"that theCompetition Act 1998 has been in force since March 2000; that it prohibits anti-competitive agreements and abuses of a dominant position; and that in the event of a breach of either prohibition measures may be taken, including the imposition of a fine. The Group has no locus to consider, and has consequently reached no view on, whether either prohibition has been breached by any of the companies mentioned in this report."
"The Secretary of State has asked the [Director] to keep the market under close review with regard to potential infringements of the prohibitions in the [1998 Act]."
"The OFT has accepted informal interim assurances from Robert Wiseman & Sons Ltd and Robert Wiseman Dairies PLC relating to milk sales in the Highlands of Scotland. Wiseman has agreed to cover its cost of supply to each middle-ground customer it sells to in the Highlands of Scotland. (Middle-ground customers are those customers who are not supermarkets or supplied by doorstep deliveries.) This will ensure that competition for those customers is protected while the OFT carries out its current investigation under the Competition Act. In order that Wiseman's competitors can bid for each customer's Highlands outlets, the interim assurances mean that some national contracts may have to be split."
"I am writing to let you know the position on the Office of Fair Trading investigation into Robert Wiseman and Sons Ltd and Robert Wiseman Dairies PLC ("
"Investigation on suspected Chapter II infringement Substantial efforts and resources have been devoted to the investigation. It focused on the following alleged behaviours: 1) predatory behaviour; 2) price discrimination (involving exclusionary or excessive pricing); 3) exclusive dealing. We asked Wiseman to provide a substantial amount of information under several section 26 requests. We analysed data on the monthly prices charged, since November 1998, for a selection of 10 products, to 800 customers across Scotland. We also looked at costs and other information for the same products and customers covering over 1000 individual outlets for three different months (May 2000, November 2000, and May 2001). The sample for this data was carefully selected from a complete set of Wiseman's customers (including postcodes of individual outlets and details of products and volumes delivered). This was to ensure we had sufficient information to obtain statistically valid results, and as far as possible to minimise the burden on Wiseman. Evidence was also collected during a section 28 on site investigation on the premises of Wiseman and section 26 notices were sent to farmers in the Highlands. As well as the extensive information collected from Wiseman, we also used other independent sources of information, when available, to compare the results of the analysis of the data. The analysis of the information collected for the investigation required us to allocate an additional Principal Case Officer to the case from January. Specialised software was also bought to carry out the quantitative analysis. Consultations inside and outside the OFT with expert practitioners have been held at different stages of the investigation to assess the quality of the information and of its analysis. We investigated several dimensions of price patterns: -- price evolution over the period November 1998 – September 2001, factoring for raw milk input price, by customer and by product type; -- geographical pattern of prices in Scotland, by customer outlet for May and November 2000 and May 2001; -- price differentials between customers and according to volume; -- comparison of price policy with other milk suppliers; We also investigated total and variable costs for each outlet served: -- comparison of costs to selected measures of cost for other producers; -- the schedule of price-cost margins in relation to volumes, by product type and on aggregate; -- the price-cost margin differentials by customer/outlet; -- the geographical pattern of outlets' price-cost margins in Scotland and in selected areas (by product type and on aggregate); -- regression analysis of price-cost margins on volume, customer types, product type, and other factors. Market Definition We did find persuasive evidence that the middle-ground market in Scotland is a distinct market. On the basis of the information provided in the Competition Commission's Scottish Milk Report, of interviews with competitors, and price patterns, we would have been likely to conclude that Wiseman had a dominant position on this market. Alleged Abuse Our investigation into Wiseman's alleged infringement of the Chapter II prohibition looked for sufficiently persuasive evidence of the alleged abusive behaviours. We are not however reaching any final view on these points because the evidence gathered during the investigation is not sufficiently persuasive as to the existence or absence of an infringement. Predation We found instances of pricing below total cost, for example in the health care/hospital sector, which appeared to diverge from regular pricing behaviour, but we do not think these instances could support a conclusion that Wiseman had engaged in predatory behaviour. Furthermore, the investigation did not uncover sufficiently persuasive evidence either way of intent to exclude competitors. Price discrimination (involving exclusionary or excessive pricing) We also found evidence of price discrimination. However, there is insufficient evidence to come to the conclusion as to whether or not it departs significantly from normal competitive behaviour, it constitutes targeted pricing or has exploitative or exclusionary effects. Exclusive supply contracts As you know these were covered in detail by the Competition Commission and we studied its report and conclusions carefully. We also looked closely at the material you provided in your letter of 5 August in relation to Aberness. We looked at whether the all Scotland deals entered into by Wiseman were in fact based on exclusive supply either explicitly or by offering additional incentives for exclusivity. However, the evidence available was not sufficiently persuasive to lead us to think that we would be able to make a finding as to whether or not the contracts in question amounted to abuse. I hope you find the above constructive and helpful. I shall be on leave for the next three weeks. If you have any queries please feel free to call Donald Mason or any of the case officers in my absence or get back to me on my return."
