Controversies about Capitalism and Apartheid in South Africa: An Economic Perspective¶
Nicoli Nattrass (1991). Controversies about Capitalism and Apartheid in South Africa: An Economic Perspective. Journal of Southern African Studies, Vol. 17, No. 4. Taylor & Francis.
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | synthesis review |
| Period studied | 1940s–1990 |
| Regions | South Africa |
| Themes | business and state, comparative theory, labour markets wages, politics ideology elections, origins segregation, methods data sources, transition post apartheid |
| Citation key | nattrass1991 |
Argument¶
The article reviews the liberal-radical (race-class) debate on apartheid and capitalism from an economist's standpoint, focusing on the neglected role of economists. Its central claim is that confusion arose from two things: different models of wage determination, and failure to separate the short-term interests of capitalists from the long-run interests of capitalism. Both camps are said to have read their conclusions off their theoretical paradigms rather than from empirical evidence. Neo-classical liberals relied on marginalist tools that take the market as given, while radical structuralists relied on untestable functionalist arguments. Nattrass traces a convergence since the late 1970s: radicals moved toward contingency and agency, liberals toward power and class, and both toward social democracy, with the prospect of a negotiated transition helping this along.
Evidence and method¶
A historiographic survey of secondary literature: Hobart Houghton, Horwitz, Wilson, Bromberger, Lipton, Johnstone, Wolpe, Legassick, Davies, Greenberg, Posel, Hindson, Saul and Gelb, and others. No new primary data are presented. The author refers to her own unpublished Oxford D.Phil. on wages and profits.
Place in the debates¶
- Liberal vs radical (revisionist) debate on apartheid and capitalism (see Liberal vs revisionist: was apartheid functional or dysfunctional for capitalist growth?). Tries to mediate. Both sides are criticised for deriving conclusions from theory rather than evidence. She is more critical of radical functionalism but also faults neo-classical liberals for taking the status quo as given.
- Cheap labour thesis (see Cheap labour thesis and the function of the reserves). Notes the hypothesis was reformulated in relative terms, which makes it untestable. She contrasts the radical subsistence-wage view with the liberal supply, demand and bargaining view, and argues that recent work has undermined the early radical presumptions.
- Apartheid: functional for capital or a fetter on growth (see Liberal vs revisionist: was apartheid functional or dysfunctional for capitalist growth?). Distinguishes what is good for capitalists from what is good for capitalism. Argues that the liberal claim of net harm to growth cannot be proven empirically, and that the radical functionalist claim is also unfalsifiable.
- Whether economic growth would erode apartheid (see Liberal vs revisionist: was apartheid functional or dysfunctional for capitalist growth?). Shows liberals were divided (O'Dowd optimistic; Horwitz and Wilson pessimistic), so the 'central liberal thesis' is not representative.
- Were black wages too high or too low (internal liberal debate) (see Racial wage gaps, wage convergence and wage determination). Reports the 1950s-70s division between those wanting minimum wages (Katzen, Burrows) and those saying wages tracked productivity (Spandau, Viljoen). She does not adjudicate.
Findings¶
-
Liberals separated the interests of capitalists from those of capitalism; under-payment was good for the former, bad for the latter. (p. 7)
'Exploitation' or under-payment in the conventional economic perspective was regarded as good for capitalists but bad for capitalism.
-
Radicals are said to have collapsed this distinction, committing a fallacy of composition. (p. 7)
The analytical distinction between the short-term interests of capitalists and the dynamic interests of capitalism was unfortunately lost on most radical critics.
-
The liberal claim that apartheid slowed growth rests on theoretical faith rather than empirical proof. (p. 8)
is in the final analysis derived more from a faith in the efficiency of the market mechanism than from empirical findings
-
Radical structuralism's central methodological flaw is functionalism, which is untestable. (p. 16)
The main weakness of the radical structuralist approach was its functionalist methodology.
-
Functionalist explanation is unfalsifiable because almost any policy can be read as serving capital. (p. 16)
It is difficult to imagine any policy which could not be construed as 'functional' in some respect
-
The relative reformulation of the cheap-labour hypothesis cannot be tested. (p. 18)
Such a position, which reformulates the cheap-labour hypothesis in relative rather than absolute terms, is, as Black and Dollery have argued, impossible to test as it involves counterfactual evidence which is obviously unavailable.
-
Liberal and radical views of wage determination yield opposite conclusions about exploitation. (p. 18)
The exploitation of labour can only occur when workers get paid less than their marginal product, that is, when markets are not functioning perfectly.
-
Empirical work by Posel and others undermined the image of apartheid as a monolithic plan serving capital. (p. 21)
Posel for example shows the limits of the view of apartheid as monolithic 'Grand Plan'
-
The 1980s crisis and business disaffection undermined the view that apartheid was functional for accumulation. (p. 22)
By the 1980s it had become abundantly clear that the engine of growth which drove the 1960s boom had run out of steam.
-
Saul and Gelb's profit-squeeze account lacked empirical support, and recent work contradicts it. (p. 24)
They simply asserted that profit rates soared because they believed that apartheid policies sufficiently depressed the wage share.
