Liberal vs revisionist: was apartheid functional or dysfunctional for capitalist growth?¶
Did apartheid and racial policy serve capitalist accumulation, or did it conflict with market growth so that growth would erode it?
Works engaging the debate: 81
Positions¶
- Liberal/modernisation: apartheid was economically irrational and growth would erode it (Houghton, Horwitz, O'Dowd, Lipton, Moll, Feinstein)
- Revisionist/Marxist: apartheid and capitalism were mutually reinforcing (Legassick, Wolpe, Johnstone, Saul and Gelb, Davies et al.)
- Mixed or periodised: functional until the 1960s, dysfunctional afterwards (Nattrass, Posel, Seekings and Nattrass, Freund)
Evidence base¶
Much of this debate was argued from theory and from compilations rather than from primary series. Both camps leaned on the same reference apparatus. The SAIRR annual Surveys of Race Relations (Horrell (1968); Horrell et al. (1972)) supplied re-tabulated Hansard replies, Department of Statistics releases, Mining Statistics and Bureau of Market Research household surveys. The financial press (Financial Mail, Rand Daily Mail) carried employer statements. The commissions (Fagan UG 28/1948, Tomlinson UG 61/1955, Riekert RP 32/1979, Wiehahn) supplied policy intent. The early revisionists built their case by close reading of Hansard 1947-71, Board of Trade and Industries and Van Eck reports, and Fagan evidence (Legassick (1974)). They also used Chamber of Mines tariff-cost estimates, budget figures and mining wage and employment statistics for 1921-32 (Davies et al. (1976)), and Frankel's mining revenue figures (Trapido (1970)). The later Poulantzian work rested on press quotations, April 1977 vacancy statistics and ILO occupational shares (Davies (1979a); Davies (1979b)). Several liberal and synthetic reviews present no data of their own (Kantor and Kenny (1976); Nattrass (1991); Freund (1987a); Posel (1983)). Mine labour arguments on all sides reuse Wilson's gold-mine earnings series at second hand (Harvey (1974); Lemon (1984a); Labor Markets in South Africa... (n.d.)).
The quantitative turn came from a small set of works. Moll (1991b) (also Moll (1991c); Moll (1991d)) benchmarks South Africa four ways. It compares peak-to-peak GDP growth before and after 1948, using Du Plessis and South African Statistics series with a Chamber of Mines ore-milled adjustment. It ranks South Africa against 20 middle-income countries for 1950-85 using World Bank World Tables and Summers-Heston selection. It tracks UNCTAD manufactured-export shares for 1955-85. It sets thesis-based TFP for 1950-73 against Maddison's comparators. Nattrass (1990) decomposes manufacturing, mining and core-sector wage shares, profit shares and output:capital ratios for 1948-55 and 1955-60 from CSS and SARB national accounts, alongside Wage Board reports. Nattrass (1977) reconstructs racial income shares for 1970-75 from Statistical Year Book wage series, Manpower Surveys 8 and 11, and homeland national accounts. Mariotti (Mariotti (2008); Mariotti (2009a); Mariotti (2009b)) uses 1970 and 1980 population census microdata, biennial Manpower Surveys for 1969-85 (271 occupations by race, ranked by the SEO index) and a manufacturing-census industry panel of 327 observations. Fedderke and Simkins (2004) combines Union Statistics for Fifty Years and censuses with sectoral growth accounting, SARB series and a political-instability index for 1935-97. Fine and Rustomjee (1996) reclassifies the 1984 and 1988 input-output tables and JSE ownership data. Archival industry and regional studies add depth on particular cases. Crankshaw (1990) uses Censuses of Industrial Establishments 1942-50, Industrial Tribunal reports and union records. Coupe (1990) uses TUCSA, Garment Workers Union, Assocom and Germiston Industries records. Minkley (1990) uses Border Chamber of Industries files, company archives and worker interviews. Greenberg (1987) draws on 1982 interviews with labour bureau and TEBA officials. Torchia (1988) reads FCI annual reports and Industry and Trade. Kerby (2014) compiles homeland wages from SALDRU press clippings.