"I think that it must go without saying that the OFT will consider all complaints about and evidence of abuse under theCompetition Act 1998 that is submitted to us. However, I think it only right to express my view that in this case, given its history over the last three to four years, we would need to have persuasive if not compelling evidence of abuse before we would be likely to devote significant administrative resources to further investigation of Wiseman's behaviour in the market."
"42. We took two measures of variable costs. We took as the low measure of variable costs the sum of ingredients, drivers' wages, fuel costs and tyres as we believed that those costs are directly related to the supply of the product and can reasonably be avoided in the short term. For the sake of robustness, a high measure of variable costs was also computed, adding packaging and processing costs to the previous measure. This high measure includes some elements of fixed costs, but likewise, most of these costs can be avoided in the short term. Neither measure includes the variable costs of depots, but we can assume these to be mostly unavoidable, with a significant proportion of fixed costs. Besides, depot costs are relatively low and do not vary significantly. Taking the May 2001 cost data, we obtained for each measure 2646 measures of price cost margins (1150 outlets, each taking several products, with some gaps in the data). These were analysed on a number of different bases and I attach as Annexes A to V to RBL1 pages 116-118 a number of maps, graphs and tables that were produced for the analysis and which we found particularly illuminating, along with an explanation of each. 43. The low measure of variable costs led to a finding that no negative price cost margins were found at the outlet level. The high measure of variable costs led to a finding of negative price cost margins in 33 instances. This was a negligible proportion of the total of the 2646 observations, and the majority belonged to small customers so that the volume affected was lower than that proportion would suggest (0.8 per cent). We did not feel able to build any robust conclusion on that result. 44. However, we took the matter further by looking to see whether there was any customer for whom supplies were made at prices below variable costs on the high measure. The lowest margins were for customers in the health sector as illustrated by Annexes H and I to RBL1 pages 116-118. All these customers were located outside the area of Express/Claymore's principal concern in the Highlands, most of them being close to Wiseman dairies and to population centres. See Annex K to RBL1 pages 116-118 which shows the location of Wiseman's health sector customers in the sample. 45. Overall, therefore, the examination of variable costs uncovered hardly any evidence of below cost pricing. Instances of very low, possibly below cost level, were recorded at outlet level but we found it impossible to regard these as significant in the overall context of the case. When examined at the customer level, prices charged to customers covered average variable costs in every instance and on both measures. 46. We then examined whether there was pricing below total costs. 47. To do this, we repeated the steps described above in respect of total costs. We adopted the approach of aggregating every one of the costs requested from Wiseman, adding 3 per cent to reflect financial administration and central selling costs, and using that measure as a working proxy for total cost (although some elements of total cost – possibly up to 5 per cent - were left out of that proxy) ... 48. On that basis, we found, in respect of May 2001, 495 instances of below total cost pricing at the outlet level for 3039 records (1150 outlets, each buying several products). This number seemed relatively significant and raised some cause for concern. As our measure of cost was probably a slight underestimate of the true total costs, we considered this evidence as a first indication of possibly significant pricing below total costs. In addition, the relative high frequency of below or near total cost pricing seemed to indicate either intense competition or possibly anti-competitive behaviour. 49. However, when we started to perform a qualitative analysis of the instances of below cost pricing in terms of type of outlet and geographical location, it soon became clear that it would not be possible to identify any pattern that indicated a predatory strategy against Claymore based on any commercial logic that we could identify to the standard of evidence needed to show a Chapter II infringement. A breakdown of average price/total cost margin by customer group is shown in the table at Annex J to RBL1 pages 116-118. Our analysis showed that negative values of the price/total cost margin were concentrated in customers belonging to the health sector (hospitals, health boards, health care). Those customers are relatively large customers, but not the largest, and at around 5 per cent of the total volume sampled, accounted for a relatively small share of supply. It is difficult to see any reason why Wiseman would adopt a strategy of trying to exclude Express/Claymore from the market by targeting the health sector. It should also be noted that the instances of selling below total cost in the health care sector were all outside the Highlands area, the area of Express/Claymore's principal concern. 