South Africa in a Comparative Study of Industrialization¶
Stanley Trapido (1970). South Africa in a Comparative Study of Industrialization. Journal of Development Studies.
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | theoretical or conceptual |
| Period studied | 1830s–1970 |
| Regions | South Africa, Witwatersrand, Kimberley, border industrial areas, Imperial Germany, Tsarist Russia, American South |
| Themes | origins segregation, mining migrant labour, agriculture land, manufacturing industrial policy, bantustans removals influx, labour markets wages, business and state, politics ideology elections, comparative theory |
| Citation key | trapido1970 |
Argument¶
Trapido treats South Africa as a 'labour-repressive' economy, in the sense of Barrington Moore, and compares it with Imperial Germany, post-emancipation Russia and the pre-Civil War American South. He argues that coercion of African labour in agriculture, in the mining compounds and later in urban industry has held African wages almost static for eighty years while capital accumulated rapidly. The mining compound, an alliance of gold and maize and heavy taxation of gold together produced a large surplus that financed manufacturing, the state and successive groups of beneficiaries. The four phases are commercial agriculture, diamond and gold mining, English-speaking manufacturing and Afrikaner finance and industry. He rejects the liberal expectation that growth will dissolve racial controls, and sees apartheid's homelands and border-industry policy as the latest, most systematic stage of repression, supported by the interests of white consumers, Afrikaner capital and foreign investors.
Evidence and method¶
Secondary literature (Moore, Bendix, Gerschenkron, Blainey, Nevins); Frankel's figures on gold output, dividends and state revenue (including Table 1); official reports and commissions (Smit, Fagan, Stallard, Mine Natives Wages Commission); figures on border-industry investment and IDC finance. No archival or survey work is cited.
Place in the debates¶
- Liberal vs revisionist views on whether industrial growth erodes racial controls and raises African wages (see Liberal vs revisionist: was apartheid functional or dysfunctional for capitalist growth?). Revisionist. He criticises white liberals' 'economic determinism' and argues that labour repression and the economy reinforce each other, with job fragmentation holding African wages down.
- Cheap labour thesis / role of labour repression in accumulation (see Cheap labour thesis and the function of the reserves). Strongly in favour. Depressed African wages produced the surplus that, via taxation, tariffs and railways, financed mining profits, the state and manufacturing, and they attract foreign investors.
- Causes of the Jameson Raid and the South African War (see Pre-Union colonial economy, the mineral revolution and the South African War). Following Blainey, he is convinced that the need to rationalise control of mine labour and cut costs was a main cause of the Raid.
- Whether Western worker 'embourgeoisement' rests on extracting surplus from non-industrial nations. Rejects the premise. Returns to foreign investors are highest in mature economies, and South Africa is an advanced industrial economy.
Findings¶
-
The distinctive feature of South Africa is stagnant African labour costs despite capital accumulation and growing skills. (p. 2)
The unique feature of the South African economy is that in the last eighty years the cost of African labour has changed little, both in agriculture and in mining.
-
He draws a political parallel between German iron and rye and South African gold and maize. (p. 3)
has its South African counterpart in the uneasy union of ‘gold and maize’.
-
Mine labour control is tied to the Jameson Raid. (p. 4)
convinced that the need to rationalize control of the labour force—as a major part of the overall need to cut costs—was one of the main causes of the Jameson Raid
-
Mining profitability would have allowed higher African wages in the 1935-50 period, yet the Chamber of Mines resisted them and the state repressed the 1946 strike. (p. 7)
Even if we accept the imperative of the Chamber of Mines cost structure, therefore, it would have been possible to raise African wages significantly some time between 1935 and 1950.
-
State revenue from gold mining grew far faster than output or dividends between 1913/14 and 1936-37 (index numbers from Frankel). (p. 7)
the corresponding figures for 1936-37 were 217, 196, 203 and 1,191
-
Gold taxation funded the infrastructure of manufacturing, such as iron and steel, and was in effect made possible by low mine wages. (p. 8)
Revenues from gold mining provided the state with the means of establishing the infrastructure of manufacturing industry.
-
Liberal expectations are wrong: when Africans take over white jobs, the work is fragmented and wages stay unchanged. (p. 9)
But, as unorganized African workers come to do work previously undertaken by Whites, job fragmentation takes place and African wages remain unchanged.
-
By 1945 English-speaking industry no longer needed repression for accumulation, but the 1948 Nationalist victory blocked reform. (p. 9)
repression of African workers was no longer necessary for capital accumulation.
-
The homelands ideology legitimates a new, more systematic phase of labour repression. (p. 10)
The homelands ideology, therefore, provides policy-makers with a terminology and an implicit justification for the most recent and systematic phase of labour repression.
-
Border industries received substantial private capital in 1960-68, alongside Industrial Development Corporation support and lower African wage determinations. (p. 10)
Private capital, some of it redirected from the older industrial areas, estimated to be R.314 million, was invested in the border areas in the years 1960-68.
