Motives for mechanisation in South African agriculture, c1940–1980¶
Stefan Schirmer (2004). Motives for mechanisation in South African agriculture, c1940–1980. African Studies, 63(1). Routledge (Taylor & Francis).
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | mixed |
| Period studied | 1940–1980 |
| Regions | Transvaal (incl. Western Transvaal maize triangle), Orange Free State, Natal, Cape (Western and Eastern), South Africa (national) |
| Themes | agriculture land, labour markets wages, business and state, bantustans removals influx, mining migrant labour, politics ideology elections |
| Citation key | schirmer2004 |
Argument¶
Schirmer argues that many white farmers mechanised not mainly for cost savings but to gain control over, and reduce dependence on, African farm workers they saw as unreliable and insubordinate. He follows Marglin's claim that control can drive mechanisation where competition is weak. First, state marketing boards, soft Land Bank credit and cooperatives shielded farmers from competitive pressure, which left room for non-economic motives. Second, he shows that paternalist labour relations, the labour tenant system and compound recruiting generated endemic conflict and constant farmer demands for coercive control. Third, he draws on farmer statements, Farmers Weekly, commissions and De Klerk's survey to show that farmers themselves cited labour problems as the reason to mechanise. He concludes that this probably distorted resource allocation and held back productivity gains.
Evidence and method¶
Largely qualitative: Farmers Weekly letters and articles, Native Affairs Department and South African Agricultural Union (SAAU) minutes (Central Archives Depot), government commission reports (1941, 1960-61 labour tenant inquiry, 1957-60 rural depopulation commission), and farmer testimony. Also an interview and a few quantitative series (tractor and combine counts from Reports on Agricultural and Pastoral Production, credit figures, De Klerk's 1980s questionnaire on maize harvesting), plus secondary literature.
Place in the debates¶
- Control vs cost-efficiency as motive for mechanisation (Marglin vs Landes, applied to farms) (see Farm labour: mechanisation, labour shedding, wages and labour control). Sides with Marglin: labour control was a significant, sometimes independent motive, made possible by state protection from competition. He does not claim that cost was irrelevant.
- Transition from paternalist to capitalist farm labour relations (e.g. Van Onselen) (see Farm labour: mechanisation, labour shedding, wages and labour control). Stresses continuity: paternalism and the conflict it produced persisted throughout 1940-1980, rather than giving way to wage relations.
- Causes of farm mechanisation (labour shortage/wage push vs state subsidy and credit) (see Farm labour: mechanisation, labour shedding, wages and labour control). Combines both: cheap state credit and marketing-board prices made machines affordable, while farmers' dissatisfaction with labour supplied the motive. He argues that the 1970s shortage was minor and that farmers over-reacted.
- Productivity effects of mechanisation under state protection (see Farm labour: mechanisation, labour shedding, wages and labour control). Argues that improved productivity was often an unintended consequence and that control-driven mechanisation reduced productivity gains, citing World Bank and Thirtle et al. on modest growth. He does not test this himself.
Findings¶
-
The paper's thesis: farmers across the country and over a long period mechanised because of dissatisfaction with labour relations. (p. 2)
many farmers, across the country and for a long period of time, were motivated to adopt more and better mechanical implements
-
State credit to cooperatives expanded very rapidly after the National Party took power, financing machinery purchases. (p. 5)
Loans to cooperatives increased from 7.1 million in 1945 to 27.8 million in 1950 before reaching 65.1 million in 1955
-
Farmers explicitly saw tractors as a tool against black labour rather than as a means of replacing white labour. (p. 12)
Let the Government see to it that we get tractors … and you need not worry about the white man, rather about the black.
-
Tractor adoption accelerated sharply after the Second World War. (p. 12)
In 1937 white farmers who had controlled over 543 000 ploughs had only used 6019 tractors. In 1946 ownership of tractors had jumped to just over 20 000 and by 1950 to almost 48 500
-
Tractor growth far outpaced growth in cultivated area, so tractors were used mainly to save labour rather than to expand output. (p. 13)
between 1945 and 1960 the area under cultivation increased by 4.95 per cent, while the number of tractors increased by 82.97 per cent.
-
The apartheid state built a labour bureau system that channelled Africans to farms and eased farm labour shortages. (p. 14)
The state’s new bureaucratic system spread rapidly and by 1957 there were 234 local and 278 district labour bureaux.
-
Mechanised stock grew markedly in the Transvaal and Orange Free State between 1960 and 1979, while Natal lagged. (p. 18)
The number of tractors and combines in the Orange Free State expanded from 31 087 in 1960 to 50 894 in 1979. In the Transvaal the expansion was from 43 538 to 69 271.
