The contribution of entrepreneurs to the emergence of manufacturing in South Africa before 1948¶
Stefan Schirmer (2008). The contribution of entrepreneurs to the emergence of manufacturing in South Africa before 1948. South African Journal of Economic History, 23(1/2). Economic History Society of Southern Africa.
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | mixed |
| Period studied | 1818–1948 |
| Regions | Cape, Natal, Transvaal/Witwatersrand, Port Elizabeth/Eastern Cape, South Africa |
| Themes | manufacturing industrial policy, origins segregation, business and state, comparative theory |
| Citation key | schirmer2008 |
Argument¶
Schirmer challenges the view that South African manufacturing was small, unsophisticated and a dependent creation of state protection (especially the 1925 Pact tariffs) and of mining. Aggregate and sectoral statistics for 1918-1929 show steady growth that does not track tariff changes closely. Using descriptive case histories of firms and industries, he argues that entrepreneurs built local capabilities through learning by doing and through knowledge networks linking them to European expertise and capital. He concludes that post-1925 blanket protection, extended after 1948, displaced a more organic growth path, and that government should have supported rather than merely protected existing capabilities, as Moll suggested.
Evidence and method¶
Cape census of 1891 and De Kock's data; Union Statistics Office 'principal manufacturing industries' series in the South African Year Books (1924, 1929, 1930); Feinstein's aggregate output figures; trade journals (Industrial South Africa, South African Journal of Industries, South African Commerce and Manufacturers' Record); Who's Who in Business; Holloway and Cullinan Commission reports; a Board of Trade and Industry submission in the Central Archive Depot; and secondary firm histories (Kaplan, Frankel, Mendelsohn).
Place in the debates¶
- Did manufacturing depend on state protection? (state-creation vs entrepreneurial-capabilities view) (see Industrialisation, protection and industrial structure). Rejects dependence on tariffs. Argues the tariff-value added correlation is weak and that supply-side entrepreneurial capabilities mattered at least as much as protection.
- Was 1924/25 (Pact government) the turning point for manufacturing? (see Industrialisation, protection and industrial structure). Disputes this. Output growth was 8.76% p.a. in 1916-24 and 6.56% in 1924-29, which suggests continuity rather than a break.
- Did manufacturers capture the state in 1924, and were they creatures of mining? (see Industrialisation, protection and industrial structure). Sceptical of the capture thesis (Yudelman) and of dependence on mining. Stresses Cape roots and internal dynamics while acknowledging Witwatersrand mining demand and the WWI boost.
- Did the apartheid-era economy fail, and was protectionism the cause? (see Did the apartheid economy fail? Growth, productivity, investment and the post-1970s crisis). Follows Moll in blaming blanket protectionism and the focus on 'advanced' sectors for post-war underperformance and the collapse of manufactured exports.
Findings¶
-
Cape manufacturing was already substantial before Union, with output comparable to the leading export sectors. (p. 3)
The value added in manufacturing in the Cape during 1891 was by comparison a very respectable £2 300 000.
-
Aggregate output growth in the years before and after 1924/25 does not show a break, contrary to Feinstein's own figures. (p. 5)
output increased at an average rate of 8,76 per cent during the period 1916 to 1924, and at a rate of 6,56 per cent during the period 1924 to 1929.
-
The 20 'principal industries' were sizeable in output and employment. (p. 7)
these often denigrated sectors produced a net output of over £176 million and employed, on average, over 60 000 people during a given year.
-
Correlation between value added and tariff changes is weak, undermining the dependence-on-protection view. (p. 7)
The correlations between the value added changes and tariff levels are not strong enough to support those views that see manufacturing as entirely dependent on artificial protection.
-
Even the large 1925 tariff shift coincided with only modest changes in growth trends. (p. 8)
tariff levels changed by more than 142 per cent.
-
Firms deepened capital and absorbed European know-how, as in biscuit making. (p. 16)
The ratio of capital to labour within the sector as a whole increased from £42 per worker in 1918 to £732 in 1929.
-
Local brewers dominated the domestic market well before the protective tariff of 1925. (p. 21)
97,5 per cent of the ale, beer and stout consumed in South Africa during 1909 was produced by local brewers.
-
Plywood firm example: a firm seeking tariff relief but not dependent on it, undercutting importers. (p. 29)
charging their customers 25 to 40 per cent less than the prices charged by the importers.
-
Entrepreneurial capabilities, combining local conditions with European knowledge and capital, shaped the industries, which were not mere creations of policy. (p. 31)
they utilised local conditions, in combination with European sources of knowledge and capital, to establish industries and firms that built up their capabilities over time.
