Apartheid and Economic Growth: Craft Unions, Capital and the State in the South African Building Industry, 1945-1975¶
Owen Crankshaw (1990). Apartheid and Economic Growth: Craft Unions, Capital and the State in the South African Building Industry, 1945-1975. Journal of Southern African Studies, Vol. 16, No. 3. Taylor & Francis.
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | archival narrative |
| Period studied | 1945–1975 |
| Regions | South Africa, Transvaal, Orange Free State, Cape Province, Natal, Witwatersrand |
| Themes | labour markets wages, unions strikes, business and state, manufacturing industrial policy, politics ideology elections, origins segregation |
| Citation key | crankshaw1990 |
Argument¶
Crankshaw argues that the building industry shows how apartheid and capitalist growth interacted. Its white and 'Coloured' craft unions began the 1940s strong enough to control training and the supply of skilled labour themselves. From the 1950s the state, through the Native Building Workers Act and the racial restructuring of the unions, together with a skill shortage and an influx of non-apprenticed labour, eroded that control. The 1960s boom accelerated fragmentation of the skilled trades into semi-skilled work done by Africans. The unions turned to the state for job reservation just as the state, under pressure from capital and the need to house the urban African working class, was quietly abandoning it. Employers at first had ambiguous views on job reservation but opposed it as the labour shortage deepened. The argument rests on Industrial Tribunal reports, trade union and TUCSA archives, the trade press and censuses.
Evidence and method¶
Industrial Tribunal reports on work reservation (1960, 1961, 1963); Industrial Council Agreements and Government Gazette notices; TUCSA and Hartwell archive collections (Wits); union journals (The Woodworker, ASW Monthly Report); The South African Builder; newspapers (Rand Daily Mail, The Star); industrial and construction censuses; Manpower Surveys; SAIRR Surveys; Department of Labour reports; Scheiner's 1976 MA thesis; secondary literature (Lewis, Horrell, Alexander and Simons, O'Meara).
Place in the debates¶
- Liberal vs revisionist views on apartheid and capitalist growth (whether racial labour policy served or hindered capital) (see Liberal vs revisionist: was apartheid functional or dysfunctional for capitalist growth?). Sides with a revisionist, class-based reading. Labour apartheid, craft unionism and growth coincided and conflicted. Capital increasingly opposed job reservation, and the state steadily dismantled it in practice while still protecting white privilege rhetorically.
- Whether state labour legislation (e.g. Native Building Workers Act) simply advanced white worker privilege (see The job colour bar, white labour and its erosion). Disputes this reading. The Act, though it looks racist and protective, bypassed craft union control of training and wages and served capital and state housing needs.
- Cheap labour thesis / role of white craft workers in the racial division of labour (see The job colour bar, white labour and its erosion). Shows white craft unions were undercut by cheaper African and 'Coloured' labour and became dependent on the state. The racial division of labour was contested, not simply imposed by white labour or by capital.
Findings¶
-
The building industry was capital-poor in the late 1940s, which helped sustain craft unionism. (p. 3)
In 1946, for example, capital intensity in the building industry was only one quarter that of the manufacturing sector in general.
-
1950s mechanisation in building replaced unskilled rather than skilled tasks, so it did not directly threaten the artisans. (p. 4)
So, although capital-intensity in the building industry doubled during the 1950s, the pattern of this mechanisation was quite unlike that of most other industries.
-
The Native Building Workers Act was seen by the unions as an attack on artisans because it created a training channel outside union control. (p. 7)
However, as the trade unions of the time pointed out, this legislation was a deadly blow for building artisans
-
The Act trained a significant number of African building workers by 1960. (p. 8)
By 1960, almost 3000 African building workers had received training in the skilled trades, almost 15% of the total number of white artisans employed in the industry.
-
Exemptions under the Act allowed severe wage undercutting of white artisans. (p. 8)
allowed building contractors and owner-builders to employ African labour on skilled work at one quarter of the cost of white artisans.
-
Incentive bonus schemes had limited reach but acted as a wedge against craft control. (p. 12)
ensured that by the late 1950s only 11% of building firms were applying any form of incentive payment.
-
The 1960s boom marked a sharp acceleration in the growth of building output. (p. 17)
During the 1950s the growth rate for the average year was only 2.5%. After the Sharpeville recession, however, the annual rate of growth leapt to an average of about 10%
-
Employment in building expanded dramatically by the mid-1970s. (p. 17)
by the mid-1970s, employment had grown to a staggering 47,600; an increase of almost 300%.
-
Employers won legal fragmentation of the skilled trades in the Transvaal in 1966. (p. 18)
the building contractors finally gained their first major victory in 1966 when they won the right to legitimately employ cheaper African labour on semi-skilled work in the Transvaal.
-
The ratio of African to non-African workers rose gradually, then jumped after the late-1960s fragmentation agreements. (p. 20)
In the period between 1945 and 1961, the ratio of African workers to non-Africans increased from 1.7 to 2.1.