-
Mechanisation shifted farm labour toward seasonal, female and child workers who were easier to control; men's share in harvesting teams collapsed. (p. 18)
In 1968 in the Western Transvaal it was found that all but a few of the harvesting teams included men. Thirteen years later only 40 per cent of the teams had adult males in their ranks.
-
Combine harvesting in the maize triangle was adopted in the mid-1970s, when mine wages rose rapidly. (p. 19)
The switch from hand harvesting to machine harvesting occurred in the section of the maize triangle surveyed by De Klerk mostly during the period 1973–1977.
-
Rural unemployment rose steeply in the mid-1970s, yet mechanisation continued, which Schirmer takes to show farmer over-reaction rather than a genuine labour shortage. (p. 20)
it has been estimated that unemployment amongst rural black work-seekers rose from 25 000 in 1970 to 315 000 in 1975.
-
Mechanisation continued despite cheapening labour, suggesting the motive was not purely cost-driven. (p. 20)
The process of mechanisation in no way responded to this cheapening of labour and continued unabated.
-
State protection from competition let farmers act on perceptions of black unreliability. (p. 21)
Farmers did not face the competitive pressures that would have forced them to make optimal use of their inputs
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Report on Agricultural and Pastoral Production, 1945/46, 1960, 1976, 1979 (tractor, combine, reaper/binder and cultivated-area figures; Appendices A and B) | official statistical series | 1945-1979 | Provides tractor numbers and cultivated area used to show rapid mechanisation after 1945 and the compared growth of tractors and cultivated area. |
| Report on Agricultural and Pastoral Production, 1945-46 (reapers and binders) | official statistical series | 1945 | Shows Cape farmers owned 62 per cent of all reapers and binders by 1945, evidencing early mechanisation in Cape wheat. |
| Houghton, The South African Economy (tractor and plough statistics 1937-1960) | secondary literature compilation | 1937-1960 | Supplies national counts of ploughs and tractors (6019 in 1937, ~20,000 in 1946, ~48,500 in 1950, 119 thousand in 1960). |
| Inter-Departmental Committee of Enquiry in connection with the Labour Tenant System (1960 hearings; report 1961) | commission or official report | 1960-1961 | Testimony from farmers' associations and individual farmers shows continued dissatisfaction with labour relations and support for mechanisation as a solution in 1960. |
| Third Interim Report of the Industrial and Agricultural Requirements Commission (1941) | commission or official report | 1941 | Quoted to show wartime official advocacy of mechanisation to raise per capita output and reduce numbers on the land. |
| Social and Economic Planning Council (1944) | commission or official report | 1944 | Cited with the 1941 commission as keeping alive the idea that mechanisation could radically reduce labour requirements. |
| Report of the Commission of Enquiry into European Occupation of Rural Areas (1960) | commission or official report | 1957-1960 | Documents the 1957 enquiry into rural depopulation and the shift towards opposing land subdivision. |
| Du Plessis Commission 1963 and Subdivision of Agricultural Land Act 1970 | legislation gazette | 1963-1970 | Evidence of a policy shift against small farms by 1970. |
| Report into Agricultural Credit (1961) and Wickens 1983 (Land Bank loans to cooperatives) | commission or official report | 1945-1955 | Provides Land Bank loan totals to cooperatives showing the scale of state-subsidised credit. |
| Commission of Enquiry into Matters Referring to the Coloured Population Group (1976) | commission or official report | 1976 | Source for Western Cape wage and paternalist labour conditions c1940. |
| Union of South Africa Assembly Debates, 12 May 1960 | parliamentary debates | 1960 | Cited for the growing state bias against small farms and subdivision. |
| Union of South Africa Department of Agriculture reports (1913, 1921, 1915, 1934, 1939) | commission or official report | 1910-1939 | Cited for the state's protection of farmers from competition and for farmers' complaints to commissions in the 1930s about labour tenancy. |
| Central Archives Depot (CAD), Native Affairs Department (NTS) files, e.g. NTS 3/371 vol. 9257; 222/280 vols 2095, 2229; 10/280 vol. 2010; 463/280 vol. 2229 | state archives | 1938-1948 | Farmer complaints about labour, SAAU/TAU congress resolutions, the agricultural officer's memo on machinery, and minutes of NAD-SAAU meetings on labour bureaux. |
| CAD, LDB R3633, Memoranda Issued by the Department of Agriculture, c1930; CAD MLD 14/20 vol. 19 | state archives | c1930; 1955 | Source for the Department's statement that African labour supply depends on land and grazing, and for 1950s protection of weak farmers. |
| Farmers Weekly (1930-1952 articles and letters) | newspapers periodicals | 1930-1952 | Main source of farmers' own statements linking tractors and machinery to labour troubles, e.g. Luther Uys, the Rustenburg farmer, Saaiman, Dell. |
