This document is an excerpt from the EUR-Lex website
Document 62004CC0095
Opinion of Advocate General Kokott delivered on 23 February 2006. # British Airways plc v Commission of the European Communities. # Appeals - Abuse of dominant position - Airline - Agreements with travel agents - Bonuses linked to growth in sales of that airline's tickets over a given period in comparison with a reference period - Bonuses granted not only for tickets sold once the sales target achieved, but for all tickets sold during the period in question. # Case C-95/04 P.
Opinion of Advocate General Kokott delivered on 23 February 2006.
British Airways plc v Commission of the European Communities.
Appeals - Abuse of dominant position - Airline - Agreements with travel agents - Bonuses linked to growth in sales of that airline's tickets over a given period in comparison with a reference period - Bonuses granted not only for tickets sold once the sales target achieved, but for all tickets sold during the period in question.
Case C-95/04 P.
Opinion of Advocate General Kokott delivered on 23 February 2006.
British Airways plc v Commission of the European Communities.
Appeals - Abuse of dominant position - Airline - Agreements with travel agents - Bonuses linked to growth in sales of that airline's tickets over a given period in comparison with a reference period - Bonuses granted not only for tickets sold once the sales target achieved, but for all tickets sold during the period in question.
Case C-95/04 P.
European Court Reports 2007 I-02331
ECLI identifier: ECLI:EU:C:2006:133
OPINION OF ADVOCATE GENERAL
KOKOTT
delivered on 23 February 2006 1(1)
Case C-95/04 P
British Airways plc
v
Commission of the European Communities
other party to the proceedings:
Virgin Atlantic Airways Ltd
(Appeals – Abuse of a dominant position (Article 82 EC) – British Airways – Agreements with travel agents – Commission and other benefits dependent upon turnover in British Airways tickets)
I – Introduction
1. This case originates with a competition proceeding of the Commission concerning certain commissions and benefits which the British airline British Airways plc (‘BA’) granted to travel agents dependent on their turnover in BA tickets. The Commission held that BA had abused its dominant position on the market (Article 82 EC), and fined it EUR 6 800 000.
2. That decision of the Commission of 14 July 1999 (2) (‘the contested decision’) was confirmed in its entirety by the judgment of the Court of First Instance of 17 December 2003 in Case T-219/99 (3) (‘the contested judgment’).
3. The Court of Justice now has before it an appeal by BA against that judgment at first instance. What needs to be determined, essentially, is under what circumstances the granting of bonuses by a dominant undertaking may be regarded as an abuse within the meaning of Article 82 EC.
II – Legal context
4. The legal context of this case is defined by Article 82 EC, which reads:
‘Any abuse by one or more undertakings of a dominant position within the common market or in a substantial part of it shall be prohibited as incompatible with the common market in so far as it may affect trade between Member States.
Such abuse may, in particular, consist in:
a) directly or indirectly imposing unfair purchase or selling prices or other unfair trading conditions;
b) limiting production, markets or technical development to the prejudice of consumers;
c) applying dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage;
d) making the conclusion of contracts subject to acceptance by the other parties of supplementary obligations which, by their nature or according to commercial usage, have no connection with the subject of such contracts.’
III – Facts and procedure
A – Facts
5. The underlying competition case originates with complaints by Virgin Atlantic Airways Ltd (‘Virgin’), a competitor of BA. (4) It concerns the market for air travel agency services in the United Kingdom, on which, according to the Commission’s findings, BA was the dominant purchaser. (5)
6. On that market, travel agents supply airlines with services consisting of promoting the air transport services provided by airlines, helping travellers choose the appropriate services, and undertaking the administrative work of issuing tickets, collecting money from travellers and remitting it to the airline. In return for those services, the airlines pay commissions to the agents based on the sales of tickets made through those agents. (6)
7. BA concluded agreements with travel agents established in the United Kingdom and accredited by the International Air Transport Association (IATA) entitling them to a basic standard commission on their sales of BA air tickets. Between 1976 and 1997, that commission amounted to 9% on sales of international tickets and 7.5% for ticket sales on domestic flights. (7) Thereafter, it was replaced by a new single commission of 7% on all air tickets sold in the United Kingdom. (8)
8. In addition to that basic commission system, BA concluded agreements with IATA travel agents comprising three distinct systems of financial incentives: ‘marketing agreements’, ‘global agreements’ and, finally, a ‘performance reward scheme’. (9)
9. Concerning marketing agreements and global agreements the Court of First Instance held: (10)
‘6 The first system of incentives established by BA consisted of “marketing agreements”, which enabled certain IATA travel agents established in the United Kingdom to receive payments in addition to their basic commission, namely:
– a performance reward, plus certain special bonuses, based on the volume of sectors flown on BA;
– cash sums from a fund for travel agents to use for staff training;
– cash sums from a business development fund established by BA with a view to increasing its revenue and the resources of which were to be used by each agent for financing promotional projects in favour of BA.
7 The marketing agreements also required the United Kingdom travel agents not to accord less favourable treatment to BA than that which they accorded to any other airline, particularly in relation to the display of their fares, products, brochures and timetables.
8 Those marketing agreements, concluded for one year at a time, were in principle reserved for United Kingdom IATA travel agents with more than GBP 500 000 annual sales of BA tickets (“flown revenue”). Travel agents with annual flown revenue exceeding GBP 500 000 but below GBP 10 million were offered a standard marketing agreement. Those with a flown revenue exceeding GBP 10 million entered into a marketing agreement individually negotiated with BA.
9 The performance reward was calculated on a sliding scale, based on the extent to which a travel agent increased the value of its sales of BA tickets. In addition to the general performance reward, certain routes qualified for a special performance bonus.
10 Payment of the performance reward or the special bonus was subject to travel agents increasing their sales of BA tickets from one year to the next. Although, as a general rule, neither of those two bonuses was paid in respect of sectors flown on BA domestic services within the United Kingdom, those sectors were counted in determining whether sales objectives were achieved, since those objectives were calculated in terms of global flown revenue, including long-haul, short-haul and domestic flights.
11 In addition to the marketing agreements, BA concluded a second type of incentive agreement (“global agreements”) with three IATA travel agents. For the 1992/93 winter season, BA set up global incentive programmes with three travel agents, entitling them to receive additional commissions calculated by reference to the growth of BA’s share in their worldwide sales.’
10. Concerning the performance reward scheme which BA applied from 1998, the Court of First Instance held: (11)
‘14 On 17 November 1997, BA sent all travel agents established in the United Kingdom a letter in which it explained the detailed operation of a third type of incentive agreements, consisting of a new system of performance rewards, applicable from 1 January 1998 (“the new system of performance rewards”).
