Skip to content

Capital intensive industrialisation and comparative advantage: Can South Africa do better in labour demanding manufacturing?

Literature › Works

Anthony Black; Reviva Hasson (2012). Capital intensive industrialisation and comparative advantage: Can South Africa do better in labour demanding manufacturing?. Towards Carnegie III conference, University of Cape Town, 3-7 September 2012 (draft of 31 July 2012). University of Cape Town (Carnegie III conference).
Type conference paper
Scope Post-1994 South African economy
Method synthesis review
Period studied 1961–2012
Regions South Africa, Thailand (comparison), Eastern Cape (Coega), Richards Bay
Themes manufacturing industrial policy, business and state, labour markets wages, inequality poverty unemployment, comparative theory, transition post apartheid, macro fiscal finance debt
Citation key black2012

Argument

South Africa's industrial policy, before and after 1994, has been interventionist in the wrong direction: it has favoured capital- and energy-intensive, resource-based heavy industry and thereby hindered labour-demanding manufacturing, employment and growth. South Africa's 'revealed' comparative advantage in capital-intensive products is partly the product of this distortion, along with skills shortages, upstream market power and import-parity pricing, and cheap energy and capital subsidies. The authors document the bias through cheap electricity, tax and investment allowances, DTI incentive schemes (37E, SIP, 12i, CIP, IDZs) and IDC finance. They conclude that policy should tilt incentives toward employment-intensive growth, subsidising labour and training instead of capital, electricity and infrastructure for heavy industry.

Evidence and method

Secondary literature and published data: Levy (1992) cross-country capital intensity comparison, Pollin et al. employment-intensity data, World Bank investment climate surveys (Clarke et al. 2007), OECD/IEA coal and electricity prices, van Horen and van Heerden electricity studies, DTI/IDC and SIP project data (Roberts, Mondi and Roberts), Burton's dissertation. A 2009 Durban Automotive Cluster remuneration survey and a Thai–South African automotive benchmarking study (Barnes and Black 2012) are used for labour-cost comparisons. No new archival work.

Findings

  • Employment intensity of output fell sharply after 1994, especially in manufacturing and mining. (p. 5)

    From 1994-2001 employment in relation to output fell by 28% and the sharpest declines in employment intensity over the whole period have been in manufacturing and mining during the 1990s.

  • Post-1994 liberalisation sharply cut protection, but the export structure stayed resource-based, with heavy sectors keeping roughly 60% of merchandise exports. (p. 6)

    Apart from a dip in 1999 and 2000, minerals, basic metals, basic chemicals and pulp & paper have maintained a share of around 60 per cent of total merchandise exports since 1994.

  • Skilled and managerial labour, not production labour, is the main cost disadvantage for manufacturing, shown in the Thai–SA automotive comparison. (p. 7)

    the ratio of production workers’ wages in South Africa compared to Thailand was nearly three to one, for professionals 6:1 and for artisans an incredible 12:1

  • Cheap, state-driven electricity expansion underpinned the minerals-energy complex and energy-intensive beneficiation. (p. 9)

    According to Winkler and Marquard (2009:52) electricity prices in South Africa have averaged around 40% of the US price over the last four decades.

  • Eskom's investment drive in the 1970s and early 1980s was a very large share of national investment. (p. 9)

    Eskom undertook massive over investment in the 1970s and early 1980s increasing its capacity threefold between 1970 and 1982 (Burton, 2011: 41). In 1977, Eskom’s investment accounted for no less than 12% of total GDFI

  • Subsidised power for the planned Coega smelter would have created few jobs relative to investment, and the project was cancelled. (p. 11)

    The R21bn greenfield investment would have employed just 800 people, with the product expected to be almost entirely exported in primary form.

  • The SIP tax programme favoured highly capital-intensive projects. (p. 14)

    The average capital-intensity of SIP supported projects was R3.7million per employee, amongst the most capital-intensive in the manufacturing sector (Roberts, 2008:24).

  • IDC finance historically perpetuated the capital-intensive bias, continuing under the new government into the 1990s. (p. 15)

    Much of the IDC finance in the second half of the 1990s continued to be oriented to large, capital-intensive, resource-based activities.

  • The 37E tax incentive for resource projects was undermined because firms avoided domestic sales, limiting labour-intensive downstream linkages. (p. 14)

    many firms avoided this requirement by simply not selling to domestic consumers, thus hampering downstream more labour-intensive linkage effects.

  • Policy conclusion: the incentive structure should be tilted toward labour and training rather than capital and heavy industry. (p. 19)

    Incentives should subsidise labour and training rather than capital investment, electricity and infrastructure for capital intensive firms.