-force on the gold mines

Evidence ledger

Sources used

Source Kind Period How used
Chamber of Mines annual reports, statistical tables and employment series by source area and mining group, 1960-1993 chamber of mines or recruiting records 1960-1993 Establishes foreign versus South African employment shares, retrenchment patterns by origin, and lay-offs by mine and mining house (Tables 2-6).
Anglo American Corporation accommodation figures for black miners, 1987 and 1993 (Table 1) company records or reports 1987-1993 Shows the share of workers in hostels, married housing, home-ownership and living-out arrangements.
Moroney 1982 on mine married quarters and Witwatersrand tin-town surveys, 1913 secondary literature compilation 1900-1920 Supplies counts of married quarters and tin-town residents showing early stabilisation experiments.
Survey of 40 gold mines, March 1974, and Chamber/AAC married housing figures (via Crush, Jeeves and Yudelman 1991) secondary literature compilation 1950-1974 Shows only 0,7 percent of black miners in married quarters and the failure of the 3 percent rule to be approached.
Chamber of Mines testimony to government commissions of inquiry (1950s) and Gemmill evidence of 1952 commission or official report 1940s-1950s Documents the Chamber's defence of migrancy and its stabilisation targets (12 percent, then 3 percent).
Rand Daily Mail, 3 February 1953 newspapers periodicals 1953 Cited for Verwoerd's public rejection of Oppenheimer's family-village plan.
Union legislation: Native Labour Regulation Act and Mines and Works Act (1911); related Acts legislation gazette 1911-1930s Shows the legal entrenchment of migrancy and the colour bar as a political compromise.
Historical commissions on mine labour (Buckle 1914, Holloway 1932, Fagan 1948, Froneman, Du Randt, Riekert, Wiehahn) commission or official report 1914-1980 Referred to as the historic debate on migrancy, with no data drawn from them.
Teba Manpower Data Centre records and Teba estimates of employment histories chamber of mines or recruiting records early 20th century-1990s Illustrates the surveillance capacity of the computerised record system.
Crush's interviews with Transkei miners on computerised ID cards interviews oral history c.1990-1993 Supports the claim that miners see more advantages than disadvantages in the ID system.
De Vletter 1986/1990 comparative skills survey of Anglo American foreign and South African miners secondary literature compilation 1986 Evidence that foreign miners are concentrated in higher skill grades.
Government Mining Engineer annual reports RP 25/1983 and RP 46/1987; Marais Committee reports commission or official report 1982-1990 Used for white employment trends and the Kinross disaster characterisation.
Chamber of Mines reviews, newsletters and annual reports (1988-1993) and Gold Fields Mineral Services Gold 1991/1993 company records or reports 1981-1993 Supplies the series on employment, profitability, gold price, NUM membership and world production used for the profit-squeeze argument.
Gold Fields Mineral Services, Gold 1993 (world gold production) other 1981-1992 Shows the fall in South Africa's share of world gold production.
Flynn 1992 account of the Kinross disaster and Gencor documents secondary literature compilation 1980-1986 Supports the claim that the 1986 Kinross fire was preventable and cost-driven.
Human Sciences Research Council hostel study for the Goldstone Commission, 1993 commission or official report 1992-1993 Counts hostels countrywide and mine hostels by region and owner (Appendices A and B).
Department of Local Government and National Housing (1991) and CSIR (1992) hostel bed counts official statistical series 1991-1992 Estimates the number of hostel beds and notes that residents cannot be counted precisely.
Goldstone Commission evidence on the President Steyn mine violence, 1991 commission or official report 1991 Source for NUM allegations about firearms smuggling and security complicity.
HSRC/Goldstone hostel interviews at three mine hostels, 1992-1993 (Payze) interviews oral history 1992-1993 Provides residents' views of hostel conditions and accommodation preferences.
McNamara and Schlemmer 1988 attitude survey, Evander gold mines author survey or fieldwork 1987-1988 Shows Basotho and Pondo attitudes after the 1987 gate violence.
Interviews by Vivienne Ndatshe with Xhosa-speaking women, Vaal Reefs area, 1990 interviews oral history June-December 1990 Supplies the life histories on town women, country wives and housing preferences.
