The Poverty of Neo-Marxism: The Case of South Africa¶
B. S. Kantor; H. F. Kenny (1976). The Poverty of Neo-Marxism: The Case of South Africa. Journal of Southern African Studies, Vol. 3, No. 1. Taylor & Francis.
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | theoretical or conceptual |
| Period studied | 1900–1975 |
| Regions | South Africa, Katanga, Northern Rhodesia |
| Themes | comparative theory, mining migrant labour, labour markets wages, business and state, politics ideology elections, bantustans removals influx, manufacturing industrial policy, origins segregation |
| Citation key | kantor1976 |
Argument¶
Kantor and Kenny, economists at UCT, attack the neo-Marxist (Wolpe, Legassick, Trapido, Johnstone) account of South Africa on three fronts. First, it rests on a labour theory of value and assumes migrant labour and influx control served capital, an inference drawn from a single case. Second, it cannot explain conflicts between the state and capitalists, among capitalist groups, or between white and black workers, because it treats politics as derivative of economics. Third, its claim that growth and white supremacy are mutually reinforcing ignores the long-run logic of industrialisation. The authors argue instead that migrant labour is inefficient, that the 'functionality' of the system must be tested rather than asserted, and that a choice between supremacy and industrialisation may eventually be forced. They also criticise liberal stage-theory determinism (O'Dowd).
Evidence and method¶
Secondary literature and published texts: Wolpe, Legassick, Johnstone, van der Horst, Wilson, Jeeves, Berger on Katanga and Northern Rhodesia, Bell on decentralisation, the Keiskammahoek Rural Survey, a 1974 House of Commons report on wages, Blumer, Landes and Arrow. There is no original data or archival work.
Place in the debates¶
- liberal vs revisionist on apartheid and growth (see Liberal vs revisionist: was apartheid functional or dysfunctional for capitalist growth?). Rejects the revisionist view that growth and white supremacy are mutually reinforcing. Argues industrialisation has structural requirements that conflict with white supremacy over the long run. Also criticises naive liberal stage theory (O'Dowd).
- cheap labour thesis / functionality of migrant labour to capital (see Cheap labour thesis and the function of the reserves). Sceptical. Migrant labour is 'not always functional to capital'. Influx control restricts the labour supply and raises its cost. The mines gained but manufacturers and many employers did not, and farmers benefited most from influx control.
- the capitalist state and intercapitalist conflict (see The state and capital: instrumentalism, autonomy and fractions of capital). The state mediates a plurality of interests, mostly white, and pursues political goals. Class analysis is therefore of limited use. This is a pluralist critique of the instrumentalist view of the state.
- Reserves subsidising urban/mine wages (see Cheap labour thesis and the function of the reserves). Notes the argument is older than Wolpe and Legassick (Houghton, Macmillan). Van der Horst argued the Reserves raised bargaining power. The authors call the Wolpe thesis internally inconsistent, since a declining subsidy should push wages up.
Findings¶
-
The authors characterise the neo-Marxist position as a theory in which the state is an epiphenomenon of capitalist interests. (p. 2)
The purpose of the state, logically enough, is to make South Africa safe for capitalism. By its very nature it is an instrument of capitalist rule.
-
They dismiss the labour theory of value as an explanation of racial income gaps and note that alternative explanations involve weighing several factors. (p. 4)
The analytical difficulties arise, of course, not in suggesting explanations for racial income differences, but in allocating appropriate weights to separate influences.
-
Influx control restricts labour supply and so raises labour costs, which cuts against the supposed capitalist interest in it. (p. 4)
Any con- trived scarcity of labour naturally implies more expensive labour.
-
Employers' main difficulty until the 1930s was attracting labour, and Indian and Chinese labour is cited as evidence of this shortage. (p. 5)
at least until the depression of the 1930s, the evidence strongly suggests that the employers' major difficulty was to attract sufficient supplies of labour.
-
Industrial decentralisation largely failed, which undercuts Wolpe's claim that it was a cheap-labour mechanism. (p. 8)
In spite of strong official preference for decen- tralization, the policy has substantially failed.
-
Legassick's account of capital intensity is held to be logically inconsistent. Excess capital intensity is better explained by the migrant system preventing factor proportions from expressing themselves. (p. 9)
excessive capital intensity has been the result of the inability of the factor propor- tions to express themselves as a result of the migrant labour system.
-
Katanga and the Northern Rhodesian Copperbelt show that capital did not uniformly favour migrant labour, so the neo-Marxist inference rests on one case. (p. 10)
Had Wolpe and Legassick looked slightly further north they would have found highly successful examples of capitalist development where migrant labour was deliberately rejected
-
The only defensible conclusion is limited and conditional. (p. 11)
All we are entitled to say is that 'migrant labour is not always functional to capital'.
-
Labour market segmentation by residence rights gives urban permanent workers advantages and makes migrant labour a costly, inefficient way to organise labour. (p. 12)
The high information and trans- action costs inherent in the migrant labour system tend to make such labour less attractive to urban employers, other things being equal
-
The mining industry shows conflicting interests both within capital and between capital and the state, including a conflict between mines and manufacturers over migrancy. (p. 14)
there is a clear conflict of in- terest between mine owners and manufacturers with respect to the desirability of the migrant labour system.
