Agriculture and Technology - Four Case Studies¶

Literature › Works

Debbie Budlender (1984). Agriculture and Technology - Four Case Studies. Second Carnegie Inquiry into Poverty and Development in Southern Africa, Carnegie Conference Paper No. 23. SALDRU, University of Cape Town.


Type	conference paper
Scope	Economics and economic history of apartheid
Method	synthesis review
Period studied	1960–1983
Regions	Kirkwood (Eastern Cape), Western Transvaal, Stellenbosch (Western Cape), Natal sugar belt, reserves/homelands (Transkei, KwaZulu, Bophuthatswana)
Themes	agriculture land, labour markets wages, mining migrant labour, bantustans removals influx, demography urbanisation, inequality poverty unemployment
Citation key	`budlender1984`

Argument¶

Mechanisation in South African agriculture should not be reduced to tractor/labour ratios; it is a linked set of innovations (irrigation, chemicals, harvesting, handling) in which one innovation implies another. Through four case studies (Kirkwood citrus, Western Transvaal maize, Stellenbosch wine, Natal sugar) the paper shows that technology choice is driven by labour availability, control and farm size and capital rather than by relative factor prices alone. The result is labour-saving change, a shift to female and child casual labour, and displacement whose costs fall on the reserves rather than being visible as retrenchments on white farms. The author argues for ground-level studies rather than aggregate statistics.

Evidence and method¶

Secondary case material: De Klerk's 1983 UCT MA thesis on the Western Transvaal (1968-82); Nelson's 1983 BSc thesis based on interviews with 108 Natal cane farmers (1972-82); Scharf's information on seven Stellenbosch wine farms; a single Kirkwood farm study; Young's SALDRU working paper on two Anglo American farms; African Studies Institute oral history interviews; Research Station and industry personal communications.

Place in the debates¶

Causes of farm mechanization (labour scarcity/cost vs relative factor prices vs other drivers) (see Farm labour: mechanisation, labour shedding, wages and labour control). Labour availability and control are the main determinants. Sugar farmers said labour availability mattered more than relative factor prices. Mine wage increases in the 1970s are linked to the Western Transvaal and Natal shifts, with farm size and capital access also important.
Visibility of labour displacement (retrenchment vs attrition, farm consolidation and shifts to female labour) (see Farm labour: mechanisation, labour shedding, wages and labour control). Mechanisation displaces workers indirectly, through farm consolidation, attrition, loss of casual reserve labour and substitution by women. The cost is therefore hidden and felt in the reserves.
Economic rationality of advanced mechanisation (invention vs innovation) (see Farm labour: mechanisation, labour shedding, wages and labour control). Notes Young's questioning of the economic advisability of Anglo's mechanisation. Argues the gap between invention and innovation reflects capital access and scale rather than technical unavailability.
Whether technology benefits or harms workers (Luddite vs optimistic positions) (see Farm labour: mechanisation, labour shedding, wages and labour control). Rejects a Luddite position. Technology can ease arduous work, but under current economic and political conditions it is chosen for farmer benefit, with harmful effects on health, skills, ecology and employment.

Findings¶

Mechanisation must be understood as a set of interlinked innovations, not tractorisation alone. (p. 4)

we cannot view mechanisation as tractorisation alone, and secondly, the decision to introduce new technology in one operation depends on decisions with regard to other operations.
Early tractorisation converted independent tenants into wage labourers. (p. 3)

Both these trends encouraged the (forced) conversion of the more independent forms of rural existence to wage labour.
In Kirkwood, microjet irrigation cut irrigation labour from 21 workers to one, though the trigger was water scarcity rather than labour. (p. 7)

Instead of 21 workers being tied up for two weeks at a stretch, the total irrigation staff was reduced to one person
Western Transvaal combine adoption in the 1970s was tied to farm size and recruiting difficulty linked to rising mine wages. (p. 9)

De Klerk attributes the decisi,on to mechanise to (1) the growing size of the farms and (2) the increasing difficulty in recruiting labour.
Combines cut labour demand at the peak periods, so farms could rely on on-farm labour. (p. 9)

they cut the labour req~irements precisely at'its peak periods, i.e. the times when casual labour previously had to be brought in.
Weedicide adoption in maize was near-universal by 1982 but hand-hoeing persisted as a way of keeping on-farm women busy. (p. 10)

De Klerk found that 95% of his sample used weedicides in 1982.
Wine farms mechanised irrigation and spraying but harvest peak demand stayed high, so they relied on Herschel contract migrants plus women and children. (p. 14)

These men constitute about 70% of the labour force.
In Natal sugar, cutting remained manual and farmers said labour problems must be very severe before mechanising pays. (p. 18)

Nelson found that 99% of the farmers interviewed cut their cane by hand.
Sugar farmers' technique choices tracked the labour market: labour-saving in the tight mid-1970s, slowed or reversed in the recession. (p. 21)

in the earlier period there was a definite move towards labour-saving methods, while in the latter period this move was slowed down or even reversed.
Displacement is concealed: agricultural employment fell while farmers denied retrenching, and effects are felt in the reserves. (p. 24)

There has also been an absolute decrease in the number of agricultural workers in south Africa since 1970.
Mechanisation encourages cultivation of marginal land and over-cultivation of maize. (p. 23)

mechanisation presents a problem in that it tends to encourage the ever increasing use of marginal land.

