Trade and the structure of South African production, 1984–97

Literature › Works

Lawrence Edwards (2001). Trade and the structure of South African production, 1984–97. Development Southern Africa, Vol 18, No 4. Routledge / Development Bank of Southern Africa.

Type	journal article
Scope	Post-1994 South African economy
Method	descriptive quantitative
Period studied	1984–1997
Regions	South Africa
Themes	manufacturing industrial policy, labour markets wages, sanctions divestment, transition post apartheid, methods data sources, comparative theory
Citation key	edwards2001

Argument

Using a Chenery–Syrquin input–output decomposition of gross output growth across the 1984, 1988, 1993 and 1997 tables, Edwards argues that trade liberalisation has not deindustrialised South African manufacturing. Export growth has matched and exceeded import-induced losses, and the net trade effect has been positive since 1993, in line with the international evidence in Chenery et al (1986). Exports served as a 'vent for surplus' during the stagnant 1984–93 years. Strong capital-intensive exports together with import penetration in ultra-labour-intensive sectors have shifted production towards capital-intensive sectors. He attributes this shift mainly to domestic factors rather than liberalisation: capital-biased supply-side support, wage and labour-market institutions, technological change, and long-standing weaknesses in ultra-labour-intensive industries. He stresses that these conclusions are only suggestive because the data are aggregate.

Evidence and method

South African input–output tables for 1984, 1988, 1993 and 1997 (the last two RAS-updated from 1988), aggregated to 29 SIC-4 sectors and deflated with National Productivity Institute gross-output deflators. Trade and capital-intensity classifications follow Bell & Cattaneo. Comparison with the Chenery et al (1986) multicountry decomposition results. Supporting material comes from World Bank discussion papers (Belli et al, Tsikata, Fallon), DTI and IDC reports, and secondary literature on protection, factor prices and labour legislation.

Findings

• Headline result: imports have risen but exports have more than offset the losses, so there has been no deindustrialisation. (p. 1)

  Although import penetration has risen, export growth has matched and exceeded the potential import-induced losses in domestic production.

• The manufacturing structure shifted toward capital-intensive sectors between 1984 and 1997. (p. 2)

  Of interest, however, is the rise in capital-intensive sector output as a share of manufacturing output, from 37,6 per cent in 1984 to 42,4 per cent in 1997.

• Exports offset weak domestic demand in 1984–8, acting as a vent for surplus. (p. 7)

  As a result, exports accounted for 123,9 per cent of the rise in total gross output between 1984 and 1988, with manufactures and services contributing to most of this.

• Domestic demand drove the post-1993 recovery. (p. 7)

  Between 1993 and 1997, improved domestic demand led the recovery, contributing 60,5 per cent of the increase in total gross output.

• Net trade has contributed positively to output growth since 1993, whereas it was negative in 1988–93. (p. 9)

  In contrast to the 1988–93 period, net trade has affected growth positively since 1993 and accounted for 9,7 per cent of the rise in total gross output between 1993 and 1997.

• In manufacturing, exports compensated for collapsing final demand before 1993, but falling domestic demand cut investment and capital stock, which limited export capacity. (p. 10)

  Declining domestic demand affected investment negatively, resulting in a 10 per cent decline in total capital stock between 1984 and 1988, and a 27,8 per cent decline

• Poor performance of ultra-labour-intensive sectors predates liberalisation, which may only have accelerated their decline. (p. 12)

  suggests severe quality and competitive problems within many of the subsectors (leather products, footwear and wearing apparel) that existed prior to trade liberalisation.

• Cross-country comparison: domestic demand expansion normally dominates manufacturing growth. (p. 14)

  In 23 of the 28 cases, domestic demand expansion accounted for over 50 per cent of the change in manufacturing output.

• South Africa's post-1993 trade pattern resembles other economies that shifted to more open regimes. (p. 14)

  South Africa is no exception and displays very similar trade patterns between 1993 and 1997 to Israel between 1965 and 1972.

• Export structure is moving toward capital and skill intensity, which is paradoxical in a labour-abundant economy. (p. 16)

  It appears that South Africa is, paradoxically, moving towards a high skill and capital-intensive export structure.

• Industrial support programmes favoured capital-intensive projects, and job creation was costly. (p. 16)

  employment generated through the spatial development initiatives of the Department of Trade and Industry has been at a cost of between R1,2 million and R1,7 million a job.

• Competing explanations for rising capital intensity: Kaplinsky attributes it to political factors rather than relative factor prices. (p. 17)

  argues that the increase in capital intensity is not due to relative factor price changes, but to political factors that reduced private investment in labour-intensive sectors.