The EC’s SoC Predicament: Is the “Secondary Market” a “Relevant Market”?
January 13, 05Acting on complaints from ex-Sightholders, the European Commission’s Competition Authorities are reaching the final phases of their reopened investigation into the Supplier of Choice (SOC) marketing system of De Beers. They are facing, however, a major stumbling block –the resolution of which may make a decisive difference in the EC’s final findings. Within the context of the present EC investigation, the EC doesn’t endeavor to review all aspects of the SOC arrangements again, but it rather focuses on specific complaints – and it must rule whether these are valid or not. The focal point of the major complaints is the charge that De Beers, in its efforts to make a more efficient market and in order to get maximum visibility of the final (retail) destination of the resultant polished from its rough sales, is in the process of “liquidating” the traditional role of rough dealers and minimizing the size of the secondary market. Moreover, by expanding the role of Diamdel, it is charged that De Beers is cornering the secondary rough market for itself, something which may constitute an abuse of dominant power, and thus an infringement of Articles 81 and 82 of the EC Treaty.
The EC Competition Authorities (also called DGIV) are, however, at a loss to understand whether from an economic and legal perspective, using the parameters of competition law, there is such a thing as a “secondary rough market” in general or in DTC sight goods in particular. From an EC perspective it is crucial that the descriptions and operations of what we call a “secondary rough market” indeed meet the test of constituting a separate market from a competition policy perspective.
The findings may be as crucial as the infamous glove in the O. J. Simpson murder trial, when the defense attorneys shouted to the jury “if the glove doesn’t fit, you must acquit.” If there is, from a DGIV perspective, no separate secondary rough market, then the charge that De Beers has illegally cornered it will automatically fall away.
It is for that reason that, in its questionnaires to Sightholders and non-Sightholders, the DGIV posed the question: Do you consider that there exists a ‘secondary market’ for rough? And in the event the question was answered in the affirmative, the DGIV added a number of follow-up questions:
If so, in what manner is this ‘secondary market’ distinct from a would-be ‘primary’ market? Describe in detail the differences between these markets explaining for example whether the diamonds sold on the ‘secondary market’ are substitutable for diamonds sold on a ‘primary market’ in terms of quality/range and price.
If there is a price difference for the same stones (please give concrete examples) between these markets? According to you what explains that difference? Specify whether the customers are different, in what respect?
The very fact that this question is being asked, and the very fact that the EC asks for detailed replies, underscores both the importance of the establishment of the definition of relevant market as well as the uncertainty felt by the DGIV team in this respect. The question posed asks for a very specific, detailed and convincing response – otherwise the comparison with O.J. Simpson’s glove becomes relevant.
After having approved an arrangement, DGIV needs convincing arguments to justify a reversal. From the feedback (and queries) we received from those obligated by law to answer the question, it seems that the present picture that is emerging in Brussels is a confusing one. In this issue of Diamond Intelligence Briefs – the text of which will also be submitted as a document to DGIV – we want to shed light on the issue and try to come up with some answers.
Moving from Theory to Practice: What Market?
The EC has been asked by various complainants to rule that De Beers is abusing its dominant position in the market through threatening the existence of a competitive sub-market (i.e. the secondary market) in rough diamonds. It is important to recognize at the outset that, in competition law, the issue as to how markets should be defined has never been answered conclusively. Economic theorists generally take the market as a given – we really don’t think about it too much, we assume that we all agree that we know what a market is. However, when one engages in empirical analysis, it becomes necessary to make judgments on what products and sellers constitute a “relevant market”.
A “relevant market” has both a “product” and a “geographic” component. In terms of rough diamonds, the product’s consumers (the diamond manufacturers, cutters, and polishers) represent one worldwide market. An Indian or Israeli manufacturer will get his rough from anywhere geographically speaking and is not confined to a specific source location. Rough suppliers can offer their rough to clients anywhere and there are no geographic barriers preventing anyone from participating in that market. The major problem facing the EC is defining the product market for rough diamonds and, specifically, whether there exist sub-markets which differentiate sufficiently to deserve being considered a separate relevant market.
To understand the EC’s dilemma it is useful to consider a non-diamond example. In competition law, every beginner’s textbook will cite the famous (U.S. versus Du Pont) case where the outcome of a court case hinged upon whether the product market was defined as “cellophane”, in which case the defendant’s market share would have been 75%, or “flexible wrapping paper”, in which case the share would have been 20%. The court addressed the question as whether flexible wrappings were reasonably interchangeable. It decided on the broader market definition, by relying on the high cross-elasticity between cellophane and other wrapping materials.
