---

COMPETITION TRIBUNAL

REPUBLIC OF SOUTH AFRICA

Case Number: 31/IR/Apr04

In the matter between

Nuco Chrome (Pty) Ltd Applicant

and

Xstrata South Africa (Pty) Ltd 1^st Respondent

Rand York Minerals (Pty) Ltd 2^nd Respondent

Reasons

Introduction

This is an application for interim relief in terms of section 49C of the Competition Act, 1998 as amended (“the Act”). The application was heard on the 14^th May 2004. On the19th of May 2004 we dismissed the application.

The product at the center of this dispute is chrome sand. It is generated in the process of mining chrome ore and is a key input in the production of ferrochrome. The lion’s share of chrome sand is utilized in the production of ferrochrome – accordingly, the vast bulk of chrome sand is not traded but is rather produced and consumed in the vertically integrated process of mining chrome ore and producing ferrochrome.

However, a residual quantity of chrome sand is available to users other than the vertically integrated ferrochrome producers. Although the precise size of this residual is a function of the fluctuating requirements of ferrochrome production, it appears that there is an established market in chrome sand which, in addition to its role as an input in the production of ferrochrome, is also used in certain foundry processes as well as in the production of certain chemical and pharmaceutical products.

This residual trade in chrome sand has created an opportunity for middlemen who, in purchasing chrome sand for on-sale to the end users, add a minimal value, largely in the form of drying and bagging the product. It appears that some of these middlemen act as wholesalers, while others act as agents of the ferrochrome producers. There is also evidence that the ferrochrome producers sell directly to end-users.¹ Be that as it may, there is clearly an active trade in chrome sand, with both domestic and international end-users active purchasers of chrome sand produced in South Africa.

The applicant, Nuco Chrome (Pty) Ltd (“Nuco Chrome”), a middleman in the purchase and sale of chrome sand, has, since 1997, purchased all of its chrome sand requirements in bulk from the first respondent, Xstrata South Africa (Pty) Ltd (“Xstrata”), a large, vertically integrated ferrochrome producer, and, hence, producer of chrome sand. Xstrata extracts chrome ore from its mines and processes the ore through beneficiation plants to yield lumps, pebbles and fines. It then further processes the fines into three grades of chrome sand namely metallurgical grade sand, which is used in the production of ferrochrome, chemical grade sand, which is mostly used in the chemical and pharmaceutical industries, and foundry grade, which foundries use to line moulds and tap-holes in the casting process.² It appears that the other suppliers of chrome sand in South Africa are Samancor, SA Chrome, Bayer, Groot Marico, Dilikon ASA Metals, National Manganese, Assore and Angloplats.

Nuco Chrome dries and re-packs the chrome sand purchased from Xstrata in its own bags and re-sells it to its local customers in South Africa. The second respondent, Rand York Minerals, another middleman, also purchases all of its supplies of chrome sand from Xstrata.

We have not been told much about Nuco Chrome’s other competitors, local or foreign, except that Xstrata’s largest competitor, Samancor, sells its chrome sand to Mineralloy, a competitor of Nuco Chrome in the local market. Nor were we told much about Nuco Chrome’s downstream customers. All we know is that Nuco Chrome supplies chrome sand to local customers only, while it appears that Rand York focuses largely on the export market.

Refusal to supply

Historically Xstrata has supplied Nuco Chrome with all three grades of chrome sand. It is now common cause that Xstrata no longer produces chemical grade sand, and Nuco Chrome advised us at the hearing that it will no longer persist in its claim in respect of that grade. We need only concern ourselves then with metallurgical and foundry grade sand.

The history of the alleged refusal is murky. When it began, why it began, and whether it amounted to a total refusal or a scaling back in supply, are not only a matter of dispute between the parties but also a matter of inconsistency in the version presented by Nuco Chrome.

