The Poverty of Coloured Labour Preference: Economics and Ideology in the Western Cape¶
Ian Goldin (1984). The Poverty of Coloured Labour Preference: Economics and Ideology in the Western Cape. SALDRU Working Paper No. 59. Southern Africa Labour and Development Research Unit, University of Cape Town.
| Type | working paper |
| Scope | Economics and economic history of apartheid |
| Method | mixed |
| Period studied | 1795–1982 |
| Regions | Western Cape, Cape Peninsula, Cape Town |
| Themes | bantustans removals influx, labour markets wages, business and state, unions strikes, politics ideology elections, inequality poverty unemployment, origins segregation, demography urbanisation |
| Citation key | goldin1984 |
Argument¶
Goldin argues that the Coloured Labour Preference Policy (CLPP) is rooted in a long colonial and segregationist history, and is driven mainly by political-ideological aims: allying with 'Coloured' people against the black majority, a 'white homeland' idea, and apartheid ideology. It has not improved the position of the bulk of Coloured workers, especially the unskilled. Its main economic effect has been to intensify influx control and to fragment African labour into 'permanent', 'contract' and 'illegal' strata, which employers exploit. Employers backed influx control but disliked the CLPP's rigidity and the way it blocked African upward mobility. The argument rests on archival and parliamentary sources, employer and union interviews, census data and a 1981/82 survey of 383 hostel dwellers. Goldin finds that workers see the legal divisions as unimportant and are united in their interests.
Evidence and method¶
Hansard, Theron, Riekert and Wiehahn commission reports and evidence, Cape Chamber of Industries yearbooks, SAIRR race relation surveys, census data for the Cape Peninsula 1904-1970, interviews with employers (a grid sample of 20 firms employing 22,000 workers) and unionists (78% of the unionised Western Cape workforce), a survey of 383 African hostel dwellers, and Schneier's township household survey for comparison.
Place in the debates¶
- Origins and purpose of the Western Cape preference policy (Lacey's labour-supply explanation vs a political/ideological one) (see Coloured Labour Preference policy in the Western Cape). Goldin calls Lacey's view that strict control served to shore up Eastern Cape labour for the mines 'incomplete'. He stresses Hertzog's Coloured alliance, the Depression, and the National Party's strategic and ideological aims.
- Whether the state and employers act as a unified conspiracy (cheap labour thesis and structuralist readings of apartheid) (see Coloured Labour Preference policy in the Western Cape). Goldin rejects 'conspiratorial reified' interpretations and emphasises contradictions between employers and the state, while still reading labour control as maintaining a cheap and divided workforce.
- Did the CLPP benefit Coloured workers? (see Coloured Labour Preference policy in the Western Cape). It did not benefit the unskilled bulk; Coloured wages relative to whites fell and Coloured workers were confined to the working class.
- Are permanent and migrant African workers divided by interest? (see Worker consciousness, strikes and conflict on the mines and shop floor). His survey found no significant perceived division, challenging the 'insider/outsider' divide-and-rule hypothesis.
Findings¶
-
Contract labour in the Peninsula fell sharply under the CLPP, with the contract share of registered African workers dropping to about a third. (p. 16)
The number of contract workers has fallen from a peak of 47 554 in 1960 to around 18 000 at the present time.
-
The state froze black employment in Western Cape firms and then ordered annual reductions. (p. 13)
On the 31st August 1966 the black labour complement in all industrial and commercial enterprises in the Western Cape was frozen.
-
Employers backed influx control but objected to its uneven enforcement and its effects on costs. (p. 18)
What worried employers was not influx control itself but its uneven enforcement throughout South Africa.
-
Employers preferred migrants for their insecurity, which made them docile and cheap. (p. 25)
contract blacks give even less trouble than local blacks because they're so scared of getting sent back to the Transkei or Ciskei or starving to death.
-
The CLPP did not lift the bulk of Coloured workers, and the unskilled faced bleak prospects. (p. 46)
it is clear that the Coloured Labour Preference Policy, has not advanced the position of the bulk of the 'Coloured' population.
-
The employer survey shows a stark racial division of labour: most African workers were unskilled and most whites in managerial or skilled grades. (p. 46)
eighty-seven percent of Africans are employed in unskilled grades, seven percent in semi-skilled, six percent in skilled and a negligible proportion in the managerial grades.
-
The wage differential gave employers a financial incentive to hire Africans over unskilled Coloured workers. (p. 51)
the wage differential between 'Coloured' and African workers in the unskilled occupations which is of the order of 1.46:1.