-
Illegal employment of Africans on skilled work was widespread by 1973. (p. 20)
They revealed that more than 60% of building sites were illegally employing African labour on skilled work
-
The state refused blanket exemptions for Africans in the major centres and left employers and unions to negotiate, then ratified the outcome. (p. 24)
In the major centres, the state refused to grant blanket exemptions to job reservation, and instead insisted that the building contractors fight it out with the unions themselves.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Census of Industrial Establishments, 1942-1943, U.G. 20/46, Table 1b | industrial mining census | 1942-1943 | Establishes that capital intensity in building in 1946 was only a quarter of that in manufacturing generally. |
| Censuses of Industrial Establishments 1944-45 (U.G. 37/48), 1945-46 (U.G. 50/49), 1946-47 (U.G. 63/50), 1947-48 to 1949-50 (U.G. 30/54), Table 12 | industrial mining census | 1944-1950 | Shows that almost 60 percent of building contractors employed fewer than 20 workers in the late 1940s. |
| Censuses of Industrial Establishments 1944-45 to 1949-50; Summary of Reports Con.1 and 2, Industrial Censuses 1950-51 to 1960-61 (Report No. Con.3); Censuses of Construction 1965-66, 1967-68, 1969-70, 1972 (Report Nos. 05-41-01 to 05-41-05) | industrial mining census | 1944-1972 | Supplies the series on African and non-African employment used to compute the racial ratio of labour and measure fragmentation. |
| Report of the Department of Labour for the Year Ended 31st December 1960, RP6/1961 | commission or official report | 1960 | Gives the number of African building workers trained under the Native Building Workers Act (almost 3000 by 1960). |
| Report of the Department of Labour for the Year Ended 31st December 1951, UG 45/1953 | commission or official report | 1951 | Cited for the decline in registered apprenticeships after the Second World War. |
| Industrial Tribunal Report on Reservation of Work in the Building Industry, Orange Free State and Transvaal, House of Assembly Annexure No.211, 1960 | commission or official report | 1958-1960 | Source for the racial composition of artisans, the African/non-African labour ratio, regional union positions and undercutting in the Transvaal and OFS. |
| Industrial Tribunal Report on Reservation of Work in the Building Industry in the Cape Province and Natal, House of Assembly Annexure No.136, 1963 | commission or official report | c.1960-1963 | Evidence on closed-shop enforcement, Coloured and Indian artisans, union divisions and the mild job reservation determination in the Cape and Natal. |
| Industrial Tribunal Report on the Labour Position in the Western Cape Province, House of Assembly Annexure No.365, 1961 | commission or official report | 1961 | Supports the claim that the post-war boom opened less exacting tertiary jobs for whites, worsening skill shortages. |
| Social and Economic Planning Council Report No.10 (U.G. 37-46) | commission or official report | 1946 | Cited for severe cyclical fluctuations in the building industry as a cause of low investment. |
| Native Building Workers Act (No. 27 of 1951) and Proclamations and Regulations thereunder | legislation gazette | 1951-1955 | Establishes the legal content of the Act, including exemptions that let contractors employ Africans on skilled work. |
| Government Gazette (Extraordinary) notices on Industrial Council Agreements and job reservation determinations, 1953-1975 | legislation gazette | 1953-1975 | Documents the successive fragmentation agreements (1953, 1959, 1966-69, 1975) and ratifying exemptions that redefined skilled tasks. |
| Manpower Survey No.8, Volume 2 (1969) | official statistical series | 1969 | Gives the semi-skilled share of the workforce (12%) in 1969, contrasted with 1% in 1958. |
| South African Statistics 1982 | official statistical series | 1982 | Cited in a footnote to the discussion of the racially exclusive Cape craft unions. |
| Debates of the House of Assembly (Hansard), Volume 103, 1960 | parliamentary debates | January-May 1960 | Cited for craft unions calling on the state for protection by the end of the 1950s. |
| TUCSA records, Church of the Province of South Africa Archives, University of the Witwatersrand (AH 1426), incl. Conference proceedings, correspondence with AUBTW and ASW, Government Legislation (Da 38) | union or movement records | 1947-1969 | Evidence of union complaints about undercutting, internal splits over job reservation, and relations with TUCSA. |
| Hartwell Labour Collection (AD 1456), Building Trade Unions (1.1) incl. ASW Memorandum to Industrial Tribunal, 18 July 1958; Miscellaneous Correspondence (3.9) | union or movement records | 1958-1961 | Source of union submissions to the Industrial Tribunal and evidence on exemptions and union positions on job reservation. |
| Amalgamated Society of Woodworkers Monthly Reports (nos. 422, 424, 426, 447, 449) and The Woodworker (1959-1972) | union or movement records | 1956-1972 | Evidence on rank-and-file resistance to incentive schemes, undercutting by Africans, regional splits and defection to the WBWU. |
| The Building Worker (WBWU/AUBTW periodical), 1961 | union or movement records | 1961 | Reports the replacement of a Pietermaritzburg union delegate and the expulsions of members over job reservation. |