Contested evidence¶
The growth record itself is read in opposite ways. Revisionists simply assert that growth coexisted with intensified discrimination (Legassick (1974)). Sources of Growth in the... (1990) uses Nattrass and Kuznets comparisons to call growth of about 5 per cent a year comparable to the fastest-growing countries. Moll (1991b) finds post-1948 growth no faster than before on the ore-milled measure, a tied 13th rank among 20 peers, falling manufactured-export shares and near-zero TFP. Mariotti and Fourie (2014) and Posel (2011) adopt Moll's reading, and Fedderke and Simkins (2004) interprets its instability and rights indices as favouring the liberal case. Profitability is the sharpest empirical clash. Saul and Gelb's soaring apartheid profits and Davies's post-1974 profitability crisis are asserted without profit-rate series (Davies (1979a); Davies (1979b)). Nattrass (1990) finds the manufacturing profit share rising in 1948-55 because black real wages fell, which partly vindicates the radical claim. She also finds the manufacturing rate of profit falling by about 2 per cent a year in both 1948-55 and 1955-60, because the output:capital ratio fell. Her series are not reproduced, so the result cannot be audited.
The erosion of job reservation is a second case where the same chronology is assigned different causes. Davies (1979a) credits a capital imperative after the mid-1970s. Crankshaw (1990) cites building-industry ratios of Africans to non-Africans rising from 1.7 in 1945 to 2.8 in the early 1970s, together with state exemptions, as evidence of capital and the state overriding craft unions. Mariotti (2008) uses Manpower Surveys and manufacturing wage bills to date desegregation before the 1973 strikes and attributes it to white workers. All of them lack Industrial Council determinations. Employer preferences are argued from incompatible scraps. Lipton's liberal reading is countered by Freund (1987a) and Freund (1987b) with Webster's foundry-manager poll and Bernstein's finding that 45 per cent favoured some controls. Torchia (1988) and Coupe (1990) quote industrialists blaming influx control for poor productivity. Influx control's effectiveness depends on which urbanisation figures are used. Hindson (1987) relies on Simkins's declining urbanisation rates. Posel (2011) gives urban African growth of 58 per cent in the 1950s and 28 per cent in the 1960s and says the 1950s preference policy largely failed. Greenberg (1987) infers breakdown from interviews. Lemon (1984a) calls the system an efficient adaptation for capital, while Curtis (1984) calls its effects contradictory, with the same bureau allocation counts. Racial income shares also diverge. McGrath's roughly constant white share of about 70 per cent for 1920-70 (Sources of Growth in the... (1990); Steward (2004a)) and Randall's data (Smith (1972)) imply stasis. Nattrass (1977) shows a fall from 74 to 68 per cent in 1970-75, but only by assuming all property income accrues to whites. Wilson's flat black mine earnings are read as monopsony (Labor Markets in South Africa... (n.d.)), as labour repression (Trapido (1970)) and as shared exploitation (Harvey (1974)).
Underused and missing sources¶
The debate's central quantities, namely profits by firm and sector, black and white wages by occupation, and enforcement intensity, are rarely measured from primary series. Chamber of Mines wage, cost and recruiting records are almost never used directly. Most authors take them at second hand through Wilson, Spandau or Lipton (Lemon (1984c); Rich (1990); Kantor and Kenny (1976); Mazanderani (2012)). Labour bureau registers, pass-law arrest and endorsement-out series, and Department of Native Affairs files would settle the influx-control dispute (Greenberg (1987); Frankel (1979); Posel (2011)). Industrial Council agreements and Industrial Tribunal determinations would test the white-worker mechanism behind desegregation (Mariotti (2008); Mariotti (2009a)). FCI, Assocom and Chamber of Commerce internal minutes would test claims about business attitudes (Torchia (1988); Freund (1987a)). Company accounts would test Davies's profitability claim. Census microdata exist only for 1970 and 1980 and exclude the independent homelands (Mariotti and Fourie (2014); Kerby (2014)).