Market Research Africa 1986 Anglo American home-ownership survey; NUM 1990 Final Housing Survey; Urban Foundation 1990 household or labour survey microdata 1986-1990 Used to show worker interest in, and affordability barriers to, home ownership.
Molapo interviews and screening at Western Deep Levels; Employee Assistance Centre 1992 report; medical station monthly report author survey or fieldwork 1992-1993 Documents job stress, health perceptions and presenting health problems among Basotho miners.
Employee Assistance Centre Annual Report 1992 (Western Deep Levels) company records or reports 1992 Source for the ranking of alcohol, STDs, HIV and depression as presenting problems.
Epidemiological Comments (national TB notification and HIV antenatal surveys) 1991-1993 official statistical series 1971-1992 Provides national TB notification rates and age-specific incidence by race, and HIV prevalence.
Chamber of Mines Annual Health Department report 1992 and mine TB in-patient series chamber of mines or recruiting records 1974-1992 Gives mine TB incidence trend (10,6/1 000 in-patients in 1992).
Pathaut autopsy study, National Centre for Occupational Health other 1975-1991 Autopsy data used to correlate years of service with TB prevalence.
Interviews with mine medical officers, August 1993 (Packard and Coetzee) interviews oral history August 1993 Establishes screening practices and silicosis-diagnosis practices that distort the data.
Malawi AIDS Secretariat reports and Malawian interviews (Chirwa) interviews oral history 1985-1994 Interviews with 163 returned migrants and official Aids data underpin the discussion of migrancy and HIV risk.
Malawi AIDS Secretariat/hospital case reports official statistical series 1985-1993 Used to show Aids was entrenched in Malawi before the 1987 South African ban.
Teba/Wenela staff interviews, telex orders and Teba fee structure (Head) interviews oral history 1987-1994 Supports claims that gate hiring is replacing Teba recruitment and that fees disadvantage foreign labour.
Anglo American Corporation 1986 and 1993 workforce surveys (De Vletter 1986; AAC 1993 tables) company records or reports 1986-1993 Used to compare age, education and length of service of Mozambican and South African workers.
Interviews with more than 60 ex-miners in Lesotho, Setsabi et al 1992, Sechaba Consultants 1994, Lesotho Bureau of Statistics interviews oral history 1992-1994 Establishes experiences of retrenchment, benefit denial and rural conditions among Basotho ex-miners.
Setsabi, Lesaoana and Molefi 1992 study of Basotho miners and Sechaba Consultants 1994 secondary literature compilation 1986-2001 Source of unemployment, mineworker share of male population and remittance statistics.
Interview with Puseletso Selae, NUM representative, Maseru, 10 January 1994 interviews oral history January 1994 Supports claims about NUM Maseru work and alleged non-necessity of retrenchments.
Lesotho Bureau of Statistics national income accounts 1980-1991 and Cobbe's migrant earnings series national accounts macro 1980-1991 Annual series used for OLS regressions of GDP/GNP on migrant earnings and investment.
World Bank data on Lesotho growth and World Development Report 1992 international trade or finance data 1965-1991 Gives comparative GNP and GDP growth rates for the region.
Seidman survey of 80 households in Mandela Park, Virginia and interviews in Virginia and Maseru, 1992-1993 author survey or fieldwork 1992-1993 Documents the local social impact of down-scaling at Harmony mine.
Seidman snowball sample of retrenched Harmony workers in Maseru interviews oral history 1993 Shows the plight of ex-Harmony workers in Lesotho.
NUM Development Unit records on retrenchments, co-operatives and micro-loans union or movement records 1987-1994 Source for NUM job-creation programme and Finsch retrenchment case.
Interviews with leaders and members of Lesotho co-operatives, May 1994 interviews oral history May 1994 Evidence on income, problems and prospects of ex-miner co-operatives.
South African Department of Home Affairs deportation figures and UNHCR Maputo interviews official statistical series 1988-1993 Used as the only available indicator of clandestine migration.
Lesotho Labour Force Survey 1985/6 (Lesotho Bureau of Statistics 1990) household or labour survey microdata 1985-1986 Shows wage employment and migrant shares of the economically active population.
Teba operations manager interviews and Teba capitation data (Davies and Head) interviews oral history 1993-1994 Shows the gap between mine strength and men processed by Teba.
Teba internal recruitment and payment statistics (Rowett) chamber of mines or recruiting records 1988-1994 Documents the decline of recruitment and Teba's changing functions.