15 In addition to the new basic flat commission rate of 7% to be applied thenceforth to all tickets sold in the United Kingdom, each travel agent could earn an additional commission of up to 3% for international tickets and up to 1% for domestic tickets. The size of the additional variable element for domestic and international tickets depended on the travel agents’ performance in selling BA tickets. The agents’ performance was measured by comparing the total flown revenue arising from the sales of BA tickets issued by an agent in a particular calendar month with that achieved during the corresponding month in the previous year.
16 Under the new system of performance rewards, every percentage point of improvement in performance level over a benchmark of 95% earned the travel agent an additional variable element of 0.1% by way of extra commission on the sale of international tickets and in addition to the basic commission of 7%. For sales of domestic tickets, the variable element was 0.1% for every 3% increase in sales over the 95% benchmark. The maximum variable element payable to travel agents under the new performance rewards system was 3% for international tickets and 1% for domestic tickets for a performance level of 125% or above in both cases.
17 For example, if an agent’s performance level for a particular calendar month was 112%, the variable element for international tickets was 1.7% [(112 – 95) × 0.1%] of the international reward revenue for that month. On the other hand, at that level of performance the variable element for domestic tickets was 0.5% [(112 – 95) ÷ 3 × 0.1%] of the domestic reward revenue for that calendar month. Payments of the variable elements under the performance rewards system took place every month.
18 The new system of performance rewards was initially intended to last until 31 March 1999. For the month of December 1997, BA established a transitional scheme whereby the new performance reward system was applied on top of the pre-existing standard commissions of 9% and 7.5% for international and domestic tickets respectively. On 8 February 1999, BA announced that that system would not be renewed for the year 1999/2000.’
11. According to the Commission’s findings, quoted by the Court of First Instance, (12) the effect of the commission schemes described may be summarised as follows:
‘(29) The commission schemes for travel agents described above all have one notable feature in common. In each case meeting the targets for sales growth leads to an increase in the commission paid on all tickets sold by the agent, not just on the tickets sold after the target is reached. In the [marketing agreement] schemes the cash bonus per ticket paid to the travel agent increases for all tickets sold. In the [performance reward scheme] the percentage commission paid increases for all ticket sales by the travel agent. This means that when a travel agent is close to one of the thresholds for an increase in commission rate selling relatively few extra BA tickets can have a large effect on his commission income. Conversely a competitor of BA who wishes to give a travel agent an incentive to divert some sales from BA to the competing airline will have to pay a much higher rate of commission than BA on all of the tickets sold by it to overcome this effect.
(30) An example will illustrate this effect of the BA commission schemes. Assume a travel agent’s sales of international air tickets amounted to [GBP] 100 000 a month in the benchmark year. If the travel agent sells [GBP] 100 000 worth of BA international air tickets a month it will earn the basic commission of 7% and a “performance reward” of 0.5% [(100 – 95) × 0.1%], giving a total commission income on international air ticket sales of [GBP] 7 500 [100 000 × (7% + 0.5%)]. If the travel agent diverted 1% of its international ticket sales to a competitor of BA, its “performance reward” would decrease to 0.4% [(99 – 95) × 0.1%] and this reduced rate would be applied to all of the agent’s sales of BA tickets. The agent’s commission income from the sale of international BA tickets would drop to [GBP] 7 326 [99 000 × (7% + 0.4%)]. A reduction of [GBP] 1 000 in sales of international BA tickets leads to a drop of [GBP] 174 in commission income. The “marginal” commission rate can be said to be 17.4%. In practical terms, this means that a competitor to BA that could offer flights that would replace [GBP] 1 000 of the travel agent’s sales of BA tickets would have to offer a commission of 17.4% on these tickets to compensate the travel agent for its loss of BA commission revenue. Although BA also has to offer this high marginal rate of commission to increase its sales of tickets, it is at an advantage over the new entrant which must offer this high rate of commission on all of its sales. ...
This effect increases if the number of tickets in question is a smaller percentage of the travel agent’s benchmark sales of BA tickets. This effect is also increased if the travel agent in question is not only earning extra commissions under the performance reward system but can also earn bonuses under a marketing agreement.’
B – The contested decision
12. The Commission held in the contested decision that, by applying the marketing agreements and the new system of performance rewards (13) to air travel agents established in the United Kingdom, (14) BA had abused its dominant position on the United Kingdom market for air travel agency services.
13. Both commission schemes (15) gave travel agents a financial incentive to maintain or increase their turnover of BA tickets rather than selling their services to BA’s competitors, because those rewards were not dependent on the amount of the travel agents’ sales of BA tickets in absolute terms. (16) In addition, they imposed on the travel agents in question dissimilar conditions to equivalent transactions. (17) Finally, the Commission takes the view that BA’s abusive conduct on the United Kingdom market for air travel agency services has the effect of distorting competition between BA and other airlines on the United Kingdom markets for air transport services. (18)
14. The operative part of the contested decision reads:
‘Article 1
British Airways plc infringed Article 82 of the Treaty by operating systems of commission and other incentives with the travel agents from whom it purchases air travel agency services in the United Kingdom, which, by rewarding loyalty from the travel agents and by discriminating between travel agents, have the object and effect of excluding BA’s competitors from the United Kingdom markets for air transport.
Article 2
For the infringements referred to in Article 1, a fine of EUR 6.8 million is hereby imposed on British Airways plc. …’
C – Proceedings before the Court of First Instance
15. On 1 October 1999, BA brought an action before the Court of First Instance, claiming that the contested decision should be annulled in its entirety and the Commission ordered to pay the costs of the proceedings. The Commission contended that the action should be dismissed and BA ordered to pay the costs.
16. By order of 9 February 2001, Virgin was authorised to intervene in support of the Commission. However, an application by the French airline Air France to intervene in support of British Airways was dismissed.
17. In the contested judgment, the Court of First Instance dismissed BA’s application and ordered it to pay the Commission’s and Virgin’s costs in addition to its own.
18. In its appeal, lodged at the Court Registry on 26 February 2004, BA now claims that the Court of Justice should:
– set aside the judgment under appeal in whole or in part;
– annul or reduce the amount of BA’s fine by an amount considered appropriate by the Court in the exercise of its discretion;
– take any other measures that the Court deems appropriate.
19. The Commission contends that the Court should:
– dismiss the appeal in its entirety; and
– order BA to pay the costs.
20. Virgin contends that the Court should:
– declare the appeal inadmissible or, in any event, clearly unfounded and dismiss it by reasoned order pursuant to Article 119 of the Rules of Procedure of the Court of Justice;
– (in the alternative) dismiss the appeal and uphold the judgment under appeal in its entirety; and
– (in any event) order BA to pay the costs of the appeal, including Virgin’s costs.
