Seasons That Will Never Return: The Impact of Farm Mechanization on Employment, Incomes and Population Distribution in the Western Transvaal¶
Michael De Klerk (1984). Seasons That Will Never Return: The Impact of Farm Mechanization on Employment, Incomes and Population Distribution in the Western Transvaal. Journal of Southern African Studies, Vol. 11, No. 1. Taylor & Francis.
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | survey fieldwork |
| Period studied | 1968–1981 |
| Regions | Western Transvaal, Bophuthatswana, Transkei |
| Themes | agriculture land, labour markets wages, bantustans removals influx, inequality poverty unemployment, demography urbanisation |
| Citation key | klerk1984b |
Argument¶
Based on a 1982 survey of 61 maize farms, the article shows that combine harvesting, bulk grain handling and chemical weed control spread rapidly in the 1970s, especially 1973-77. This cut seasonal employment sharply and shifted the remaining seasonal jobs from workers living in black rural areas to families of resident farm workers, and from men to women and children. Farmers cite labour shortage, but De Klerk argues that the key stimulus was a 'rural labour push': combines were cheaper on nearly all farms once available, and farm enlargement and men's reluctance to take farm jobs triggered adoption. Mechanization also increased farmers' control over the labour process. Women in black rural areas bore the brunt of the job losses. State policy and relocation were only minor causes; if anything, mechanization acted as an informal part of the relocation programme.
Evidence and method¶
Survey of 61 maize farms in six Western Transvaal magisterial districts (Feb/March 1982), with farmer recall data for 1968, 1973, 1977 and 1981. A small, informal Transkei survey of seasonal workers was used to check the farmers' answers. Also uses Department of Agriculture wage estimates, population and farm censuses, Marais Commission report and Surplus Peoples Project records.
Place in the debates¶
- Causes of farm mechanization (labour shortage / urban labour pull vs technology and scale push vs state policy) (see Farm labour: mechanisation, labour shedding, wages and labour control). Accepts farmers' labour-shortage claims only partly. Concludes the 'rural labour push', with economies of scale and new foreign technology, outweighed the 'urban labour pull'. Gives state policy and relocation little weight, though real wages rising and real interest rates falling in the 1970s are treated as contributing factors.
- Farm labour control and the labour process (see Farm labour: mechanisation, labour shedding, wages and labour control). Argues mechanization was partly a way for farmers to gain control, by using smaller teams, women, children and resident families, and by undercutting the bargaining position of reserve communities.
- Relocation, influx control and rural depopulation (see Farm labour: mechanisation, labour shedding, wages and labour control). Treats farm mechanization as an informal, unintended extension of the relocation programme. Finds influx control did not always immobilize farm workers.
Findings¶
-
Combine harvesting rose from a minority to nearly all of the maize area, mostly in 1973-77. (p. 6)
between 25 and 30 per cent of the area planted with maize was being harvested by combine. By the end of the period, in 1981, this had risen to about 95 per cent.
-
Farms grew substantially in size while the area planted to maize stayed roughly constant, implying greater concentration of control. (p. 9)
the average gross surface area of farming units in the survey grew by almost 75 per cent, from 664 ha. in 1968 to 1155 ha. in 1981.
-
Seasonal and permanent harvest jobs declined sharply per unit of land. (p. 12)
the decline in the number of jobs is seen to have been almost 70 per cent for seasonal and 50 per cent for permanent workers.
-
The number of seasonal workers in the surveyed districts more than halved. (p. 12)
it appears to have been approximately 105 000 in 1968 and 40 000 - 45 000 in 1981.
-
Seasonal harvest jobs moved from external workers to families of resident farm workers, partly because of their lower wages. (p. 10)
In 1968, intemal' workers held only 9 per cent of such jobs. By 1981, this had risen to 48 per cent.
-
Real wage gains were small for seasonal workers and larger for permanent workers. (p. 12)
For seasonal workers, between 1969 and 1981, it is no more than 20 per cent. That is, from 39 cents per day in 1969, the real wage rose to only 47 cents in 1981
-
Combines were cheaper at very low harvest tonnages, so the switch was a matter of time and stimulus on all but the smallest farms. (p. 18)
the critical harvest tonnage above which it was cheaper to use combines was only 250 tonnes.
