Structural Changes in Manufacturing Industry 1916 to 1975¶
Marais, G. (1981). Structural Changes in Manufacturing Industry 1916 to 1975. South African Journal of Economics, vol. 49, no. 1.
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | descriptive quantitative |
| Period studied | 1916–1975 |
| Regions | South Africa |
| Themes | manufacturing industrial policy, business and state, labour markets wages, inequality poverty unemployment, macro fiscal finance debt, methods data sources |
| Citation key | marais1981 |
Argument¶
Marais traces South African manufacturing from 1916/17 to 1975 using a stage-of-development framework. He tests it against five markers of industrial maturity: manufacturing's share of GDP, import and export ratios, inter-industry linkages and technology-intensive growth. He argues that manufacturing grew far faster than GDP, mainly through state-promoted import replacement (tariffs from 1925, import control from 1948, tariff protection again in 1956-67). It moved from consumer goods towards intermediate goods and processed minerals. Gold, foreign trade and government policy shaped this path, and the colour bar and a shortage of black skills limited it. The paper closes by arguing that future growth depends on capital-intensive, export-capable mineral processing and on better training of black workers, and that manufacturing alone cannot absorb black unemployment.
Evidence and method¶
Aggregate national-accounts and trade statistics on output, imports and exports by industry group, converted to 1956/57 constant prices (Table 1, Fig. 1, Appendix A classification). Official reports (Board of Trade and Industries, Tariff Commission, Commission on the Export Trade, Economic Development Programme 1978-87) and earlier studies (Schumann, Du Plessis, Ariovich, Meyer, Kleu) plus the author's own earlier papers.
Place in the debates¶
- Import replacement vs export-led industrialisation (see Industrialisation, protection and industrial structure). Credits import replacement with building the existing industrial structure. Expects its future scope to be limited, so growth must shift to manufactured exports, especially capital-intensive protected industries that become 'forced exporters'.
- Capital-intensive vs labour-intensive industrial strategy and employment (see Industrialisation, protection and industrial structure). Favours capital-intensive mineral processors on comparative-advantage grounds, arguing capital deepening may raise productivity with a lag. Concedes manufacturing alone cannot solve black unemployment, which requires training and removal of labour-market discrimination.
- Stages of economic development (agricultural-mining-manufacturing) (see Industrialisation, protection and industrial structure). Adopts the Schumann/Du Plessis stage framework, dating the shift to a manufacturing stage to after 1925 (or around 1939). Treats gold mining as having suppressed manufacturing's share early on.
Findings¶
-
Manufacturing's share of GDP rose about fourfold over the period. (p. 3)
the share of manufacturing in South Africa's GDP rose from 6,2 per cent in 1916/17 to 25,1 per cent in 1975
-
Manufacturing outgrew the economy in real terms. (p. 3)
manufacturing output increased at the average annual rate of 6,9 per cent compared with 4,5 per cent for GDP in real terms
-
Import penetration fell sharply as local production replaced imports. (p. 3)
Expressing imported manufactures as a percentage of domestic manufacturing production, a sharp decline is observed from 104,6 per cent in 1916/17 to 34,8 per cent in 1975
-
Tariff protection after 1925 coincided with rapid growth in manufacturing's share of GDP. (p. 5)
The contribution of manufacturing to GDP increased from 8,1 per cent in 1926/27 to 12,9 per cent in 1936/37
-
Wartime import replacement drove growth in 1936-46. (p. 5)
The value of manufactured imports as a percentage of local production declined from 101,4 per cent to 62,5 per cent
-
The colour bar and the two-tier employment system depressed productivity in labour-intensive manufacturing and limited exports. (p. 6)
Because of factors like the colour bar (job reservation) and lack of training on the part of most black workers, the prevailing two-tier system of employment led to low productivity in labour-intensive manufacturing industry
-
Growth in 1956-67 was mainly state-induced import replacement. (p. 7)
During the period 1956-67 growth resulted mainly from import replacement, actively promoted by the government by means of tariff protection and in some instances by import control.