50. We also took the view that there was no clear and compelling direct evidence of intent by Wiseman to eliminate Express/Claymore from the market. As I have said above, we did regard some of the documents obtained from Wiseman during the section 28 investigation on10 May 2001 as showing that Wiseman tended to categorise Express/Claymore differently from other competitors and hence as lending some weight to a suspicion of exclusionary intent (see paragraph 37 of the section 35(1) notice at RBL1 pages 1-31). However, those documents fell well short of demonstrating any such exclusionary intent; it is hardly surprising that Wiseman should categorise the competitive challenge from Express — one of England and Wales' major dairies and with which it had been waging a fierce battle for market share in England and Wales — rather differently from established competition from smaller local competitors; in short, it is hardly surprising that Wiseman would want to keep a particularly close eye on Express/Claymore's progress. Wiseman may also have wanted to keep a close eye on Express/Claymore because Express/Claymore had made a complaint about it to the competition authorities. We did of course take account of the statement of Alan Wiseman made to the CC at the joint hearing, to which Express/Claymore drew our attention on19 June 2002 (page 501 of the Application, paragraph 2.23, and set out at greater length in paragraphs 5.155 to 5.156 of the Application), but we did not regard that statement as in itself, even taken together with the evidence from Wiseman documents, as providing clear and compelling evidence of intent to exclude Express/Claymore from the market; in essence, what it showed was that Wiseman categorised Express/Claymore differently from its other competitors. … 53. We therefore took the view that there was little prospect of what we had providing clear and compelling evidence of predatory pricing by Wiseman."
"54. The OFT also considered the question of whether Wiseman may have infringed Chapter II by engaging in price discrimination aimed at eliminating Express/Claymore from the market. We had in mind the principles outlined in cases such as Compagnie Maritime Belge and Irish Sugar to the effect, in broad terms, that in certain circumstances selective non-cost related discounts or price reductions targeted on particular customers with exclusionary intent or effect could be regarded as an abuse of a dominant position, even where the prices remained above the total or variable costs of supply. Our task was to examine whether the facts as we could establish them indicated that there would be a real prospect of establishing an infringement along those lines. 55. As I said above, we took as our starting point the CC report. The CC concluded that Wiseman had engaged in price discrimination. In particular, it relied on the analysis of the graph at figure 4.17 of the CC report at page 284 of the Application, which shows a scatter plotting of prices across volume. The graph demonstrates that higher volumes are correlated with lower prices, which is not itself surprising, and, more interestingly, that for lower volumes, price differentials that are not explained by volume differences appear; that appears to emerge from the spread of prices at volumes between 6 and 8 on the x-axis of that table. 56. Price discrimination is generally understood to involve price differences that are not explained by cost factors. It was clear from the CC findings that extensive price differentials were observed, but the CC's analysis does not in itself provide clear and compelling evidence of price discrimination since costs may alter considerably between customers taking equal volumes. Nonetheless, the CC's analysis suggested that price discrimination might be occurring, particularly in the light of the other characteristics of the industry, such as the fact that delivered prices are charged and distribution costs are spread over all customers. A distinct possibility was that such discrimination might be driven by the desire to target some competitors and price them out of the market. 57. The examination of potential targeting was carried out chiefly by comparing Wiseman's data on price and price/cost margins using as benchmarks: Wiseman's own conduct across market segments (geographic, customer, products), and across time. Competitors' conduct (focusing mostly on pricing policies)."
"64. In conclusion, we found evidence of price discrimination at low to mid volume levels, but limited price discrimination at higher volume levels. In our view, these findings were more robust than the CC's conclusion at paragraphs 4.330-4.336 based on its table 4.17. However, the pattern of price discrimination did not provide clear and compelling evidence that Wiseman had been targeting particular customers in order to exclude Express/Claymore. Wiseman's discriminatory behaviour does not appear to be markedly different from the pricing behaviour towards middle-ground customers observed elsewhere (Express/Claymore and Lordswood). Insofar as there is some suggestion that Wiseman's margins are lower in the Highlands of Scotland, that may be explained by the cost of delivering milk there and by the presence of Express/Claymore, to which some reduction in margin would be a normal competitive response. Similarly low margins are observed elsewhere in Scotland far from Express/Claymore's area of activity. 65. As far as exclusionary intent is concerned, I refer to paragraph 54 above where I explained why we did not believe that there was sufficient evidence of such intent to make a sustainable finding in that regard."
