South Africa loses billions in export misinvoicing
“This suggests that export underinvoicing is not due to underreporting of the true value of gold exports, but rather to pure smuggling of gold out of the country. In other words, virtually all gold exported by South Africa leaves the country unreported.”
This is the sensational conclusion of a very recent study undertaken by UNCTAD on the issue of trade misinvoicing – a key method for illicitly exporting capital out of a country with severe consequences for financing much needed development, reducing poverty or in the case of South Africa, dealing with debilitating mass unemployment.
The study, « Trade Misinvoicing in Primary Commodities in Developing Countries: The cases of Chile, Cote d’Ivoire, Nigeria, South Africa and Zambia (UNCTAD/SUC/2016/2) http://unctad.org/en/PublicationsLibrary/suc2016d2.pdf, » is by Professor Lonce Ndikumana, University of Massachusetts, Amherst for UNCTAD’s Commodity division, and was released in Nairobi Kenya on 18 July.
The study provides empirical evidence on the magnitude of trade misinvoicing in the case of primary commodity exports from five natural-resource-rich developing countries: This sample comprises four resource-dependent developing countries and South Africa – a more diversified exporter of primary commodities.
Estimates of trade misinvoicing have been primarily based on bilateral trade data published in the Direction of Trade Statistics (DOTS) of the International Monetary Fund (IMF), providing aggregate values of imports and exports between a country and its trading partners. However, this study breaks new ground by providing an analysis at more disaggregated sector and product levels. The analysis additionally uses data published in the United Nations Commodity Trade Statistics (UN Comtrade) Database, which provides time series on imports and exports broken down by product, country and trading partner.
There are substantial levels of trade misinvoicing in all five countries covered by the study, but the patterns vary substantially across countries, products and trading partners.
Given the historical importance of gold exports to the South African economy, the study is extremely alarming. It is as if the sanction busting practices of the apartheid government before 1994 simply stayed in place. But today it is tax revenues, local investment and, not least, wages of gold mine workers that are ‘busted’.
Of course this is very much about tax revenue and the current account deficit, where more foreign currency is going out than coming in, creating dependency on borrowing and speculative financial inflows for a government that shuns capital controls. But we shall return to the issue of stakeholders – other than SA and all African governments – hurt by the staggering rip off. Let us first give some highlights from the UNCTAD study, where platinum and silver was under invoiced by $2.3 billion between 2006-2014 and, in fact, a further $2.3 billion between 2000-2005, even if exports were three times greater for that period.
Misinvoicing within SA’s gold sector is however even more staggering. Gold exports from SA, says the UNCTAD report, is not simply about undervaluation of exports. It is “rather a case of pure smuggling of gold out of the country”.
“The most striking feature of the gold sector in South Africa is the huge discrepancy between the amounts recorded in that country’s official trade statistics and those reported in its trading partners’ records. According to South Africa’s data, the country’s cumulative gold exports were $34.5 billion from 2000 to 2014, whereas according to trading partner data for that period they were more than three times higher, at $116.2 billion. This is indicative of massive export underinvoicing.” In fact, the study reports, the physical volume of exports (using the data from SA’s partners) and export underinvoicing are in “perfect correlation”. “This suggests that export underinvoicing is not due to underreporting of the true value of gold exports, but rather to pure smuggling of gold out of the country. Total misinvoicing of gold exports to South Africa’s leading trading partners was $113.6 billion over the 15-year period.” At an average exchange rate of R9 per dollar, this corresponds to over R1 trillion.
More bizarre and requiring further investigation is a phenomenon illustrated by Hong Kong. According to the South African data, Hong Kong’s share in South Africa’s gold exports to its trading partners fell from 90 per cent in 2010 to a mere 2 per cent (!) in 2011. Yet, the data from all SA’s trading partners show that Hong Kong’s share of the same trading partners gold imports from South Africa actually rose from 4.6 per cent in 2010 to 22.3 per cent in 2011!
For South Africa, the study only finds so called “overinvoicing” for iron export to Japan and the Netherlands. In such a case the value of exports from SA to a trading partner is higher than the import value recorded for SA by that trading partner.
It is only with Netherlands that the study finds a net “overinvoicing” situation. For the period 2000-2014 SA overinvoiced iron export to the Netherlands by $1.4 billion. The report laconically concludes that “a large proportion of iron ore exports from SA to the Netherlands … does not appear to have ever docked in the Netherlands”. This can be because of the iron being smuggled into the Netherlands. It can also be that the buyers’ residence is incorrectly reported: It could be that the SA company is selling the iron to partner companies in tax havens even if they report something else.
To solve the overinvoicing enigma, the report recommends studies at company level. This is currently quite difficult given the lack of access to financial records of TNC subsidiaries. Nevertheless, this study brings us nearer to this possibility, because, for the first time, we now have a study of misinvoicing, commodity by commodity for South Africa.
One motive for corporations to underinvoice the value of exports from SA can of course be to avoid tax. This is common cause. The companies in question avoid tax on profits and royalties. And indeed: individual beneficiaries who control the set-up avoid personal income taxes.
Lower taxes on profits and lower royalties – a tax on the sales regardless of profits earned – are clearly very important. But it can also be seen as a profitable side-effect of the more over-arching motive, which simply is to move massive amounts of dollars out from South Africa. This is where ‘wage evasion’, as a concept, becomes important.
Against the background of the UNCTAD report, it possible to argue that all mineworkers could be paid a decent living wage of least R12500; all that is needed is for these illicit outflows to stop.
Underinvoicing in the platinum sector amounted to $4.6 billion between 2000-2014, or $300 million per year. At an exchange rate of R12 to R14 per dollar, the misinvoicing factor alone allows for a monthly wage increase of between R1100-R1300 per month for 190 000 permanent and contract workers in this sector. We have then first simply assumed that 28 per cent of every $100 in illicit outflows was used for other purposes than wages, as per the corporate income tax rate. Even when personal income tax and VAT is taken into account, the underinvoicing potentially affects worker wages more than tax revenues.
In the gold sector between 2000-2014, $113.6 billion left the country through underinvoicing. These particular illicit gold sector outflows averaged a staggering $7,6 billion per year. In the same vain, and at an exchange rate of R12 to R14 per dollar, the misinvoicing factor alone allows for a monthly increase of R54 500 to R63 600 for every worker.
We don’t think the 100 000 gold sector workers will demand that magnitude of wage increase. Such windfalls are only for CEOs. Isn’t there a class action going on for another 200 000 former gold mining workers who became sick from silicosis? They, too, could do with a bit of this money.
Brian Ashley is director and Dick Forslund senior economist at Alternative Information & Development Centre.