-
Manufacturers face a dilemma: higher African purchasing power would widen the market but raise costs, so low wages continue to supply the surplus. (p. 11)
To raise the purchasing power of Africans would expand the internal market, but it would also raise the cost of manufacturing
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Frankel, Investment and the Return to Equity Capital in the South African Gold Mining Industry, 1887-1965 (1967) | secondary literature compilation | 1887-1965 (state revenue 1919-33 and 1935-63) | Supplies the figures comparing state revenue and investors' returns as percentages of equity capital in large gold mines. |
| Frankel, Capital Investment in Africa (1938), pp. 113-14 (Table 1) | secondary literature compilation | 1913-1936/37 | Index table showing gold output, net product, dividends and Government revenue from gold mining rising at very different rates, with Government revenue rising most. |
| Frankel's 1930 estimate of the Witwatersrand share of Government and Provincial revenue | secondary literature compilation | 1930 | Indicates that about half of government revenue came from the Witwatersrand gold mines. |
| Mine Natives Wages Commission report (1934 recommendation) | commission or official report | 1934 | Used to show that the Chamber of Mines rejected a small wage increase for African miners even when profits were rising. |
| Smit Report and Fagan Commission | commission or official report | c. 1942-48 | Cited as evidence that by 1945 non-Afrikaner industry and commerce favoured a more stable African labour force. |
| Stallard Commission Report (1922) | commission or official report | 1922 | Shows that the 'redundant Native' concept used in homelands policy had precedent in 1922. |
| Unspecified reports on urban African family incomes relative to the poverty datum line | commission or official report | c. 1930-1970 | Supports the claim that urban African wages remained near or below subsistence. |
| Figures on private capital and Industrial Development Corporation investment in border areas, 1960-68 | official statistical series | 1960-68 | Quantifies capital flowing to border industries to support the claim of a new phase of labour repression. |
| Blainey, 'Lost Causes of the Jameson Raid', Economic History Review (1965) | secondary literature compilation | 1890s | Supplies Hammond's 'gold factories' quotation and underpins the claim that labour-cost control was a cause of the Jameson Raid. |
| Bendix, Work and Authority in Industry (1956) | secondary literature compilation | post-emancipation Russia | Provides the comparative evidence on Russian internal passports, barracks and migrant labour. |
| Comparative secondary literature on Germany and the US South (Dawson 1911; Gerschenkron 1943; Bowen 1950; Nevins 1947; Moore 1967) | secondary literature compilation | 19th century-early 20th century | Supplies the comparative cases (Prussian agrarian coercion, 'iron and rye', cotton plantations) for the analogy with South Africa. |
Inference¶
Quantification: Illustrative figures.
Comparative historical-structural argument by analogy with Tsarist Russia, Imperial Germany and the American South, resting on Barrington Moore's 'labour-repressive economy' concept. It draws on selected figures from Frankel and unsourced official statistics, with no systematic data analysis.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| South African industrialisation rested on a labour-repressive system, with African wages essentially static for eighty years in mining and agriculture despite capital accumulation and rising skills. (p. 2) | Asserted from comparative theory; no wage series is presented. | Asserted |
| Gold mining supplied a disproportionate and rising share of state revenue, which financed manufacturing infrastructure such as iron and steel, and this was made possible by depressed mine wages. (p. 315) | Frankel (1938, 1967) index and percentage figures plus 1962-63 official tax calculation; the link to wage depression is asserted. | Partial data |
| The Chamber of Mines could have raised African wages between 1935 and 1950 because rising gold prices and profits allowed it, yet it refused and instead used coercion. (p. 315) | Profit-period statements, the 1934 Mine Natives Wages Commission recommendation and the 1946 strike; no figures are given for returns or wages. | Partial data |
| The Kimberley and Witwatersrand compound system and the migrant labour system were instruments of labour control, and the need to cut labour costs was a main cause of the Jameson Raid. (p. 312) | Blainey (1965) plus the author's assertion; the author acknowledges the point is controversial. | Illustrative or anecdotal |
| Gold and maize form a political alliance analogous to Germany's 'iron and rye', which blocked reform of labour controls, and the 1948 Nationalist victory entrenched labour repression. (p. 311) | Analogy to the German literature (Dawson, Gerschenkron, Bowen) and general political narrative; no election or lobbying data. | Illustrative or anecdotal |
| Homelands and border industry policy is a new systematic phase of labour repression that depresses African wages and shifts income toward Afrikaner areas. (p. 318) | Border-area investment figures (R314m private capital, IDC R21m to R65m) and the statement that wage determinations are lower there; no source is cited. | Partial data |
| Mining technology improved greatly (faster shaft sinking, drilling and drives) while the labour regime persisted, making repression more effective. (p. 314) | Unsourced technical figures for 1898, 1917 and 1960. | Illustrative or anecdotal |
The author's caveats¶
The author notes that 'this is tricky ground for the uninitiated' when using Frankel's estimates, and that the Jameson Raid causation is 'a matter of controversy'. Otherwise the article is an interpretive essay and does not discuss data limitations.
Gaps in the evidence¶
The central claims of static African wages, a surplus extracted through wage depression, and a mine wage–profit trade-off need wage series and profit data. Chamber of Mines wage and cost data, Frankel's underlying series and census and industrial-census wage data existed then and could test them. The border-industry and Afrikaner-capital claims lack cited sources. Statistical comparison with other cases was not attempted, and the comparison with Russia, Germany and the US South rests on secondary literature.
Replicability: Data poorly documented.