New data

Several chapters rest on primary fieldwork: Seidman's random survey of 80 Mandela Park households plus 10 Maseru interviews; Coplan and Thoahlane's 60+ ex-miner interviews; Chirwa's 163 migrant interviews; Payze's focus groups at three hostels; Ndatshe's interviews with Xhosa-speaking women; Molapo's 400 interviews and 800 blood-pressure screenings; Head's Teba/mine management interviews. Crush assembles Tables 2-6 on foreign labour and lay-offs by mine, and Cobbe constructs a 1980-1991 Lesotho series with a derived migrant earnings series.

Inference

Quantification: Statistical analysis.

An edited conference volume combining mostly descriptive tables from Chamber of Mines and official statistics, historical narrative based on secondary literature, qualitative interviews and focus groups, small surveys, and one set of simple OLS regressions on short Lesotho national-accounts series. Claims rest on descriptive trends, comparison across mining houses, origin areas and time, and case studies of individual mines and hostels, with policy argument layered on top.

Key claims

Claim Basis Strength
Entrenchment of migrancy on the gold mines was the outcome of a political bargain (colour bar, state-regulated recruitment), not of mining economics alone. (p. 25) Jeeves and Crush's historical narrative drawing on Moroney, Crush et al 1991, the 1911 Acts and commission evidence; no new primary data. Partial data
The foreign share of the gold mine workforce stabilised around 1980 and rose slightly in the 1980s, contradicting 'internalisation' predictions. (p. 37) Chamber of Mines employment series by origin (Figure 1, Table 2). Systematic data
Retrenchments in 1986-1992 fell disproportionately on South African miners and rural source areas rather than foreign workers. (p. 40) Mine-by-mine and mining-house employment tables (Tables 3-6) derived from Chamber of Mines data. Systematic data
Despite reforms since the 1980s, the great majority of black miners remain migrants in hostels; family housing and home ownership reach only a small elite. (p. 34) Anglo American accommodation Table 1 plus company survey and NUM housing survey figures. Partial data
The rising TB incidence among mineworkers is not simply a reflection of background TB trends; ageing, longer service, HIV and silica exposure may contribute. (p. 130) COM health series, national notification data, Pathaut autopsy data, interviews with medical officers, hospital HIV/drug-resistance data. Partial data
Retrenchment of Basotho miners since 1987 is a deliberate strategy to undermine the NUM and cut labour costs, with benefits misappropriated. (p. 164) Interviews with 60+ ex-miners and NUM officials in Lesotho; no company records. Illustrative or anecdotal
The gold industry's profitability crisis was driven mainly by falling real revenue (gold price, ore grade) rather than rising costs. (p. 186) Chamber of Mines statistics, Gold Fields data and Nattrass's own 1993 working paper. Partial data
Migration from Lesotho reduces domestic output and the GNP multiplier of migrant earnings is below one, though more migrants may raise agricultural output. (p. 178) OLS regressions on 12 annual observations of Lesotho national accounts and derived migrant earnings series. Partial data

The author's caveats

Cobbe says the statistical evidence is 'in no sense conclusive; it is suggestive, no more', the series is short with large measurement errors and no weather proxy. Crush warns against reading too much into lay-off tables without within-grade data. Packard and Coetzee note that African TB rates may be artificially low and that diagnostic practices distort silicosis data. Chirwa notes testing and reporting are unreliable. Payze's hostels are anonymised and some residents refused interviews, and Seidman cannot establish how typical her Maseru sample is. Coplan and Thoahlane concede they have no reliable statistics on abandoning wives.

Gaps in the evidence

Many claims about management motives (deliberate favouring of foreign labour, retrenchment as union-busting, benefit misappropriation) rest on interviews and aggregate lay-off tables; mine-level personnel records by job grade, union status and nationality would test them, and may exist in Teba and company archives. The Lesotho regressions need longer, consistent national accounts and household panel data on remittances and output. Systematic TB, HIV and wage data from the Chamber of Mines and mining houses, now partly available, would allow stronger tests than the aggregates and small samples used here.

Replicability: Data described but not public.