An Economic Investigation of Farms in the Maize Districts of the Orange Free State: Preliminary Report for Crop Season 1927-1928¶
J. C. Neethling (1930). An Economic Investigation of Farms in the Maize Districts of the Orange Free State: Preliminary Report for Crop Season 1927-1928. Economic Series No. 12, Division of Economics and Markets, Department of Agriculture. Government Printer, Pretoria (Union of South Africa, Department of Agriculture).
| Type | report |
| Scope | Economics and economic history of apartheid |
| Method | survey fieldwork |
| Period studied | 1927–1928 |
| Regions | Orange Free State, Kroonstad, Bethlehem, Lindley and Reitz districts, maize triangle |
| Themes | agriculture land, labour markets wages, methods data sources |
| Citation key | neethling1930 |
Argument¶
This preliminary report presents a farm-management survey of 109 white-owned farms in the north-eastern Free State maize districts for the 1927-28 crop year. Neethling argues that a typical farm combines maize with sheep, wool and some cattle. Operator's earnings depend mainly on farm size and on the share of cash receipts coming from maize, and cattle and large wool shares are associated with lower earnings. High yields lower unit costs because most per-morgen costs are roughly fixed. Extensive maize farming pays only at large scale. The report also shows how farm labour was paid, largely in kind through land, crop shares and grazing, and how little tractor mechanisation there was.
Evidence and method¶
A questionnaire survey: 114 farms visited and 109 analysed (107 for the maize enterprise, 97 with complete labour figures), in four districts. Census figures for 1927 and Department of Agriculture maize production statistics are used for comparison. Farmers' own estimates and notes (many kept for the Revenue Department) are checked by the enumerator. Some cost items are approximations or based on expert advice, such as veterinary and poultry officers. Rainfall tables and a farm map are included. The farm-by-farm Appendix I is not in the supplied text.
Place in the debates¶
- Cost-account versus farm-survey methods in agricultural economics (see Data reliability and econometric method). Prefers the survey method over complete cost accounts, since cost accounts represent only the most successful, specially trained farmers. Argues the 'law of averages' makes the survey representative.
- Whether Free State maize farming should be organised extensively at large scale or as mixed farming (see Agricultural policy, productivity, farm size and land reform). Finds a direct relationship between size and operator's earnings, and a strong positive link between the maize share of receipts and earnings. Concludes that extensive farming must be on a large scale and that cattle on a commercial scale did not pay in 1927-28.
Findings¶
-
The survey used 109 farms and relied on the survey method, with farmers' estimates verified where possible, rather than complete cost accounts. (p. 7)
For this etudy, 109 recorda were taken-a large enough number for the law of averages to function well.
-
The surveyed sample was biased slightly upward relative to the Census, with larger farms and higher yields (7.9 against 5.8 bags per morgen), so results are not strictly representative of the districts. (p. 14)
The average yield on 20,727 morgen of maize included in the survey was 7-9 bags per morgen, while the yield on 319,108 morgen included in the Census
-
Larger farms were scaled-up versions of the same mixed system: maize area, livestock and capital rose roughly in proportion to land. (p. 17)
a larger bnain_ of the 88mB type, and not a new kind of farming.
-
Regular African labour was thin: an average of about nine natives per farm cultivated roughly 31 morgen of crops each, plus livestock. (p. 20)
There was an average of 9 natives per farm, whieh mesns that 21-2 morgen of maize and 10 of other crops, or a total of 31- 2 morgen of crops, were cultivated per native.
-
Labour was the largest farm expense, and much of the wage bill took the form of in-kind payment (cash, maize lands, grazing for workers' stock). (p. 22)
It is clear that the most importent item of expense is that of labour. Therefore it is importent to analyse this item in greater detail.
-
Land given to African labourers was costed as an expense to the farm and a receipt at half price, because workers did most of the work on their own lands. This is evidence of labour-tenancy-like arrangements. (p. 23)
The product. thu.· given to the natives by the farm are 8 cost to the farm on the one hand and a receipt on the other.
-
Cattle farming on a commercial scale was unprofitable relative to other enterprises in the sampled year. (p. 26)
Appsrently it did not pay to go in for cattle farming on a commercial ecale.
-
Maize dominated the farm's crop income, supplying 47.4% of cash receipts and 90.8% of crop receipts. (p. 32)
Maize, the main enterprise, contributed 47·4 per cent. of the total cash receipts (see Table XV) or 90·8 per cent. of the cash received from crop
-
Average maize profit per farm was £373, and profit per morgen and per bag were positive in 1927-28 once land rent was included as a cost. (p. 34)
-
Higher yields lowered unit costs because most cultivation costs per morgen are fixed, so yield is the key driver of maize profitability. (p. 35)
it ill clear that a larger production per morgen will reduce COllt per onit.