In its founding papers Nuco Chrome alleges that it was expanding in the market until Xstrata stopped supplying it with metallurgical grade sand, in April 2003.³ We do not know the reason for this cessation of supply. In September 2003 Xstrata resumed its supply to Nuco Chrome. However, the amounts were far below the tonnage requested. In October and November 2003, it stopped supplying but then again resumed supply in December 2003 to February 2004 but again in an amount insufficient for Nuco Chrome's needs. Matters came to a head in December 2003 when Xstrata wrote to Nuco Chrome to inform it that it could only guarantee a maximum of 500 metric tons per month of metallurgical grade sand.⁴ On 1 March 2004 a meeting took place between Xstrata’s chairman, Mr Nienaber, and Messrs Van Zyl and Butler of Nuco Chrome. According to Nuco Chrome it was told at this meeting that Xstrata would not supply it any longer with metallurgical grade sand. Xstrata was “slowly starving it to the point that it is no longer able to carry on its business in which it was expanding”, alleges Nuco Chrome.⁵

In respect of foundry grade sand Nuco Chrome was supplied with less than it ordered.

Xstrata's version is that there has never been a refusal to supply. Rather, supply constraints have meant that it could not always meet the demands of the market. This is evidenced by letters that Xstrata wrote on 15 December 2003, as well as 2 March 2004, in which it undertook to supply Nuco Chrome with a maximum of 500 tons of metallurgical as well as foundry grade sand.⁶

In its answering affidavit Xstrata informed the Tribunal that it intended to cease supplying Nuco Chrome from July 2004 because it needed more chrome sand for its own ferrochrome smelters and also because it did not consider Nuco Chrome a suitable distributor.⁷

It is thus common cause that-

1. at the latest from July 2004, Xstrata will cease supplying Nuco Chrome with any grade of sand;

2. that for some time, at any rate at least since April 2003, supplies to Nuco Chrome have been sporadic and insufficient for its needs.

Xstrata’s refusal to supply, alleges Nuco Chrome, effectively arises from a supply agreement between Xstrata and Rand York. In consequence, it is then further alleged, this agreement contravenes Section 5(1) of the Act, the section, which proscribes anti-competitive vertical arrangements. Nuco Chrome also alleges that Xstrata is a dominant supplier of chrome sand and, hence, that the refusal to supply constitutes an abuse of that dominance, thus offending against various provisions of Section 8 of the Act, the section dealing with abuse of dominance. Finally Nuco Chrome alleges that this dominant firm – Xstrata – is supplying chrome sand to Rand York at prices lower than those at which it supplies Nuco Chrome and that this contravenes Section 9 of the Act, the section which proscribes certain forms of price discrimination.

Xstrata opposed the application. It contended that its unwillingness or inability to supply chrome sand to Nuco Chrome did not derive from its agreement with Rand York but was rather dictated by the level of its own needs as a ferrochrome producer. It furthermore denied that it was a dominant producer of chrome sand or that the alleged price differential referred to by Nuco Chrome constituted price discrimination as defined in the Act.

The second respondent, Rand York, did not participate in the hearing.

Relief sought

Nuco Chrome seeks the following relief:

“That the Respondent be ordered to:

1. Supply the Applicant with its requirements of foundry grade, and chemical grade, and metallurgical grade Chrome sand as will be ordered by the applicant from the first respondent from time to time – sections 5(1) and 8(c);

2. Not give preference in the supply of the said foundry grade and/or chemical grade and/or metallurgical grade Chrome sand to any of its customers including the Second Respondent – sections 5(1) and 8(c);

3. Charge the applicant for the foundry grade and/or chemical grade and/or metallurgical grade Chrome sand the same prices that it charges its other customers including the second respondent for similar quantities ordered – sections 8(a), 9(1)(b) and 9(1)(c)(1);

4. Charge the applicant for each of the foundry grade and chemical grade and metallurgical grade Chrome sand only reasonable prices based on the cost of production and allowing the first respondent a reasonable mark up – section 8(a);

5. Increase the price it charges the applicant for each of the foundry grade and chemical grade and metallurgical grade Chrome sand only after giving the applicant at least one full calendar months notice thereof, the increases to be no larger than is justified by paragraph 2.4 hereof – sections 8(a) and 8(c);

6. Extend to the applicant at least the same credit terms it gives to its other most favoured customer – section 9(1)(c)(iv).”