"68. It is important here to distinguish between two types of arrangement. The first type of arrangement is an exclusive agreement under which a middle-ground retailer with several outlets across Scotland agrees to take all supplies of milk for all its outlets from, say, Wiseman, that is to say, that it agrees not to purchase milk from any other supplier. This type of agreement has to be distinguished from a second type of agreement under which, say, Wiseman agrees to supply any outlet of the retailer at a particular uniform price, but where the retailer remains free to take supplies for any outlet from any other supplier. 69. In relation to the first type of arrangement – exclusive agreements – the OFT would accept that it may well be an abuse of a dominant position for a supplier to enter into such agreements in circumstances where that will have the effect of foreclosing a significant part of the market to a competitor. The difficulty in the present case is to establish that any such agreements exist. 70. The circumstances and terms of the agreements between Wiseman and Aberness and between Wiseman and CWS are discussed by the CC in its report (see in particular paragraphs 2.107 and 2.117). In neither case did the CC reach a robust conclusion that the agreements in question were exclusive, in the sense set out above, although, as I have said above, we did take the view at the start of the investigation that the CC report gave rise to a reasonable suspicion that such agreements might have been entered into. … 72. During the section 28 investigation on10 May 2001 we looked for any further indication in Wiseman's documentation that it had entered into exclusive agreements with Aberness, CWS or anyone else. We found no such evidence. 73. We received no further indication during the course of the investigation about any further allegedly exclusive agreements. Nor had we been alerted that the terms of any of the supply agreements alleged to be exclusive had been renegotiated until the very last minute when, by fax on5 August 2002 , Express/Claymore told us that one of the contracts in question (Aberness) had been re-awarded to Wiseman on the same terms as before. However, the material relied on by Express/Claymore fails to demonstrate that an exclusive agreement has been entered into. Aberness refers to Wiseman as the "principle" [sic], rather than "exclusive", supplier to its own stores, i.e. AR Gray stores (Application, page 605), and even Express/Claymore accepts that it is free to bid for supply to those stores on whose behalf Aberness negotiates but which it does not own. The new material made no difference to the analysis and conclusion we had already come to. There was no indication in our correspondence with other smaller dairies, see paragraph 32 above, that exclusive agreements involving Wiseman were a feature of the market. 74. In the absence of any sufficiently compelling evidence of exclusivity, we were unable to conclude that the "all-Scotland" arrangements entered into by Wiseman infringed Chapter II of the Act. We did not believe that merely by committing itself to offer a single price for milk to all outlets of a particular retailer Wiseman had infringed Chapter II. There is no clear or compelling evidence that Wiseman had ever made supply in any area or to any outlet conditional upon supply on an all Scotland or regional basis. In such a case, it is difficult to see that we could have maintained a case to the standard of proof required that Express/Claymore is precluded by agreements of the type alleged from competing for the business of outlets in the Highlands area, and from gaining that business where it is able to offer a more competitive price."