With that series, Nattrass's unreproduced 1948-60 wage-share and profit decomposition could be rebuilt, Mariotti's industry panel extended back before 1950, and Crankshaw's racial employment ratios checked beyond building.
Strength of key claims¶
| Basis | Share of key claims |
|---|---|
| Systematic data | 9% |
| Partial data | 46% |
| Illustrative or anecdotal | 23% |
| Asserted | 23% |
Comparative figures for all debates are on the Literature overview.
Best-evidenced works¶
Moll (1991d), Nattrass (1990), Mariotti (2008), Fedderke and Simkins (2004), Fine and Rustomjee (1996), Crankshaw (1990)
Most-used primary sources¶
Holdings are matched by machine against the archive's file manifest; a dash means no match was found, not that the project certainly lacks the source.
Works and their positions¶
| Work | Position |
|---|---|
| Atkinson (2007) | Endorses Greenberg's argument that capitalist development eventually outgrew the need for the racial labour apparatus. Applies it to the shift from labour shortage to surplus after about 1970. |
| Becker (1987) | Presents both views; notes forces (boycotts, cost of discrimination, competition from firms hiring cheaper nonwhites) that threaten apartheid, while noting National Party efforts to hold the black middle class back. |
| Carter (1958) | An early liberal-leaning view: apartheid is meant to preserve racial divisions in a dynamic economy, but dependence on African labour makes strict separation doubtful. It does not use the later 'cheap labour' framework. |
| Cooper (2007) | Cooper says the book takes both traditions seriously, and moves beyond generalised conceptions of the logic of capitalism and its relationship to race towards a systematic study of income and welfare and their relationship to institutions and programmes. He endorses this approach. |
| Coupe (1990) | Treats English-speaking business liberal rhetoric as partly self-serving. It highlighted white racist labour practices while downplaying mining capital's stake in African oppression. |
| Crankshaw (1990) | Sides with a revisionist, class-based reading. Labour apartheid, craft unionism and growth coincided and conflicted. Capital increasingly opposed job reservation, and the state steadily dismantled it in practice while still protecting white privilege rhetorically. |
| Crush (1984) | Crush criticises Western's 'orthodox, and dated, liberal' race-centred reading and favours a class analysis linking race to the social relations of production. |
| Curtis (1984) | Criticises its omission of power, conflict, force and class. It can explain apartheid's persistence only through the 'abnormal psychology' of racist whites. |
| Davies (1979a) | Revisionist. He holds that the racial division of labour accorded with capital's interests until the mid-1960s and that capital now seeks to change it because of crisis and class struggle. He rejects the view that change flows from the normal growth of capitalism alone. |
| Davies (1979b) | Revisionist: the racial division of labour first served capital, then ceased to do so when profitability fell and black militancy rose. Reform is a bourgeois strategy to restructure and secure exploitation, not a victory for liberal rationality. |
| Davies et al. (1976) | Marxist-structuralist. Apartheid is the outcome of struggle among capital fractions to restructure relations of exploitation, chiefly agriculture's labour crisis. It is not an irrational racial departure from capitalist logic, and it is not simply the uniform will of a unified capital. |
| De Kiewiet (1941) | Writing before the revisionist school, he takes the liberal economists' line: the colour bar and civilized-labour policy are economically irrational, costly protection of white privilege, and race discrimination has psychological and social roots as well as economic ones. He also stresses frontier ideology and poverty, and is sympathetic to white fears. |
| Fedderke and Simkins (2004) | Sides with the liberal view: political instability harmed and property rights helped capital accumulation, and political rights show no growth-supporting effect. It calls the wider debate 'tired and hackneyed'. |
| Fine (1992) | Fine treats both as functionalist and top-down, explaining apartheid by the needs of capital while ignoring popular struggle. He says the new unions showed their premises to be wrong. |
| Fine and Rustomjee (1996) | Treats both as sharing a race-class dichotomy that neglects the forces of production and sectoral economic structure. The authors draw on revisionism but break with it by arguing that the class structure shifts and that policy cannot be read off class power directly. |
