Contradiction and Uneven Development in South Africa: The Constrained Allocation of African Labour-Power

Literature › Works

Fred Curtis (1984). Contradiction and Uneven Development in South Africa: The Constrained Allocation of African Labour-Power. The Journal of Modern African Studies, Vol. 22, No. 3. Cambridge University Press.

Type	journal article
Scope	Economics and economic history of apartheid
Method	theoretical or conceptual
Period studied	1948–1981
Regions	South Africa, Homelands/Bantustans, urban townships, Ciskei
Themes	labour markets wages, bantustans removals influx, business and state, comparative theory, inequality poverty unemployment, mining migrant labour, agriculture land, macro fiscal finance debt
Citation key	curtis1984

Argument

Curtis argues that the post-war 'cheap labour system' of labour bureaux, pass laws and influx control was neither simply irrational and profit-reducing (the neoclassical view) nor uniformly functional and profit-enhancing (the traditional Marxian view). Using a Resnick-Wolff Marxian value framework, he treats the state's allocation of African labour-power as a 'subsumed class process'. It lowers wages but costs surplus value in taxes, supervisors and lower productivity. He examines five aspects: labour bureaux, taxation of African wages, state-subsidised housing and transport, in-kind food and housing from employers, and gender differences. Each is shown to have uneven and contradictory effects across groups of African workers. He closes by suggesting that the rising costs of repression and 'exporting poverty' to the Homelands contributed to increased foreign borrowing.

Evidence and method

Secondary statistics, mainly from the Survey of Race Relations, Wilson's gold-mine data, Savage on the costs of enforcing apartheid, Perlman on subsidies, and Lapchick and Urdang on women's occupations. Also draws on the author's own dissertation. Presented in six tables on wages versus the poverty datum line, racial wage gaps, sectoral earnings, subsidies, in-kind farm wages and women's occupations.

Place in the debates

• Neoclassical vs Marxian views of apartheid and the cheap labour system (see Cheap labour thesis and the function of the reserves). Rejects both. Against the neoclassical view, he holds that apartheid is not simply irrational or profit-reducing. Against traditional Marxian analysis, he holds that it is not uniformly functional. Its effects are uneven and contradictory.
• Cheap labour thesis (Wolpe, Legassick, Magubane) (see Cheap labour thesis and the function of the reserves). Accepts that labour-power was cheapened but extends the thesis. The mechanisms carry surplus-value costs, can reduce productivity, and affect migrants and non-migrants, miners and farmers, and men and women differently.
• Liberal/neoclassical view that firms will erode apartheid constraints on mobility (see Liberal vs revisionist: was apartheid functional or dysfunctional for capitalist growth?). Criticises its omission of power, conflict, force and class. It can explain apartheid's persistence only through the 'abnormal psychology' of racist whites.

Findings

• Labour bureaux allocated rapidly growing numbers of Africans and, together with pass laws and influx control, are what is usually meant by the cheap labour system. (p. 7)

  from I59,494 in 1952 to I,I37,162 in I966 and 4,018,733 in I9811 - reinforced by pass laws and influx controls which regulate the movement of Africans

• The bureau system was costly to operate, absorbing part of the surplus value it helped extract. (p. 8)

  It has been estimated that for I974-5 the costs of operating the labour bureaux were as much as RII2 million.

• Reserve army size has no automatic inverse effect on wages; effects are mediated by economic, political and cultural factors. (p. 8)

  Thus, the size of the reserve army does not have automatic inverse effects on the wage. Rather, its effects are mediated by economic, political, and cultural factors.

• Rising taxes on African wages, enforced by mandatory employer withholding from the mid-1960s, cut real consumption and so productivity. (p. 10)

  Further, these had a greater impact after tax-withholding out of African wages was made mandatory for employers in the mid-Ig60s.

• State subsidies for urban African transport and housing partly offset the fall in disposable income, and were won through protest against fare and rent rises. (p. 12)

  Between 1952 and I977, the total state subsidy of bus and train transportation specifically for urban Africans was R352 million

• Employers in mining and agriculture paid substantial in-kind wages, which raised the value of labour-power and reduced its cheapness. (p. 13)

  the cost of the food provided for African gold miners was equivalent to 29 per cent of their average earnings

• Women were allocated to a narrow range of low-wage jobs and benefited least from subsidies and in-kind payments. (p. 15)

  78 per cent of African women were estimated to have been not economically active in I 980

• African manufacturing wages rose from 56% of the poverty datum line in 1950 to 94% in 1981. (p. 3)

  1950 35'5 20 56 1958-9 48-4 28 58 I97I 64-4 56 87 I978 i6II9 I49 92 1981 2717I 255 94

• Late-1970s policies of relocating the unemployed to the Homelands and cultivating a black middle class required more state spending on police and military, which pushed South Africa toward foreign borrowing. (p. 17)

  it attempted to 'export poverty' to the so-called Homelands by relocating virtually all unemployed Africans (particularly women) there.