-
Success varied enormously: the ten most successful farms earned £1,861 operator's earnings against a loss of £339 for the ten least successful, with maize share of receipts and yield per morgen being the main differences. (p. 36)
The Opor_'s Earning. of the successful farms was £1,861, that for ell farms £372, and for the least successful minus £339.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Farm management survey of 109 maize farms (114 visited), Orange Free State, crop year 1927-28 | author survey or fieldwork | Crop year 1927-28 | Main dataset: officer-administered questionnaires supply farm size, capital, livestock, receipts, expenses, labour and profits for all tables. |
| Farmers' own notes on receipts and expenditure (kept for Revenue Department returns) | company records or reports | 1927-28 | Used to verify questionnaire answers and estimates given by farmers. |
| Union Census figures for 1927 (farms of 100 morgen or larger; area, size groups, maize morgen and yields by district) | agricultural census | 1927 | Benchmark to test representativeness of the sample by average farm size, size-group distribution and maize yield per morgen. |
| Census figures for 1926 (European population of towns) | population census | 1926 | Shows nearby towns are too small to provide markets for perishable farm products. |
| Maize production by province and district, 1918-27 (Tables I-III) | official statistical series | 1918-1927 | Establishes the concentration of maize production on European farms and justifies choosing the Orange Free State districts. |
| Estimate of value of agricultural production 1926-27 by W. A. Horrocks, Division of Economics and Markets | official statistical series | 1926-27 | Gives maize as 14.4 per cent of estimated agricultural production value of £75m. |
| Maize cost-of-production study by Parish (School of Agriculture, Glen) and C. J. Uys | secondary literature compilation | 1920 onward | Cited as the cost-account method contrasted with the survey method; its data are not used. |
| Expert consultations: Division of Veterinary Services (pig feeding) and Principal Poultry Officer, Department of Agriculture | interviews oral history | c. 1928-29 | Supply feeding-cost assumptions for pig and poultry enterprise profit calculations. |
New data¶
Farm management survey of 114 farms visited (109 analysed, 107 for maize enterprise) in adjacent parts of Kroonstad, Bethlehem, Lindley and Reitz districts, crop year 1927-28, covering capital, livestock, labour, receipts, expenses and profits; individual farm figures given in Appendix I.
Inference¶
Quantification: Descriptive tables.
Descriptive farm-management analysis of a single-year cross-section of 109 farms: averages, grouping of farms by size, capital, income mix and yield, and comparison of the ten most and ten least successful farms. Sample representativeness is checked against census averages; there are no statistical tests or controls.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| The surveyed farms were larger than the district average and slightly higher-yielding, but the author argues the sample is fair. (p. 13) | Comparison of survey averages with 1927 census figures (Tables VI-VIII). | Partial data |
| A higher share of cash receipts from maize is associated with higher operator's earnings whatever the size of farm; cattle income is associated with lower earnings. (p. 27) | Cross-tabulations of survey data, Tables XXII-XXV, one year. | Partial data |
| Maize dominates cash receipts from crops and is the leading farm enterprise, with sheep and wool the main livestock income. (p. 31) | Survey receipts tables XV and XXX. | Systematic data |
| Most successful farms had much better operator's earnings because of larger maize area, higher yields and lower unit costs. (p. 36) | Comparison of the ten most and ten least successful farms with the median and mean (Table XXXVI). | Partial data |
The author's caveats¶
Preliminary report on a single crop year (1927-28) described as unfavourable; results are not conclusive, and the author plans to visit the same farms for four or five years. Survey area is deliberately small and uniform. Farmers' valuations of livestock and home-used products and some cost allocations are estimates; the 5 per cent interest rate was set arbitrarily; livestock profits are gross only; a few cost figures are marked as approximations. No soil survey existed. Survey farms are on average larger than the census average.
Gaps in the evidence¶
The argument rests on one year of data from a non-random sample of 109 farms, so it cannot separate structural advantages of scale from a single season's rainfall and prices. Multi-year panel data on the same farms, which the author intended to collect, and a fully random sample drawing on census farm lists would test robustness. Tests of causation (for example, size versus maize share) and an analysis of labour terms and share-cropping would need the farm-level records (Appendix I) or later survey rounds. Some of this material may survive in departmental archives.
Replicability: Data described but not public.