21. The parties submitted their written arguments, and on 15 December 2005 presented oral argument, before the Court of Justice.
IV – Assessment
22. In its appeal, BA no longer refers to all the issues before the Court of First Instance, and particularly not the Commission’s findings on the delimitation of the market and the dominant position of BA. Instead, its five pleas on appeal are concerned solely with the statements of the Court of First Instance concerning abuse of its dominant position on the market within the meaning of Article 82 EC, appearing in paragraphs 227 to 300 of the contested judgment.
A – Preliminary observations
23. Within the scope of the application of Article 82 EC, a dominant undertaking is subject to certain limitations that do not apply to other undertakings in the same form. Because of the presence of the dominant undertaking, competition on the market in question is weakened. (19) Therefore – whatever the causes of its dominant position – that undertaking has a particular responsibility to ensure that its conduct does not undermine effective and undistorted competition in the common market. (20) A practice which would be unobjectionable under normal circumstances can be an abuse if applied by an undertaking in a dominant position. (21)
24. Thus, for example, a dominant undertaking is entitled, according to consistent case-law, to protect its own commercial interests if they are attacked, and may also take such reasonable steps as it deems appropriate to protect them. (22) In particular, it may use the methods of normal competition in products and services in the sense of competition on the merits; however, a business practice which deviates from normal market behaviour and is capable of weakening existing competition is an abuse within the meaning of Article 82 EC and therefore prohibited. (23) Not every kind of price competition is therefore permissible under Article 82 EC. (24)
25. In the area of rebates and bonuses it is particularly clear that, in individual cases, it is difficult to draw the line between legitimate conduct and the prohibited abuse of a dominant market position.
26. The Community courts have held on several occasions that the granting of certain rebates or bonuses by a dominant undertaking can be an abuse within the meaning of Article 82 EC. (25) In particular, loyalty rebates and loyalty bonuses can in practice bind business partners so closely to the dominant undertaking (the ‘fidelity-building effect’), that its competitors find it inordinately difficult to sell their products (‘exclusionary’, or ‘foreclosure’ effect), with the result that competition itself can be damaged and, ultimately, the consumer can suffer.
27. Dispute remains, however, as to the precise circumstances in which the granting of rebates or bonuses by a dominant undertaking will be abusive within the meaning of Article 82 EC. The present case gives an opportunity to clarify a number of questions in that regard:
– Under what circumstances are rebates or bonuses granted by a dominant undertaking to be regarded in general as an abuse? (first ground of appeal)
– Is it necessary in that context to investigate the concrete effects of the dominant undertaking’s rebates or bonuses on its competitors and consumers? (second, third, and fourth grounds of appeal)
– Under what circumstances are the rebates or bonuses of a dominant undertaking to be regarded as discrimination between trading parties, thereby placing them at a competitive disadvantage? (fifth ground of appeal)
28. In this context it is immaterial how the Commission intends to define its competition policy with regard to Article 82 EC for the future. (26) Any reorientation in the application of Article 82 EC can be of relevance only for future decisions of the Commission, not for the legal assessment of a decision already taken. Moreover, even if its administrative practice were to change, the Commission would still have to act within the framework prescribed for it by Article 82 EC as interpreted by the Court of Justice.
B – The first ground of appeal: General criteria for the assessment of commission schemes
29. BA’s first ground of appeal takes up by far the greatest amount of space in the appeal. It concerns paragraphs 272 to 298 of the contested judgment, in which the Court of First Instance – like the Commission before it – held first that the bonuses granted by BA had a ‘fidelity-building’, and therefore exclusionary, effect, and, secondly, that they were not economically justified. (27)
30. In this ground of appeal, BA essentially raises the question under what circumstances the rebates or bonuses of dominant undertakings can in general be regarded as abusive. In its second, third and fourth grounds of appeal, to be examined thereafter, BA turns to the legal requirements in determining the effects of such rebates or bonuses on competitors and consumers.
1. Essential arguments of the parties
31. The parties are essentially arguing as to whether the Court of First Instance correctly applied the case-law of the Court of Justice, particularly in the Hoffmann-La Roche and Michelin I judgments, (28) to the present case.
32. BA argues that the Court of First Instance erred in law by judging its commission schemes by a false criterion. It argues that, in examining whether such systems were ‘fidelity-building’, the Court of First Instance did not differentiate between customer loyalty secured by abusive exclusionary conduct and customer loyalty arising from legitimate price competition. It argues that it is in the nature of legitimate price competition for an undertaking to have the freedom to grant higher rebates to its business partners than do its competitors. In BA’s submission, the criteria laid down by the Court of First Instance lead to considerable legal uncertainty concerning the scope of legitimate price competition and have a deterrent effect on undertakings; they thereby undermine the fundamental purpose of Community competition law.
33. In this case, BA argues, the Court of First Instance should have applied subparagraph (b) of the second paragraph of Article 82 EC and examined whether BA had in fact limited the sales opportunities of its competitors and whether consumers were thereby prejudiced. Such a limitation of competitors’ sales opportunities requires more than just the granting of generous bonuses. It is conceivable only in two sets of circumstances, neither of which is relevant here:
– first in cases where the granting of bonuses is made dependent on the recipient dealing exclusively or predominantly with the dominant undertaking, or being supplied by it, (29) and
– secondly in situations where the recipient of the bonuses cannot choose freely between the dominant undertaking and its competitors, for example because it is only by a predominant commercial tie to the dominant undertaking that it can expect to make profits, or because the dominant undertaking practises predatory pricing and its competitors cannot withstand the pressure.
34. The Commission and Virgin, on the other hand, are unanimously of the opinion that the criteria set out by the Court of First Instance are correct and in accordance with the previous case-law. There is, they maintain, no error of law in the examination by the Court. Virgin further argues that, if BA’s approach to Article 82 EC were to be upheld, it would lead to an alteration in the case-law on a comparable scale with the judgment in Keck and Mithouard. (30)
2. Assessment
35. As for whether, as BA argues, the Court should have been guided by the criteria in subparagraph (b) of the second paragraph of Article 82 EC, suffice it so say that that provision gives merely an example of the abuse of a dominant position. (31) Rebates and bonuses by dominant undertakings can infringe Article 82 EC even if they do not correspond to any of the examples contained in that second paragraph. (32) The Court of First Instance cannot therefore be accused of an error of law.
36. In terms of content also, the Court of First Instance has not in any way misconstrued the legal position as set out in the previous case-law of the Court of Justice.
a) No closed categories of abusive rebates and bonuses
37. It is true that, in both the judgments discussed by BA, the Court of Justice established that certain rebates granted by two market dominant undertakings were abusive in character.