Evidence ledger

Sources used

Source	Kind	Period	How used
Survey of Race Relations in South Africa (SAIRR annual), various years, incl. 1982 edition	official statistical series	1950-1981	Supplies African and white wage series, PDL comparisons, farm in-kind payments, corporate tax figures and beer/liquor profits used in Tables 1-3, 5 and the tax discussion.
Poverty Datum Line estimates for Johannesburg (Potgieter; Horrell; SAIRR)	secondary literature compilation	1950-1981	Benchmark for judging whether African manufacturing wages were below subsistence (Table 1).
Curtis, 'The Value of African Labor-Power in South Africa: 1948-1978', Ph.D. dissertation, Amherst, 1983	secondary literature compilation	1948-1978	Source of wage data for Tables 2 and 3, plus supervisor and tax material.
Jill Nattrass, The South African Economy (1981) – GDP series	national accounts macro	1950-1979	Establishes GDP growth from R4.4bn to R15.5bn in constant 1970 prices.
Labour bureau registration and allocation figures (van der Merwe 1969; Survey of Race Relations)	official statistical series	1952-1981	Documents the number of labour bureaux and the growth in Africans allocated through them.
Michael Savage, 'Costs of Enforcing Apartheid and Problems of Change' (African Affairs, 1977)	secondary literature compilation	1974-75	Supplies the estimated R112 million cost of operating the labour bureaux.
J. Perlman, The State of the African Working Class in the Pretoria-Odi Area (1982)	secondary literature compilation	1952-1977	Source of transport subsidy totals and state housing spending in Table 4.
Irving Kaplan et al., Area Handbook for the Republic of South Africa (1971)	secondary literature compilation	to 1971	Documents African poll tax and special levies.
Francis Wilson, Labour in the South African Gold Mines 1911-1969 (1972)	secondary literature compilation	1911-1969	Used to compute that food for gold miners equalled 29 per cent of average earnings and to show employer rationale.
Alide Kooy, 'Farm Labour in the Karoo' (1977)	secondary literature compilation	1970s	Evidence that employers gave housing to farm labourers to compensate for the rural-urban wage gap.
Legislation on urban African housing, transport and levies (as summarised in Horrell, Laws Affecting Race Relations)	legislation gazette	1950s-1970s	Establishes the statutory basis for transport levies and the beer/liquor revenue subsidies.
Kraak, 'Financing of Worker Accommodation in Cape Town' (SARDU Working Paper 35, 1981)	secondary literature compilation	1981	Evidence on housing finance and the effect of the 1976 beerhall destruction on rents and fares.
Lapchick and Urdang, Oppression and Resistance (1982) – occupations of African women	secondary literature compilation	1960-1970	Table 6 data on the occupational distribution of African women.
Yawitch, 'The Incorporation of African Women into Wage Labour, 1950-1980'	secondary literature compilation	1950-1980	Source for estimate that 78 per cent of African women were not economically active in 1980.

Inference

Quantification: Descriptive tables.

Theoretical, Marxian value-theory argument (Resnick-Wolff subsumed class concepts) illustrated by secondary and published figures in descriptive tables. It uses a qualitative case-by-case examination of five aspects of the cheap labour system and offers no statistical testing.

Key claims

Claim	Basis	Strength
African wages were below the poverty datum line and only a fraction of white wages, though the gap to the PDL narrowed over 1950-1981. (p. 381)	Survey of Race Relations series for Tables 1 and 2, with PDL for Johannesburg.	Partial data
Labour bureaux allocated rapidly growing numbers of Africans, mostly to mining and agriculture, depressing wages there and reversing the 1940s urban drift. (p. 386)	Bureau allocation counts from van der Merwe and the Survey; wage and employment in Table 3.	Partial data
The allocation system was costly and drained surplus value, with operating costs of up to R112 million in 1974-5. (p. 387)	Single estimate from Savage (1977).	Partial data
Taxation of African wages rose in the post-war period and cut real consumption, productivity and surplus value extraction. (p. 389)	Descriptive account of taxes (Kaplan; Survey); no measurement of the productivity effect.	Asserted
State subsidies for urban African housing and transport partly offset wage cuts for workers in metropolitan areas, and were won through struggle. (p. 391)	Perlman subsidy totals in Table 4 and anecdotal accounts of boycotts and protests.	Partial data
In-kind food and housing formed a large part of miners' and farm labourers' remuneration, which raised the value of labour-power and reduced cheapness. (p. 392)	Wilson for the 29 per cent mining food cost; Survey data for Table 5 (valuation unspecified).	Partial data
African women bore the system's effects more heavily than men: restricted job allocation, lower wages, less subsidy and fewer in-kind benefits. (p. 395)	Table 6 occupations and Yawitch, Shapiro and Cock; no wage series by sex.	Illustrative or anecdotal
The cheap labour system had uneven and contradictory effects on capitalism, neither purely irrational nor purely functional. (p. 395)	Conceptual argument from the five aspects; no direct measurement of profit or surplus value.	Asserted

The author's caveats

Notes that Table 5 data vary by reporting area and the valuation of in-kind use-values is not specified. Assumes wage equals value equals price of labour-power and omits racial wage differentials and non-capitalist aspects for reasons of length.

Gaps in the evidence

The argument's central quantities (value of labour-power, surplus value, rate of surplus value, productivity effects) are never measured; the claimed contradictions rest on theoretical assertion and ceteris paribus reasoning. Direct evidence would include firm and mine-level wage and productivity data, Chamber of Mines and farm records, and household expenditure and nutrition surveys, much of which existed at the time or is now archived. The sex-differentiated claims lack wage series by sex, and the tax and subsidy effects lack incidence estimates from government budgets and Hansard.

Replicability: Data poorly documented.