Historical Roots of Inequality in South Africa

Literature › Works

Francis Wilson (2011). Historical Roots of Inequality in South Africa. Journal of Economic History of Developing Regions, Vol. 26(1), introductory essay to special issue on inequality (this copy: paper circulated for the 'Towards Carnegie III' conference, University of Cape Town, 3-7 September 2012).

Type	journal article
Scope	Economics and economic history of apartheid
Method	synthesis review
Period studied	1652–2010
Regions	South Africa, Cape Colony, Kimberley, Witwatersrand/Transvaal, Eastern Cape, Bantustans/reserves, Southern Africa labour catchment (Malawi, Tanzania, Zambia)
Themes	origins segregation, mining migrant labour, agriculture land, bantustans removals influx, labour markets wages, inequality poverty unemployment, education human capital, transition post apartheid, demography urbanisation
Citation key	wilson2011

Argument

Wilson's introductory essay argues that present South African inequality, poverty and unemployment cannot be understood without their deep history. Racially biased accumulation began with conquest of land and water, then mineral wealth that flowed to whites, and was entrenched by the compound and migrant labour system. The same dynamics that created wealth in the white-controlled economy created poverty in the reserves/Bantustans. A century of racial discrimination in human capital, through colour bars, Bantu Education and barriers to black farmers and entrepreneurs, embedded inequality that will take generations to undo. The argument is narrative and synthetic, drawing on secondary literature, a few statistics and illustrative case studies (Healdtown/St Andrew's, the Bokwe-Matthews family).

Evidence and method

Mostly secondary literature (Penn, Turrell, Bundy, Plaatje, Trollope, Carnegie Commission, Wilson's own earlier work); statistics from the SALDRU PSLSD 1993 baseline survey, Pillay on education spending, a Commonwealth Expert Group report and OECD income distribution work; a household income decile diagram from Wilson's Dinosaurs, Diamonds & Democracy; anecdotes and family history.

Place in the debates

• Cheap labour thesis / revisionist view of migrant labour and racial capitalism (see Cheap labour thesis and the function of the reserves). Sides with the revisionist view: the migrant labour system, developed on the diamond and gold mines and later the model for apartheid urban labour policy, kept black wages static in real terms and generated poverty in the reserves as it generated wealth in the cities.
• Causes of rural poverty in the Bantustans (see Homelands, separate development, decentralisation and removals). Argues that reserve poverty was generated by the dynamics of white-controlled industrialisation (labour drain, declining rural investment, forced removals), not pre-existing or incidental.
• Legacy of apartheid and persistence of inequality after 1994 (see Long-run legacy and persistence of apartheid). Holds that the democratic government has done little to shift inherited poverty, unemployment and inequality, and that human capital deficits (Bantu Education) make the racial gap slow to close.

Findings

• Income is highly concentrated: the top decile earns more than the rest combined. (p. 4)

  the richest 10% --with income over R400 000 per household---alone earns more than the other 90% combined.

• The compound/migrant system from Kimberley and the gold mines became the template for apartheid's organisation of urban African labour; about half of urban black workers were oscillating migrants by 1972. (p. 8)

  By 1972 it was estimated that one out of every two black workers in the urban areas was an oscillating migrant, housed in town on a temporary basis.

• Migrant labour and the pass laws blocked normal urbanisation and kept women out of town, depriving black families of household income sources and urban infrastructure. (p. 8)

  by making it extremely difficult---and in the case of the mines impossible---for black women to follow their men to town the system ensured that a major source of income

• Wealth creation and poverty creation were simultaneous products of the migrant system; Bantustan poverty is historically produced. (p. 9)

  the very processes that were generating wealth in the country were simultaneously generating poverty.

• The mine wage gap widened because foreign recruitment held black wages static in real terms until the 1970s. (p. 9)

  the widening gap between average black and white earnings on the mines from approximately 12:1 in 1936 to 21:1 thirty five years later

• State education spending on African pupils was a tiny fraction of that on whites in 1976 and still under a fifth in 1986/87. (p. 11)

  the per capita expenditure by the state on African pupils was a mere 7,5% of the expenditure on pupils in the all-white schools.

• Racial gaps in schooling were stark on the eve of democracy: only 11% of blacks aged 18+ had 12 years of school versus 61% of whites. (p. 12)

  Whilst nearly two-thirds (61%) of whites had completed school only one in nine (11%) of blacks had done so.

• Black people were severely underrepresented in skilled professions in 1989. (p. 12)

  In 1989 1% of architects; less than

• Colour bars originated on the mines and were consolidated in the Mines and Works Act of 1911, then spread via job reservation, civilised labour policy and pass laws. (p. 13)

  The first legal colour bar, protecting white workers, was enacted before the outbreak of the South African war

• State support for white farmers alongside obstruction of black farmers amounted to a century of discriminatory action, producing the commercial/reserve agricultural divide. (p. 14)

  It is the visible consequence of a century of deliberately discriminatory action against one group and in favour of another.

• Unemployment remains very high under narrow and expanded definitions, and the post-1994 government has made little headway. (p. 4)

  anything between 25% [narrow definition] and 41% [inclusive definition] of those wanting work are unable to find it.

