The Economics of Apartheid: An Introduction¶
Martine Mariotti; Johan Fourie (2014). The Economics of Apartheid: An Introduction. Economic History of Developing Regions, Vol. 29(2). Economic History Society of Southern Africa.
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | synthesis review |
| Period studied | 1911–2014 |
| Regions | South Africa, Bantustans/homelands, Malawi, Mozambique |
| Themes | methods data sources, business and state, labour markets wages, bantustans removals influx, sanctions divestment, mining migrant labour, inequality poverty unemployment, politics ideology elections, comparative theory |
| Citation key | mariotti2014 |
Argument¶
The introduction to a special issue surveys the literature on the economics of apartheid. It argues that the ideological Marxist-liberal debate has given way to quantitative, econometric and public-choice analysis, and that apartheid's legacy and the new digitised data make the field timely. It reviews the main syntheses (Lipton, Moll, Feinstein, Nattrass and Seekings) and new work by young scholars using the homelands and the 1974 mine-labour shock as natural experiments. It then summarises the issue's papers, which stress the pragmatic, fluid character of apartheid economic ideology and policy.
Evidence and method¶
Secondary literature review; summaries of working papers and special-issue contributions. It cites Feinstein (2005), Nattrass and Seekings (2011), Moll (1991) and Kaempfer and Lowenberg (1988), as well as in-progress research using censuses, Malawian census data, ward-level election data and archival sources. It also comments on the lack of apartheid-era data.
Place in the debates¶
- Marxist vs liberal interpretations of capitalism and apartheid (who benefited; cheap labour thesis) (see Liberal vs revisionist: was apartheid functional or dysfunctional for capitalist growth?). 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).
- Did apartheid growth derive from cheap labour, and why did the system fail? (see Did the apartheid economy fail? Growth, productivity, investment and the post-1970s crisis). They endorse Moll's finding that early rapid growth reflected the post-war global rebound rather than cheap labour, and that growth was slow by comparative standards. They also endorse the view that apartheid collapsed under its own weight (Feinstein): gold profitability, industrial inefficiency and external pressures.
- Sanctions effectiveness (see Sanctions, disinvestment and the anti-apartheid movement). They pose the question of whether sanctions force regime change or strengthen the oppressor. They report Kaempfer and Lowenberg's conclusion that the domestic costs of running apartheid, rather than sanctions, made the system untenable.
- Who benefited from job reservation (see The job colour bar, white labour and its erosion). Reporting Lundahl, they say that white unskilled labour benefited most and capitalists were constrained, contrary to the Marxist claim.
Findings¶
-
Lack of apartheid-era data has deterred economists; the censuses excluded many blacks and inclusive surveys came only in the 1990s. (p. 5)
the decadal censuses of 1970 and 1980 excluded a large proportion of blacks, and even where they were enumerated, more limited information was obtained than for coloureds, Indians and whites.
-
Moll argues that early apartheid growth reflected the world post-war rebound rather than cheap labour, and was slow by comparison. (p. 4)
the rapid growth of the South African economy during the early years of apartheid was more likely to have been due to the worldwide rebound after World War II than to cheap labour
-
Kaempfer and Lowenberg's public choice model attributes the system's demise to domestic enforcement costs more than sanctions. (p. 4)
it was the costs of running apartheid at home, particularly as resistance was increasing in the 1980s, that ultimately made the system untenable
-
Feinstein attributes the collapse to three causes that prompted De Klerk's 1990 reforms. (p. 4)
the gold mines’ loss of profitability, industry’s low efficiency and high production costs, and adverse external economic and political changes
-
Nattrass and Seekings find mechanization replaced much unskilled labour with less semiskilled labour, and within-group inequality rose while the overall Gini was stable. (p. 5)
while South Africa’s Gini coefficient remained roughly the same between 1970 and 1993, within-group inequality increased significantly.
-
Job reservation induced capital-intensive manufacturing with weak unskilled job creation. (p. 2)
The shortage of semi-skilled workers caused by the apartheid regime’s statutory job reservation policy obliged manufacturers to overinvest in capital technology
-
Wilse-Samson links the loss of foreign mine labour to mine and farm mechanization and political change. (p. 7)
the reduced supply of foreign mine labour increased local demand for black workers but the slow local response also increased mine and farm mechanization.
-
Mariotti finds white upward mobility eased job reservation, and that employers treated black and white workers as substitutes. (p. 7)
as white workers shifted into increasingly skilled jobs, the pressure to reserve semi-skilled jobs decreased and blacks began to move into those jobs, even before the formal end of statutory job reservation.
-
Nattrass (in this issue) rejects the claim that cheap unskilled labour drove rapid apartheid growth. (p. 9)
the data simply do not support the notion that cheap unskilled labour was the cause of rapid economic growth during apartheid.