-
The authors argue that the coexistence of apartheid and rapid growth need not imply causation, and that a future choice between supremacy and industrialisation may be forced. (p. 21)
The co-existence of white supremacy and rapid industrial development does not necessarily imply a causal relationship except of the post hoc, propter hoc variety.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| House of Commons Report on Wages and Conditions of African Workers employed by British Firms in South Africa (1974), Para. 32 | commission or official report | 1974 | Cited for its summary of the causes of racial earnings differences (occupational stratification, skill shortage, union and political rights) as an alternative to the labour theory of value. |
| Keiskammahoek Rural Survey, Vol. II: The Economy of a Native Reserve (Houghton and Walton, 1952) | author survey or fieldwork | 1952 | Used to show that migration was typically undertaken by one family member to bring family income up to subsistence level, so cash earnings supplemented land income. |
| Debate on the 1958 Native Taxation Bill (Hansard), as quoted by Legassick | parliamentary debates | 1958 | Quoted to show that the state recognised the burden of urban services on the state rather than employers, undermining the thesis that the state serves capital. |
| Urban Areas Act of 1921 | legislation gazette | 1921 | Cited as evidence that municipalities and the central government, not employers, bore the costs of urban African housing and welfare after the 1918 influenza epidemic. |
| S. T. van der Horst, Native Labour in South Africa (1942) | secondary literature compilation | to c.1942 | Relied on for the debate on the Reserves' effect on wages and the effects of urban services and segregation on labour supply and real wages. |
| F. Wilson, Labour in the South African Gold Mines (1972) and C. R. Diamond MA thesis, University of Cape Town, 1968 | secondary literature compilation | late nineteenth and early twentieth century | Cited for the effect of climatic factors on rural output and mine wage-setting, and for the alternatives of active recruitment versus waiting for workers to present themselves. |
| T. Bell, Industrial Decentralization in South Africa (1973) | secondary literature compilation | to 1973 | Supports the claim that decentralization policy substantially failed, with border-area development confined to areas adjacent to established centres. |
| E. L. Berger, Race, Labour and Colonial Rule (1974) | secondary literature compilation | inter-war period | Source for the Katanga and Northern Rhodesian comparison cases showing migrant labour was not always functional to capital. |
| A. H. Jeeves, 'The control of migratory labour on the South African gold mines in the era of Kruger and Milner', JSAS 2(1), 1975 | secondary literature compilation | Kruger era and reconstruction period | Cited for the finding that the mines–state relationship was complex and variable rather than simple domination. |
| South African Institute of Race Relations, A Survey of Race Relations in South Africa, 1969 | official statistical series | 1969 | Cited for organised industry's and commerce's criticism of the Physical Planning Act as evidence of conflict between employers and the state. |
| Kooy and Robertson, 'The South African Board of Trade and Industries; the South African Customs Tariff and the development of South African industries', SAJE 1966 | secondary literature compilation | 1920s | Cited for the claim that 1920s industrial protection aimed partly at providing white employment. |
| Statutory job reservation (instituted 1956) and pass law contraventions, referred to without citation | legislation gazette | 1956 onwards; post-war | Referenced in general terms to argue that job reservation lost significance and pass offences reveal labour market pressure; no figures given. |
Inference¶
Quantification: None.
Theoretical and logical critique of the neo-Marxist (Wolpe, Legassick, Johnstone) writings, using orthodox economic reasoning, selective historical illustration (Katanga, Northern Rhodesia, gold mines-state relations) and appeals to other scholars' work (Landes, Blumer); no systematic data or tests.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Neo-Marxist reliance on the labour theory of value is unhelpful because it makes any wage–output gap exploitation by definition and cannot weight alternative causes of racial income gaps. (p. 21) | Theoretical argument plus the UK House of Commons 1974 report quoted on the causes of earnings differences. | Asserted |
| Migrant labour is not always functional to capital, as shown by Katanga and Northern Rhodesia, where companies or administrations favoured or accepted stabilised labour or fluid arrangements. (p. 29) | Berger's Race, Labour and Colonial Rule used as a secondary case comparison. | Illustrative or anecdotal |
| Influx control and migrant labour do not straightforwardly serve all capitalists; employers generally want a wider labour pool, and the main beneficiaries of influx control were farmers. (p. 33) | Economic reasoning, the Physical Planning Act criticism cited from SAIRR 1969, and the history of mine recruitment. | Illustrative or anecdotal |
| Reserves' subsidy of urban wages is complex and may not hold; employers' main pre-1930s difficulty was attracting labour, and wages were shaped by supply and demand with the Reserves giving workers bargaining power. (p. 23) | Van der Horst, the Keiskammahoek survey, Wilson and Diamond, and the author's reasoning. | Partial data |
| Industrial decentralization has largely failed because industrialists preferred to stay in the towns despite state inducements, contradicting Wolpe's cheap-labour interpretation. (p. 26) | Bell (1973) on border industries and Rosslyn. | Partial data |
| There is no unified capitalist interest; the state, mines, manufacturers and farmers have conflicting interests, which neo-Marxism cannot accommodate. (p. 33) | Jeeves, SAIRR 1969 on the Physical Planning Act, and general knowledge of mining taxation and subsidy policy, mostly without cited data. | Illustrative or anecdotal |
| Industrialization and white supremacy are not mutually reinforcing in the long run; post-war colour bars have eroded and Africans have moved into skilled and semi-skilled jobs, though the compatibility is only temporary. (p. 39) | Theoretical authorities (Landes, Blumer) and unquantified assertions about the post-war occupational structure. | Asserted |
The author's caveats¶
The authors note that there is no reliable index of rural un- and under-employment, and that some of their explanations are 'tentative'.
Gaps in the evidence¶
The argument never tests its claims with systematic data: no wage series by race, occupational census data, migrant labour numbers, or Chamber of Mines recruiting and cost records, which existed and could have tested whether Reserves subsidized wages or employers bore costs. Claims on decentralization, capital intensity, job-reservation decline, and African upgrading are asserted without figures from the population and industrial censuses or Department of Labour statistics. The comparative cases rest on a single secondary source (Berger) and Hansard or archival evidence on employer lobbying is not examined, though it is available.
Replicability: Not applicable.