The EC questionnaires received by Sightholders specifically asks about “substitutability”. In competition law a product market is composed of products that are reasonably substitutable from the point of view of buyers. The economic tool most commonly used in determining a product’s market is cross-elasticity of demand. Cross-elasticity of demand is a measure of the substitutability of products from a buyer’s perspective. In a more technical sense, it measures the responsiveness of the demand for one product to changes in the price of a different product.
Competition law is concerned with monopolistic pricing. The test in substitutability is at what point consumers (in our case diamond manufacturers) react to price rises by switching from one product to another.
We want to state at the outset that there is no “substitutability” in the rough diamond market. If the market price – set by the monopolist market leader – is too high, the diamond manufacturer has no other product to turn to. [At this point synthetic diamonds are not an available substitute; they might be five years or so down the road. If such synthetics would not be produced by the same dominant seller of natural diamonds, this may impact the competition analysis.]
Thus there should be no illusion: the rough diamond manufacturer does not have the option of substitution. He may close down his business and create a different business (or become a shoe manufacturer), but that is not the type of “substitution” that is relevant under competition law when considering the power of a monopolist.
In any event, in a “sub-market” of rough diamonds, if there exists a sub-market, the manufacturer finds the same diamonds.
Any hypothetical argument that the diamonds in the “primary market” or the “secondary market” are different diamonds will not meet any empirical test. The only differentiation might be “price” – and, looking at the questions asked by the EC, this is something the EC wants to examine in great detail.
Let’s forget for a moment the esoteric legal arguments and look at industry practice. Thereafter, we’ll try to find a “connect” – if there is one – between this practice and competition issues.
A Historical Perspective on the Primary and Secondary Markets
Functionally, there is one market (called the “rough market”) which facilitates the raw materials mined by a handful of producers to reach the sole users of these raw materials, who are the polished diamond manufacturers. That market consists of DTC controlled goods (55% in 2003) and non-DTC goods (45% in 2003). It has become common usage to refer to direct sales from mining conglomerates as “primary market” sales and the re-sale of these goods as taking place on “the secondary market”. Basically, these are perceived as two levels, two phases, in the diamond pipeline.
It is hard to find the origins of the term “secondary market” as used in the diamond industry. According to a De Beers official “the term was coined by De Beers’ directors in reference to non-De Beers goods in the market.” Dating back to the days that De Beers, through the Central Selling Organisation (CSO), controlled some 80% of the rough market, the “primary” market referred to the first sale of CSO goods to the market, and the secondary market referred to the resale of this CSO-rough to other dealers or manufacturers.
The market of non-CSO goods was traditionally called the “open market” and/or “outside goods”, i.e. goods which are traded outside the traditional cartel framework. These goods consisted almost totally of alluvial African goods from digging operations in the DRC, Sierra Leone, Angola, etc. In the past decade, however, the situation in the market has changed.
Major mining conglomerates (Rio Tinto, BHP Billiton and, to some extent, Alrosa) initially became part of the “outside market”. As they adopted basically similar sight-based distribution policies as the CSO as to which part of the rough is allocated either directly to the consumers of rough (i.e. the manufacturers) or to dealers for further distribution, Rio’s and BHP’s sales directly to manufacturers are now also seen as a “primary market” activity.
The time-honored distribution system pioneered by De Beers, and subsequently adopted (with some minor variations) by the other main suppliers, simultaneously followed both a one-level and a two-level model to move the rough from the mining company to the rough diamond manufacturer. There is no doubt about the existence of these arrangements.
The market views that “second level” as a “secondary market” so, from a diamantaire’s perspective the EC question could be answered in the positive – yes, at least semantically and functionally, there is a “secondary market.” Indeed, the latest definition of “secondary market”, as understood by both De Beers and Alrosa is implicit (though not defined) in the De Beers-Alrosa Trade Agreement Commitments, which, under the heading “purchases on the secondary market” contains the undertaking that De Beers will not purchase Alrosa diamonds “from parties other than Alrosa”. So Alrosa’s sales to any non-De Beers entity is seen as sales to the “secondary market” and De Beers will not make purchases of such rough on the secondary market. [How anyone thinks that this can be monitored is a different question; however, that’s not our present concern.] This wording reinforces the view that De Beers, Alrosa – and the market participants – view the secondary market as the “second distribution level” in the sales by producers.