38. The Hoffmann-La Roche case concerned rebates the granting of which was, for the most part, expressly linked to the condition that, during a reference period – normally a year or half a year – the business partner in question was to cover all its requirements for particular vitamins, or at any rate the predominant part of those requirements, with Hoffmann-La Roche. (33) The Court of Justice regarded such a rebate system as the abuse of a dominant position (34) and stated that ‘the granting of fidelity rebates intended to give the purchaser an incentive to obtain his supplies exclusively from the undertaking in a dominant position [is] incompatible with the objective of undistorted competition within the common market’. (35)
39. The Court of Justice also saw abuse of a dominant market position as proven in Michelin I. (36) Unlike in Hoffmann-La Roche, the business partners in that case of the dominant undertaking Michelin were not obliged to obtain all, or a specified part, of their supplies from that undertaking. (37) Nevertheless, the annual rebates granted by Michelin took the form of ‘target rebates’: In order to enjoy them, Michelin’s business partners had to attain individual sales targets. Those sales targets were determined according to the turnover in Michelin tyres which the relevant business partner had achieved the previous year. (38) The Michelin I case was further distinctive for a whole series of factors which, taken together, led the Court of Justice to regard the rebate system introduced by Michelin as the abuse of a dominant market position. In particular, the disputed rebate system was based on a ‘relatively long reference period’ of one year, (39) the way in which the system functioned was untransparent for the business partner, and the differences between the market shares of Michelin and its main competitors were great.
40. Contrary to what BA maintains, however, this case-law does not indicate any closed categories of abusive bonus and rebate schemes. It cannot in any way be inferred from those judgments that bonuses or rebates granted by an undertaking in a dominant position are abusive only in the circumstances described in detail in those cases. That would be to ignore the fact that individual sectors of the economy and markets can differ significantly from one another and that, moreover, economic circumstances are subject to constant change, which, not least, can also entail new business practices.
41. The decisive factor is rather the underlying thoughts which have guided the previous case-law of the Court of Justice and which can also be transposed to a case such as the present.
42. It therefore needs to be examined first whether the rebates or bonuses granted by an undertaking in a dominant position can give rise to a foreclosure effect, i.e. whether they are capable of making it difficult or impossible for the competitors of the dominant undertaking to have access to the market and for the business partners of the dominant undertaking to choose between various sources of supply; secondly, it needs to be examined whether there is an objective economic justification for the rebates or bonuses granted. (40)
43. Whilst, of course, objective economic justification becomes relevant only if the rebates or bonuses granted do give rise to a foreclosure effect, both steps in this examination are designed to distinguish abusive from lawful conduct and thus ensure that legitimate price competition does not fall foul of Article 82 EC.
b) First step in the examination: Foreclosure effect
44. The Court of First Instance was right to proceed on the basis that rebate schemes may infringe Article 82 EC even if they are not, as in Hoffmann-La Roche, linked to a condition of exclusivity, i.e. they do not require the contractual partner to obtain all or a specified part of its supplies exclusively from the dominant undertaking, or that it provides its own services exclusively, or as to a specified proportion, for the dominant undertaking. (41) Even where there is no such conditionality, the foreclosure effect of a rebate or bonus scheme may arise from the other circumstances of the particular case. (42) Nor, contrary to what BA argues, is the decisive factor whether the contractual partners of the dominant undertaking can still choose freely between various sources of supply. Application of Article 82 EC is in no way deferred until there is practically no effective competition left in a market. Its purpose is rather to protect existing competition in a market, weakened by the presence of the dominant undertaking, (43) and its scope thus extends much further, beyond the second category described by BA. (44)
45. Whether, therefore, rebates or bonuses granted by a dominant undertaking may be regarded as having a foreclosure effect depends solely on an examination of all the circumstances of the individual case. (45) That approach ensures at the same time that no rebates or bonuses are caught which could be regarded as a part of legitimate price competition on the relevant market.
46. The starting-point for the examination is the respective criteria and rules for the granting of a rebate or bonus. (46) It further needs to be examined whether the relevant rebate or bonus scheme as a whole is capable of making it difficult or impossible for the dominant undertaking’s competitors to have access to the market, or for its contractual partners to choose between various sources of supply or business partners. (47)
47. Even if the case-law cannot exhaustively define cases of rebate and bonus schemes with foreclosure effect, it can supply indications of when such a foreclosure effect will be present in the normal course of events. There are three aspects which are particularly relevant in determining, in accordance with the case-law, that the rebates or bonuses granted by a dominant undertaking represent more than just a particularly favourable market offer.
48. First of all, a foreclosure effect may arise from rebates and bonuses the granting of which is linked to the fulfilment of individually defined sales targets (‘target rebates’ or ‘target bonuses’). (48) If, for example, a contractual partner is offered a rebate or bonus for achieving over a given reference period the same, or a higher, turnover in the dominant undertaking’s products than in the comparable period of the preceding year, (49) even a partial switch to products of competitors can become less attractive for him. In this way, the dominant undertaking can exercise pressure on its contractual partners and bind them to itself. It can thus consolidate, and even extend, its market position. (50)
49. According to the findings of the Court of First Instance, the commission schemes applied by BA are based on precisely such individual sales targets, and were indeed dependent on the growth in turnover in BA tickets achieved by the travel agents over a particular period. (51)
50. Secondly, the binding of the contractual partner to the dominant undertaking and the pressure exercised upon it are as a rule particularly strong if a rebate or bonus refers not just to the increase in turnover for the reference period, but extends back to the whole volume of turnover in the dominant undertaking’s products achieved by the contractual partner during that period. In that way even relatively small changes – upwards or downwards – in the turnover in the dominant undertaking’s products can have disproportionate effects for the contractual partner. If, for example, he sells only slightly fewer products of the dominant undertaking than in the period with which comparison is made, he is already running the risk of no longer coming, overall, within the benefit of a rebate or a bonus, or at least of enjoying only a reduced rebate or bonus. If, on the other hand, he sells only slightly more of such products than in the comparable period, he may benefit from a higher rebate or bonus, and moreover do so retrospectively in respect of his whole turnover volume in those products, not just for the future in relation to the products additionally sold. (52) In that way, the contractual partner is in a state of uncertainty during the reference period as to how high his profit margin with the dominant undertaking’s products will finally be; that is a strong incentive to him not to switch to competitors – even in part.