During the course of the hearing Nuco Chrome indicated that it was abandoning prayers 1 and 4 above, and that, consequently, the words “the said” in the first line of prayer 2 should be deleted and the words “the increases to be no larger than is justified by paragraph 2.4 hereof” should be deleted in prayer 5. Thus par 2, 3, 5 and 6, as amended, remain as the relief sought by it. The relief sought thus reads as follows:

That the Respondent be ordered to:

1. Not give preference in the supply of foundry grade and/or chemical grade and/or metallurgical grade Chrome sand to any of its customers including the Second Respondent – sections 5(1) and 8(c);

2. Charge the applicant for the foundry grade and/or chemical grade and/or metallurgical grade Chrome sand the same prices that it charges its other customers including the second respondent for similar quantities ordered – sections 8(a), 9(1)(b) and 9(1)(c)(1);

3. Increase the price it charges the applicant for each of the foundry grade and chemical grade and metallurgical grade Chrome sand only after giving the applicant at least one full calendar months notice thereof – sections 8(a) and 8(c);

4. Extend to the applicant at least the same credit terms it gives to its other most favoured customer – section 9(1)(c)(iv).

In response to queries regarding the above prayer 1 and, in particular, in relation to the meaning of ‘not give preference’, Nuco Chrome explained that this was to ensure fair treatment in relation to all of Xstrata’s customers. Hence, were Xstrata to have only say 10000 tons of chrome sand available for sale, and it received two separate orders exceeding the total amount available, it would pro rate the supply to each customer.

Xstrata nevertheless insisted that the relief sought is vague and imprecise. However, since the application for interim relief has failed we need not address this. We now turn to the reasons for dismissing the application.

Requirements for interim relief

The requirements for interim relief in terms of section 49C are set out in section 49C(2)(b) and are similar to the requirements for an interim interdict at common law:

The Competition Tribunal may grant an interim order if it is reasonable and just to do so, having regard to the following factors:

1. The evidence relating to the alleged prohibited practice;

2. The need to prevent serious or irreparable damage to the applicant; and

3. The balance of convenience

The standard of proof required is less exacting than the normal burden of a balance of probabilities in civil litigation for permanent relief.⁸

While the requirement of Section 49(2)(b) that we ‘have regard’ to the three factors listed above suggests that a strong positive finding on two factors may outweigh a lesser or possibly a negative finding on the third, we would, as we have observed elsewhere, be extremely reluctant to uphold an application for interim relief in the absence of evidence confirming the restrictive practice alleged.⁹

In this matter we have found that Nuco Chrome has not made out a case that a prohibited practice has occurred, and it is therefore not necessary for us to deal with the remaining factors in section 49C(2)((b)(ii) and (iii).

Evidence of a prohibited practice

The allegations in this matter effectively focus on a refusal to deal, and in particular on a refusal by a supplier to provide a customer with an input that the latter requires for the conduct of its business. As already noted, Nuco Chrome alleges that the refusal to deal derives from a supply agreement between Xstrata and Rand York that had the effect of excluding Nuco Chrome from the market. For this reason, argues Nuco Chrome, the agreement falls to be prohibited in terms of Section 5, the section of the Act that proscribes restrictive vertical agreements.

Nuco Chrome also argues that the alleged refusal is the unilateral act of a dominant firm and, as such, contravenes Sections 8(c) which prohibits a ‘dominant firm’ from engaging in an ‘exclusionary act’ where the anti-competitive effect of that act outweighs its technological, efficiency or other pro-competitive gains.