Undoubtedly, some lawyers will argue that the fact that De Beers makes a commitment to refrain from certain activities in that market surely strengthens the view that there exists such a separate market. Indeed, the trade agreements dictate different rules to apply to the primary market and the secondary market. It limits the access by Alrosa to the secondary market and/or limits the DTC’s own ability to purchase on that market. This raises intriguing legal issues. Moreover, for the EC to agree to those commitments may also imply that it agrees that there exists a secondary market, something which might strengthen the case of those who claim that De Beers “is cornering” or “jeopardizing” that market through an (illegal) use of dominant power. [This may sound like semantic nit-picking; however, if I would have been advising the De Beers lawyers, I would have substituted the phrase “sales to the secondary market” simply with “sales to the rough diamond market.” Why provide free ammunition to adversaries?]
DTC Changes Distribution Models
By reducing sales to dealers, by turning dealers into manufacturers, and by selling more goods through Diamdel, the DTC is changing its very own marketing model – and, it is claimed, the DTC’s marketing is therefore more efficient. These policies do not affect the “open market” or the “outside” goods – the marketing models of these goods are not directly impacted by SOC, though, indirectly, many of the SOC policies have been incorporated in the marketing practices of other producers (BHP Billiton, Rio Tinto, Alrosa) who are selling an ever larger part of their productions directly to manufacturers and not to rough dealers.
Empirically, the EC question whether there is a “secondary market” must be answered in the affirmative, or, more precisely, that there has always been one, provided we refer mostly to a “secondary market in DTC goods”. A De Beers sponsored publication, over 30 years old, described the role of the Central Selling Organisation (CSO) and its clients as follows: “There are 270 clients of the CSO. These are the mainstays of the marketing structure. They are men of substance: among them there are master-cutters who by their skills and their long-standing reliability are leaders in the world industry, and there are dealers who have the reputation and financial strength to command international contacts. These clients must be “active” clients.
They must “work” the goods they get from the CSO. In other words, they must get the goods “moving” through the marketing network. It is fundamental to the success of this marketing machinery that these clients push their diamonds out through the network to the “end-users”. The CSO is a central wholesaler: its clients are virtually sub-wholesalers, but they begin the process of mixing and re-mixing which is the essence of distributing the diamonds to those who want special qualities and special shapes and special sizes.”
So, yes, there always existed a secondary market in DTC goods as an integral and functional part of the De Beers marketing system. The dealers, supplying that market, fulfilled specific tasks – which were clearly defined by the CSO.
It may be argued that under Supplier of Choice, with its specific client-tailored assortments, many of these tasks that were previously in the hands of dealers are now fulfilled by the DTC itself. When the aforementioned quoted words were written, in 1970, the CSO controlled some 80% of the rough market. It was the first channel through which the cartel moved its goods into the rough market. We would argue that it was the very marketing model of the CSO and its contracted partners that created a “secondary market in DTC goods”. This was not a market developed by economic forces. It was not a “competitive market”. It was truly an inherent part of the then existing rough marketing system.
For many decades, the CSO would sell the largest sight allocations to rough dealers whose task it was to distribute the rough through smaller specialized dealers or sell the rough to manufacturers. So, again, there can be no doubt that there existed a secondary market for mostly DTC rough diamonds.
Inevitably, the dealers operating in these markets (including Diamdel) would try to widen their selling assortment by also purchasing and reselling goods from the “outside market”, goods which did not originate from the CSO (or, as it is now called, DTC). The dependence on “dealers” also had practical reasons: producers need to sell all qualities and sizes, they didn’t have the skills to make the precise assortments needed by the thousands of players in the diamond industry, and they needed dealers to “move the goods”.
What has changed through Supplier of Choice (SOC) is not “the market”, but the DTC’s “marketing model”, and the changes were forced upon it to a large extent by its gradual loss of market share. Aiming eventually at a diamond industry dominated by not more than some 30-40 major players (in London in-house they refer to the “supertankers”), the need for the dealer function in that model will gradually disappear. Other producers are following a similar model and they tend to sell their rough directly to the “consumers of rough”, i.e. to diamond manufacturers.