-
Havemann finds that the multiple exchange rate system failed to stop capital flight or attract investment. (p. 9)
the regime was unable either to halt politically inspired capital flight during the 1980s or to encourage foreign investment.
-
Van der Berg shows that the 1970s shift in social spending towards blacks was well targeted. (p. 9)
the initial spending shifts were so well channelled that post-apartheid social spending is now extremely well targeted.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Presidential State of the Nation addresses, 1994-2014 (South Africa Government Online) | official statistical series | 1994-2014 | Counts how often 'apartheid' appears in the addresses (20 of 25) and quotes Mbeki and Zuma to show apartheid's continuing salience. |
| Decennial population censuses of 1970 and 1980 | population census | 1970, 1980 | Described as a data limitation: the censuses excluded many blacks, and collected less information on those they did enumerate. |
| PSLSD 1993 household survey (Project for Statistics on Living Standards and Development) | household or labour survey microdata | 1993 | Identified as the first inclusive household survey, which marks the start of comprehensive microdata. |
| 1996 population census | population census | 1996 | Described as the first inclusive census; Bastos & Bottan use it to document income gaps inside and outside the homelands. |
| Official apartheid-era publications used by Nattrass & Seekings (2011) | secondary literature compilation | 1970-1993 | Summarized for findings on mechanization, Gini coefficients and within-group inequality. |
| Apartheid-era primary archival sources on homeland industrial zones (Kerby 2014) | state archives | c. 1960s-2000s (over 40 years) | Used by Kerby to trace the development and failure of industrial incentive zones. |
| Ward-level election returns, 2000-2011 (De Kadt & Larreguy 2014) | election returns | 2000-2011 (five elections) | Used in a difference-in-differences estimate that traditional leaders deliver about 6.5 points of ANC vote share. |
| Malawian census data (Dinkelman & Mariotti 2014) | population census | Post-1967 and post-1974 cohorts | Used to estimate the long-run schooling effects of exposure to labour migration to South African mines. |
| Survey and census data on drought exposure and disability (Dinkelman 2013) | secondary literature compilation | Not specified in text | Cited for the finding that early childhood drought exposure raises later disability by 3.7-5.8%, with double the effect in the homelands. |
| Wiehahn Commission recommendation on job reservation | commission or official report | late 1970s | Used in the Mariotti and Van Zyl-Hermann paper, as summarized, on the decline and abolition of statutory job reservation. |
Inference¶
Quantification: None.
A survey and review essay. It describes the literature and introduces the special issue's papers by summarizing other authors' methods and findings (natural experiments, difference-in-differences, models, historical analysis). It runs no analysis of its own.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Apartheid-era data are scarce, mainly because the homelands were excluded from the censuses and macroeconomic data, and this delayed economists' work on the period. (p. 117) | Description of the 1970 and 1980 censuses and the first inclusive survey (1993) and census (1996); no systematic audit of the data. | Partial data |
| South African growth in the early apartheid years owed more to the post-war worldwide rebound than to cheap labour, and was slow by international comparison. (p. 116) | Relies on Moll (1991), Feinstein (2005) and Nattrass & Seekings (2011); no new data. | Partial data |
| Within-group inequality rose sharply after 1970 even though the overall Gini stayed roughly constant between 1970 and 1993. (p. 117) | Nattrass & Seekings (2011), based on official publications. | Partial data |
| Statutory job reservation pushed manufacturers toward capital-intensive technology, which reduced unskilled job creation. (p. 114) | Asserted in the introduction with no data or source cited. | Asserted |
| Labour migration from Malawi to the South African mines raised schooling among those left behind. (p. 119) | Dinkelman & Mariotti (2014), using Malawian census data and two historical shocks. | Systematic data |
| Traditional leaders gave the ANC an electoral advantage of about 6.5 percentage points in wards shaped by the Bantustan borders. (p. 119) | De Kadt & Larreguy (2014), difference-in-differences on ward-level election returns, 2000-2011. | Systematic data |
| The data do not support the view that cheap unskilled labour caused rapid growth under apartheid. (p. 122) | Nattrass's paper in the special issue on declining capital productivity, summarized without detail. | Partial data |
The author's caveats¶
The authors stress that apartheid-era data are scarce: the censuses excluded the homelands, little macro data were published on the homelands, and black wages are poorly known. They add that archives hold large official collections still being digitized.
Gaps in the evidence¶
The introduction summarizes others' findings but supplies no data of its own and gives no assessment of how reliable the underlying sources are. Testing its claims, such as the capital-intensity effect of job reservation or the sources of growth, would need firm- or industry-level manufacturing census series, wage data by race and homeland, and archival records on job reservation. Much of this exists in South African archives and government publications and is only now being digitized.
Replicability: Not applicable.