Skip to content

A Firm Level Analysis of Trade, Technology and Employment in South Africa

Literature › Works

Lawrence Edwards (2002). A Firm Level Analysis of Trade, Technology and Employment in South Africa. TIPS Annual Forum 2002, Glenburn Lodge, Muldersdrift. Trade and Industrial Policy Strategies (TIPS).
Type conference paper
Scope Post-1994 South African economy
Method econometric
Period studied 1994–1999
Regions South Africa, Greater Johannesburg Metropolitan Area, national manufacturing
Themes labour markets wages, manufacturing industrial policy, transition post apartheid, education human capital, methods data sources, comparative theory
Citation key edwards2002a

Argument

Using two firm-level surveys (the national NE survey and the World Bank/GJMA large-firm survey), the paper tests whether post-1994 tariff liberalisation drove South Africa's employment decline and rising skill intensity. Cross-tabulations and labour demand functions (Cobb-Douglas and CES) show a negative link between liberalisation and employment only among large firms. The overall effect is judged small, and export gains did not translate into jobs. Technology (computer investment, imported inputs, foreign ownership) raised skill intensity, with some evidence of Wood's 'defensive innovation' among import-hit firms. Results are suggestive rather than conclusive because the data cannot establish causation.

Evidence and method

National Enterprise (NE) survey of 941 manufacturing firms (late 1999/early 2000, covering from early 1998, 5 occupational categories, binary employment-change data) and the World Bank/GJMA large-firm survey of 325 Gauteng firms (1999, covering 1994-98 with recall employment data). Firms were self-classified as significantly, moderately or little/not affected by tariff cuts. Also draws on aggregate studies (Bell and Cattaneo, Edwards 2001a, Bhorat).

Findings

  • Large firms were much more affected by import competition (market share loss) than small firms. (p. 23)

    51 % of large firms were either significantly or moderately affected compared to 24 % of small firms.

  • Among large firms, employment decreases are more common among firms hit by liberalisation; no such pattern among small firms. (p. 27)

    In all skill categories the share of firms decreasing employment was higher amongst firms significantly affected by trade liberalisation than firms not affected.

  • In the GJMA survey the median firm affected by tariff cuts shrank while unaffected firms grew, but the relationship is weak. (p. 27)

    the median firm not affected by lower import duties/taxes since 1994 raised employment by 7.7 %. In contrast the median firm that was affected reduced employment by 8.9 %.

  • Firms hit by liberalisation account for only a minor share of employment decline, so the overall impact is small. (p. 28)

    firms significantly or moderately affected by tariff reductions account for 22 % of the decline in total employment excluding the food & beverages sectors between 1994-98.

  • Exporters, despite better competitiveness, were more likely to cut employment. (p. 47)

    exporters, relative to non-exporters, had a higher probability of reducing employment of unskilled (-0.61) and highly skilled (-0.45) workers since the beginning of 1998.

  • Very few firms invested in order to cut the workforce, and investment motives are unrelated to liberalisation severity. (p. 34)

    Very few firms (less than 6 %) invested in order to reduce wage costs or labour conflict by cutting the workforce.

  • Large firms that cut unskilled labour and lost market share invested disproportionately in efficiency-raising technology, consistent with defensive innovation. (p. 35)

    51.5 % of firms that reduced unskilled labour and experienced significant or moderate negative impacts on market share from trade liberalisation invested in new technology in order to raise efficiency and product quality.

  • Computer investment and imported raw materials raise skill intensity, evidencing skill-biased technological change and trade-related technology transfer. (p. 47)

    the share of investment in computers and the imported share of raw materials are positively related to the relative demand for high skilled labour within the NE survey results.

  • The estimated elasticity of substitution between skilled and less skilled labour is below one. (p. 48)

    The elasticity of substitution of skilled and less skilled workers ranges between 0.408 to 0.47.

  • Foreign and export firms are larger, more skill- and capital-intensive and more productive, especially among large firms. (p. 19)

    Large, foreign and export firms are more skill intensive, produce more output, are more capital intensive and employ more labour which is also more productive