Economic Sanctions Against South Africa¶
Charles M. Becker (1987). Economic Sanctions Against South Africa. World Politics, Vol. 39, No. 2. The Johns Hopkins University Press.
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | theoretical or conceptual |
| Period studied | 1960–1986 |
| Regions | South Africa, Botswana, Lesotho, Swaziland (BLS), Southern Africa, Zimbabwe, Malawi |
| Themes | sanctions divestment, macro fiscal finance debt, labour markets wages, inequality poverty unemployment, mining migrant labour, comparative theory, politics ideology elections |
| Citation key | becker1987 |
Argument¶
Becker disputes the widely held belief that effective sanctions would necessarily devastate poor black South Africans. Using a dualistic (Lewis-Hansen) model of the economy and a general-equilibrium framework adapted from his India model, he argues that sanctions on high-technology exports and on gold and diamond imports would shift demand towards labour-intensive sectors (agriculture, import substitutes) via devaluation. This would limit losses for unskilled nonwhites while hurting whites, who own the capital and mining rents. An oil embargo, by contrast, would be severe for blacks and neighbours. Divestment would yield windfall gains for white buyers, so a ban on new investment is preferable, and aid to South Africa's neighbours is needed. The analysis is a qualitative, comparative-statics exercise and explicitly supplies only surmises about the direction of net effects.
Evidence and method¶
No new empirical data set. It draws on secondary sources (chiefly Lipton 1985, Cooper 1984, Lundahl, Hansen's model, Christiansen and Kydd on Malawi, Barclays Business Briefs), on the author's own computable general equilibrium model of India, and on congressional debate statements. Figures on migrant labour, trade dependence, foreign debt and nonwhite occupational shares come from these secondary sources.
Place in the debates¶
- Sanctions effectiveness and who bears the costs (would sanctions hurt the black poor most?) (see Sanctions, disinvestment and the anti-apartheid movement). Argues against the claim that all effective sanctions would greatly hurt poor blacks; carefully designed sanctions (high-tech export ban, gold/diamond boycott) could spare unskilled nonwhites in the medium run, though an oil embargo would not.
- Divestment versus ban on new investment (see Sanctions, disinvestment and the anti-apartheid movement). Critical of forced divestment, which would give white South Africans windfall gains and could worsen conditions for black workers; prefers a ban on new investment.
- Evolutionist versus sceptic views on whether growth and a nonwhite middle class will erode apartheid (see Liberal vs revisionist: was apartheid functional or dysfunctional for capitalist growth?). 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.
- Employment codes (Sullivan Principles) by multinationals (see Sanctions, disinvestment and the anti-apartheid movement). Sees clear gains for skilled nonwhites and the expanding middle class but possible unskilled unemployment; supports them as part of a package.
Findings¶
-
The paper's central thesis is that effective sanctions need not greatly harm unskilled blacks if they induce expansion of labour-intensive sectors. (p. 2)
it is likely that bans on exports of high technology to South Africa and imports of South African gold and diamonds would cause labor-intensive sectors to expand
-
The South African economy is characterised as a dual economy with a white-owned modern sector and a traditional homeland sector that sanctions would affect only indirectly. (p. 5)
Consequently, sanctions other than an oil embargo cannot affect the traditional economy directly.
-
A gold and diamond boycott would cause a crisis that is not an equilibrium: devaluation would stimulate agriculture and manufacturing and stabilise labour demand. (p. 13)
The trade imbalance would force a devaluation of the rand, making exports more attractive in other sectors, particularly modern agriculture and manufacturing.
-
Whether unskilled blacks gain depends on relative labour intensity and speed of expansion, and the author concedes it is an empirical question not yet studied. (p. 13)
Unskilled nonwhite workers will gain if the increase in labor demand from the expanding sectors exceeds the decreases in the declining industries.
-
A high-tech export ban might produce medium-run gains for unskilled nonwhites by making production more unskilled-labour-intensive. (p. 16)
an embargo on exports of key intermediate goods to South Africa might generate real income gains for the unskilled non-white population in the medium run
-
An oil embargo is the one measure likely to severely hurt unskilled nonwhites and neighbouring states. (p. 17)
An effective oil embargo would be far more likely to reduce real incomes of unskilled nonwhite South Africans than O.E.C.D. export restrictions on high-tech goods
-
South Africa's foreign debt rose sharply during stagnation, reflecting gold-price-driven balance-of-payments strains met by borrowing. (p. 18)
South Africa's foreign debt rose from $18.0 billion in mid-1981 to $23.7 billion in late 1985—a period of stagnant GNP.
-
Forced divestment would mainly benefit white South African buyers and could harm black employees where Sullivan-code firms sell out. (p. 19)
Forced sales of the multinationals' South African subsidiaries would attract only a small number of (largely white South African) buyers, who would then be in a position to reap windfall gains.
-
Sanctions campaigns may work through expectations of rand depreciation and capital flight even without changing actual investment decisions. (p. 20)
If holders of rand-denominated savings merely anticipate that such campaigns will have some success, they may also anticipate a devaluation of the rand
-
South Africa's dependence on foreign mine labour fell sharply between 1973 and 1982. (p. 24)
This share peaked at 79% in 1973 (297,000 of 379,000), but fell to 42% by 1982 (190,000 of 448,000).
