How Botswana Vanquishes the African Diamond Trade-Indepth

This paper considers the loss to Namibia’s fiscus from the peculiar nature of the Southern African Customs Union revenue sharing formula. It is argued that Namibia loses considerable amounts of customs revenue by virtue of the trade that has developed as a direct result of the De Beers- GRB (Government of the Republic of Botswana)  diamond marketing agreement whereby  diamond aggregation has been shifted from London to Gaborone. As a direct result, all of the De Beers production in Namibia, some ZAR 11 billion in 2016 was exported to Botswana which was previously exported to the UK. This in turn meant that based on the customs component of the SACU Revenue Sharing formula, Botswana share of intra-SACU imports increased substantially and Namibia and South Africa’s share decreased as a direct result of the De Beers-Botswana agreements. This in effect meant that Namibia and South Africa were subsidizing the Botswana policy of domestication of diamond trade. The paper estimates the losses and suggests ways of resolving these unintended consequence of the  2011 De Beers –GRB marketing agreement.
This paper begins by considering the relevant marketing provisions of  the De Beers-GRB  2004 and 2011 agreements and the effect that this has had on intra-SACU trade.  The SACU Revenue Sharing Formula is then considered in the second section of the paper. The diamond trade figures are then considered and the losses in revenue to Namibia from the agreement are estimated. In the final section various policy options available to Namibia are considered so that Botswana, the world’s largest diamond producer by value can continue to aggregate without the unintended consequence that Botswana’s neighbors are unintentionally subsidizing its policy of diamond aggregation and beneficiation.
The Legal Relationship between De Beers and the GRB
Two sets of agreements govern the relations between Botswana and De Beers. The first is the  agreement that was reached when the De Beers permits were extended for a period of 25 years in 2004. The second was the 10 year Marketing Agreement between De Beers and Botswana (2011) which  together created the  economic arrangement which  saw Botswana become the aggregation centre for De beers diamonds globally. It was not just SACU members but Canadian diamonds are also aggregated in Botswana. The relationship was however more complicated by virtue of the fact the parcels that were needed for  beneficiation were aggregated in Botswana and hence exported to Namibia and South Africa. This has created a ZAR 28 billion intra- SACU trade in 2014/15 which has substantially changed the value and distribution of SACU Customs revenues.
2004 De Beers –Botswana Agreement
In 2004 the GRB signed an agreement with its partner, De Beers to renew the mining leases of the latter. The agreement between GRB, which owns 15% of De Beers,  and De Beers set the stage for the movement, post-2011 for the movement of aggregation operations to from London to Gaborone and was to set . This agreement and the subsequent 2011 marketing agreement, which remains in force were to have significant effects on the nature of the trading relationship in SACU
a) Marketing Arrangements – DTC and DTCB
Under the arrangement between GRB and De Beers, all diamonds were to be marketed through the De Beers group. This is central to the market power that De Beers would exercise in the global market and was true both pre and post 2000 when the CSO came to a formal  end. However, this direct control of diamonds excluded special stones. The agreement between the GRB and De Beers   in the privatization arrangements in 2000 was that Debswana and the GRB would remain part of the De Beers group as long as all the diamonds were marketed through the company. However, this provision did not apply to the high value stones which could be marketed outside the group. It is unknown whether De Beers had, during the 1990’s, a separate marketing channel for the high value stones and where the revenues from these stones went.
In 2004 Botswana negotiated a marketing arrangement with De Beers (which was eventually formalised in 2006): trading was to continue through DTC London for five years until 2011. Following this, DTC was expected to set up  DTC Botswana in the capital, Gaborone, at a reported cost in excess of US$83 million. It was agreed that DTC would return to DTC Botswana all the sights that were meant for the purposes of beneficiation. These particular sights would be sold at international (De Beers) book prices and would contain the same international mix of products. In other words, the agreement eliminated the possibility of a two tiered two pricing for the purpose of beneficiation. De Beers also agreed to establish a jointly- owned trading company (DTCB- 50/50) which would sort and value all Debswana diamonds. It was expected that the role of DTCB would be to facilitate the development of a local processing industry.
The 2004 agreement stipulated that DTC would also make available a certain volume of diamonds for the purposes of beneficiation. It was expected that the supply of rough diamonds for the purposes of beneficiation would be gradually increased from USD100 million in 2006 to some USD550 million by the end of the agreement. Over the same period the agreement stipulated that employment in diamond cutting was expected to rise from 1,600 to 2,500. However, in order to facilitate the use of high technology equipment the employment obligations of the 2004 agreement fell away in 2011. This was to have serious ramifications on industry employment levels during the diamond market crisis of 2016.  To become a DTCB sight holder it was necessary to beneficiate. Therefore to gain access to rough diamonds became dependent, inter alia upon local beneficiation:
The criteria used to determine individual supply levels included global competitiveness criteria, which assesses technical; marketing and distribution efficiency; as well as local criteria, which sought to promote skills development; job creation; and local manufacturing.
