Industrial policy in developing economies: developing dynamic comparative advantage in the South African automobile sector

Abstract

Mainstream economics and the Washington Consensus caution against industrial policies that target sectors, firms and regions. At the most they favour cross-sectoral policies which address generalized market failures. This paper analyses the success of an industry-specific policy, South Africa's Motor Industry Development Programme. It documents significant learning processes and shows the impact of the sector's growth on macroeconomic performance. It also addresses the ‘costs’ of industrial policy and shows how well-designed scale-enhancing selective policies can provide domestic consumers with global-quality products at global-price levels, without subsidy from the exchequer. The conclusion addresses the relevance of such selective policies to other developing economies, arguing the case for intelligent and appropriately crafted industrial policy.

Keywords:

• Industrial policy
• Market failure
• Automobile and components sector
• Washington Consensus
• South Africa

Acknowledgements

The authors are grateful to Anthony Black, David Kaplan and Sean Ellis for their comments and help both before and during the course of the drafting of this work; and to Frank Flatters, the CENTRIM Reading Group, Hubert Schmitz, Clive Williams and two anonymous referees for their comments on an earlier draft. The usual disclaimers apply.

Notes

This judgement is not an ideological one, since one of the authors of this paper was deeply involved in the definition of the post-1994 policy regime. For background to this see Joffe et al. (1995).

The reasons for this anomaly in policy are complex. But a significant factor was the power of the two trades unions involved. Not only had they influenced pre-Transition policy, but the new government was, in effect, a coalition of a political movement (the ANC) and a trade union movement (COSATU). COSATU, in turn, was heavily influenced by two of its most powerful constituent members, NUMSA (the metalworkers union, covering the auto sector) and SACTWU (the union organizing clothing and textile workers).

Data provided by David Kaplan and drawn from South African Statistics and Revenue Services databases.

For an analysis of the auto components sector in emerging markets, see Humphrey (2001).

The Fiat Uno and Toyota Tazz are face-lifted variants of successful late 1980s models, whilst the CitiGolf is an upgraded Golf 1, introduced in the late 1970s. All three vehicles remain strong sellers in South Africa, with the Toyota Tazz the top-selling model in South Africa in both 2001 and 2002. Whilst not meeting current European emission-, economy- and ride-standards, they offer far more space than their European comparators.

http://europa.eu.int/comm/competition/car_sector/price_diffs/.

Price-deflation in the UK was between one and two percent for all categories of cars, expect for the budget class (−7.8%) – www.carpriceindex.com.

http://www.jda.com/presspass/pr/images/200204bfull.gif.

An additional adjustment which can be made is to correct for the 6.5 percent import duties levied on South African auto imports into the UK. This is a debatable exercise, since South African imports serve only a portion of the European market. But even if this is factored into the equation, South African auto prices are below UK prices by 6.8 percent for the Golf, 11.2 percent for the Toyota, 16.2 percent for the Mercedes Benz and 20.7 percent for the BMW. It makes no sense to adjust for the Clio and the Alfa Romeo, since these are not exported from South Africa.

The basic salary of an operator in the South African automotive industry in 2002 was US$1.22 per hour (approx. $210 per month), whilst a line function manager earned an average of approximately US$2,060 per month (Department of Trade and Industry 2002).

These benchmarking clubs operate as continuous improvement networks for automotive component manufacturers, with member firms benchmarked internationally and against one another on an annual basis. They are a clear example of firms taking advantage of the horizontal industrial policies created post-1994 (see www.bmanalysts.com).

The way in which the MIDP has differentially facilitated the assemblers integrating local operations into their global production strategies is discussed in Barnes and Morris (2004).