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Merle Lipton, Capitalism and Apartheid (1985) – tabulated data on homeland population, mine labour force and nonwhite occupational shares | secondary literature compilation | 1960-1982 | Supplies the homeland population shares, foreign share of mine labour, nonwhite occupational growth and income-share figures that the argument relies on. |
| J. H. Cooper, 'South Africa and the Threat of Economic Sanctions', South African Journal of Economics 52 (1984) – trade, dependency and petroleum data | secondary literature compilation | 1976-1983 | Source for the shares of South African manufactured exports sold to Africa, SACU trade and revenue linkages and petroleum dependence of neighbours. |
| Colclough and McCarthy, The Political Economy of Botswana (1980); Cobbe, 'The Changing Nature of Dependence: Economic Problems in Lesotho' (1983) | secondary literature compilation | mid-1970s to 1983 | Provides migrant labour and remittance magnitudes showing the dependence of Lesotho and Botswana on South African employment. |
| Christiansen and Kydd, 'The Return of Malawian Labour from South Africa and Zimbabwe' (1983) and internal World Bank documents | secondary literature compilation | 1966-1977 | Evidence that Malawi absorbed returning miners while smallholder output per head rose, used as a precedent for repatriation after sanctions. |
| Barclays Business Briefs, 'Adjustment and Recovery: 1981-1987' (Johannesburg, January 1986) | newspapers periodicals | 1981-1986 | Used for South Africa's foreign debt growth and the weak manufacturing export response to the 1983 exchange decontrol. |
| Congressional Digest, October 1985 (President Reagan's Executive Order; congressional debate on the Anti-Apartheid Act of 1985) | parliamentary debates | 1985 | Documents the political positions (Reagan, Kemp, Wolpe) that sanctions would hurt poor blacks. |
| Becker, Mills and Williamson, 'Modeling Indian Migration and City Growth, 1960-2000' (1986) and Kelley and Williamson (1984) – computable general equilibrium simulation model | secondary literature compilation | 1960-2000 (simulation) | The Indian CGE model is adapted as the analytical basis for the qualitative simulations of sanctions' sectoral and wage effects in South Africa. |
| Fallon and Layard (1975) cross-country multi-skill production function estimates; Lluch, Powell and Williams (1977) household demand patterns | secondary literature compilation | 1970s | Empirical grounding for the claims that skilled labour is complementary to capital and that the poor spend little on import-intensive goods. |
Inference¶
Quantification: Illustrative figures.
Theoretical, comparative-statics reasoning (Hansen-Lewis dualism, Heckscher-Ohlin/Stolper-Samuelson) informed by an adapted CGE model of India. Each sanction is traced qualitatively through sectors and factors, supported by scattered statistics from secondary works; no estimation or formal simulation of South Africa is reported.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Effective sanctions need not greatly hurt poor blacks: sanctions on gold, diamonds and high technology could expand labour-intensive sectors and limit the loss to unskilled nonwhites. (p. 148) | Theoretical reasoning (Stolper-Samuelson, Hansen dualism) and an adapted Indian CGE model; no South African estimates. | Asserted |
| Whether expanding sectors offset contracting ones in labour demand is an empirical question that has not been studied, so the results are guesses. (p. 159) | Author's own acknowledgment; no studies cited. | Asserted |
| An oil embargo would hurt unskilled nonwhites and neighbouring states severely, because petroleum is an intermediate used in virtually all sectors and South Africa supplies the BLS states' oil. (p. 162) | Structural argument plus Cooper's data on neighbours' petroleum dependence. | Partial data |
| South Africa's neighbours are tightly dependent on South African trade, labour migration and revenue, and so are more vulnerable to sanctions than the Republic. (p. 151) | Cooper, Cobbe, Colclough and McCarthy and Lipton figures on migrants, remittances and trade shares. | Partial data |
| Forced divestment would yield windfall gains to white South African buyers, whereas a ban on new investment would not. (p. 164) | Theoretical reasoning on asset-sale demand; no data on divestment sale prices. | Asserted |
| South Africa's reliance on foreign mine labour fell sharply, with the foreign share peaking at 79% in 1973 and falling to 42% by 1982, which reduces neighbours' exposure to hostage-taking. (p. 169) | Lipton's compiled mine labour figures. | Partial data |
The author's caveats¶
The author states that only empirical work can produce precise estimates and that the paper provides only surmises of the direction of net effects. He also notes that no empirical studies exist on the sectoral labour-demand question.
Gaps in the evidence¶
The argument needs South African input-output tables, sectoral employment and labour-intensity data, and import-content and consumption-basket data for unskilled households to test whether expanding sectors would absorb displaced mine workers. An estimated CGE or econometric model calibrated to South African data, rather than an adapted Indian model, would also be needed. Such data (Census of Manufacturing, Agricultural Census, national accounts, household expenditure surveys) existed in the 1980s, and later experience of sanctions episodes now allows ex-post testing.
Replicability: Not applicable.