Origins of Leading Sector Technology in South Africa: Coal-Based Synthetic Fuels and Cyanide-Based Gold Extraction¶
Peter Hilsenrath; Thomas Pogue (2012). Origins of Leading Sector Technology in South Africa: Coal-Based Synthetic Fuels and Cyanide-Based Gold Extraction. University of the Pacific, Eberhardt School of Business (preliminary draft, April 6, 2012). University of the Pacific.
| Type | working paper |
| Scope | Economics and economic history of apartheid |
| Method | archival narrative |
| Period studied | 1886–2011 |
| Regions | South Africa, Witwatersrand, Sasolburg, Secunda, Germany |
| Themes | mining migrant labour, manufacturing industrial policy, business and state, sanctions divestment, comparative theory, origins segregation, macro fiscal finance debt |
| Citation key | hilsenrath2012 |
Argument¶
Using the resource-curse literature (Van der Ploeg 2011), the paper asks whether two South African natural-resource technologies were blessings or curses. Coal-based synthetic fuels (Sasol) were transferred largely from Nazi Germany via Anglovaal and then built as a state-backed parastatal. They survived thanks to subsidy, low or negative real interest rates in the 1970s-80s, and a weak rand. Cyanide-based gold extraction, developed on the 1890s Witwatersrand through an open community of metallurgists, the Chamber of Mines, and professional societies, underpinned rapid gold-mining growth and 'crowded in' other capital. The authors conclude that both technologies created development opportunities that would not otherwise have existed, and that South Africa is a 'mixed blessing'. They note the gold industry rested on an exploitative labour structure and that post-apartheid reinvestment of resource rents has been weak. The method is descriptive history drawing on the authors' two dissertations.
Evidence and method¶
Authors' dissertations (Hilsenrath 1985; Pogue 2006); Proceedings of the Chemical and Metallurgical Society of South Africa and other technical journals; Chamber of Mines metallurgy handbooks; patent application table; secondary histories (Lang, Kubicek, Bozzoli, Emden); World Bank and Van der Ploeg resource-rent and genuine-savings data; Sasol annual report 2011; gold output series 1886-1902.
Place in the debates¶
- Resource curse vs blessing (see Industrialisation, protection and industrial structure). Argues that South Africa is a mixed case. Gold-cyanide and Sasol generated knowledge spillovers and industrial capacity, but the gold industry rested on exploitative labour and post-1994 reinvestment of rents has been poor.
- Economic efficiency of Sasol / synfuels subsidy (see Industrialisation, protection and industrial structure). Concedes oil-from-coal is normally uncompetitive and was subsidised, but argues that opportunity cost is low in a capital-abundant, high-unemployment economy. The case is judged 'not clear' rather than clearly inefficient.
- Open innovation and cooperation among Rand mining-finance groups (see Industrialisation, protection and industrial structure). Follows Kubicek in seeing cooperation in operations alongside competition for finance. Stresses Marshallian 'knowledge in the air', the fixed gold price, and COMSA coordination.
- Role of Afrikaner nationalist/pro-German sympathies in the synfuels technology transfer (see Industrialisation, protection and industrial structure). Suggests such sympathies may have helped but notes the patents went to Anglovaal, a Jewish-owned, non-Afrikaner firm.
Findings¶
-
The German synfuels patents were taken up by English-speaking mining capital (Anglovaal), not Afrikaner business. (p. 12)
Yet it was not an Afrikaans business interest that took possession of the synfuels patents, but rather a Jewish one, Anglovaal.
-
The 1947 Liquid Fuel and Oil Act gave Anglovaal the exclusive position it would have had under the German agreement, making the state the enabler. (p. 15)
The Liquid Fuel and Oil Act effectively gave Anglovaal what it would have had with the German agreement, exclusive rights for synfuels production and marketing.
-
After Anglovaal withdrew, the state's IDC created Sasol in 1950. (p. 15)
In 1950 the South African, Coal, Oil and Gas Company also known as South African Synthetic Oil Limited (Sasol) was established.
-
The IDC financed the large Sasol II and III expansion with state equity, in a scheme likened to Goering's plan. (p. 16)
The Industrial Development Corporation provided much of the equity capital in a scheme not unlike that of the Goering Plan.
-
Even in the mid-1980s, Sasol did not cover coastal areas, which remained exposed to sanctions-hit imports. (p. 16)
By the mid 1980’s, Sasol provided for most petroleum needs in the interior of the country but coastal areas, including Cape Town and Durban, still depended on imports.