Finally, Nuco Chrome alleges that Xstrata’s pricing policy contravenes the Act on two counts. Firstly, it alleges that Xstrata is charging ‘excessive’ prices in contravention of Section 8(a) of the Act. Secondly, it alleges that Xstrata is engaging in discriminatory pricing in contravention of Section 9 of the Act.

In order to sustain a claim under both sections 8 and 9, it is necessary to establish that the alleged perpetrator is dominant in the relevant market.

Section 5

Section 5(1) of the Act reads:

An agreement between parties in a vertical relationship is prohibited if it has the effect of substantially lessening or preventing competition in a market, unless a party to the agreement can prove that any technological, efficiency or other pro-competitive gain, resulting from that agreement outweighs that effect.

In order to sustain a claim under this section of the Act, a number of elements have to be established. An agreement must be in existence, it must be between vertically related parties, and its effect must be to substantially lessen competition in a market. If each of these elements is established then the participants to the agreement are entitled to avail themselves of a pro-competitive defense.

We accept that there is an agreement in existence between Xstrata and Rand York and that they are in a vertical relationship to one another. Nuco Chrome insists that the agreement effectively precludes Xstrata from supplying distributors other than Rand York, because of the quantities that it has committed to supply and because of the lengthy duration of the agreement.

However, the evidence and assertions regarding the status and terms of the agreement are in dispute. Nuco Chrome alleges that Xstrata has entered into a written agreement with Rand York, the material terms of which is that it has undertaken to supply Rand York with 10 000 tons of chrome sand per month and that the agreement is of three years duration.

Xstrata denies that a written supply agreement exists. It alleges that at one time it intended entering into a fixed-term supply agreement for three years with Rand York, which was meant to commence in January 2002 and continue until the end of December 2004, but the agreement was never signed. The reason for this, it explains, was that it did not want to commit itself to supplying 120 000 tons of foundry grade sand per annum in the current market conditions. What it has done instead was to supply Rand York on a quarterly basis, the terms of which were confirmed by letter each quarter, in advance. A typical example of this is a letter dated 15 December 2003, (see page 172 of the record), which states:

“ The new price will come into effect for all orders placed from 1 January 2004 to 31 March 2004. The price will be again reviewed during March 2004.

"Xstrata South Africa (Pty) Ltd guarantees 500 metric tons maximum per month.”

Nuco Chrome in reply has not been able to dispute Xstrata’s version of the nature of the agreement nor has it been able to put up any version of what it considers to be the contract in operation. For this reason we must prefer the version of Xstrata.

Having placed the status of its ‘agreement’ with Rand York in this context, Xstrata argues that the attenuation of Nuco Chrome’s supply is a function of its policy in respect of the amount of chrome sand it wishes to make available to the market. In other words, it is not the “agreement” between it and Rand York that is responsible for a reduction in Nuco Chrome’s supply of chrome sand, but rather a unilateral decision, thus placing the reduction in supply beyond the bounds of Section 5. The reason for this decision is the current boom in the international commodities market, which is reflected in an increase in demand for ferrochrome, which is an input in the production of stainless steel. Xstrata thus needed to retain the chrome sand that it produced for its own ferrochrome smelters.

In addition to the issue of the agreement, our dismissal of this claim rests on Nuco Chrome’s failure to prove – even under the lower standards associated with an application for interim relief – that the agreement has substantially lessened competition in a market.

Nuco Chrome has not clearly identified the market in question. Its assertions on this subject suggest that the middleman is an essential aspect of its functioning. Hence it has defined the market as that for ‘the supply of chrome sand to middlemen in South Africa for supply by them to the local market for chrome sand and for export.’