The Secondary Market May Eventually Disappear
From a functional perspective, the secondary market exists – and there is no doubt that the present distribution policies of the major producers will lead to a reduction in its size, a reduction of the number of players, and, eventually, this market may virtually disappear. As BHP Billiton, Rio Tinto and others are relatively small producers and they are not dominant in any way or shape, they cannot be accused of an abuse of dominant power. So their behavior is beyond the purview of the EC authorities.
The fact that De Beers is a dominant supplier doesn’t need to be argued – that is self-evident and has been confirmed by the EC. The question which the EC should then ask itself is basically whether the DTC’s new business model that will make many dealers redundant and that may lead to a removal of the “secondary market” as a separate phase in the diamond pipeline, represents an abuse of dominant power or not?
Competition law doesn’t penalize dominancy per se – to the contrary, if a firm is so successful that it expands and produces most of a particular product (such as diamonds) making huge profits, this only will encourage new entrants into the market. The authorities would intervene, however, if these profits are created due to the monopolist’s success in establishing barriers to new entrants on the supply side, in creating obstacles to prevent the expansion of competitors, and/or in avoiding the emerging of adequate product substitutes on the demand side.
There are many elements in the SOC policy, including its policy on clients’ material interests in diamond mining and the Alrosa-De Beers trade agreements, which might be successfully challenged on competition grounds. The functional removal of the secondary market in DTC goods may not be undesirable from a competition perspective. Indeed, competition law will take a different view to the one the market players may subscribe to.
Economists and governments generally view antitrust as a set of laws designed to promote competition and, therefore, economic efficiency. Indeed, in a world of limited resources, the consumer is believed to benefit most from a free and competitive market in which it secure products at a most desirable level. In theory, when the producers eliminate the “secondary market” and move the goods directly to the consumers of rough (i.e. the manufacturers), it may claim that it is enhancing efficiency. The EC will like such an argument – it will like it very much.
Conceivably, one could raise valid counter arguments. The fewer the phases in the diamond value chain, the greater the impact of the DTC on the behavior of the chain, and the greater the ability of De Beers to optimize its own revenues. The consumer doesn’t necessarily get the benefits of lower prices. This is, however, conjecture. One may argue – and the argument is true – that since the launch of Supplier of Choice the end consumer prices paid for polished diamonds have not really gone up; they certainly have not been increased by higher prices of rough diamonds on both the primary and secondary rough markets. What we have seen is a reduction in margins by retailers and greater efficiencies in the diamond value chain. Each and every diamond manufacturer is producing polished today in a far more cost-effective manner than a few years ago.
It is unfortunate that the DTC is “mumbling” – it constantly repeats statements about the importance of secondary market, about the role of dealers, etc. I don’t believe these assertions and I am not sure whether DTC executives don’t privately hold views that stand in contrast to their official position. Functionally we expect that the changing distribution models will make the secondary market extinct. As an efficiency issue – the EC will probably welcome it.
But this is not the subject of this article. We want to establish whether this endangered species, i.e. the secondary market, does exist under competition law.
Supply and Demand Side Substitution
We have already argued that there is no demand side substitution: if the monopolistic price is too high, the diamond manufacturer doesn’t really have the alternative of buying a different product. Diamonds are unique. There remains a question whether there is substitution on the supply side - whether competing producers could increase production, or move to other products, or whether the high price could induce new players to enter into the market. Here too the answer is mostly negative.
There is no substitution in the rough market – and that makes it irrelevant whether that is the determination of one single rough market or a rough market that has several “sub-groups” (such as a secondary market). If we view the diamond “primary market” and “secondary market” as “two markets”, there is no doubt that there is full product substitutability – because the same products are traded in each of these two markets. Isn’t that, by itself, the overriding proof that there is only one market?
We not only have the same product in each “market”, but we also have the same players, both in terms of buyers and sellers. Virtually each and every diamond manufacturer who purchases diamonds from De Beers (or from Diamdel) also purchases rough diamonds from other sources. So most “consumers” of rough (measured by value) are active in both the primary and secondary market. [SOC, by reducing the number of Sightholders, is increasing the number of players which have no access to primary supplies – but SOC is not reducing the total supplies by De Beers to the market.]
Diamdel, which distributes De Beers' diamonds, is also active in the non-De Beers market, and makes purchases on the “open market”. We would estimate, for example, that about a third of the 2004 rough diamond sales of Diamdel in Israel represent non-DTC goods.