Evidence ledger

Sources used

Source	Kind	Period	How used
Leibbrandt, Woollard, Finn & Argent, Trends in South African Income Distribution and Poverty Since the Fall of Apartheid 1993–2008 (OECD, 2010)	secondary literature compilation	1993-2008	Supports the claim that 70% of the population earns only 17% of total income.
Wilson, Dinosaurs, Diamonds & Democracy (2009), p. 117 – household income by decile diagram	secondary literature compilation	c. 2000s	Diagram of average household income per decile illustrates the extreme income concentration in the top decile.
Statistics SA official and expanded unemployment definitions (Lehohla, Business Report, 14 Oct 2004)	official statistical series	2004	Source for the definitions behind unemployment rates of 25% (narrow) to 41% (expanded).
Project for Statistics on Living Standards & Development (PSLSD), South Africans Rich & Poor: Baseline Household Statistics, SALDRU, 1993/94, Tables 3.5-3.7	household or labour survey microdata	1993	Provides educational attainment by race (years of schooling) to show embedded racial inequality in human capital in 1993.
Beyond Apartheid: Human Resources in a New South Africa, Report of a Commonwealth Expert Group (1991), table 5.1	commission or official report	1989-1991	Source for black shares of professions (architects, engineers, computer programmers) in 1989 and the maths-denial under Bantu education.
P.N. Pillay, 'The development and underdevelopment of education in South Africa' (1990)	secondary literature compilation	1976-1987	Provides per capita state expenditure ratios for African versus white pupils in 1976 and 1986/87.
Carnegie Commission, The Poor White Problem in South Africa (1932)	commission or official report	1932	Cited for the importance of factory jobs for working-class white women in supporting families, contrasted with pass-law restrictions on black women.
Transvaal Indigency Commission 1907 (evidence of Hofmeyr)	commission or official report	1907	Cited for Hofmeyr's view that racism's roots lay in slavery.
Mines and Works Act of 1911; Land Act of 1913; Group Areas Act; pass laws	legislation gazette	1911-1960s	Legislation invoked as instruments of the colour bar, land dispossession and barriers to black entrepreneurs.
Anthony Trollope, South Africa (1878), travel account	other	1877	Contemporary observer evidence that racism pre-dated mining competition for jobs.
Sol Plaatje, Native Life in South Africa	other	1913-14	Contemporary documentation of the Land Act's effects on black farmers.
Francis Wilson, Labour in the South African Gold Mines 1911-1969 (1972) and International Migration in Southern Africa (SALDRU WP 1, 1976)	secondary literature compilation	1936-1971	Implicit source for mine wage gap rising from 12:1 to 21:1 between white and black earnings and for foreign recruitment holding black wages static.
Frieda Bokwe Matthews, Remembrances (1995) and Lovedale Missionary Institution (1884)	other	1824-1990s	Family memoir used for the Bokwe-Matthews-Pandor case of intergenerational human capital accumulation.
Newspaper reports (Cape Times 9 Feb 2011 on Essa Moossa; Argus on Masibambane High School)	newspapers periodicals	2011	Anecdotal illustrations of Group Areas Act effects and current school achievement.

Inference

Quantification: Illustrative figures.

Narrative historical synthesis, an introductory essay drawing on secondary literature and a few published statistics, with two illustrative case studies (Healdtown vs St Andrew's; the Bokwe-Matthews family) to argue that land, mineral wealth, migrant labour, colour bars and human-capital discrimination generated racial inequality.

Key claims

Claim	Basis	Strength
Income inequality is extreme: the richest 10% of households earn more than the other 90% combined, and 70% of the population earns 17% of income. (p. 3)	Wilson's decile diagram and OECD (Leibbrandt et al.) income distribution study.	Partial data
The migrant labour system, developed in diamond and gold mining, generated wealth in cities while generating rural poverty in the reserves/Bantustans. (p. 8)	Relies on earlier work (Turrell, Van der Horst, Wilson 1972, Wilson & Ramphele 1989); no data presented here.	Asserted
The black–white earnings gap on the mines widened from 12:1 in 1936 to 21:1 because of recruitment of foreign labour keeping black real wages static. (p. 8)	Cites the author's earlier work on mine labour and migration; figures stated without table.	Partial data
State education spending per black pupil was a small fraction of that per white pupil (7.5% in 1976; 19% in 1986/87). (p. 10)	Pillay (1990) secondary compilation of state expenditure data.	Partial data
By 1993 human capital was deeply racially unequal: 61% of whites versus 11% of blacks had completed 12 years of schooling. (p. 11)	PSLSD 1993 baseline household survey tables 3.5-3.7.	Systematic data
Black South Africans were severely under-represented in skilled professions in 1989 (1% of architects, under 2% of engineers, 13% of computer programmers). (p. 12)	Commonwealth Expert Group report, Beyond Apartheid (1991), table 5.1.	Partial data
Racially biased accumulation of mineral wealth resulted from land ownership, capital access and state control by whites. (p. 6)	Historical narrative from secondary literature and legal arrangements of the Transvaal republic; no primary data.	Asserted
Legal and social colour bars, land laws and discriminatory state support deliberately blocked black skills acquisition and entrepreneurship, including farming. (p. 13)	Legislation (Mines and Works Act 1911, Land Act 1913, Group Areas Act), Plaatje, Bundy, Roux, Richardson; illustrative anecdote.	Illustrative or anecdotal

The author's caveats

Notes that a single issue cannot cover all aspects of inequality and that the essay offers examples of drilling down; acknowledges that per-pupil expenditure need not exactly measure quality differences and that matters were more nuanced than the synopsis suggests.

Gaps in the evidence

The argument links historical mechanisms to present inequality without time-series data on wages, land holdings, education spending or income by race across the 20th century. Census data, Chamber of Mines wage series, and departmental education expenditure series existed and could test the causal chain. Several figures (e.g. 1972 migrant share, schooling in the OFS) lack citations, and some footnotes are incomplete (page placeholders).

Replicability: Data poorly documented.