In our view the market is more accurately defined as the market for chrome sand. The applicant and other middlemen effectively provide the participants in this market with a distribution service which they achieve by inserting themselves between the buyers and sellers, by, in other words, performing a wholesaling function. However, should they attempt to extract a higher price for their service – this would effectively mean either compelling the sellers to accept a price reduction or the buyers a price increase – there is no reason to expect the wholesaling mode of distribution to remain inviolate. There are other modes of distribution.¹⁰ Above all, there is always the prospect that the sellers and buyers engage directly with each other in transacting their business. This is, after all, a business in which competition between sellers of chrome sand appears virtually non-existent and in which there is a small number of end users. In circumstances such as these, should the middlemen attempt to extract an undue return for their minor contribution, they will be quickly by-passed.¹¹

But, assuming for the moment that the wholesalers are an essential element in the distribution of chrome sand and that the effect of the agreement is to remove Nuco Chrome from the market, let us examine the claim that by eliminating wholesalers from the market competition is substantially lessened.

Nuco Chrome rests its case on one simple assertion: it avers that it, in common with Rand York, relies on Xstrata for its supply of chrome sand. Hence, the argument continues, the refusal removes it, a competitor, from the market for the supply of chrome sand.¹² It claims that users of chrome sand (it presumably means, South African users) have only three possible sources of the product, these being Nuco Chrome and Rand York, who rely on Xstrata for their supply of chrome sand, and a company called Mineral Alloys, which receives its supplies of chrome sand from Samancor, the largest producer of Ferrochrome in South Africa.¹³

The evidence relied upon by Nuco Chrome in support of its various allegations concerning the relevant market is conspicuously sparse and confusingly presented. Hence, although Nuco Chrome appears to insist that there are only two local sources of chrome sand, namely Xstrata and Samancor, evidence brought by Nuco Chrome contradicts this assertion – a media report, attached to its Heads of Argument, states that ‘South Africa boasts 10 large producers of chrome sand’, noting that ‘many of these producers do not supply to small volume clients’.¹⁴

Samancor is clearly South Africa’s largest producer of chrome sand. While it appears to market its residual chrome sand output through a single middleman, Mineral Alloys, we do not know whether or not this is an exclusive arrangement. Indeed, Nuco Chrome averred that it could not access Samancor chrome sand because of a commercial dispute between itself and Samancor, rather than because of the operation of an exclusive dealing arrangement with Mineral Alloys. Accordingly, on the evidence presented we cannot properly assess whether Nuco Chrome, a middleman, has alternative sources of supply of chrome sand.

We know even less of the downstream market, the market in which the end users purchase chrome sand from the middlemen. As already noted, of the three South African middlemen identified, at least one, Rand York, appears to on-sell exclusively to customers outside South Africa. Nor do we know whether all of the residual supply of chrome sand is marketed through middlemen or whether, either currently, or, pertinently, in the absence of an effective middleman, chrome sand would be traded in a direct exchange between the producer of chrome sand and the various end users.

Not a shred of evidence has been submitted regarding these end users, the downstream customers. We do not even know the identity of any of them. All that we have been told that is that chrome sand is used in certain foundry processes and in the production of some unspecified chemical and pharmaceutical products. Moreover, we do not know whether it is substitutable by any other intermediate products. Again, our assumption, given South Africa’s large position in the world chrome market, is that, the vagaries of the ferrochrome market notwithstanding, South Africa produces chrome sand in excess of local requirements and, hence, the attraction of the international market for a middleman like Rand York.

Finally, no coherent evidence has been submitted regarding the pricing of chrome sand. Our assumption – and this was put to the hearing and not contradicted – is that the price of chrome sand is largely derived from the ferrochrome price. As already noted, there is a very low value added in the distribution process, i.e. by middlemen. In fact, Nuco Chrome says in its papers that it can only compete with other middlemen on service and price.¹⁵ However, we are not told what the competitive market price for chrome sand is and how Nuco Chrome’s customers would react to non-competitive pricing in the market.¹⁶ Would and do Nuco Chrome’s customers have the choice of moving to other suppliers if it increased its prices? We were not supplied with any of this information.¹⁷

All this is, needless to say, precisely the sort of evidence that would help establish whether or not the agreement had indeed diminished competition. Why is Nuco Chrome not able to access alternative supplies of chrome sand? If Nuco Chrome is forced to exit the market what impact will this have on the supply and price of chrome sand to the various end users? Indeed, noting that we have already gone through a period when Nuco Chrome’s supplies from Xstrata have been reduced, has this had a discernible impact on the availability and price of chrome sand in South Africa?