-
Mechanization is linked to the technology and push from the rural side rather than urban pull. (p. 19)
This suggests that the reduction in farm employment was due mainly to a 'rural labour push' rather than to an 'urban labour pull'.
-
State tax policy on machinery came too late to explain most mechanization. (p. 20)
the tax provision which enabled farmers to write off the entire cost of new machinery in the year of purchase, thereby reducing both their tax liability and the effective cost of new machinery. But this was introduced only in 1977
-
The black rural population of the survey districts stopped growing in the 1970s, indicating net outflow. (p. 21)
between 1970 and 1980 the black rural population fell by 4 per cent. The expected natural increase of about 30 per cent was more than offset by a substantial outflow of residents.
-
Women in black rural areas were most affected, as alternative jobs were scarce for them. (p. 22)
Thus women living in black rural areas have borne the brunt of technological change on farms.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Author's survey of 61 maize farms in six Western Transvaal magisterial districts (Coligny, Delareyville, Koster, Lichtenburg, Schweizer-Reneke, Wolmaransstad), Feb/March 1982 | author survey or fieldwork | 1968-1981 (recall data; survey Feb/March 1982; analysis on 1968, 1973, 1977, 1981) | Main source for mechanization, employment, wages, team composition, worker residence and farmers' reasons for mechanizing. |
| Follow-up survey of seasonal workers in the Transkei | interviews oral history | 1982 (workers employed in 1981 and earlier years) | Checks the 'class bias' of farmer-only data by corroborating employers' employment and wage answers with workers. |
| Population censuses 1951, 1960, 1970 and 1980 (black rural population in five Western Transvaal districts) | population census | 1951-1980 | Shows black rural population growing 1951-70 then falling 4 per cent in 1970-80, interpreted as outflow linked to mechanization. |
| Department of Agriculture wage data for seasonal maize-harvest workers, 1969-1976 | official statistical series | 1969-1976 | Provides the benchmark average cash daily wage of seasonal workers (39 to 77 cents) against which the survey's 1981 wage is compared. |
| Department of Agriculture estimates of average all-inclusive annual wage of permanent farm workers, 1970-1977 | official statistical series | 1970-1977 | Series (R220 to R653) extended with the author's 1981 estimate (R1 777) to trace permanent worker wages. |
| Department of Agriculture time studies of threshing/harvesting labour | official statistical series | 1960s | Basis for calculating the share of employment decline attributable to reorganisation of hand-harvesting. |
| M. J. de Klerk, 'The Incomes of Farm Workers and Their Families', Second Carnegie Inquiry Conference Paper No. 28, 1984 | author survey or fieldwork | 1968-1981 | Companion study that supplies wage composition, estimation method and the correlation of mechanization with family incomes. |
| M. J. de Klerk, unpublished M.A. thesis, University of Cape Town, 1983, 'Technological Change and Employment in South African Agriculture' | author survey or fieldwork | 1968-1981 | Underlying research on which the article is based. |
| Commission of Inquiry into Agriculture (Marais Commission), Second Report, RP 84/1970 | commission or official report | 1970 | Quoted to support the argument that ease of control, not only profit, motivates farmers to adopt machinery. |
| Surplus Peoples Project records on removals from white farms | secondary literature compilation | 1970s | Used to claim few black families were removed from white farms in the region, so relocation was a minor cause of mechanization. |
| Abolition of Labour Tenancy notices (state proclamations) | legislation gazette | 1960s-1970s | Notes only one of six survey districts was covered, indicating labour tenancy was already absent. |
| National and regional wage and employment statistics for mining, manufacturing and construction | official statistical series | 1970s | Used to infer that urban industrial jobs became more attractive and absorbed displaced male farm workers. |
| Recruitment figures for Bophuthatswana workers in mining | chamber of mines or recruiting records | 1970-1976 | Shows rapid rise in mine recruitment from the main seasonal labour source region. |
| BENBO, 'Bophuthatswana Economic Report' and RSA Chief Director of Surveys and Mapping 1:250 000 sheets | maps spatial data | 1980 | Base for Figure 1 map of farming units surveyed. |
| L. H. Fisher, The Harvest Labor Market in California (1953) | secondary literature compilation | 1953 | Analogy for interpreting farmers' complaints of labour shortage. |
New data¶
Author-conducted 1982 survey of 61 Western Transvaal maize farms in six magisterial districts, with retrospective data for 1968-81 (analysed for 1968, 1973, 1977, 1981) on mechanization, employment, wages, team composition; plus a small Transkei survey of seasonal workers (four farms' workers).