Evidence ledger¶

Sources used¶

Source	Kind	Period	How used
African Studies Institute Oral History project interviews (former labour tenants, incl. Schweizer-Reineke district)	interviews oral history	early-to-mid 20th century recollections (c. 1920s onwards)	Testimony of former labour tenants that tractors and threshing machines displaced tenants and forced sale of cattle.
Author's case study of a Kirkwood citrus farm (farmer interview/farm study)	author survey or fieldwork	c. 1969-early 1980s	Describes the shift from flood to microjet irrigation, labour reduction and growing reliance on female labour on one farm.
Aires, A. 'The Changing Patterns of Labour Relationships in the Sundays River Valley of the Eastern Cape', SALDRU Farm Labour Conference, 1976	secondary literature compilation	1976	Quoted for the view that labour is harder to handle than chemicals (control motive for mechanisation).
Population census (last census figure for African residents on the Kirkwood farm)	population census	c. 1980/1970s (unspecified 'last census')	Cited to show a large African resident population (500) on the studied farm.
De Klerk, M. 'Technological Change and Employment in South African Agriculture', MA Economics, UCT, 1983 (unpublished)	secondary literature compilation	1968-1982	Main data source on mechanisation of delivery, harvesting and weeding in maize, including sample finding of 95% weedicide use and farm-size correlation.
Young, G. 'Labour in Transvaal Agri-Business: A Study of Two Farms', SALDRU Working Paper No 16	secondary literature compilation	earlier than 1984	Comparison case showing Anglo maize farm mechanised earlier than individual farmers, distinguishing invention from innovation.
Scharf (Wilfred) researcher information and findings on seven Stellenbosch wine farms	author survey or fieldwork	1940s-1983	Basis of third case study on irrigation, chemical weeding, trellis vines and labour composition on seven wine farms.
Nelson, J. 'A Survey of Technical Change and Employment in the South African Cane Industry', BSc Agric. Economics, Univ. of Natal, 1983 (unpublished), interviews with 108 farmers	author survey or fieldwork	1972-1982	Primary data for the sugar case study on mechanisation by process and the labour-market explanation of changing technique.
South African Sugar Association Research Station, Mount Edgcombe, officers and other industry sources (personal communication)	interviews oral history	early 1980s	Corroborates Nelson's mechanisation percentages and supplies technical and irrigation information.
De Beer, A.G. 'Semi-mechanized Sugar Harvesting Systems for a Developing Country', Agricultural Engineering, 1980(?)	secondary literature compilation	c. 1980	Cited for the view that fully mechanised cane harvesting is not viable until manual harvesting becomes less attractive.
Farm wage index (industry sources)	official statistical series	1977/8-1980/1	Shows fall in the wage index to support the claim that labour became more plentiful in the 1980s.
Science, 'Pesticide Office Demands New Safety Studies', Vol 221, p. 442; official pesticide poisoning figures	newspapers periodicals	1983	Supports claims about laxity of pesticide testing and under-recording of poisonings.
Macro statistics on agricultural employment, farm size and population change	official statistical series	since 1970	Asserted to show absolute fall in agricultural workers and displacement into reserves.

New data¶

Limited author-collected qualitative material: a single Kirkwood citrus farm study and interviews with farmers and informants; other case studies draw on others' unpublished theses (De Klerk, Nelson) and Scharf's work on seven wine farms.

Inference¶

Quantification: Illustrative figures.

Comparative descriptive case studies of four farming sectors (citrus, maize, wine, sugar), drawing on secondary theses and interviews; mechanisation drivers are inferred by comparing processes and periods, with explicit rejection of aggregate statistics.

Key claims¶

Claim	Basis	Strength
Tractorisation in the first half of the century converted labour tenants and squatters into wage labour by extending cultivation and forcing stock reduction. (p. 2)	Oral history interviews with former labour tenants (African Studies Institute).	Illustrative or anecdotal
In Kirkwood citrus, microjet irrigation drastically cut labour needs (21 irrigation workers to one) and expanded potential use of female labour. (p. 7)	Single-farm case study by the author via farmer interview.	Illustrative or anecdotal
Mechanisation of maize delivery, harvesting and weeding in the Western Transvaal was driven by farm size and labour recruitment difficulties linked to rising mine wages; larger farms mechanised earlier. (p. 9)	De Klerk's 1983 MA thesis sample of Western Transvaal farms, relayed second-hand.	Partial data
On Stellenbosch wine farms, mechanisation in irrigation, weeding and trellising cut labour but harvest peak demand persisted, leading to greater use of Herschel contract migrants, women and children. (p. 14)	Scharf's research on seven wine farms, relayed by the author.	Illustrative or anecdotal
In Natal sugar, mechanisation is limited by terrain and cheap labour, and trends towards labour-saving methods slowed or reversed in the second sub-period as labour became plentiful during recession. (p. 21)	Nelson's 108-farmer interview survey 1972-82, SASA officers, and a wage index.	Partial data
Mechanisation has displaced workers, mainly via farm consolidation, attrition and shifting to female casual labour, with effects felt in the reserves rather than being admitted as retrenchments. (p. 24)	Farmer interviews, De Klerk's informants and unspecified macro statistics.	Asserted
The shift towards female labour is probably accompanied by lower wages. (p. 25)	Inference only; author states casual wages cannot be estimated.	Asserted

The author's caveats¶

The author notes that Nelson's study is too small to disaggregate by area and focuses almost exclusively on employment; that the reasons for individual farmers' decisions at Kirkwood cannot yet be answered; and that casual farm wages are impossible to estimate accurately.

Gaps in the evidence¶

The argument rests on a handful of case studies, mostly relayed from unpublished theses, with no systematic national data on tractor numbers, farm employment by sex or wages by process. Agricultural census series (1970s-80s) and Department of Agricultural Economics data on machinery, employment and farm consolidation existed and could test the claims about displacement, concentration and feminisation of labour. The claim of displacement into the reserves needs matched farm-level or census data on removals and resident populations, and the causal ordering (mechanisation versus labour supply) needs time series on wages and labour supply.

Replicability: Data poorly documented.