Did the Apartheid Economy 'Fail'?

Literature › Works

Terence Moll (1991). Did the Apartheid Economy 'Fail'?. Journal of Southern African Studies, Vol. 17, No. 2. Taylor & Francis.

Type	journal article
Scope	Economics and economic history of apartheid
Method	descriptive quantitative
Period studied	1920–1985
Regions	South Africa
Themes	manufacturing industrial policy, macro fiscal finance debt, comparative theory, methods data sources, business and state, labour markets wages, transition post apartheid
Citation key	moll1991a

Argument

Moll challenges the widespread claim, made on both the right and the left, that the apartheid economy grew 'exceptionally rapidly' until the 1970s. He tests this against four criteria: growth after 1948 relative to before, growth relative to comparable middle-income developing countries, export performance (especially manufactured exports), and total factor productivity growth. South Africa's record is mediocre on output growth and poor on manufactured exports and TFP, even though its resource endowment and the post-war world boom were highly favourable. He concludes that the economy 'failed' to reach its potential, partly because of the apartheid superstructure and partly because of short-sighted state industrial, trade and exchange-rate policy. This leads him to side with the liberal critique over the radical 'cheap labour' view, while denying that a free-market alternative follows. He argues instead for better-quality, directed state intervention after apartheid.

Evidence and method

Published national accounts (South African Statistics, Du Plessis/Du Piesanie series for 1920-48), with real GDP recalculated using gold-bearing ore milled in place of ounces of gold. World Bank World Tables and Summers-Heston ICP data for a 20-country developing comparison group and a 10-country developed group. UNCTAD trade statistics for manufactured export shares. TFP figures for South Africa taken from Moll's 1990 Cambridge PhD thesis and compared with Maddison's TFP estimates for Korea, Taiwan, Japan and four Latin American countries. Secondary literature on the capital goods sector (Kaplan), import-substitution policy, and Reynders Commission evidence.

Place in the debates

• Did the apartheid economy grow exceptionally fast (the 'apartheid boom')? (see Did the apartheid economy fail? Growth, productivity, investment and the post-1970s crisis). Rejects it. The post-1948 growth rate was not higher than before 1948, and South Africa's rank among comparable developing countries was consistently mediocre.
• Liberal vs radical/revisionist views on apartheid and capitalism (the 'cheap labour' thesis) (see Liberal vs revisionist: was apartheid functional or dysfunctional for capitalist growth?). 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.
• Export orientation vs import-substitution industrialisation as growth strategy (see Industrialisation, protection and industrial structure). Faults inefficient import-substituting protection and the neglect of labour-intensive manufactured exports. He also argues that East Asian-style directed state intervention, not free markets, is the lesson.
• Causes of the slow growth: external vs domestic supply constraints (sanctions, demand, income inequality) (see Did the apartheid economy fail? Growth, productivity, investment and the post-1970s crisis). Attributes the poor export performance to domestic supply-side and policy factors rather than external demand. Sanctions only mattered from the late 1970s, and the limited home market was mainly a consequence of poor export policy.

Findings

• Growth after 1948 was no faster than before, and relative to the world boom it was weaker. (p. 7)

  there is certainly no sign of an economic surge after 1948

• Within a group of comparable middle-income developing countries, South Africa's growth was consistently mediocre. (p. 10)

  It can thus be concluded that South Africa's overall growth-record is consistently mediocre within this group of countries.

• Even the 1960s boom looks weak next to the fast-growing newly-industrialising countries. (p. 10)

  South Africa's 5-6 per cent per annum growth rate during the 1960s cannot compare, for example, to the average of 7.6 per cent per annum achieved by four large

• South Africa's share of world manufactured exports more than halved between 1955 and 1985. (p. 12)

  South Africa's share of world manufactured exports fell by more than half between 1955 and 1985, from 0.8 per cent to 0.3 per cent.

• Policy favoured inefficient import substitution rather than exports, with apartheid-specific features. (p. 14)

  economic policy in South Africa was directed instead towards an inefficient import-substitution industrial model in the 1950s and 1960s, with local touches due to the apartheid system.

• A foreign-exchange constraint, rooted in weak exports, repeatedly cut booms short. (p. 14)

  Many studies conclude that a continual constraint on economic growth in South Africa in the 1950s and 1960s was a shortage of foreign exchange.

