Trade creation is new trade from a Free Trade Agreement (FTA) partner which would not have existed otherwise, and that, generally, is good. But often much of this may be just trade diversion away from other, non-preference partners, and that may be bad – bad in the sense that it has resulted from an artificial advantage under the FTA, meaning that one is not buying from the world’s lowest cost supplier.

Starting from 2005 imports and the 2006 Southern African Customs Union (SACU) tariff schedule and using the standard formulas for this type of analysis we find that the Trade, Development and Cooperation Agreement (TDCA) increases imports from the European Union (EU) into South Africa by some R4.3 billion, but R2.7 billion of this is trade diverted from other sources. Next, simulating the impacts of the Mercosur/SACU agreement as it now stands we found that it has limited impacts upon imports from Mercosur: trade creation of R93 million but diversion of R70 million for an overall gain of only R23 million.

We then sequentially applied the TDCA tariff preferences in turn to China, the US and then India, using the calculated post-TDCA trade flows from all the different import sources. Granting China these preferences increased Chinese imports by R4.1 billion, with R2.4 billion diverted and therefore R1.7 billion in new trade. Then, bringing a SACU/US FTA at TDCA preferences into operation increased the imports from the United States (US) by R0.7 billion, with R0.5 billion of this being trade diversion. Next, granting India the same preferences led to trade creation of R325 million but some R 260 million of this in trade diversion limited the overall increase to only R65 million.

Finally, to ‘level the playing field’ somewhat and examine what a moderate outcome from the Doha Development Agenda (DDA) of the World Trade Organization (WTO) may do we reduced bound tariff rates by 30 percent in all cases except heavy motor vehicles and applied these rates to the imports from all those import sources excluded from preferences to date. The trade creation here was another modest R453 million, but with only R58 million in trade diversion and therefore an overall import increase of R395 million. Thus, a Doha outcome would make limited difference to South Africa’s import profile under these conditions where much (64.75%) of the imports were already entering under preferences.

What is significant with respect to the SACU Customs revenue pool is that this revenue reduces by an estimated 29.7 percent from a non-preference base, and, given that Lesotho and Swaziland rely on this revenue for half or more of their total government revenues, this will have serious consequences for them. Otherwise, at the end of all this creating and diverting, the total imports into South Africa increase by only around one percent. There will, of course, be general welfare effects that ripple through the economy, and these are not considered. As always, several assumptions have been made, and there are limitations to the study that are acknowledged.