"76. By April 2002, for the reasons I have explained, we had therefore begun to take the view that the evidence gathered during the investigation was not conclusive on the existence or absence of infringement in any of the separate strands of the investigation. At a meeting on 14 March Wiseman had been able to provide alternative explanations for its pricing behaviour which we felt were logical and would undermine any attempt to prove that Wiseman's actions departed from a normal competitive response to Express's entry. Wiseman explained the mechanisms through which prices were arrived at for different classes of customer. As we understood it, for the smaller middle-ground retailer customers, this was done without specific reference to the cost associated with the particular run to which that customer might end up being allocated, although the overall depot cost would be a factor. This explained why our observations showed a number of cases of below total cost pricing. For example, retailer A might be allocated by Wiseman to a high cost run (e.g. where the other customers served were all small and widely-dispersed outlets), while retailer B — perhaps identical in all respects and even quite close to retailer A geographically — might be allocated to a run serving a few high volume customers. The run costs for retailer A would obviously be much higher than for retailer B, but Wiseman would not wish to charge them different prices only on the basis of their allocation to different runs. It followed that retailer A might be regarded for the purposes of our observations as being supplied at below total cost. … 78. … I should make it clear that we did not regard the work we had done as in any way demonstrating that Wiseman had not abused what we took to be its dominant position in any of the above ways. We would also accept that in theory further work could have been done on the case to test further the allegations that Wiseman was infringing the Chapter II prohibition. We took the view that on the basis of the evidence we had, such further work could not be justified in terms of the OFT's priorities. In effect, we decided to cut our losses on the case and move on to more productive work. It follows that our analysis of the issues in the case is not conclusive and contains some rough edges. Since we were not proceeding to a rule 14 notice, or issuing a detailed decision to the effect that Wiseman was not infringing the Chapter II prohibition, we did not feel that it was necessary to develop the case any further… 80. Secondly, as far as the allegations of exclusive contracts is concerned, it is, I suppose, possible that we could have pursued the allegation in respect of Aberness by, for example, using powers under sections 26 to 28 of the Act to obtain documents and information about its arrangements with Wiseman from Aberness. We could also, I suppose, have discussed with other retailers whether they were party to exclusive arrangements with Aberness. However, we bore in mind that in its evidence before the CC, Aberness had stated in terms that it was clear to Aberness that its arrangements with Wiseman were not exclusive (see paragraph 2.103 of the CC report, Application page 174). For reasons I have already explained, we did not regard the position in this respect as being affected by the material enclosed in Express/Claymore's letter of5 August 2002 (Application, attachment 21)…"
"Ground 1 In failing to identify the relevant time period of the abuse (in particular for the purposes of the identification of elements of avoidable cost) the Respondent erred in law and/or in its assessment in respect of pricing below AVC. … Ground 2 The OFT erred in its approach to and assessment of AVC generally. … Ground 3 The OFT erred in its approach to and assessment of Average Total Costs. … [2] [3] Ground 5 The OFT erred in its approach and/or failed to provide any adequate reasons in respect of the relevance of incremental customers and/or outlets. … Ground 6 The OFT erred in its approach and/or failed to provide any adequate reasons in respect of its appraisal of below cost pricing generally. … Ground 7 The OFT erred in its appraisal of the evidence of intent generally. … Ground 8 The OFT misdirected itself in law in requiring evidence of intent to eliminate Claymore from the market. … Ground 9 The OFT misdirected itself in law in requiring clear and compelling "direct" evidence of intent to eliminate Express/Claymore from the market and that relevant documents should "demonstrate" such intent. … Ground 10 The OFT erred in its appraisal and/or failed to provide any adequate reasons in respect of its assessment of the available evidence of targeted price discrimination. … Ground 11 The OFT erred in law in that it failed to address its mind to whether the significant evidence of price discrimination amounted to abusive conduct in the context of all the circumstances of the case and, in particular, whether it tended to exclude Claymore. … Ground 12 The Respondent erred in law and approach in assessing Wiseman's conduct in committing itself to single price all of Scotland contracts. … Ground 13 The investigation undertaken by the Respondent was inadequate and/or the Respondent erred in its approach to the all of Scotland arrangements generally. … Ground 14 The OFT erred in law and approach in failing to consider whether Wiseman's reaction to Express' investment in Claymore in the form of the all of Scotland contracts was reasonable and proportionate and complied with the special responsibility placed upon dominant undertakings not to weaken the structure of the market generally. … Ground 15 The OFT's methodology, approach and standards of care in respect of data gathering and analysis failed to comply with minimum adequate standards. … Ground 16 The OFT erred in law by failing to assess the overall impact of Wiseman's conduct upon the structure of the relevant market. … Ground 17 The OFT erred in law and approach in its failure to pay any or any adequate regard to the evidence or findings made in connection with the Chapter I cartel investigation. … Ground 18 The OFT erred in law and approach by examining the three heads of investigation that were investigated as separate strands of inquiry rather than paying due regard to the cumulative effect of the three abuses upon each other and the market as a whole. … Ground 19 The OFT erred in law and approach in relation to the application of the standard of proof."