51. That has also been established by the Court of First Instance in this case in relation to BA’s commission schemes. The favourable rates of commission extended retrospectively to all BA tickets sold by the travel agent in question, not just to those sold after reaching the turnover target; for the commission income of a travel agent as a whole, therefore, it could be of decisive importance whether, after reaching a certain turnover, he sold even relatively few BA tickets in addition or not. (53) It is precisely this that the Court of First Instance refers to when it points to the commission schemes’ ‘very noticeable effect at the margin’, and emphasises the radical effects which even a slight decrease of turnover in BA tickets could have on a travel agent’s rates of performance rewards. (54) It is not necessary to determine in this context whether the less favourable commission rate feared could meaningfully be compared with a ‘penalty’; whatever the wording, it is clear what the Court of First Instance wished to express: even a small decrease in the turnover in BA tickets could, in the opinion of the Court, lead to significant financial losses for a given travel agent and thereby effectively restrain him from switching to competitors.
52. Thirdly, it is particularly difficult for competitors of the dominant undertaking to outbid such whole-turnover-based rebates or bonuses. Because of its much higher market share, a dominant undertaking is normally, so far as other market participants are concerned, an unavoidable trading partner. (55) Also, whole-turnover-based rebates or premiums granted by a dominant undertaking will, in absolute terms, regularly weigh more strongly in the balance than anything which even more generous offers from competitors could normally achieve. In order to draw the contractual partners of the dominant undertaking onto their side, or at any rate to receive a sufficient volume of orders from them, its competitors would have to offer them disproportionately higher rebates or premiums, (56) which even for equally efficient competitors is often uneconomic.
53. In this case also, according to the findings of the Court of First Instance, BA’s market share was significantly higher than the shares of its five main competitors in the United Kingdom; those competitors were therefore not in a position to offer travel agents the same advantages as BA. (57)
54. The Court of First Instance therefore reasoned on the basis of the previous case-law and made the findings which normally have to be made when examining the commission schemes applied by BA for their foreclosure effect (exclusionary effect).
55. As regards the assessment of the market conditions which were determined and the competitive situation, it is not the task of the Court of Justice in appeal proceedings to substitute its appraisal for that of the Court of First Instance. Apart from the question of distortion of the facts or evidence, which has not been alleged in this case, those are not legal questions, which is all that appeal proceedings are concerned with (Article 225(1) EC and Article 58(1) of the Statute of the Court of Justice). (58) No account should therefore be taken of BA’s objection that its competitors were financially perfectly able to make the travel agents competitive counter-offers. The same applies to BA’s argument that the Court of First Instance overestimated the ‘very noticeable effect [of the commission schemes] at the margins’. For, in essence, BA is thus second-guessing the factual and evidential assessment of the Court in the proceedings at first instance, which is not permissible in appeal proceedings.
c) Second step in the examination: Objective economic justification
56. Following its examination of foreclosure effect (exclusionary effect), the Court of First Instance rightly turned to the question whether BA’s commission schemes could be objectively justified in economic terms.
57. Not all rebates and bonuses which a dominant undertaking grants to its contractual partners and produce a foreclosure effect are necessarily abusive and therefore prohibited under Article 82 EC. According to consistent case-law, such rebates and bonuses are to be regarded as abusive only if they are not based on an economic transaction which justifies them. (59) If there is a discernible objective economic justification for the rebates or bonuses, they are not to be regarded as abusive, despite their foreclosure effect.
58. In order to illustrate the difference between rebates or bonuses which are objectively justified in economic terms and those which are abusive, quantity rebates and fidelity rebates are often contrasted. (60) In Hoffmann-La Roche, for example, (61) the Court of Justice stated: ‘The fidelity rebate, unlike quantity rebates exclusively linked with the volume of purchases from the producer concerned, is designed through the grant of a financial advantage to prevent customers from obtaining their supplies from competing producers.’ In the same judgment, the Court saw it as characteristic of an economically justifiable quantity rebate that it is granted having regard to objectively determined amounts applicable to all purchasers, and, unlike a fidelity rebate, is not based on individual sales targets, variably determined according to the contractual partner and its absorption capacity. (62)
59. Independently, however, of the use of the terms ‘quantity rebate’ and ‘fidelity rebate’, the question of the economic justification for such rebates or bonuses is always to be determined by reference to all the circumstances of the individual case. The determining factor is whether the foreclosure effect of the rebates or bonuses, which is detrimental to competition, can be compensated for, or more than compensated for, by efficiency advantages which also demonstrably enure to consumers. (63) It is thus ultimately a question of balancing the advantages and disadvantages for competition and consumers against one another. If the foreclosure effect of a dominant undertaking’s bonus or rebate scheme bears no discernible relation to advantages for competition or consumers, or if it goes beyond what is necessary to ac hieve those advantages, that bonus or rebate scheme is to be regarded as abusive.
60. For example, a rebate which is based on objective sales quantities that apply in the same way to all contractual partners can normally be explained by cost savings which the producer can achieve by producing larger quantities. (64) It will normally be different in the case of a rebate which depends on the attainment of individually defined sales targets by the contractual partner in question, and which is primarily intended to bind that partner to the dominant undertaking and restrain him from switching to competing undertakings.
61. In this case, the Court of First Instance correctly followed these criteria derived from the previous case-law. It considered in detail the question of the economic justification of BA’s commission schemes. (65) It correctly declined to stick to the purely schematic classification of those schemes as quantity or loyalty rebates and also comprehensively took a position on BA’s arguments, particularly on the significance of fixed costs and capacity utilisation in air transport. On the basis of its assessment of the circumstances of the individual case, the Court came to the conclusion that BA’s commissions had no objective economic justification.
62. In this connection, it must be repeated that it is not one of the tasks of the Court of Justice on an appeal to substitute its appraisal of market conditions and the competitive situation for that of the Court of First Instance. (66) In particular, therefore, no account should be taken of BA’s renewed arguments on the significance of fixed costs and capacity utilisation in air transport. In essence, BA is thereby second-guessing the assessment of facts and evidence by the Court at first instance, which is not permissible in appeal proceedings.
d) Intermediate result
63. It follows that the Court of First Instance did not make any error of law in its examination of BA’s commission schemes in the light of the relevant criteria. The first ground of appeal should therefore be dismissed as unfounded.