-
The gold boom pushed down interest rates when Sasol was financed, easing its high capital cost. (p. 17)
High gold and other resource prices led to windfall export earnings and an abundance of liquidity. This pushed down interest rates at just the time Sasol was financed.
-
Rand depreciation raised the rand price of crude about thirtyfold, making Sasol cost-competitive in the mid-1980s. (p. 17)
The rand price of a barrel of Saudi light crude increased from R1.85 in July 1973 to R54.71 in November 1984.
-
Efficiency critiques of Sasol are tempered by high unemployment and abundant capital. (p. 17)
What is the opportunity cost of a country awash in capital with very high levels of unemployment?
-
By 1898 the Witwatersrand produced almost a third of world gold output. (p. 19)
Gold produced on the Witwatersrand accounted for nearly 30% of annual global production by 1898.
-
Mine employment expanded roughly sevenfold in the decade after cyanide was introduced. (p. 19)
Employment on the Witwatersrand mines grew just over 20,000 in 1890 when the cyanide process was initially introduced to over 150,000 in 1898
-
Cyanide extraction transformed Rand mining from shallow outcrop working into capital-intensive deep-level industry. (p. 27)
Cyanide‐based extraction thereby initiated an unprecedented process of industrialization of gold mining.
-
The gold industry's growth rested on exploitative labour arrangements, making it a partial curse as well as a blessing. (p. 32)
underpinning its growth was an exploitive productive structure that would have profound impacts on the lives of native Africans across southern Africa for decades
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Table 1: Applications for Patents in the Union of South Africa (Ruhrchemie/Anglovaal synfuel patent applications, 1931-1936) | legislation gazette | 1931-1936 | Shows the German gasification and liquid-fuel synthesis patents transferred to Anglovaal. |
| Anglovaal–Ruhrchemie agreements, 1936 (contract of July 25, 1936 at Oberhousen-Holten and a second agreement) | company records or reports | 1936 | Documents the transfer of German Fischer-Tropsch patents and the production and marketing rights for southern Africa. |
| Liquid Fuel and Oil Act of 1947 and 1949 production licence | legislation gazette | 1947-1949 | Shows state support and the exclusive rights and ash-content licence conditions that favoured indirect liquefaction. |
| Petroleum Products Act of 1977 (Statutes of the Republic of South Africa: Trade and Industry) | legislation gazette | 1977-1978 | Evidence that Sasol and oil data were shrouded in secrecy. |
| Sasol 2011 Integrated Annual Report: Pursuing Responsible Growth | company records or reports | fiscal 2011 | Supplies current employment and revenue figures to show Sasol's present economic importance. |
| Rand price of Saudi light crude, July 1973–November 1984 (author series; Hilsenrath 1985; Stokke) | official statistical series | 1973-1984 | Shows rand depreciation raised the domestic crude price and made Sasol output cost-competitive in the mid-1980s. |
| Hilsenrath (1985) dissertation, Coal-based synthetic fuels: The South African experience | secondary literature compilation | 1936-1985 | Source of the oil-from-coal account (descriptive history of Sasol). |
| Pogue (2006) dissertation, The Evolution of Research Collaboration in South African Gold Mining 1886-1933 | secondary literature compilation | 1886-1933 | Source of the cyanide-extraction account and details on the structure of technology development. |
| Proceedings of the Chemical and Metallurgical Society of South Africa (CMSSA) 1894-1899, incl. von Gernet 1894, Williams 1897, Carter 1898 | union or movement records | 1894-1899 | Chronicles the collaborative development of slime treatment and zinc-lead and electrolytic precipitation. |
| Gold Metallurgy in South Africa (Adamson ed., Chamber of Mines, 1972) and Gold Metallurgy on the Witwatersrand (King ed., 1949) | chamber of mines or recruiting records | 1949, 1972 | Technical data on amalgamation recovery rates, pulp ratios, tube-mills and the precipitation process. |
| Gray and McLachlan (1933) history of the MacArthur-Forrest cyanide process and patent battle (J. Chem. Met. Min. Soc. S. Afr.) | newspapers periodicals | 1933 (covering 1890s) | Evidence on cyanide's cost advantage over chlorination and on the patents ruled invalid in 1896. |
| Contemporary reminiscences and technical papers: MacArthur (1908), Caldecott (1908), Webb (1903), Thurlow and Prentice (1928), Johnson (1913), Janisch (1986) | newspapers periodicals | 1903-1986 | Document the first pilot plants (Salisbury, Robinson), diffusion timing and the absence of trade secrets. |
| Witwatersrand gold output 1886-1902 (Figure 1) and employment figures | official statistical series | 1886-1902 | Shows rapid growth of gold output and employment after the cyanide process. |
| Chamber of Mines of South Africa (COMSA) institutional history (Patent Committee 1892, concentration sub-committee 1893), via Lang (1986) | chamber of mines or recruiting records | 1889-1896 | Shows the Chamber's role in coordinating technology diffusion and contesting patents. |
| World Bank (2006) Where is the Wealth of Nations? (resource rents, genuine savings) | international trade or finance data | 1970-2000 | Context on South Africa's resource rents share and poor reinvestment/genuine savings rate. |
| Van Der Ploeg (2011) JEL survey data on resource rents in GDP | secondary literature compilation | 1970-2000 | Shows South Africa is not highly resource dependent by resource rents in GDP. |
| German synfuel wartime production data (US Department of Energy 2006; Becker 1981; Sauter 1958) | secondary literature compilation | 1936-1944 | Describes German synfuel production and the Goering plan as background to technology transfer. |
Inference¶
Quantification: Illustrative figures.