"38. The analysis of the team is that the present situation could be the result of an inefficient entry by Express in the Scottish market that would have triggered a Stackelberg-warfare type of situation. This would mean that Express's strategy in entering in the market would have been a mistaken move, the outcome of which would be mutual aggression. Views in the industry have tended to find Express strategy suboptimal, as entry in the Central belt sounded like the legitimate move. The "illegitimacy" of Express in Scotland could stem from its location in Nairn. 39. Wiseman recognised Express as a serious contender for leadership and due to the oligopolistic situation in the market had to react to Express' all-Scotland strategy."
"350. The cases cited above demonstrate, in our view, that the question whether a certain pricing practice by a dominant undertaking is to be regarded as abusive for the purposes of the Chapter II prohibition is a matter to be looked at in the round, taking particularly into account (i) whether the dominant undertaking has had "recourse to methods different from those which condition normal competition in products or services on the basis of the transactions of commercial operators" ( Hoffmann-La Roche , cited above, at paragraph 91); and (ii) whether such conduct has the effect of weakening or distorting competition in the relevant market, having regard to the special responsibility of a dominant firm not to impair genuine undistorted competition. In our view, these principles apply particularly to the case of a dominant firm facing new entry, where retaliatory measures going beyond what is reasonable and proportionate are likely to require close scrutiny under the Chapter II prohibition. 351. Within that framework, the cases of AKZO , Tetra Pak II and Compagnie Maritime Belge , cited above, give further guidance as to when prices below costs are likely to be regarded as abusive. As the Director points out at paragraphs 151 to 152 of the decision, AKZO (at paragraph 71) and Tetra Pak II (at paragraph 41) show that pricing below average variable costs by a dominant firm is normally to be regarded as an abuse. "
"71. Prices below average variable costs (that is to say, those which vary depending on the quantities produced) by means of which a dominant undertaking seeks to eliminate a competitor must be regarded as abusive. A dominant undertaking has no interest in applying such prices except that of eliminating competitors so as to enable it subsequently to raise its prices by taking advantage of its monopolistic position, since each sale generates a loss, namely the total amount of the fixed costs (that is to say, those which remain constant regardless of the quantities produced) and, at least, part of the variable costs relating to the unit produced. 72. Moreover, prices below average total costs, that is to say, fixed costs plus variable costs, but above average variable costs, must be regarded as abusive if they are determined as part of a plan for eliminating a competitor. Such prices can drive from the market undertakings which are perhaps as efficient as the dominant undertaking but which, because of their smaller financial resources, are incapable of withstanding the competition waged against them."
"47. To [arrive at total costs] we repeated the steps described above in respect of total costs. We adopted the approach of aggregating every one of the costs requested from Wiseman, adding 3 per cent to reflect financial administration and central selling costs, and using that measure as a working proxy for total cost (although some elements of total cost – possibly up to 5 per cent – were left out of that proxy) 8 ."
" 8 In order to check whether we had a measure of costs giving a reliably close approximate of costs as a whole, we compared the measure of costs reported in the CC report with the equivalent measure as calculated by us from the information collected from Wiseman. CWS is one customer account studied extensively in the CC report. We had available as part of our sample full data relating to the costs of supplying CWS during our sample periods. The average total costs to supply CWS in para. 4.342 of the CC report is […][ C ] ppl for May 2000. Our measure of the average total cost for May 2000 for the same account is […][ C ] ppl ([…][ C ] ppl when the 3 per cent mark-up is applied). The difference between these two measures is only about 5 per cent which we believed was acceptable for the purpose of what was an initial investigation into costs."
"74. In the absence of any sufficiently compelling evidence of exclusivity, we were unable to conclude that the "all-Scotland" arrangements entered into by Wiseman infringed Chapter II of the Act. We did not believe that merely by committing itself to offer a single price for milk to all outlets of a particular retailer Wiseman had infringed Chapter II. There is no clear or compelling evidence that Wiseman had ever made supply in any area or to any outlet conditional upon supply on an all Scotland or regional basis. In such a case, it is difficult to see that we could have maintained a case to the standard of proof required that Express/Claymore is precluded by agreements of the type alleged from competing for the business of outlets in the Highlands area, and from gaining that business where it is able to offer a more competitive price."
"89. An undertaking which is in a dominant position on a market and ties purchasers – even if it does so at their request – by an obligation or promise on their part to obtain all or most of their requirements exclusively from the said undertaking abuses its dominant position within the meaning of Article [82] of the Treaty, whether the obligation in question is stipulated without further qualification or whether it is undertaken in consideration of the grant of a rebate."