C – The second ground of appeal: Effects of the commission schemes on competitors
64. BA’s second ground of appeal is closely connected with the first. It relates to paragraphs 293 to 298 of the contested judgment and concerns once again the findings of the Court of First Instance on the effects of BA’s commission schemes. The Court proceeded on the basis that it was not necessary to demonstrate a concrete effect on the markets concerned, (67) those commission schemes being in any case clearly likely to have a restrictive effect on the United Kingdom markets for air travel agency services and air transport and the Commission having demonstrated such an effect in a concrete way. (68)
1. Essential arguments of the parties
65. BA accuses the Court of First Instance of ignoring altogether that Article 82 EC requires an examination of the effects of commission schemes on the market, or at any rate of using incomplete and contradictory arguments in that regard. In the first place, the Court allowed itself to be satisfied by way of proof of the restrictive effects of the commission schemes with the fact that the conduct of the dominant undertaking ‘is capable of having, or likely to have, such an effect’. (69) Secondly, it assumed from the mere fact that 85% of air tickets sold in the United Kingdom during the period in question were sold through travel agents that BA’s commission schemes ‘[could] not fail’ to have their foreclosure effect. (70) It saw it as immaterial whether the commission schemes actually achieved their anti-competitive object. Finally, it disregarded evidence to the contrary, showing that BA’s commission schemes had no substantial foreclosure effect on its competitors; BA’s market share fell during the period in question, while that of its competitors increased.
66. Virgin regards this ground of appeal as inadmissible, and the Commission regards it as unfounded.
2. Assessment
67. The main part of BA’s second ground of appeal concerns the question whether, in order to find an abuse under Article 82, it is necessary to prove that the conduct of the dominant undertaking had actual and material effects on its competitors. That is a legal question, which may legitimately be raised in an appeal.
68. The starting-point here must be the protective purpose of Article 82 EC. The provision forms part of a system designed to protect competition within the internal market from distortions (Article 3(1)(g) EC). Accordingly, Article 82 EC, like the other competition rules of the Treaty, is not designed only or primarily to protect the immediate interests of individual competitors or consumers, but to protect the structure of the market and thus competition as such (as an institution), which has already been weakened by the presence of the dominant undertaking on the market. (71) In this way, consumers are also indirectly protected. (72) Because where competition as such is damaged, disadvantages for consumers are also to be feared.
69. The conduct of a dominant undertaking is not, therefore, to be regarded as abusive within the meaning of Article 82 EC only once it has concrete effects on individual market participants, be they competitors or consumers. Rather, a line of conduct of a dominant undertaking is abusive as soon as it runs counter to the purpose of protecting competition in the internal market from distortions (Article 3(1)(g) EC). (73) That is because, as already mentioned, a dominant undertaking bears a particular responsibility to ensure that effective and undistorted competition in the common market is not undermined by its conduct. (74)
70. Significantly, BA itself states (75) that it is not necessary in each case to establish actual anti-competitive effects of a rebate or bonus scheme on competitors. The burden on competition authorities, courts, and, in some cases, private complainants, in even attempting to establish it would in many cases be entirely disproportionate.
71. What is to be proved is, rather, the mere likelihood of the conduct in question hindering the maintenance or development of competition still existing in the market by means other than competition on the merits, thereby prejudicing the goal of effective and undistorted competition in the common market. With regard, therefore, to rebates and bonuses of a dominant undertaking, it has to be proved that they are capable (76) of making it difficult or impossible for that undertaking’s competitors to have access to the market and its business partners to choose between various sources of supply. (77)
72. It is true that when doing so, as already mentioned in relation to the first ground of appeal, (78)all the circumstances of the individual case must always be assessed. Those circumstances, particularly the criteria and rules for granting rebates or bonuses, or particular market circumstances, may reveal that the conduct of a dominant undertaking was not in fact capable of hindering competition on the market in question.
73. It depends, in other words, whether the rebates or bonuses of the dominant undertaking were capable not only in the abstract but also in the particular case of making it difficult or impossible for its competitors to have access to the market and its business partners to choose between various sources of supply and various business partners.
74. Whether, on the other hand, abusive pricing practices of a dominant undertaking have successfully produced their anti-competitive effect by hindering or even excluding competitors can, if necessary, play a part in determining the amount of a fine to be imposed. (79) In this appeal, however, BA has not accused the Court of First Instance of any errors of law in determining the fine.
75. In this case, the Court of First Instance correctly let itself be guided by the principles set out in paragraphs 67 to 73 above and was correctly satisfied with proof that the abusive conduct of the dominant undertaking ‘tends to restrict competition, or, in other words, that the conduct is capable of having, or likely to have, such an effect’. (80)
76. BA’s criticism of the Court’s reference to ‘capable of having’ and ‘likely to have’ is not convincing. It is too strongly bound to the wording of a single passage in the judgment, and is, moreover, based on a purely semantic distinction. The actual yardstick laid down by the Court in this case is expressed by the formula ‘tends to restrict competition’, which the Court of Justice has already relied on, particularly in the Michelin I judgment. (81)
77. If one looks, moreover, at a number of other passages in the contested judgment, it becomes clear that, in this case, the Court of First Instance did not restrict itself to a purely abstract examination of BA’s commission schemes, but assessed the likelihood of their hindering competition in a quite concrete way by reference to the circumstances of the case. In particular, the Court took account of the concrete market circumstances, such as the development of BA’s and its competitors’ market shares and the fact that, at the material time, 85% of all air tickets sold in the United Kingdom were sold through travel agents. (82)
78. All in all, therefore, the Court of First Instance did not disregard the legal requirements in relation to proof of the likelihood of the dominant undertaking’s conduct restricting competition. The first part of the second ground of appeal is thus admissible, but unfounded.
79. In its further arguments under the second ground of appeal, BA is essentially claiming that the Court of First Instance gave insufficient weight to contrary evidence such as the reduction in BA’s market share, which showed that its commissions did not have an effect on competitors. The Court also wrongly relied in its argument on the fact that, at the material time, 85% of all air tickets sold in the United Kingdom were sold through travel agents.
80. Suffice it to say here that it is not the function of the Court of Justice on an appeal to substitute its appraisal of market conditions and the competitive situation for that of the Court of First Instance. That is because assessment of facts and evidence is – save in the event of distortion, which has not been claimed here – solely the task of the Court of First Instance and cannot be re-examined on an appeal. (83) Where the Court of First Instance has established or assessed the facts, the Court of Justice is empowered under Article 225 EC only to review the legal classification of those facts and the legal consequences which the Court of First Instance drew from them. (84)
81. The question whether, in view of the involvement of travel agents in 85% of all air ticket sales, BA’s commission schemes were likely to have a foreclosure effect forms just as much part of the assessment of concrete market circumstances as the conclusions which the Court drew from the reduction in BA’s market share during the relevant period. The considerations raised by the Court in that regard concern not the legal qualification of BA’s conduct as abusive but preliminary questions of a factual nature for that qualification.
82. This part of BA’s arguments under the second ground of appeal can thus already be seen as inadmissible.
83. It would be different if, in its statements on the reduction of BA’s market share, the Court of First Instance had obviously infringed the laws of logic. A complaint of infringement of the laws of logic must, like a claim of distortion of the facts or evidence, be admissible in an appeal. Even if, however, one were to interpret the submissions of BA in this sense, they would in any event be unfounded. As the Court of First Instance has correctly pointed out, (85) one cannot exclude the possibility that, without BA’s commission schemes, the market shares of its competitors would have grown even more strongly. The reduction which was found in BA’s market share did not necessarily, therefore, have to be regarded as an indicator that its commission schemes were ineffective.