• Political-economic interests may have deterred an export strategy, which would have accelerated black proletarianisation and urbanisation. (p. 16)

  It is not impossible to suspect that such a spectre discouraged active industrial exports policies in South Africa.

• Total factor productivity growth in 1950-73 was the lowest among the comparison countries, close to zero. (p. 17)

  In the earlier period, South Africa's near-zero productivity growth is the lowest shown

• Counterfactual: Brazil-level TFP growth would have made the 1973 economy about 50 per cent larger. (p. 19)

  its GDP growth-rate would have been 1.9 per cent per annum higher, and by 1973 the economy would have been 50 per cent larger.

• Moll reinterprets the 1980s 'crisis' as part of a long slump masked by South Africa's resource wealth. (p. 21)

  the South African economy has suffered from a prolonged economic slump since the 1940s

• Aggregate evidence supports the liberal critique and qualifies the radical cheap-labour argument. (p. 21)

  The aggregate evidence discussed above appears to support the liberal critique of apartheid

• A free-market alternative is not implied; the quality of state intervention is what matters. (p. 22)

  In other words, it is not simply the fact of state intervention in South Africa, but its quality and intentions, that should be questioned.

Evidence ledger

Sources used

Source	Kind	Period	How used
Du Plessis (Du Piesanie), 'Die Bepaling en die Gebruik van Kapitaalopbrengsverhoudings' (M.A. thesis, Pretoria, 1968), chapter 4 – real GDP 1920-48 at 1958 prices	secondary literature compilation	1920-1948	Supplies the pre-1948 real GDP series used in Table 1 to compare growth before and after 1948.
South African Statistics 1976 (Pretoria, 1976), p. 21.7 – real GDP 1948-69 at 1963 prices	official statistical series	1948-1969	Official national accounts GDP series for 1948-69 used for conventional growth rates in Table 1.
South African Statistics 1982 (Pretoria, 1982), p. 21.7 – real GDP 1969-74 at 1975 prices	official statistical series	1969-1974	Extends the conventional real GDP series to 1974 for Table 1.
Chamber of Mines of South Africa, 1988 Statistical Tables (Johannesburg, 1989), p. 12S – gold-bearing ore milled	chamber of mines or recruiting records	1920-1974 (and later)	Tons of ore milled replace ounces of gold as the gold-sector output indicator in the 'recalculated' GDP.
Moll, 'Output and Productivity Trends in South Africa: Apartheid and Economic Growth' (Ph.D. thesis, Cambridge, 1990), chapters 3-6	author survey or fieldwork	1920-1985	Source of the recalculated GDP series, TFP estimates, elasticity estimates, sectoral employment shares and methodological detail, many available only on request.
Summers and Heston, 'Improved international comparisons of real product and its composition: 1950-1980', Review of Income and Wealth (1984) – UN International Comparison Project data	international trade or finance data	1960 base year	Population and PPP-based real per capita GDP in 1960 used to select the 20 comparison middle-income developing countries.
World Bank, World Tables 1976, 1983 and 1987 (Economic Data Sheet 1; country tables) – real GDP growth	international trade or finance data	1950-1985	Source of national-accounts-based GDP growth rates used to rank South Africa within the comparison group (Table 2).
World Bank, World Development Report 1980 (Washington, 1980), p. 111	international trade or finance data	1980	Justifies excluding Iran and Iraq from the comparison sample as capital-surplus oil exporters.
South African Statistics (Pretoria), most recent issues – used to recalculate South African growth for Table 2	official statistical series	1950-1985	Provides the numbers used to derive the gold-recalculated South African growth rates by adjusting World Bank figures.
Chung-In Moon, 'The future of the Newly Industrialising Countries', Table 7.2 (NIC growth rates)	secondary literature compilation	1960s-1970s	Supplies the NIC average growth rates (7.6% in 1960s, ~8% in 1970s) compared to South Africa.
UNCTAD, Handbook of International Trade and Development Statistics 1976, 1983, 1988, Tables 4.1, A6, A9, A10	international trade or finance data	1955-1985	Source of manufactured export values used to compute South Africa's share of world and developing-country manufactured exports (Table 3).
Report of the Commission of Enquiry into the Export Trade of the Republic of South Africa (Reynders Commission, RP69/72)	commission or official report	early 1970s	Quoted to show that foreign exchange shortage constrained growth in the 1950s-60s and that more exports were needed.
Social and Economic Planning Council, Economic Aspects of the Gold Mining Industry (Report II, UG32/48), p. 51	commission or official report	1948	Cited as evidence that devaluation was felt to be inflationary and ineffective, explaining policy reluctance.
Maddison, The World Economy in the 20th Century (Paris, 1989), pp. 81, 91 – TFP, GDP, labour, capital for comparator countries	secondary literature compilation	1950-1984	Provides comparator countries' growth accounting data in Table 4 against which South Africa's TFP growth is compared.
Population Censuses (South Africa) – educational stock figures, 1951 Population Census	population census	1951, 1960-85	Educational stock numbers modified from Population Censuses used as a robustness check on the labour quality adjustment in the TFP calculations.
Kaplan, 'The South African capital goods sector' (1989) and 'The limited development of the South African machine tool industry' (1987)	secondary literature compilation	1980s	Relied on for the finding that capital goods exports were poor and technological capabilities low and declining.
Moll, sectoral employment shares of white and black workers 1950 and 1970 (calculated from figures described in thesis pp. 196-199)	secondary literature compilation	1950-1970	Establishes that most white workers were in services versus fewer black workers, supporting the political-economy argument about opposition to devaluation.