For reasons best known to itself Nuco Chrome has chosen not to submit basic evidence essential to the adjudication of the claims that it has itself placed before the Tribunal. Section 5(1) clearly places the onus for establishing the anti-competitive effect of the agreement complained of on the applicant, and its failure to discharge this onus must lead us to reject the application.

Section 8(c)

Section 8(c) reads:

It is prohibited for a dominant firm to –

(c) engage in an exclusionary act, other than an act listed in paragraph (d), if the anti-competitive effect of that act outweighs its technological, efficiency or other pro-competitive gain

In order to sustain a claim under Section 8, the applicant must establish that the perpetrator of the alleged abuse is a dominant firm. Section 7 of the Act provides that

A firm is dominant in a market if –

1. It has at least 45% of that market;

2. it has at least 35%, but less than 45%,of that market, unless it can show that it does not have market power; or

3. it has less than 35% of that market, but has market power.

Again, we have not been presented with evidence that enables us to establish the existence of dominance.

As indicated at the outset, the lion’s share of chrome sand that is produced never enters the market but is instead consumed in the vertically integrated process of mining chrome and producing ferrochrome. The market for chrome sand then consists of that residual quantity that is not utilized in the production of ferrochrome and that is sold to other users of the product, principally for use in foundry processes and in the production of certain chemical and pharmaceutical products.

We have no way of knowing whether Xstrata, the alleged perpetrator of the abuse complained of, is a dominant firm as defined. No coherent evidence has been presented regarding its market share. We do know that it is not the largest supplier of chrome sand to the market, a position occupied by Samancor.

In fact, the allegation of dominance appears to be based on the argument that, because Nuco Chrome depends on Xstrata for its supply of chrome sand, the latter is accordingly, dominant in relation to Nuco Chrome.¹⁸

However dominance is not based on so solipsistic a view of the world. It is an objective measure. That Nuco Chrome is not able to procure supplies of chrome sand from Samancor, either because, as it is variously alleged, Samancor has an exclusive distribution arrangement with Mineral Alloys, or because it is engaged in a commercial conflict with Nuco Chrome, does not remove Samancor’s supply of chrome sand from the market. It merely removes it from the grasp of Nuco Chrome (who may then be able to justify a refusal to supply claim against Samancor) but it does not justify arbitrarily placing Samancor’s supply of chrome sand beyond the boundaries of the market. It is a decidedly peculiar argument that asserts that there are, for example, five suppliers in the market, each of whom possesses a 20% market share. Four of these refuse, for whatever reason, to supply a willing purchaser. For that purchaser the market then effectively shrinks to one supplier who, by virtue of the refusal of the other four to supply, becomes a dominant firm, indeed a monopolist, so that when it too refuses to supply becomes vulnerable to an abuse of dominance claim.¹⁹

This does not establish the basis for a claim of dominance. Nor, we should add, would this be sufficient to sustain a claim under Section 8(c) – the abuse of that dominance. For the purpose of Section 8(c), the exclusionary conduct would have to be established.

Conclusion

In view of the fact that Nuco Chrome has failed to make out a prima facie case that the agreement between Xstrata and Rand York has the effect of substantially lessening or preventing competition and has failed to present sufficient evidence that Xstrata is a dominant supplier of chrome sand, we find that Nuco Chrome’s claims under sections 5(1) and 8(c) and 9(1) cannot succeed.

The application is accordingly dismissed.

Costs

We order that Nuco Chrome pay Xstrata’s costs in the application on a party and party scale, including the costs of two legal representatives.

_____________ 18 August 2004

D.H. Lewis Date

Concurring: N.M. Manoim; L.P. Reyburn

---