Sightholders cutting diamonds in Botswana were not required to cut all their diamond allocations and were permitted to export upwards of 70% of their sights for cutting in other, usually lower- cost destinations such as India or Thailand. They also began to import large volumes of rough and semi-processed diamonds. Botswana was in effect establishing what is called in trade parlance a ‘trade related investment measure’ which could be challenged at the WTO if beneficiation ever became a significant portion of trade.  There has also been considerable discussion as to whether there was a cross- subsidization of Botswana’s beneficiation program by the De Beers group at the time; an allegation that De Beers has always vigorously denied.
2011 Marketing Agreement
The marketing agreement that was negotiated in 2004/6 as part of the suite of agreements  was of a five year duration expiring on 31st December 2010 when the substantive new marketing agreement of ten year duration entitled ‘Agreement in Respect of Diamond Sorting, Valuing, Marketing and Sales’ was finally completed in September 2011. The agreement contained several important and novel components. These included a marketing arrangement of rough diamonds, an independent marketing window of Debswana diamonds and commitments to a source of supply of rough diamonds for the purposes of beneficiation. Broadly, what the arrangement signalled was the move of Botswana away from its role as purely a source of mined diamonds to one of marketing, and at least in theory, the production of polished diamonds.
Marketing Arrangements- Botswana as a trading centre
Perhaps the most important and enduring development was the agreement between the parties to localize the aggregation and sale of rough diamonds in Botswana away from the De Beers headquarters in London. This was to be completed in December 2013. The sales of rough diamonds in Botswana began in August 2012 and all aggregation activities were now undertaken in Botswana after 88 years of operation at Charterhouse in London. This meant that diamonds from Canada, Namibia and South Africa would be brought to Botswana for aggregation. This agreement saw a transfer of some 90 positions to Botswana and De Beers   has indicated that the migration of aggregation and sales business to Botswana would add approximately USD6 billion to the local economy by the end of 2013. There were expected to be 10 sites per year where all then 76 De Beers sightholders would gather in Gaborone instead of London. It is this provision that has the single largest impact on SACU customs revenues as it created the basic conditions for intra-SACU imports of ZAR 28 billion.
Marketing Window – Okavango Diamonds
The marketing arrangement that had previously existed between De Beers and the GRB needs to be unpacked in order to understand the GRB’s desire to have its own marketing arm. Under all the previous marketing arrangements, Botswana was paid by the De Beers Diamond Trading Company International at a 10% discount to the Standard Selling Value (SSV). This was in effect the margin paid to De Beers for marketing all of Botswana’s diamonds. If one considers the unit production value of Botswana diamonds as reported in the Kimberley Process statistics, it was almost invariably 10% below the unit export value. In 2013 the unit production value of Botswana diamonds was USD156/carat while unit export values were USD173/carat. As a result, there was a long held view in diamond industry circles in Botswana that once De Beers ceased to be a monopoly after 2000 the marketing function could readily be performed by a local firm.
Under the terms of the marketing agreement, Botswana was to obtain 10% of Debswana production beginning in 2011 which was to increase to some 15% of Debswana production by 2014. For many years parts of the GRB had wanted a marketing window for reasons of price discovery i.e. to determine the market accuracy of the De Beers Price Book. To that end the government established Okavango Diamonds in 2012 as a state trading arm which auctioned approximately 3Mcts per annum. Indeed, the very purpose of the auctioning system was to test the market valuation, as well as to claw back the 10% margin obtained by De Beers for marketing Botswana’s diamonds.
However, it must be noted that the diamonds sold by Okavango first went through DTCB, and it can be reasonably assumed that they were sold at a discount to the SSV. The precise commercial relationship and the price are not in the public domain. Thus the De Beers price will act as the floor price though there appears to be considerable volatility around that price. Other small diamond companies like Lucara Diamond Corporation also market their diamonds through auctions although the reserve price is established by the SSV. As long as the DTCB price book acts as the commercial benchmark through reserve prices for these auctions it can be argued that the outcomes of the auction  were in no way a definitive measure of market value. Thus, despite the splintering of its direct control over rough diamond prices, De Beers retained considerable control through the SSV and its affiliation to related companies.
In 2014, the then Minister of Minerals Energy and Water Resources, Hon Kitso Mokaila indicated his intention to increase Okavango Diamonds’ share of Debswana production to 25%. The difficulty with this approach was that Okavango Diamonds, unlike De Beers, did not oblige buyers to cut and polish diamonds in Botswana.  In other words Okavango is there to trade and De Beers to mine and produce. It, along with other small producers, were free to auction their diamonds and buyers were under no obligation to undertake any local processing. This in turn meant that De Beers sightholders had a differential cost disadvantage through their obligation to cut and polish diamonds domestically. In a bear market for rough diamonds, as existed in 2015 and the beginning of 2016, the commercial advantage of being a DTCB sightholder i.e. access to increased supply of diamonds vanished because diamonds without any beneficiation obligations were readily available through the state marketing arms and on the secondary market.
As can be seen from the chart above the margin between unit export value and unit production value varied from a low of 1.8% in 2003 to a high of 27% in 2011 although the average margin was 9.8% over the period. The percentage marketing margins appear to have widened significantly over the 2010-2013 period.
Extract from a recently launched research paper by Professor Roman Grynberg, Dr Teresa Kaulihowa and Dr Jacob Nyambe