84. The second ground of appeal should therefore be dismissed in its entirety.
D – The third ground of appeal: Prejudice to consumers within the meaning of subparagraph (b) of the second paragraph of Article 82 EC
85. BA’s third ground of appeal is likewise closely connected with its first. In it, BA accuses the Court of First Instance of an error in law in not examining whether BA’s conduct prejudiced consumers within the meaning of subparagraph (b) of the second paragraph of Article 82 EC.
86. As already mentioned, (86) Article 82 EC is not designed only or primarily to protect the immediate interests of individual competitors or consumers, but to protect the structure of the market and thus competition as such (as an institution), which has already been weakened by the presence of the dominant undertaking on the market. Accordingly, Article 82 EC applies not only to conduct which can directly prejudice consumers, but also to conduct which can prejudice them indirectly in that it is detrimental to a state of effective competition for the purposes of Article 3(1)(g) EC. (87)
87. It is thus sufficient to demonstrate that the rebate or bonus scheme of a dominant undertaking is likely to make it difficult or impossible for its competitors to gain access to the market and its business partners to choose between various sources of supply or business partners, unless there is an objective economic justification for it. Where there is such a hindrance to remaining competition, it can be assumed that, indirectly, consumers are also disadvantaged.
88. That is not altered by the fact that subparagraph (b) of the second paragraph of Article 82 EC expressly refers to the limitation of production, markets or technical development to the prejudice of consumers. That is because the provision gives only an example of the abuse of a dominant position, (88) on which neither the Commission nor the Court of First Instance have relied in this case. Rebates and bonuses of dominant undertakings can infringe Article 82 EC even if they do not fall within any of the examples given in the second paragraph. (89)
89. But even if subparagraph (b) of the second paragraph of Article 82 EC were to be applied in a case such as this, it would be sufficient for a finding of prejudice to consumers to prove that, without objective economic justification, the rebates or bonuses of the dominant undertaking make it difficult or impossible for its competitors to compete with it. (90) Here also, an indirect detriment to consumers can be assumed where it is shown that the conduct of a dominant undertaking is likely adversely to affect the structure of competition, unless there is an objective economic justification for it.
90. The Court of First Instance used precisely that line of argument in the contested judgment. (91)
91. In these circumstances, there is no discernible error of law by the Court of First Instance. Therefore, the third ground of appeal should also be dismissed as unfounded.
E – The fourth ground of appeal: Length of the reference period and failure to quantify the effects of the commission schemes on competitors
92. BA’s fourth ground of appeal is in two parts, the first being concerned with the differences between the marketing agreements and the new performance reward scheme, whereas the second turns once again to the requirements for proving the foreclosure effect of these commission schemes.
93. In the first part of its fourth ground of appeal, BA complains that the Court of First Instance wrongly ascribed the same effects to the marketing agreements and the performance reward scheme. It argues that each was subject to separate conditions and that at least one of them, the performance reward scheme, could not have had foreclosure effect on account primarily of its short reference periods.
94. It is true that the length of the reference period for the rebates or bonuses granted by the dominant undertaking can have an influence on their possible foreclosure effect. (92) The longer the reference period lasts, the greater the uncertainty can be for the contractual partner as to whether, at the end of the period, he will have achieved sufficient turnover to enjoy a rebate or bonus. (93) Up to that point, he also has no certainty what net price per item he has to pay for the products in question during the reference period and therefore no certainty how large his own profit margin is.
95. What determines whether rebate and bonus schemes are likely to have a foreclosure effect on the market, however, is – as already mentioned – an overall assessment of all the circumstances of the particular case. (94) As the Commission has correctly pointed out, what counts is not just the duration in absolute terms of the reference period in which turnover must be achieved, but also how far back the relevant period for comparison lies. One cannot exclude the possibility that even a system in which reference is made month by month to periods lying one year back will, on account of the continual incentive it gives to increase turnover, result in a long-term binding of the contractual partner to the dominant undertaking, making it hard for him to switch to the competition.
96. In this case, the Court of First Instance expressly took the view that the new performance reward scheme also had a ‘fidelity-building effect’, (95) although in relation to the reference periods – like the Commission before it – it had clearly pointed out the differences between it and the marketing agreements. (96) What was decisive, however, according to the findings of the Court of First Instance was not so much the duration of the respective reference periods as the fact that both schemes, on account of their ‘very noticeable effect at the margin’ could unleash exponential changes in commission rates from one period to the next (97) and the fact that, on account of their much smaller market share, BA’s competitors were not in a position to offset the absolute effect of these commissions by counter-offers. (98) The Court of First Instance found those similarities of the two commission schemes the decisive factor in this case.
97. The assessment thus carried out of the circumstances of the individual case belongs to the evaluation of facts and evidence and is the task of the Court of First Instance alone. As already stated, it is not a function of the Court of Justice on an appeal to substitute its appraisal of market conditions and the competitive situation for that of the Court of First Instance. (99) Nor, therefore, can it substitute its assessment as to the duration of the reference periods in this particular case, and of their significance for the foreclosure effect of BA’s commission schemes, for the assessment of the Court of First Instance.
98. Since, therefore, no error of law has been established, the first part of the fourth ground of appeal is unfounded.
99. In the second part of its fourth ground of appeal, BA accuses the Court of First Instance of not quantifying its statements on the foreclosure effect of its commission schemes and therefore not examining all the circumstances of the individual case. It confined itself to making general allegations such as the ‘very noticeable effect at the margin’ and the possibility of commission rates ‘rising exponentially’ from one reference period to the next. (100)
100. Contrary to what the Commission claims, this argument cannot be dismissed as out of time under Article 42(2) in conjunction with Article 118 of the Rules of Procedure on the ground that BA failed at first instance to challenge the relevant part of the Commission’s decision, namely the calculations made in the 30th recital in the preamble to the decision. That is because BA’s complaint now is directed not against the Commission’s arithmetical examples as such but against the statements of the Court of First Instance on BA’s commission schemes which BA is criticising. The second part of the fourth ground of appeal is therefore admissible.
101. The content of BA’s argument is not convincing, however. The Court’s statements of which BA complains are to be viewed in connection with the arithmetical examples given by the Commission, which the Court cites expressly in the contested judgment from the 30th recital in the contested decision and reproduces verbatim. Viewed in that way, the statements by the Court of which BA complains are sufficiently quantified. The complaint that they are too imprecise cannot therefore succeed.