| The South African Builder (trade journal), 1950-1972 | newspapers periodicals | 1950-1972 | Source for the Department of Labour 1951 survey, employment growth, employer attitudes, incentive schemes and exemptions. |
| Newspapers: Rand Daily Mail, The Star, Natal Mercury, Financial Mail, Evening Post, Pretoria News, Sunday Times, Cape Times, The Friend | newspapers periodicals | 1955-1974 | Evidence on African painters, illegal employment on skilled work, builders' estimates and union statements. |
| Survey of Race Relations in South Africa, 1967, 1971, 1974, 1975, 1979 | secondary literature compilation | 1967-1979 | Documents the sequence of exemptions from job reservation, the 1973 inspections of Witwatersrand sites, and abandonment of reservation in 1979. |
| Scheiner, P., Unpublished MA thesis, University of Cape Town, 1976, 'The Occupational Mobility of Black Workers in the Witwatersrand Building Industry' | secondary literature compilation | 1960-1976 | Relied on for evidence on illegal employment, employer federation positions and the spade/flat-iron anecdote, and for cited BIFSA memoranda. |
| Building Industries Federation of South Africa (BIFSA) Memorandum to Industrial Tribunal, 1973 (cited in Scheiner) | employer association records | 1973 | Shows employers pressing the state to insist the unions negotiate fragmentation themselves. |
| Lewis, J., Industrialisation and Trade Union Organisation in South Africa, 1924-55 and 'The South African Labour Movement and the Building Worker's Strike of 1947' | secondary literature compilation | 1924-1955 | Used for rank-and-file union organisation, the 1947 strike and the 11% of firms using incentive payments. |
Inference¶
Quantification: Illustrative figures.
Historical narrative case study of one industry (building), triangulating official industrial censuses, tribunal reports, union archives, trade journals and press across regions (Cape/Natal vs Transvaal/OFS) and time (1940s-1970s). Claims are built from documentary evidence and a few headline statistics. There is no formal statistical analysis.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| The Native Building Workers Act of 1951 undermined craft union control by creating a training and supply channel for African semi-skilled labour outside apprenticeship, with widespread exemptions enabling undercutting. (p. 508) | Act text, Department of Labour 1960 report, TUCSA and Hartwell union memoranda, and press reports. | Partial data |
| The state weakened mixed craft unions through the Suppression of Communism Act and the Industrial Conciliation Act, fostering white splinter unions such as the WBWU. (p. 511) | Legislation, Horrel and Alexander & Simons secondary accounts, and union records. | Partial data |
| Fragmentation of the skilled trades accelerated after the 1960s boom, shown by the rising African-to-non-African ratio (1.7 in 1945 to 2.8 by the early 1970s) and the semi-skilled share rising from 1% to 12%. (p. 521) | Industrial and Construction censuses, Industrial Tribunal, Manpower Survey No.8, and Industrial Council agreements in the Gazette. | Systematic data |
| Regional divisions in craft union support for job reservation reflect the different racial composition of artisans (Cape/Natal mixed vs Transvaal/OFS white). (p. 514) | Industrial Tribunal reports, ASW and AUBTW submissions, TUCSA correspondence, union reports. | Partial data |
| By the early 1970s illegal employment of Africans on skilled work was widespread, with Industrial Council inspectors finding over 60% of Witwatersrand sites in violation. (p. 521) | 1973 inspections reported in the SAIRR Survey 1975, Scheiner, newspaper reports and a builder's estimate. | Partial data |
| Capital's opposition to job reservation grew with the 1960s boom and skill shortage, while the state granted progressive exemptions and in 1979 abandoned job reservation in building. (p. 524) | Industrial Tribunal reports, SAIRR Surveys, employer federation statements in The South African Builder, and Government Gazette determinations. | Partial data |
| The building industry's low capital intensity and small firm structure sustained craft unionism into the late 1940s. (p. 506) | 1942-43 and 1944-50 Censuses of Industrial Establishments, plus secondary literature and trade journal biographies. | Partial data |
The author's caveats¶
The author notes that it is difficult to gauge with any certainty the extent to which incentive schemes undermined craft exclusion. Much evidence on undercutting rests on union complaints and press reports, and the 1950s extent of undercutting is indirectly inferred.
Gaps in the evidence¶
The argument relies heavily on partisan reports (union memoranda, employer journals, press) and a few aggregate census ratios, with no systematic occupational series by race and trade, wage data or firm-level records. Employer federation archives, Department of Labour files and Industrial Council agreement records would better test the claims about capital and state motives, and these existed and are archived. Direct wage and occupational microdata for artisans would permit quantification of undercutting.
Replicability: Data described but not public.