The main purpose of industrial policy is to speed up the process of structural change towards higher productivity activities. This paper builds on our earlier writings to present an overall design for the conduct of industrial policy in a low- to middle-income country. It is stimulated by the specific problems faced by South Africa and by our discussions with business and government officials in that country. We present specific recommendations for the South African government in the penultimate section of the paper.

Beneficiation, moving downstream, and promoting greater value added in natural resources are very common policy initiatives to stimulate new export sectors in developing countries, largely based on the premise that this is a natural and logical path for structural transformation. But upon closer examination, we find that very few countries that export raw materials also export their processed forms, or transition to greater processing. The quantitative analysis finds that broad factor intensities do a much better job of identifying patterns of production and structural transformation than forward linkages, which have an insignificant impact despite the fact that our data is biased against finding significant effects of factor intensities and towards finding significant effects of forward linkages. Moreover, the explanatory power of forward linkages is even smaller in sectors with high transport costs, and in sectors classified as primary products or raw materials, which are the most common targets of such policies. Finally, the results are the same even when only considering developed countries, meaning that colonial legacy inhibiting transitions to natural resource processing are not to blame. These results suggest that policies to promote greater downstream processing as an export promotion policy are misguided. Structural transformation favors sectors with similar technological requirements, factor intensities, and other requisite capabilities, not products connected in production chains. There is no reason for countries like South Africa to focus attention on beneficiation at the expense of policies that would allow other export sectors to emerge. This makes no sense conceptually, and is completely inconsistent with international experience. Quite simply, beneficiation is a bad policy paradigm.

This paper explores export performance in South Africa over the past 50 years, and concludes that a lagging process of structural transformation is part of the explanation for stagnant exports per capita. Slow structural transformation in South Africa is found to be a consequence of the peripheral nature of South Africa’s productive capabilities. We apply new tools to evaluate South Africa’s future prospects for structural transformation, as well as to explore the sectoral priorities of the DTI’s draft industrial strategy. We then discuss policy conclusions, advocating an ‘open-architecture’ industrial policy where the methods applied herein are but one tool to screen private sector requests for sector-specific coordination and public goods.

* See also the 7/27/07 Sciencenews article as well as the supplementary materials website.