| Frankel (1979) | Notes that some industrialists see passes as an obstacle to mobility and productivity but says this is outweighed, and that during the mid-1970s recession they accepted influx control as a means of sending unemployment to the rural areas. |
| Freund (1987a) | Takes a middle path. He accepts that the functional fit has eroded since the 1960s and that Legassick-style revisionism was too functionalist. He insists that revisionism remains a necessary foundation and that Lipton under-credits it. |
| Freund (1987b) | Takes a middle course. He accepts that Lipton's evolutionary account of declining functionality is strongly supported, and that the 1970s revisionists were too functionalist. He insists, against Lipton, that capital valued racial hierarchy and was not indifferent to colour, and that revisionism must be contextualised rather than discarded. |
| Freund (2009) | Freund traces the shift from the liberal progress narrative (Houghton, Nattrass, Lipton) to the radical view (Wolpe, Legassick) that capitalism depended on cheap labour. He treats the MEC as the next step beyond both. |
| Giliomee and Schlemmer (1985) | Nattrass proposes a three-sector interlinked model (modern, white commercial farming, black subsistence) in which political institutions and migrant labour drive rural poverty, rejecting simple dualism. |
| Greenberg (1987) | Footnote 9 says the tension between control and markets is not a reformulated growth hypothesis, because it generally produces more frenzied efforts to elaborate racial domination rather than moderation. |
| Hackland (1980) | Revisionist. Apartheid under the NP raised exploitation of black workers and profits, benefiting both Afrikaner and English capital, rather than being opposed to capitalist interests. |
| Harries (1986) | Harries criticises Jeeves for adopting liberal arguments: that the state forced migrancy and the colour bar on reluctant magnates, that Kruger's mismanagement caused the Anglo-Boer war, and that high gold recovery costs forced low wages. He argues liberal historiography reflected its own post-1948 political concerns. |
| Harris (2002) | Draws on Greenberg and O'Meara to see apartheid as racial capitalism and Afrikaner nationalism as a class-alliance response to capitalist contradictions, and not a simple racial or volk story. |
| Harvey (1974) | Doubtful. The belief that British firms are beneficial 'appears to have been largely unfounded', though he sees black employment as the area where change is most likely. |
| Hindson (1987) | Criticises the liberal view that market forces will erode racial policy. Argues that market forces and repressive state policy reinforce each other in a vicious cycle of impoverishment, so both state and economy need transformation. |
| Hindson (1989) | Accepts Wolpe's critique of reductionism but says his 'interiorised' race-in-class idea does not settle the causal weight of race and class. |
| Hindson (1989) | Argues that liberal writers first identified the contradictory relationship between capitalism and apartheid and deserve credit, and that Wolpe's account of 1970s worker and student action could be read as supporting the liberal thesis. |
| Hindson (1989) | Sympathetic to the effort to rescue CST from functionalism, but thinks empirical work on class differentiation would lead to a far more serious reappraisal of it. |
| Hirsch (1989) | Rejects the incrementalist view; economic vulnerability and weakness, not growth, can force the state into serious reform. |
| Horrell (1968) | Implicitly liberal: cites economists and business bodies arguing that racial job barriers, influx control and industrial decentralisation retard growth and fuel inflation, while noting government insists on its policy. |
| Horrell et al. (1972) | Implicitly liberal: it reports business and labour-economist views that job reservation and the colour bar cause labour shortages, inflation and constrain growth, and it gives space to Oppenheimer and Anglo American on rapid African advancement. It offers no sustained theoretical argument. |
| Iliffe (1999) | Describes both schools and their weaknesses: liberals neglect Africans and equate economic history with growth, while radicals neglect institutional history. Welcomes the shift to empirical, disaggregated research that supersedes both. |
| Innes and Plaut (1978) | Argues that Davies et al. come close to liberal historiography, in which Afrikaner interests capture the state and distort the economy. The authors place themselves in a radical Marxist tradition. |