-
The conclusion argues that the liberal/radical labels have lost meaning as debate shifts to a future mixed economy. (p. 25)
The old categories of 'liberal' and 'radical' have become blurred if not meaningless.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Johnstone, Race, Class and Gold (1976), analysis of mining capital and white labour | secondary literature compilation | early twentieth century gold mining | Treated as the main radical sectoral study whose detailed analysis of mining is discussed and then criticised for being generalised by later structuralists. |
| Wilson, Labour in the South African Gold Mines (1972) | secondary literature compilation | gold mining labour system, up to 1970s | Cited as the liberal study showing the migrant labour system, monopsonistic collusion and influx control were vital to gold mining economics. |
| Posel, Influx Control and the Construction of Apartheid 1948-61 (D.Phil, Oxford, 1987) | secondary literature compilation | 1948-1961 | Used as a detailed empirical study of state, employers and labour showing apartheid was not a monolithic Grand Plan and that differentiated labour was not the intention of policy. |
| Greenberg, Legitimating the Illegitimate: State, Markets and Resistance in South Africa (1987), study of labour bureau structures | secondary literature compilation | 1950s-1980s | Cited as radical empirical work on labour bureaux showing the controls were less monolithic and effective, prompting re-evaluation of functionalism. |
| Hindson, Pass Controls and the Urban Proletariat (1987) | secondary literature compilation | apartheid pass-law era | Used to show the radical account of the pass system relied on official pronouncements rather than evidence, and to put forward differentiated-labour reproduction. |
| Simkins, 'Agricultural Production in the African Reserves of South Africa: 1918-1969' (JSAS 1980) | secondary literature compilation | 1918-1969 | Cited as an exception to the theory-driven debate: an empirical investigation of homeland agricultural production relevant to Wolpe's analysis. |
| Bell, 'Capital Intensity and Employment in South African Industry' (SAJE 1978) | secondary literature compilation | not stated | Cited as an empirical critique of Legassick that was a rare case of evidence being brought to bear on the debate. |
| Nattrass, 'Wages, Profits and Apartheid' (unpublished D.Phil, Oxford, 1990) | secondary literature compilation | post-war | Relied on for the claim that Saul and Gelb's presumptions about soaring profit rates and the depressed wage share were incorrect. |
| Lipton, Capitalism and Apartheid: South Africa 1910-1986 (1986) | secondary literature compilation | 1910-1986 | Discussed as a major study whose conclusions about capitalist attitudes rested on what capitalists said rather than did. |
| Webster, Cast in a Racial Mould: Labour Process and Trade Unionism in the Foundries (1985) | secondary literature compilation | not stated | Cited as an example of radical empirical work distinguishing segments of labour markets. |
Inference¶
Quantification: None.
A historiographical and methodological review of the liberal-radical debate, conducted by reading the theoretical premises and empirical use in the literature. Claims rest on critical comparison of paradigms (neo-classical, Keynesian, structuralist Marxist, analytical Marxist) and quoted passages, not on original data analysis.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Both liberals and radicals derived their conclusions about apartheid and capitalism more from theoretical premises than from empirical evidence. (p. 670) | Reading of the liberal and radical literature (Horwitz, Hobart Houghton, Legassick, Wolpe, Johnstone); no data analysis. | Illustrative or anecdotal |
| The liberal claim that apartheid retarded growth cannot be proved empirically and rests on faith in the market mechanism. (p. 660) | Argument about the lack of measurable evidence, with reference to Bromberger and Hughes and Moll. | Asserted |
| Radical structuralist functionalism is untestable, since any policy can be construed as functional for capital. (p. 668) | Logical critique, supported by the Schumpeter quotation and by Lipton's critique of Legassick and Innes on the profit rate. | Illustrative or anecdotal |
| The relative reformulation of the cheap-labour hypothesis is impossible to test because it requires counterfactual evidence. (p. 670) | Legassick's postscript and Black and Dollery's critique. | Illustrative or anecdotal |
| Saul and Gelb's account of soaring apartheid-era profits lacked empirical support, and later work (the author's thesis) shows their presumptions were wrong. (p. 676) | Appeal to the author's unpublished D.Phil on wages and profits; data not shown in this article. | Asserted |
| Recent empirical studies (Posel, Greenberg, Hindson, Glaser) are eroding the functionalist view of the apartheid state and moving radicals and liberals toward a social-democratic middle ground. (p. 675) | Synthesis of the cited secondary studies of influx control, labour bureaux and industrial decentralization. | Partial data |
| Liberal economists were not united on neo-classical premises; they split over whether black wages were too high or too low and over the remedies. (p. 660) | Survey of the South African Journal of Economics literature (Burrows, Katzen, Spandau, Steenkamp, Horwood and others). | Partial data |
The author's caveats¶
The author acknowledges that the liberal and radical positions could not be tested against empirical sources and that little empirical research existed to resolve the debate. She notes that the cheap-labour hypothesis in its relative form involves unavailable counterfactual evidence.
Gaps in the evidence¶
As a historiographical review the article offers no systematic data of its own. Claims about the post-war wage share, profitability and capital intensity rest on the author's own unpublished thesis, which is not presented here. Testing these claims requires national accounts and industrial census series on wages, profits and productivity, plus archival records of employers and the state, most of which existed at the time and could now be exploited.
Replicability: Not applicable.