Humbled by the data

Roman Grynberg’s most recent diatribe in the Mail and Guardian demands a response. His main point seems to be that South Africa depends on its exports to Botswana, Lesotho, Namibia and Swaziland (BLNS) to fund its trade deficit with the rest of the world and for this reason it is prepared to go to extreme lengths to hide its trade data and pay-off the BLNS through the SACU revenue sharing agreement. Matthew Stern

November 5, 2014

Matthew Stern

Roman Grynberg’s most recent diatribe in the Mail and Guardian demands a response. His main point seems to be that South Africa depends on its exports to Botswana, Lesotho, Namibia and Swaziland (BLNS) to fund its trade deficit with the rest of the world and for this reason it is prepared to go to extreme lengths to hide its trade data and pay-off the BLNS through the SACU revenue sharing agreement.

The availability and quality of intra-SACU trade data is undoubtedly a problem, but inconsistencies between export and the corresponding import data are a universal dilemma; for this reason, trade economists usually turn to the importing country for a more accurate account of in-coming trade flows. If data on South Africa’s exports to the BLNS is questionable or not available, then it is the BLNS, and not South Africa, which should be held to account.

More importantly, given these inherent problems, it is important to analyse and use the available trade data with caution. In his article, Grynberg simply assumes that all that is reported as exports from South Africa to the BLNS, must be 100% made or manufactured in South Africa. On this assumption, if you exclude R67bn of net exports to the BLNS, South Africa’s overall trade deficit would balloon by this exact amount. This is clearly erroneous and for two main reasons.

Firstly, in today’s global economy, little is made in one country without substantial inputs (i.e. imports) from another. If one waves away a unit of exports from South Africa to the BLNS with a magic wand, as Grynberg seems to have done, you must also wave away the foreign inputs into the good that has been exported. But this form of value-added manufacturing probably accounts for a small percentage of the total value of trade between South Africa and the BLNS.

Secondly, and of far greater value, much of what South Africa reports as exports to the BLNS (or the BLNS reports as imports from South Africa), were never made in South Africa at all. The greatest benefit of being in a customs union is that goods that enter the union, say from China, can then move seamlessly and at much lower cost between member countries. Strictly speaking, these imports are re-exported from South Africa to the BLNS, but in the eyes of customs officials, they are treated as originating from within the customs union. For more on this peculiar problem see our previous article on this topic.

So unfortunately for South Africa, the “ugly facts” do not point to a boom in manufactured exports to the continent, but instead suggest that South Africa’s true exports to the BLNS are significantly lower than that which has been calculated by Grynberg, and by SACU. Legally speaking, a very large proportion of these exports (all re-exports) should be removed from SACU trade and revenue sharing calculations and the resulting payment from South Africa to the BLNS should be a fraction of the current amount. Alternatively, South Africa could require that all external trade that is destined for the BLNS be treated as transit trade, greatly increasing the cost of doing business in the BLNS. Fortunately, these legitimate policy options do not seem to be on the table.

South Africa undoubtedly benefits from its economic relations within SACU and with the continent and this is generally acknowledged (if not exaggerated) by domestic policy makers. But those countries that trade through South Africa, as part of the customs union, also accrue significant efficiency and revenue gains. Less provocation, better analysis and a more balanced dialogue is needed within the region.