-
The survey found very little union representation of contract and illegal workers. (p. 82)
seventy percent of the contract workers interviewed and seventy-seven percent of the illegal workers interviewed were unaware of any workers' orgainsation at their place of work
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Employers' postal survey (grid sample) and in-depth interviews with twenty representative firms (22 000 workers) | author survey or fieldwork | 1981/82 | Provides the racial-occupational distribution of the workforce (Table 6) and employer attitudes to Coloured, contract and illegal labour. |
| Interviews with trade unionists, employers and officials (e.g. Lewis, Fletcher, Theron, Allen, McLeod, Lawrence, de Reuck) | interviews oral history | 1979-1982 | Supplies employer attitudes to contract/illegal/Coloured labour and union views on the CLPP and racial union structures. |
| Schneier survey of random sample of houses in Langa, Nyanga and Gugulethu (non-hostel residents) | author survey or fieldwork | 1981/82 | Comparison group for legal status, education, employment and occupation against hostel dwellers (Tables 8, 9, 13, 16). |
| Population Censuses 1904-1970 for the Cape Peninsula (magisterial districts of Cape Town, Wynberg, Simonstown, Bellville) | population census | 1904-1970 | Shows population growth and changing racial composition of the Peninsula (Table 3). |
| 1970 Population Census as presented by Hendrie and Horner, SALB Vol 3 No 2 | secondary literature compilation | 1970 | Gives sectoral and occupational distribution of African and Coloured workers in the Peninsula versus the country. |
| Theron Commission Report (RP 38/1976) and its occupational statistics (Tables 4 and 5) | commission or official report | 1960-1980 (estimates) | Supplies occupational structure of the Coloured labour force, the wage ratio, the skilled/unskilled trends and the commission's conclusions on unskilled Coloured prospects. |
| Cilliers and Bekker, Die Arbeidsituasie van die Swartman in die Wes-Kaap (1980) employer survey; Bekker and Coetzee (1980) employer survey 1978 | secondary literature compilation | 1978-1980 | Used to cross-check the author's employer survey and hostel survey occupational distributions and employer preferences for African unskilled labour. |
| Simkins and Hindson, 'The division of labour in South Africa 1969-1977' and Riekert Commission data | secondary literature compilation | 1969-1977 | Supports the finding that over eighty percent of the African workforce is unskilled and provides the occupational grouping method used. |
| SAIRR Surveys of Race Relations (Horrell) 1950/51-1956/57 | secondary literature compilation | 1950-1957 | Source for numbers of Africans replaced in the civil service, women registered and endorsed out of the Peninsula. |
| Commission of Inquiry regarding the Coloured Population of the Union, UG 54/1937 | commission or official report | 1937 | Evidence that managerial ideology of Coloured unreliability is long-standing. |
New data¶
Hostels survey of 383 usable questionnaires from African hostel dwellers in Langa, Nyanga and Gugulethu (Cape Peninsula, 1981/82), plus an employer survey of twenty firms employing 22 000 workers, and interviews with unionists and employers.
Inference¶
Quantification: Descriptive tables.
Largely historical-structural narrative and argument drawing on state and employer documents, supported by descriptive cross-tabulations of the author's survey data and secondary census and commission statistics; no multivariate or causal testing, with comparisons across legal status, hostel versus house samples, and cross-checks against other surveys.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| The CLPP was first articulated as coherent policy in 1954-55 and implemented through influx control, labour bureaux, a 1966 freeze on African labour and the 1968 Physical Planning Act restrictions. (p. 10) | Verwoerd and Eiselen statements, Hansard, Bantu Labour Regulations, Physical Planning Act, SAIRR surveys. | Partial data |
| Contract African workers in the Cape Peninsula declined from a 1960 peak of 47 554 to around 18 000 and from over 60% to about 35% of the registered African workforce. (p. 16) | Table 1, compiled from unnamed sources and acknowledged to be incomplete and untrustworthy. | Partial data |
| Employers did not oppose influx control itself but resented its uneven enforcement and the CLPP's limits on African labour supply and upward mobility, while claiming Coloured workers were unsuitable for heavy manual work. (p. 18) | Cape Chamber of Industries yearbooks, commission evidence and employer interviews. | Partial data |
| The CLPP failed to advance the bulk of Coloured workers, who remain concentrated in unskilled and semi-skilled work with wages deteriorating relative to whites. (p. 46) | Theron Commission statistics (Tables 4-5), 1970 census via Hendrie and Horner, and Beinart/van der Merwe figures. | Partial data |
| A racial hierarchical division of labour exists: 87% of Africans in surveyed firms were unskilled versus 25% of Coloureds and no whites, with whites holding most managerial posts. (p. 46) | Author's employer survey of 20 firms (Table 6), cross-checked with Cilliers/Bekker and Theron data. | Partial data |
| Hostel-dwelling African workers are overwhelmingly unskilled (82.5%), poorly educated, with almost no upward occupational mobility, intra- or inter-generationally. (p. 73) | Hostels survey cross-tabulations (Tables 12-20) compared with Schneier house survey. | Partial data |
| African workers across legal categories (contract, permanent, illegal) perceive their interests as the same, contradicting divisive state strategy. (p. 84) | Open-ended question coded in hostels survey (Table 26), endorsed by union and employer interviews. | Partial data |
| Illegal African employment in the Peninsula is widespread, with estimates of 60 000 to 85 500 illegal workers. (p. 53) | Authorities' estimates, a government commission and a ministerial statement in Hansard. | Illustrative or anecdotal |
The author's caveats¶
Author states that Table 1 is derived from several sources, remains incomplete and official labour-regulation statistics cannot be trusted; census data are not comparable over time because of boundary changes, errors and subjective racial classification; the employer survey sample is small with possible statistical pitfalls; the hostel sample excludes SAR&H compounds and squatter settlements and may be biased toward workers not already settled; divergent skill classifications between management and workers; union table is tentative and incomplete.
Gaps in the evidence¶
The paper gives no systematic archival base for the historical argument, which rests on secondary works and selective quotations. The employer survey and hostel sample (non-random grid, 20 firms) are small with undescribed sampling frames and no wage data, despite wage claims. Claims of causation between the CLPP and Coloured or African occupational outcomes rest on descriptive tables without counterfactuals or time series. Official contract-worker data are incomplete. Questionnaires, coding and survey raw data are not provided. Evidence for ideology and state motives is mainly quotation.
Replicability: Data described but not public.