ColumnistsPREMIUM

GAVIN KEETON: Why the UN's damaging report is very wrong on SA’s mineral exports

Simple logic shows a report accusing mining and oil companies in Chile, Ivory Coast, Nigeria, SA and Zambia of smuggling cannot be true

Gavin Keeton

Gavin Keeton

Columnist

Picture: ISTOCK
Picture: ISTOCK

In June 2016, the UN Conference on Trade and Development (Unctad) published a report accusing mining and oil companies in Chile, Ivory Coast, Nigeria, SA and Zambia of smuggling huge amounts of capital abroad, allegedly to avoid local taxes and bypass exchange controls.

The report compared the value of each country’s mineral and oil exports against the reported value of the same commodity imports by their main trading partners.

The discrepancies were attributed to companies understating exports. This would allow them to declare reduced profits and taxes in the countries where they operate while earning full value overseas on what they actually produced.

In this way, mining and oil companies were said to have misappropriated as much as 67% of export revenue between 2000 and 2014. South African gold producers were accused of "smuggling" $78.2bn abroad. Silver, platinum and iron-ore producers were accused of underinvoicing by $24bn.

Yet simple logic shows this cannot be true. If gold producers hid 67% of revenue earned abroad as claimed, all this would be unreported (and untaxed) profit. Mines must therefore have paid workers, suppliers, taxes and dividends from just 33% of their actual revenue. Their "real" profit margins would have put them among the most profitable businesses in the world. Gold miners would be lining up to produce in SA, which is clearly not the case.

South African gold producers are all local companies. Even if they did disguise their exports, they would still have had the extra earnings in their annual financial statements.

Conscious of the enormous reputational damage being done to its members, the Chamber of Mines appointed consultancy Eunomix to probe Unctad’s claims, and fatal methodological flaws were found in the report.

Unctad had failed to realise that — in line with international reporting standards — South African trade statistics did not list gold exports by country of destination. They were reported collectively as "unclassified".

Alternative data from the Reserve Bank and Statistics SA show South African gold exports to be much higher than Unctad claimed, but still about $19.5bn less than recorded by importers.

Rather than evidence of underinvoicing, this is probably explained by the fact that Rand Refinery refines gold on behalf of other African countries. SA, correctly, does not record those sales as our exports as they originate elsewhere.

In the case of platinum, Eunomix shows that alternative export data reduce Unctad’s claimed underinvoicing to $15bn. Any discrepancy is likely to be for the same reason as for gold, since SA also refines Zimbabwe’s platinum.

The claimed underinvoicing of iron-ore exports exposed a further flaw in Unctad’s methodology. Unctad revealed large underinvoicing from 2000 to 2010 and overinvoicing from 2011 to 2014. It is nonsensical to think iron-ore producers would illegally understate exports in the earlier period and overstate them (and pay more tax) in the latter. Eunomix notes that because iron ore is a heavy, low-value product, transport costs are sometimes as much as 50% of sales value. This means the reported value of imports in the country of destination (which includes transport costs) will be far higher than the stated South African export value.

Measures of trade flows are also affected by the "Rotterdam effect". Ports such as Rotterdam are channels for trade destined for other countries. Ore used in Austria may be shipped first to the Netherlands. Austria may record imports from SA but South African exports are recorded as going to the Netherlands.

The final destinations of many metals passing through the warehouses of the London Metals Exchange are unknown at the time of sale. Indeed, metal may be bought and sold several times before final use. Comparing exports by reported destination and imports by origin is, therefore, impossible.

• Keeton is with the economics department at Rhodes University and was a consultant to Eunomix in the drafting of its reports.

Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.

Comment icon