The complaints filed with the EC assert that members of the secondary market (dealers) are denied access to the primary market. This is undoubtedly true. But what we have learned from the cellophane judgment and many other decisions since is that the authorities will take the “broader view”. The exclusion of certain players in one or more sub-groups doesn’t make these sub-groups “separate markets.” What we see today is that larger companies tend to be active on both markets (as buyers), while the smaller companies tend to be mostly active on the secondary market. However, if the only difference is the “size of the players”, this is not relevant in terms of competition law.
One Remaining Issue: Price Differentials
What may be relevant is price. In a natural pipeline or value-chain procession, each subsequent level should see an added value, and should record a higher price. If there are separate “primary” and “secondary” markets, the diamonds available on the secondary market should be more expensive than those on the primary level. That is often the case – but, especially in recent years, this is certainly not a rule. Today, most (or many) BHP Billiton and Rio Tinto goods are sold on the secondary market at a discount from the primary market. Traders are willing to take a loss just to maintain their sight eligibilities.
The situation of De Beers is, in a sense, peculiar. Through SOC and the clients’ marketing programs the DTC is supposedly selling custom-tailored boxes for clients’ requirements. However, today more DTC boxes are traded on the secondary market than in any other time in recent years. Moreover, some DTC clients have already sold their boxes forward at an advance agreed premium. Alrosa’s supplies to the secondary market (to factories) in Russia are at a discount to its selling prices to De Beers, while its open market tenders generate huge premiums (up to 30%) over its selling price to De Beers.
In 2004 we also witnessed considerable speculation: diamond dealers were hoarding boxes from the primary market in the hope to resell them a few months later at huge premiums on the secondary market. As the market took a turn southward, most of these goods were sold at a loss.
From an EC perspective, it is important to note that there is considerable arbitrage between the markets. A DTC client, who feels that he can optimize his profits by selling his DTC box and purchasing the goods he needs much cheaper on the secondary market, will do so. And this is very often the case at the moment. The price behavior in both markets is impacted by the same external and internal market forces; these forces apply to all players.
The fact that the traders may invariably belong to “both markets” and that there are ample opportunities for both trade and arbitrage would further point to the existence of only “one rough market” from an EC perspective. One cannot escape the conclusion that these two markets which functionally do exist, are likely to be viewed as “one market” by the EC authorities. It will be very difficult to argue otherwise.
DTC to Impose Service Fee
Over the past 5 years, leading up to and including the launch of Supplier of Choice (SoC), the DTC has invested extensively in providing valuable services to Sightholders in order to facilitate the objectives of SoC. These services have enabled Sightholders to create value in their own businesses by focusing on generating efficiencies, developing added value services, building effective marketing activities to help generate incremental demand for diamond jewelry and supporting trade professionalism.
However, in order that the DTC can viably maintain and re-invest in enhancing its operational services that are integral to the operation of SoC, in the coming 2005 contract period the DTC is planning to introduce a fee for these services. The DTC is still in the process of finalising the arrangements and it is expected to communicate shortly further details as to how it intends the service fee will operate, including clarification of what it covers. The market speculates that the fee will be 2% of the invoice value of the sights, but that has not yet been officially confirmed.
DTC Director of Sales Varda Shine says she “strongly believes that the formalisation of DTC’s service offer on a fee-paying basis not only reflects the reality of the many valuable operational services the DTC provides in addition to the provision of rough, but will also enable the DTC to develop its service proposition still further. It will also support DTC’s efforts to continuously deliver these services to the highest possible standards of professionalism and integrity and focus on creating value to our clients. For all these reasons, we anticipate that maintaining and enhancing DTC’s services in this way will have a positive effect upon market growth, industry efficiency, downstream transformation and sightholder profitability.”
Ms. Shine further disclosed that the DTC is currently exploring new services to offer to clients in the future. Sightholders would be able to apply for additional services from a menu of added-value services that are separate and distinct from the ones described above. These so-called Discretionary Value Added Services (DVAS) will be offered on a voluntary basis according to clearly identifiable criteria. These services are entirely separate from, and have nothing to do with, the assessment of eligibility for appointment as a sightholder or the level of goods that will be allocated to each sightholder during the term of its contract with the DTC.
DTC Demands “Absolute Access” to Client Factories