| Kantor and Kenny (1976) | Rejects the revisionist view that growth and white supremacy are mutually reinforcing. Argues industrialisation has structural requirements that conflict with white supremacy over the long run. Also criticises naive liberal stage theory (O'Dowd). |
| Kerby (2014) | Summarises both sides (capital benefited from low wages vs low wages as a by-product of protecting white workers). Argues the low homeland wages in the traditional view of the RIDP are consistent with both, and calls for a more empirical, non-ideological approach. |
| Legassick (1974) | Revisionist: rejects the liberal claim that apartheid is irrational and conflicts with capitalist growth. Argues that apartheid institutions assisted growth and the cheapening of black labour. |
| Lemon (1982) | Lemon relays the PFP and opposition claim that apartheid's costs (defence, influx control) were eroding white living standards, and notes that a dynamic economy demands skilled black labour. He treats this as a political argument rather than testing it. |
| Lemon (1984a) | Notes that job reservation and training restrictions are economically wasteful and reduce efficiency, while other controls reinforce capitalism. He holds that controls both hinder and serve growth. |
| Lemon (1984b) | Job reservation and training restrictions are wasteful and lower efficiency. The migrant and homeland system, by contrast, lets the white economy use rural labour while shifting social costs onto the homelands. |
| Lemon (1984c) | Argues that job reservation and training failures are economically wasteful, while migrant and influx controls reinforce the capitalist system by cheapening labour and externalising its costs. |
| Lewis (1990) | Sides with the 1970s class analysts (Trapido, Legassick, Wolpe, Johnstone) against liberals such as de Kiewiet and van der Horst. Also cautions that the 1970s emphasis on class over-reacted and neglected national oppression. |
| Lowenberg (1997) | Largely liberal/neoclassical: apartheid was an inefficient set of regulations that became increasingly dysfunctional for capitalist growth as manufacturing came to dominate. Whites came to bear the costs, so the system was abandoned. |
| Lundahl (1984) | Treats the white polity as a set of competing interest groups. Farmers and white skilled workers have economic reasons to oppose African skill advancement, while industrialists may favour reform. Rural-origin Afrikaner politicians give farmers disproportionate weight. |
| Malherbe and Segal (2001a) | Takes a mixed middle line: firms often benefited from apartheid protection but also bore its costs, and governments mostly served white labour and agriculture rather than capital. Defers to Lipton on the cost-benefit tabulation. Calls apartheid 'an unambiguous loss to the economy and society'. |
| Mariotti (2008) | Capital is not treated as the driving political force. Because the return to capital was lower under segregation, yet reform came late, she infers that political power lay with white workers, not capitalists. |
| Mariotti (2009a) | Argues that political power did not rest with capitalists. Capital's returns were lower under segregation, yet transformation came late. |
| Mariotti (2009b) | Denies that capitalist pressure or state intervention drove desegregation. Says capital held little political power, because the return to capital was lower under segregation and capital would otherwise have forced change earlier. Gives white workers the agency. |
| Mariotti and Fourie (2014) | The authors present Lipton as the leading liberal and Legassick as Marxist, and argue that Nattrass (1991) ended the ideological debate in favour of more scientific, data-based analysis. They note that mining benefited for a long time from pass laws and migrant labour (Feinstein). |
| Massie (1997) | Notes the debate (Lipton, Wolpe, Oppenheimer) but mainly stresses that apartheid's costs and inefficiencies taxed the economy and that business pressed for reform after Sharpeville and in 1985. |
| Mazanderani (2012) | Notes (via Iliffe) that these paradigms have dominated and paralysed the debate. Argues that both neglected finance and focused on labour and gold, and positions the thesis as filling that gap. |
| Milburn and Cole (1957) | Not engaged; this is a 1957 bibliography that predates the debate. It does separate SABRA's government views on apartheid from the SAIRR's opposing views. |
| Minkley (1990) | Takes a locally nuanced revisionist position. Capital was fractionalised and apartheid practices (influx control, labour bureau) only gradually served the needs of local manufacturing capital, which then adopted apartheid ideology to legitimise racial divisions of labour. |