-
Post-1925 blanket protection and post-1948 extension produced a dependent sector instead of an internationally competitive one. (p. 31)
the country eventually spawned a highly dependent, although technologically advanced, sector, that was, on the whole, unable to cope with external competition.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Cape Colony census of 1891 (industrial establishments), as reported via De Kock, Selected subjects in the economic history of South Africa | population census | 1891 | Establishes that about 2,000 industrial establishments and £2.3m manufacturing value added existed in the Cape by 1891, so manufacturing predated Union statistics. |
| De Kock, Selected subjects in the economic history of South Africa (1924) – wool, gold and Cape industry descriptions | secondary literature compilation | 1882-1902; 1890s | Supplies the wool export and gold output comparators and the description of the Cape industrial base (mills, wagon works, railway workshops etc.). |
| Feinstein, An economic history of South Africa (2005) – aggregate manufacturing output series | secondary literature compilation | 1916-1929 | Aggregate output growth rates (8.76% for 1916-24; 6.56% for 1924-29) used to challenge the 1924/25 turning-point view. |
| Union of South Africa, South African Year Book, 1924, 1929, 1930 (manufacturing statistics, net output/value added 1915-1929) | official statistical series | 1915-1929 | Provides net output/value added totals, 1920-21 contraction and 6.1% average growth to 1929, and the national sector totals. |
| Statistics Office series on the 'principal manufacturing industries of the Union' (20 sectors), South African Year Book 1929; 1930 | official statistical series | 1918-1929 | Basis for Tables 1 and 2 and Figure 1 on value added, employment, firm counts, capital per firm, and the correlation against tariff changes. |
| Tariff level changes series 1920-1929 (as plotted in Table 2) | official statistical series | 1920-1929 | Percentage changes in tariff levels are correlated against value-added changes to show weak association (r<0.5). |
| Holloway Commission Report (UG) | commission or official report | c.1909-1935 | Cited for soap industry supplying about half of national requirements by 1910 and for the 97.5% local share of beer consumed in 1909. |
| Cullinan Commission Report (UG) | commission or official report | c.1912 | Cited for cement/brick/pottery scale and transport-cost problems and for the tanning industry's natural advantages by 1912. |
| Industrial and Commercial South Africa / Industrial South Africa (trade periodical), 1921-1926 | newspapers periodicals | 1921-1926 | Source of descriptive accounts of industry histories, firms, factories and local industrial prospects (printing, milling, cement, Vereeniging, Estcourt, motor). |
| South African Journal of Industries (1921-1922) | newspapers periodicals | 1921-1922 | Used for descriptions of pasta, Witwatersrand Show exhibitors, fertiliser plant, boot and shoe industry and asbestos-cement. |
| South African Commerce and Manufacturers' Record (1913-1922) | newspapers periodicals | 1913-1922 | Source for firm profiles (Burmeister candles, Melachrino tobacco, Fatti macaroni, Port Elizabeth manufacturing) and the 1917 British Trade Review quote. |
| South Africa's Who's Who / Who's Who in Business (1911, 1921-22, 1923-24) | secondary literature compilation | 1911-1924 | Provides biographical and company details for printing firms, Fatti and the Transvaal Soap Company. |
| Daily newspapers: The Star and The Daily Mail (1909-1919) | newspapers periodicals | 1909-1919 | Used for reports on fertiliser, glass, tobacco, pottery and steelworks developments. |
| Central Archive Depot, MNW 729, No MM2057/24, Industry Submission to the Board of Trade and Industry, c1923 | state archives | c.1923 | Evidence on brewing export attempts and on Plywoods (SA) Ltd's founding, pricing and tariff request. |
| Firm and business histories: Frankel, Tiger Tapestry (1988); Kaplan, Jewish roots in South Africa (1986); Fridjhon & Murray, Conspiracy of giants (1986); Mendelsohn, Sammy Marks (1991); Bozzoli (1981) | secondary literature compilation | 1880s-1920s | Supply entrepreneurial narratives for Tiger Oats, SAB, Henochsberg, Vereeniging and Sanderson. |
Inference¶
Quantification: Descriptive tables.
Combines a simple correlation of annual percentage changes in value added with tariff changes for 1920-29 (aggregate and 20 'principal industries') with illustrative firm- and industry-level case narratives drawn largely from trade periodicals, to argue that capability building by entrepreneurs, not tariffs, drove early manufacturing.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Cape manufacturing before Union was substantial, with about 2,000 establishments and £2.3m value added in 1891, comparable to wool and gold. (p. 186) | 1891 Cape census figures as reported by De Kock, compared with wool and gold export/production values. | Partial data |
| Aggregate output growth did not accelerate after the 1924/25 policy shift; growth was steady through the 1920s. (p. 188) | Feinstein's aggregate growth rates and Year Book net output figures for 1915-1929. | Systematic data |
| Correlation between tariff changes and value-added changes is weak (below 0.5), so manufacturing was not entirely dependent on protection. (p. 190) | Year Book value-added series for the 20 principal industries and total sector compared with tariff level changes, Table 2. | Partial data |
| The principal industries were large in aggregate, producing over £176m net output and employing 60,000 people on average. (p. 190) | Year Book series on 20 principal industries, Table 1. | Systematic data |
| Individual industries (printing, milling, brewing, tobacco, boots, etc.) deepened capital and capabilities through entrepreneurship and absorption of European know-how. (p. 195) | Year Book firm counts, employment and capital per firm plus trade-press and business-history narratives. | Illustrative or anecdotal |
| Plywoods (SA) Ltd shows a firm that sought tariff relief but was not dependent on it. (p. 213) | Board of Trade and Industry submission MM2057/24 and a Journal of Industries report. | Illustrative or anecdotal |
| Explanations other than tariffs, namely entrepreneurial capability building, are asserted rather than tested statistically. (p. 193) | The author admits no alternative statistics; relies on qualitative description of industries. | Asserted |
The author's caveats¶
Acknowledges that no alternative statistics exist to correlate with value added and employment, so the entrepreneurial-capability explanation is speculation confirmed only qualitatively; calls the case descriptions an 'initial attempt' to be followed by regional research. Notes consistent Union statistics only start in 1916.
Gaps in the evidence¶
The tariff-output test is a bivariate correlation of annual percentage changes on about ten data points, with no effective protection rates, lags, price deflators, import data or controls for the exchange rate and the war/post-war cycle. The firm narratives come largely from promotional trade periodicals and are selected on success, with no systematic company records, archive or survivorship evidence to show capabilities caused growth. Industry-level census microdata, tariff schedules by product, and Board of Trade files exist and could test industry-specific protection against growth.
Replicability: Data described but not public.