Breaking the ‘big data’ barrier when selecting agricultural export markets: an innovative approach

A country’s comparative advantage is not only dependent on factor endowments. History, random events (wars, oil crisis, sanctions, etc.) and past government policies are important factors shaping a country’s trade patterns. Such factors are recognised in both traditional and new trade theory. Therefore the formulation (and implementation) of industrial and agricultural export policy has to be cognisant of these factors. In the search for new markets or new product opportunities in existing markets to inform both trade policy making and business decision making, a major challenge is making sense of the huge volumes of available product and market information, which is one of the manifestations of “big data”. Using an example from the South African fruit industry, this paper illustrates how the big data challenge can be tackled using the TRADE-Decision Support Model methodology. By means of this methodology an initial 1221 realistic export opportunties were identified in 107 markets. Of the overall 54 products in the fruit and nuts HS chapter 08 category, 22 have “major potential”, representing about US$3.5 billion across 102 countries. Most of the potential for “mature” products lies in “new” markets from a South African agriculture exports perspective. Of this potential 80 per cent is found in 10 products (including grapes, apples, mandarins, and lemons and limes). Some non-traditional products such as bananas, cashew nuts, kiwifruit and guavas were identified. Europe still represents approximately half of the total estimated realistic potential in the short term, estimated at US$6 billion, followed by North America (22 per cent) and Asia (21 per cent).