| Moll (1991b) | The aggregate evidence 'appears to support the liberal critique of apartheid'. Labour controls protected inefficient low-productivity, cheap-labour firms and penalised more productive ones. But he stresses that labour policy is only one factor, and that trade and industrial policy also mattered. |
| Moll (1991c) | Aggregate evidence is read as supporting the liberal critique that apartheid hampered efficient resource allocation. Controls over black labour propped up low-productivity firms, which is a challenge to the radical case. |
| Moll (1991d) | Rejects the premise of a successful 'apartheid boom'. Argues the aggregate evidence supports the liberal critique that apartheid hampered efficient resource allocation and the full use of black workers. Also faults the radical 'cheap labour' case for ignoring how controls on black workers subsidised inefficient, low-productivity firms. |
| Nattrass (1977) | Cites both views and mostly stays neutral; she urges reconsidering government policies because of their long-term costs. |
| Nattrass (1990) | Tests the radical claim empirically. It is supported only for the profit share in 1948-55, and rejected for the rate of profit and for 1955-60. Argues the liberal point (slower structural change) and the radical point (state responsive to fractions of capital) can both hold. |
| Nattrass (1991) | 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. |
| Nattrass (1991) | 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. |
| Nattrass (1991) | Shows liberals were divided (O'Dowd optimistic; Horwitz and Wilson pessimistic), so the 'central liberal thesis' is not representative. |
| Nattrass (1999a) | Notes that the TRC adopted the radical view; she regards the blanket condemnation as highly contestable. She stresses that both liberal and radical writings accept that apartheid affected sections of business differently. |
| Nattrass (1999a) | Argues culpability should be graded by power, choices and knowledge, not inferred from benefiting; criticises third-order 'guilt by association' while accepting that powerful firms bore greater responsibility. |
| Nattrass (1999a) | Treats the TRC's implicit denial that growth contributes to democracy as contestable, citing Lipset and Przeworski et al. |
| Nattrass (1999b) | Presents the divide as structuring both the business submissions and the TRC's conclusions. She does not endorse either side, but notes that both accept apartheid affected different sections of business differently, and criticises the TRC for taking the radical line wholesale. |
| Nattrass (1999b) | Argues that the TRC's third-order, benefit-based culpability implies guilt by association. She favours differentiating firms by power, knowledge and choices, as the I. G. Farben court did. |
| Nattrass (1999b) | Notes that the TRC's blanket position implied growth contributed nothing to democracy, which is highly contestable given the literature (Lipset, Przeworski et al.), though she does not simply endorse Bernstein's pro-business version. |
| Nattrass (2006) | Nattrass praises Feinstein for passing between the two camps. She reports his finding that apartheid policies hampered growth, which is closer to the liberal view, alongside a radical-informed focus on conquest and coercion. |
| Olivier (1929) | Holds that race prejudice is largely a product of the industrial employer-employee relation and of economic competition. It is not primordial, and a colour-blind state policy is the remedy. |
| Posel (1983) | Rejects the 'either-or' framing of both camps; argues class and race are interpenetrating and that class analysis is necessary but not sufficient, so class cannot be ranked invariably over race. |
| Posel (1983) | Holds that segregation and apartheid have been simultaneously functional and dysfunctional, with the balance varying by period; rejects both pure functionalism and the liberal view that growth will inevitably erode racial policy. |
| Posel (2011) | Mixed: sees deliberate racialised economic engineering within capitalism and cheap-labour dependence, but argues job reservation, Bantu education and truncated consumer markets produced skills shortages and growth-inhibiting contradictions; growth was unimpressive comparatively. |
| Rex (1975) | Rex rejects plural society theory as normally applied, because segments are dynamically tied to capitalist industrial development. He allows a revised version in which groups shaped by history play a functional role within a basically capitalist society. |