| De Klerk 1984a thesis: questionnaire survey of maize farmers, Western Transvaal 1968-1981 | secondary literature compilation | 1968-1981 | Questionnaire results on reasons for switching to combine harvesters are used to support that labour unavailability and unreliability drove mechanisation. |
| Interview with Bryna Davis, Lydenburg, 1990 | interviews oral history | 1990 | Anecdotal evidence of cooperatives handing out new tractors on easy terms after 1948. |
| Budlender 1984, Agriculture and Technology, Four Case Studies (SALDRU) | secondary literature compilation | 1970s-1984 | Case studies showing farmers use labour-replacing technology to increase power over workers and sugar research as insurance. |
| Marcus 1989, Modernising Super-Exploitation (and Antrobus study, Albany) | secondary literature compilation | 1968-1980s | Data on women's and children's share in the seasonal and permanent workforce after mechanisation. |
| Morris 1977 / Native Affairs Department figures on labour bureaux and farm placements | secondary literature compilation | 1952-1957 | Counts of labour bureaux and Africans diverted to farm labour, used to show state labour controls. |
| Native Laws Amendment Act (1952) and Natives (Abolition and Coordination of Documents) Act (1952) | legislation gazette | 1952 | Establish the labour bureaux and passbook system that controlled the farm labour supply. |
| World Bank 1994, South African Agriculture; Thirtle et al. 1993 TFP estimates | secondary literature compilation | 1947-1991 | Support the claim that productivity growth was modest and that resource allocation was distorted. |
New data¶
Author-compiled tables (Appendices A and B) of provincial tractor, combine, sprayer and cultivated-area figures drawn from the Reports on Agricultural and Pastoral Production for 1945-1979; no new primary dataset otherwise.
Inference¶
Quantification: Illustrative figures.
Qualitative reading of farmers' own statements (Farmers Weekly, archives, commission testimony) plus a few aggregate machinery statistics and De Klerk's survey, set against the argument that state-regulated product markets insulated farmers from competitive pressure, allowing non-economic motives for mechanising; no counterfactual or cost testing.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| State regulation of marketing, credit and prices insulated white farmers from competition, allowing non-cost motives to influence investment decisions. (p. 5) | Secondary literature (Bradford, Wilson, O'Meara), Land Bank loan figures, commission reports; no farm-level cost data. | Partial data |
| Farmers chronically complained of insufficient control over workers, and these labour-relations grievances persisted through the 1940s-1960s despite apartheid controls. (p. 17) | Farmers Weekly, CAD NTS files, 1960 labour tenant committee testimony. | Illustrative or anecdotal |
| Tractor adoption accelerated sharply after 1945, mainly to reduce labour needs rather than to expand cultivation. (p. 14) | Houghton and Agricultural and Pastoral Production reports (Appendix A) showing tractors up 83% versus cultivated area up about 5% between 1945 and 1960, plus farmer quotations. | Partial data |
| Farmers explicitly cited tractors and machinery as weapons against troublesome African labour in the 1940s. (p. 12) | Farmers Weekly letters and articles 1946-1951, quoted individually. | Illustrative or anecdotal |
| Mechanisation transformed the farm labour force toward seasonal, casual female and child labour and fewer resident families. (p. 18) | Marcus 1989, Antrobus study, Budlender, and localised studies; national statistics unavailable. | Partial data |
| Mechanisation in the Orange Free State and Transvaal expanded greatly from 1960 to 1979, more than in Natal, consistent with stated intentions to mechanise. (p. 19) | Appendix B provincial counts from Reports on Agricultural and Pastoral Production 1960, 1976, 1979. | Systematic data |
| Control motives, not efficiency, were a central independent driver, producing distortions and modest productivity gains. (p. 21) | Inference from the qualitative evidence plus World Bank 1994 and Thirtle et al. 1993. | Asserted |
The author's caveats¶
The author calls the evidence largely qualitative, notes that the statistics are silent on the war years, that nationwide statistics on child and women workers are unavailable, and that a full analysis of productivity effects is beyond the paper's scope.
Gaps in the evidence¶
The argument lacks farm-level cost, wage, credit and machinery price data to test the relative weight of control versus cost motives, and no comparison of mechanisers with non-mechanisers or of the relative profitability of alternatives. Farm-level surveys, cooperative and Land Bank loan records, and agricultural census microdata (some existed then, some archival now) would allow testing whether the timing of mechanisation tracks labour costs, credit and subsidies. The quotations from Farmers Weekly and union meetings reflect vocal, selected farmers and cannot show how representative they were.
Replicability: Data poorly documented.