Two descriptive historical case studies (Sasol oil-from-coal and cyanide gold extraction), drawn from the authors' dissertations and secondary literature, interpreted through the resource curse/blessing framework and innovation-systems theory. There is no formal test.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Coal-to-oil technology was transferred largely from Nazi Germany via Anglovaal's 1936 agreements with Ruhrchemie, then state-backed Sasol was created after 1947-50. (p. 12) | Patent table, the 1936 agreements, the Liquid Fuel and Oil Act and the van Dyk et al. account, as narrated from Hilsenrath's dissertation. | Partial data |
| Sasol's viability rested on state equity, very low or negative real interest rates from the gold and resource windfall, and rand depreciation in the mid-1980s. (p. 17) | Rand price of Saudi crude series and the author's earlier dissertation; the interest rate claim is asserted without data. | Illustrative or anecdotal |
| Sasol is now a profitable, significant employer with regional industrial impacts, and its inefficiency relative to imports is debatable given the low opportunity cost of capital. (p. 17) | Sasol 2011 annual report figures plus a counterfactual argument about unemployment, with no cost-benefit calculation. | Illustrative or anecdotal |
| Cyanide extraction made deep-level Witwatersrand gold viable, with output reaching nearly 30% of world production by 1898 and employment rising from about 20,000 to over 150,000. (p. 19) | Figure 1 output series and employment figures, with technical literature on recovery rates (50% by amalgamation versus cyanide). | Partial data |
| The cyanide process was developed through an open, collaborative cluster-based system (CMSSA, COMSA, labour mobility) that solved the slime treatment and precipitation problems by 1896-99. (p. 23) | CMSSA Proceedings and technical papers, and Pogue's dissertation; collaboration is inferred from the publication record. | Partial data |
| Zinc-lead precipitation diffused rapidly: by March 1899, twelve of sixteen slime plants used it, displacing electrolysis. (p. 25) | Carter (1898), Johnson (1913) and the CMSSA record. | Partial data |
| Gold mining crowded in rather than crowded out other capital, and fixed gold price and exchange rate removed Dutch disease and volatility, though the sector rested on an exploitative labour structure. (p. 27) | Argument from secondary literature (Emden, Bozzoli, Kubicek, Lang) and institutional narrative, with no quantitative test. | Asserted |
| South Africa is not highly resource dependent but reinvests resource rents poorly (genuine savings 6.9% in 2000). (p. 9) | World Bank (2006) and Van Der Ploeg (2011) cross-country data. | Partial data |
The author's caveats¶
The authors state that the methodology relies primarily on descriptive history combined with observations from the literature, drawing on their two dissertations. No other data limitations are acknowledged.
Gaps in the evidence¶
The central claims about financing costs, subsidy and opportunity cost of capital for Sasol are not backed by interest rate, subsidy or cost series, and Sasol's secrecy under the 1977 Act limits data. The cluster/collaboration claim for cyanide would be better tested with systematic data on patents, mine-level adoption dates and costs, and Chamber of Mines committee minutes, which exist in archives. The claims on spillovers, crowding-in and linkages lack input-output or sectoral data from the period; the curse/blessing verdicts are argued rather than measured, and the African labour dimension is mentioned but not analysed.
Replicability: Data poorly documented.