New data

Author-constructed recalculated GDP series (gold output measured by ore milled) for 1920-1985, and South African TFP/growth-accounting estimates for 1950-84, summarised from his Cambridge PhD thesis; the underlying series are not published in the article.

Inference

Quantification: Descriptive tables.

Four-test comparative growth assessment: comparison of South African GDP growth before and after 1948 (cycle-peak periods), rank comparison against a selected group of 20 middle-income developing countries, export-share trends in manufactured exports, and comparative TFP growth accounting against selected countries; supported by qualitative argument about policy and political economy. No regression or causal identification.

Key claims

Claim	Basis	Strength
Real GDP growth after 1948 was not materially faster than before 1948; on the gold-recalculated measure there is virtually no difference, despite a much more favourable world economy. (p. 275)	Table 1 using Du Plessis, South African Statistics and Chamber of Mines ore-milled series; compared with world boom context.	Partial data
South Africa's output growth ranks mediocre among 20 comparable middle-income developing countries (tie 13th, 1950-85). (p. 279)	Table 2 from World Bank World Tables growth rates, country selection from Summers-Heston ICP data, SA recalculated with Chamber of Mines ore data.	Systematic data
South Africa's share of world and developing-country manufactured exports fell steadily from 1955 to 1985, indicating exports were not used to stimulate growth. (p. 281)	Table 3 computed from UNCTAD Handbook data.	Systematic data
South Africa's TFP growth in 1950-73 was near zero, the lowest among comparator developing countries, despite rapid capital stock growth. (p. 286)	Table 4: author's growth accounting (thesis) compared with Maddison's figures for seven countries.	Partial data
State policy and apartheid, not lack of resources, explain the failure; South Africa was remarkably well-placed for growth after 1945. (p. 279)	Qualitative assessment drawing on Hobart Houghton, Norval and other secondary literature; no direct measurement of policy effects.	Asserted
Technological capabilities in the capital goods sector were weak and reliant on foreign technology, possibly explaining sluggish productivity. (p. 289)	Kaplan's studies of the capital goods sector; no original data.	Illustrative or anecdotal

The author's caveats

Acknowledges that the GDP numbers are questionable, 1950s-60s figures 'rather shaky' and 1930s-40s figures estimates rather than measures; GDP and TFP comparisons are 'intrinsically dubious'; TFP is uncertain as a measure; educational attainment data comparable to Maddison's could not be located; the 1980-85 ranking is partly a statistical aberration; Maddison's Latin American labour input is overestimated; and space limits mean methodological detail is available only on request.

Gaps in the evidence

The argument that policy and apartheid caused the slow growth is inferred from aggregate comparisons without testing mechanisms; firm- or sector-level data on productivity, protection rates, quota administration, and labour allocation would be needed. A counterfactual or growth-regression framework with controls for resource endowment (including gold price effects) would test the 'potential' benchmark better; such data (national accounts, tariff and trade data, industrial censuses) existed then. The thesis series underlying the recalculated GDP and TFP are not presented, so the key results cannot be verified from the article.

Replicability: Data described but not public.