-
Growth slowed markedly after 1967, when protection was eased and the world economy turned down. (p. 7)
manufacturing experienced one of its lowest average annual growth rates since 1916, namely 6 per cent, between 1967 and 1975
-
Inter-industry linkages deepened after 1960. (p. 7)
this figure increased from 18,4 per cent in 1960/61 to 34,0 per cent in 1975
-
Capital intensity rose quickly in the 1970s with only slow productivity gains. (p. 9)
between 1970 and 1977 the capital/labour ratio increased annually at the rate of 4,9 per cent
-
Industrial growth was skill-biased, leaving unskilled black workers in surplus. (p. 8)
the growth of manufacturing has resulted in the coincidence of a shortage of skilled labour with a surplus of unskilled labour
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Author's industrial output, import and export series by industrial classification, 1916/17-1975 (Table 1, Figure 1, Appendix A), at 1956/57 prices | official statistical series | 1916/17-1975 | Provides the output, import and export values and growth rates by consumer, intermediate and primary-processing groups that carry the structural-change argument. |
| Du Plessis, unpublished UNISA thesis on the industrial growth pattern 1916/17-1956/57 | secondary literature compilation | 1916/17-1956/57 | Supplies the estimate that manufacturing was about 30% below potential by 1926/27 and the stage-of-development classification in the 1930s. |
| Report of the Customs Tariff Commission 1934/35, UG 5-1936 (Holloway Commission) | commission or official report | 1934/35 | Cited for the case for industrialization and for civilized labour employment, and linked to the Tariff Commission's estimate of the cost of protection. |
| Board of Trade and Industries, Report No. 282: Investigation into Manufacturing Industries of the Union of South Africa, 1945 | commission or official report | 1945 | Supports the forecast that gold resources would decline appreciably after 1945. |
| Report of the Commission of Inquiry into the Export Trade of the Republic of South Africa, RP 69/72 | commission or official report | 1972 report; covers 1939 onward | Used to date South Africa's entry into the agricultural-mining-manufacturing stage to 1939 and to support points on relaxed import replacement and sporadic exports. |
| Report of the Commission of Inquiry into Policy relating to the Protection of Industries, UG no. 36/1958 | commission or official report | 1958 | Documents the government's stated arguments for protection and industrial growth in the 1956-67 period. |
| Economic Development Programme for the Republic of South Africa 1978-1987, vol. 1 | commission or official report | 1960-1977 | Source of the capital per worker, labour/output and productivity trend figures for manufacturing in the 1960s and 1970s. |
| Swanepoel, unpublished MCom thesis on protection of industries in South Africa | secondary literature compilation | c. 1925-1936 | Supports attributing the relative decline in imports to tariff protection. |
| Marais, unpublished Stellenbosch dissertation, statistical analysis of South African imports by product 1926-1950 | secondary literature compilation | 1926-1950 | Cited for the stage-of-development periodization after 1925. |
| Ariovich (1979) and Meyer (1979) studies of export comparative advantage and export expansion | secondary literature compilation | 1960s-1970s | Used as corroborating findings on capital-intensive export advantage and sporadic exports. |
| Kleu, Industrialization Strategy and Employment, SA Federated Chamber of Industries conference paper, 1979 | secondary literature compilation | 1979 | Source of the five reasons for the limited future scope for import replacement. |
| Industry-level export and tariff data for chemicals and rubber products, 1975/76-1978/79 (source not named) | official statistical series | 1975/76-1978/79 | Illustrates how tariff protection and export performance coincide in capital-intensive processing industries. |
| Input-output type inter-industry sales data 1960/61-1975 (source not named) | official statistical series | 1960/61-1975 | Measures inter-industry linkages among consumer, intermediate and primary-processing groups. |
New data¶
The author constructed national series of manufacturing output, imports and exports at 1956/57 prices for 1916/17-1975, reclassified into consumer, intermediate and primary-processing groups (Table 1, Appendix A); the underlying sources are not named and further tables are available from the author on request.
Inference¶
Quantification: Descriptive tables.
Descriptive time-series comparison of shares and growth rates across sub-periods and between 1916/17 and 1975. Periods are narrated chronologically, with policy and external factors assigned as causes by assertion and by reference to commissions and other scholars. There is no formal statistical testing.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Manufacturing's share of GDP rose from 6.2% in 1916/17 to 25.1% in 1975, growing faster (6.9% p.a.) than GDP (4.5%), agriculture or mining. (p. 3) | Author's constructed output series at 1956/57 prices (Table 1); sources unnamed. | Partial data |
| Manufactured imports fell from 104.6% to 34.8% of domestic manufacturing production, while the manufactured export/output ratio declined from 17.8% to 12.9%. (p. 3) | Table 1 import, export and production series. | Partial data |
| Intermediate goods and processed primary materials grew fastest, so manufacturing shifted from consumer goods towards intermediate and processed-mineral output. (p. 3) | Group growth rates in Table 1 and further disaggregated industry rates, which are not shown. | Partial data |
| Growth spurts in 1926-36 and 1960-69 were driven by tariff protection, and in 1949-60 by import control, i.e. state-induced import replacement. (p. 2) | Timing of growth rates set against policy dates, plus Schumann, Du Plessis and the commission reports; no counterfactual test. | Illustrative or anecdotal |
| Inter-industry linkages deepened, with local intermediate content in consumer-goods inputs rising from 18.4% in 1960/61 to 34.0% in 1975. (p. 7) | Unnamed inter-industry sales data for 1960/61 and 1975. | Partial data |
| Growth slowed after 1967 because protection was eased and imports rose in response to consumer demand and the world downturn. (p. 7) | Output and share figures for 1967-75 plus narrative on policy and international conditions. | Illustrative or anecdotal |
| Capital-intensive processing industries with tariff protection are the best prospect for future export-led growth, even though capital deepening has not yet clearly raised productivity. (p. 9) | Economic Development Programme figures, Ariovich, Meyer and Kleu, and a few product examples; the argument is largely inferential. | Asserted |
The author's caveats¶
The author notes that 1916/17 export figures may be abnormally high because of First World War demand, and that forced exports are difficult to identify. He also says further information on the tables may be obtained from the author.
Gaps in the evidence¶
The underlying statistical sources (the Census of Manufactures, trade statistics and national accounts deflators) are not identified, and the reclassification and construction of the 1916/17 series is undocumented. Causal claims about tariffs, import control and the colour bar are not tested against firm- or industry-level data, tariff schedules or employment series, although these existed (census of manufactures, Board of Trade reports, trade statistics). Evidence on labour use and race-based employment is limited to assertion.
Replicability: Data poorly documented.