| Rex (1975) | Rex notes that both camps use the logic of industrial capitalism. He sides with Adam against the liberals: capitalism gains from apartheid, especially from cheap unskilled labour. |
| Rich (1990) | Sides with Lipton and Dollery against the 1970s radical view that capitalism sustained apartheid. He sees the links between market economy and apartheid as tangential and the mining industry as opportunist and parasitical in adapting to the prevailing politics. |
| Schneider (2000) | Broadly revisionist and institutionalist. Labour repression and the captured apartheid state limited the domestic market and industrial development, though the pre-1948 state is seen as developmentalist. |
| Sharp and West (1984) | Argues the debate will be sterile without regional detail; in Namaqualand racial categorisation emerged from primitive accumulation on the frontier, and mining and agriculture hardened it without determining it. |
| Sitas et al. (1984) | Firmly the latter. Rejects Rostow-type stages, modernisation and the 'affluent worker' thesis; high living standards result from pressure from below and class conflict. |
| Smith (1972) | Sits between the two. He argues that separate development conflicts with economic efficiency (it encourages dispersal and creates barriers), but that white capitalists' dependence on cheap African labour means the conflict 'may not be as great as is sometimes supposed'. Growth will not by itself end racial inequality (citing Adam). |
| Smith (1972) | Allows that all races may be better off in absolute terms. He judges that most of the benefits accrue to Whites and most of the costs to non-Whites, and he rejects the 'generosity' reading of homeland expenditure as the necessary price of white survival. |
| Sources of Growth in the... (1990) | Distances itself from both camps. It argues the 1970s debate was too labour-oriented to scrutinise the growth evidence. It treats the liberal view (discrimination is dysfunctional for growth) and the neo-Marxist view (growth reinforces white supremacy) as unproven, leaning to the view that restrictions on mobility and property were anti-growth. |
| Sources of Growth in the... (1990) | Suggests racial categories in statistical tables constrain analysis and that class analysis by income or education might better explain growth. It also cautions that racial income gaps mix in many non-discriminatory differentials. |
| Sources of Growth in the... (1990) | Criticises Frank-style zero-sum 'Black poverty caused by White wealth' arguments. It says both dualism theories overstate labour's role and the isolation of the two sectors. |
| Steward (2004a) | Takes a modernisation-liberal line: market-driven urbanisation and rising incomes eroded pass laws and the Group Areas Act, and sanctions slowed these forces. It makes no reference to the revisionist view of apartheid as functional to capital. |
| Steward (2004b) | Broadly liberal. Market-driven urbanisation and growth eroded apartheid ('changing economic relationships have led to changed social relationships'), and sanctions slowed these forces. |
| Tatham (1990) | Reports both sides as paradigm-bound. Liberals saw racism as irrational and as retarding market-driven growth. Marxists saw segregation as functional to capital accumulation. The author is critical of the structuralist Marxists' politicism and neglect of struggle, and she sympathises with the social historians' emphasis on agency. |
| Torchia (1988) | Apartheid imposed growing costs on capital-intensive manufacturing, but also gave benefits (low unskilled wages, state-corporation bottleneck-breaking, a stable business environment), so manufacturers accommodated it for decades. |
| Trapido (1970) | 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. |
| Truth and Reconciliation Commission, Reparation and Rehab... (1998) | Rejects Anglo American's claim that capitalism would modernise and liberalise the country. It argues the corporation's empire was built on coerced labour and racist employment policies. |
| Van Schoor (1986) | Says segregation retards technical progress and productivity. It also says the mines subsidise the rest of the economy. |
| Van Wyk (2005) | Rejects both as inevitability theses. Draws on the liberal view that apartheid became dysfunctional for the economy and on the Marxist emphasis on structure, class and mass action, but frames apartheid as primarily political and Afrikaner-driven rather than functional to capital. |
| Van Wyk (2005) | Apartheid arguably aided growth until the late 1960s (cheap labour, negligible fiscal cost) but became counter-productive from the 1970s. |
| Zunes (1999) | Dismisses it in a note as an unrealistic scenario pushed by Western analysts. |