Inference¶
Quantification: Descriptive tables.
Descriptive analysis of a non-random retrospective farm survey with before/after comparison over four years, deflated wage series, cost threshold calculation for combines, one cross-sectional correlation in 1981, and inference from census trends and national wage/employment statistics; causal reasoning is by argument and elimination of alternatives (no regression with controls).
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Combine harvesting rose from 25-30% to about 95% of maize area between 1968 and 1981, bulk delivery was 54% in 1968 and near universal by 1977, and chemical weed control rose from 15% to 95%, mostly in 1973-77. (p. 6) | Author's 61-farm survey with recalled data for four years. | Partial data |
| Seasonal harvest and delivery jobs per 1000 ha fell almost 70 per cent and permanent jobs 50 per cent, with seasonal weeding jobs falling about 60 per cent. (p. 12) | Survey employment per farm adjusted for farm size growth and near-constant maize area. | Partial data |
| Seasonal jobs shifted from external workers (Bophuthatswana, Transkei) to internal workers on white farms, and from men to women and children. (p. 13-14) | Survey data on residence and composition of seasonal teams, 1968-81. | Partial data |
| Real wage gains were modest: seasonal real wages rose no more than 20 per cent 1969-81, while permanent workers' real wages rose about two-and-a-half-fold. (p. 12) | Department of Agriculture wage series combined with the author's 1981 survey estimate, deflated by cost-of-living. | Partial data |
| Mechanization was driven chiefly by a 'rural labour push' (economies of scale, farm enlargement, foreign technology) rather than an 'urban labour pull' or state policy. (p. 18-19) | Farmers' stated reasons, a cost calculation showing a 250-tonne break-even for combines, timing of the 1977 tax write-off, and comparison of farm and industrial wages. | Partial data |
| The black rural population of the five districts rose 32% (1951-60) and 52% (1960-70) but fell 4% in 1970-80, suggesting outflow linked to farm mechanization, though the causal share is unclear. (p. 21) | Population census figures for five districts. | Partial data |
| Mechanization correlated negatively with family incomes from on-farm sources in 1981 via fewer permanent workers per family, implying farm families became worse off. (p. 16) | 1981 cross-sectional correlation in the survey, significant at 5% level. | Partial data |
| Displaced men were probably absorbed in urban industry while women, who bore most seasonal job losses, found it difficult to get other work. (p. 22) | National/regional industrial wage and employment statistics; no direct tracing of displaced workers. | Illustrative or anecdotal |
The author's caveats¶
Sample not statistically random, with likely over-representation of large farms (biasing toward overstating mechanization and understating employment); all data came from farmers (class bias), checked only against four farms' workers in the Transkei; insufficient data for an all-inclusive seasonal wage; permanent-worker characteristics only thinly collected; the number of seasonal workers cannot be calculated accurately; the 1981 wage estimate may diverge from Department of Agriculture series; the correlation is not proof; the fate of departing workers was not directly researched.
Gaps in the evidence¶
The argument relies on farmer recall of 13-year-old employment data and lacks worker-side or farm-record data (payrolls, co-operative delivery records) that would verify employment and wages. Linking farm mechanization to rural depopulation and unemployment needs district-level census or migration data tracking individuals or households, and direct evidence on where displaced workers went (mine recruitment records, urban surveys); some of these existed then, though little exists for informal-sector and women's employment. A random sample or agricultural census farm-level data would better test representativeness and the causal weights of push versus pull.
Replicability: Data described but not public.