Innovation in South African city-regions: Can we explain it?

Thumbnail Image
Lorentzen, Jo
Mugadza, Tendeukayi
Robinson, Shirley
Associated Organizations
Organizational Unit
Organizational Unit
Supplementary to
Much innovation takes place in cities and their adjacent and linked hinterland, or city-regions. Given that city-regions can be hotbeds of innovation, it is important to understand the determinants of these processes. The bulk of analytical work and of rankings concentrates on advanced economies, especially the US and Europe. One reason for this focus is that there is rich data about (city-) regions in those areas. The same cannot be said for developing countries in general, and for Africa in particular. South Africa is no exception. This paper is an attempt to assemble data that may be relevant for innovative outcomes in South Africa’s city-regions and to make a few conjectures about the determinants of performance. To the best of our knowledge, to date no one has tried systematically to link comprehensive city-region data to innovative performance. South Africa has four major city-regions, namely the inland Gauteng city-region around the capital Pretoria and the commercial hub of Johannesburg, and Cape Town, eThekwini (Durban), and Nelson Mandela Bay (Port Elizabeth) on the coast. Together they host 60 per cent of the country’s population and generate 72 per cent of GDP. Data show that South Africa’s two relatively and absolutely most innovative city-regions (Gauteng and Cape Town) distinguish themselves from the other two (eThekwini and Nelson Mandela) through higher growth (of both economy and population), a higher employment rate, a sectorally more diverse workforce, lower poverty rates, and more investment in R&D, but not through higher per capita income, higher labour productivity, a higher share of the working in the total population, a higher growth of formal employment, a relatively more skilled workforce, and a larger degree of economic openness. Because of data limitations we run simple bivarite correlations on both patents and scientific publications. Our findings point to the fact that, R&D is good for innovation, and that a population equipped with the facilities to tune into global knowledge flows, supplied with a minimum level of services, and with very highly skilled scientists and engineers in their midst, produces more scientific articles. Due to unavailable, inaccessible, incomplete or otherwise problematic data we conclude that South Africa is not really in a position to understand innovative activities where they are generated, namely at the local level. Hence, in the short term, all that policymakers and academics can do is learn from innovation strategies in city-regions or countries elsewhere in the world in order to inform local policymaking.
Date Issued
Resource Type
Resource Subtype
Rights Statement
Rights URI