102. Both parts of the fourth ground of appeal should therefore be dismissed as unfounded.
F – The fifth ground of appeal: Discriminatory effect of the commission schemes (subparagraph (c) of the second paragraph of Article 82 EC)
103. BA’s fifth ground of appeal concerns paragraphs 233 to 240 of the contested judgment, in which the Court of First Instance confirms the findings of the Commission on the discriminatory character of BA’s commission schemes. The Court concludes in those paragraphs that BA’s commission schemes produced discriminatory effects amongst travel agents in the United Kingdom and thereby placed some of them at a competitive disadvantage within the meaning of subparagraph (c) of the second paragraph of Article 82 EC. (101)
1. Essential arguments of the parties
104. In BA’s submission, subparagraph (c) of the second paragraph of Article 82 EC does not require that all contractual partners of a dominant undertaking must have the benefit of the same prices and conditions. Such an interpretation would run counter to a rational competition policy. It argues that differences are prohibited only where compared transactions are equivalent, the conditions applied to those transactions are different, and as a result of such differences one business partner suffers a competitive disadvantage in relation to the other. Against that background, BA takes the view that the Court of First Instance wrongly applied subparagraph (c) of the second paragraph of Article 82 EC in this case.
105. In the first place, it argues that the Court of First Instance ignored the fact that the situation of travel agents whose turnover in BA tickets increased during a given period is not comparable with that of other travel agents unable to show such a growth in turnover. Essentially, BA argues that a travel agent who increases his turnover in tickets of a particular airline is particularly useful for that airline, and it is justified to reward him for that.
106. It argues in addition that, instead of applying the express wording of subparagraph (c) of the second paragraph of Article 82 EC, the Court of First Instance simply assumed a competitive disadvantage between the travel agents as ‘natural’. (102) The contested judgment contained no further examination of that competitive disadvantage.
107. The Commission and Virgin, on the other hand, unanimously take the view that BA’s commission schemes treated comparable factual situations differently without objective justification. The Commission further argues that a detailed analysis of the competitive disadvantage for the travel agents affected was not legally necessary; Virgin argues that such a disadvantage is obvious in any event.
2. Assessment
108. Subparagraph (c) of the second paragraph of Article 82 EC gives as an example of the abuse of a dominant market position ‘applying dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage’.
109. It is undisputed in this case that, in relation to travel agents doing business in the United Kingdom, BA applied different commission rates according to whether they had achieved their individual turnover targets in comparison with the period the previous year or not.
110. It remains to be determined whether the Court of First Instance was right to assume that the factual situations were comparable (‘equivalent transactions’) and whether it could, without any error of law, dispense with detailed findings as to the existence of a competitive disadvantage.
a) Equivalence of the transactions of the travel agents (first part of the fifth ground of appeal)
111. The contested judgment sets out from the premiss that two travel agents achieving ‘an identical amount of revenue’ with BA tickets during the reference period, i.e. whose turnover in BA tickets was the same in absolute terms during that period, produced equivalent transactions (‘identical services’). (103)
112. The necessary assessment of the circumstances of the individual case, from which the comparability or dissimilarity of travel agents’ services for an airline like BA may be inferred, (104) is in principle a part of the assessment of facts and evidence and thus a task for the Court of First Instance alone. As already stated, it is not within the competence of the Court of Justice in appeal proceedings to substitute its own appraisal of market conditions and the competitive situation for that of the Court of First Instance. (105)
113. The Court of Justice can, however, take a position on the criteria used by the Court of First Instance, since it is a question of law whether, in assessing the circumstances of the particular case, the Court of First Instance used permissible or impermissible criteria, or whether it may have failed to take account of criteria that legally had to be taken into account.
114. Like all prohibitions on discrimination contained in the Treaty, the particular prohibition on discrimination in subparagraph (c) of the second paragraph of Article 82 EC is an expression of the general principle of equal treatment and requires that comparable factual situations are not treated differently and different factual situations are not treated in the same way, in so far as there is no objective justification for such treatment. (106) In other words, only legitimate business considerations can justify different treatment of business partners by the dominant undertaking. (107) Such legitimate considerations can, for example, justify quantity rebates. (108) By contrast, business considerations which, in the circumstances of the particular case, represent anti-competitive behaviour cannot in any way be used as justification for unequal treatment of business partners.
115. In this case, BA essentially complains that the Court of First Instance should have taken account of the greater economic usefulness – from the airline’s point of view – of those travel agents who regularly fulfilled or exceeded their individual turnover targets.
116. The Court of First Instance was right to disregard that criterion because, according to its findings, the individual turnover targets and the incentive to increase them were, in the circumstances of this case, part of an anti-competitive business practice of BA. Therefore, BA could not legitimately attach business considerations to the attainment or otherwise of those individual turnover targets. Attainment or otherwise of the those targets, on which BA’s commission schemes were based, could not represent an objective ground for differentiation between the performances of the United Kingdom travel agents.
117. The Court would otherwise have contradicted its own finding that the commissions granted by BA had an anti-competitive exclusionary effect on account of their ‘fidelity-building effect’ and could not be objectively justified on economic grounds. (109) One and the same circumstance cannot be on the one hand branded anti-competitive and at the same time be recognised as an objective ground of differentiation. If it is an abuse to bind one’s business partners to oneself by means of particular individual turnover targets, it cannot be legitimate to differentiate between the performances of those same partners on the basis of precisely that anti-competitive criterion, i.e. according to whether or not they fulfilled the individual turnover targets set them. (110)
118. The fact that travel agents’ attainment of individual turnover targets was worth striving for from BA’s point of view and deserving of reward is irrelevant in that connection, because abusive exploitation of a dominant market position is an objective concept. (111) Accordingly, the question of discrimination amongst business partners must be determined according to objective, not subjective, criteria.
119. The Court of First Instance did not therefore make any error of law in holding that the performances of travel agents whose turnover in BA tickets was just as high in absolute terms were of equal value.
b) Requirements for determining a competitive disadvantage (second part of the fifth ground of appeal)
120. The further question arises whether it was sufficient for the Court of First Instance simply to hold that travel agents’ ability to compete was ‘naturally affected by the discriminatory conditions of remuneration’, or whether concrete proof of a competitive disadvantage was necessary.
121. In essence, the question arises whether subparagraph (c) of the second paragraph of Article 82 EC lays down a two-stage test, i.e. whether the expression ‘thereby placing them at a competitive disadvantage’ has an independent content, or whether it is merely in the nature of an explanatory addition with declaratory effect.
122. The existing case-law on this provision gives no guidance. (112)
123. The starting point for consideration of the question should be the sense and purpose of subparagraph (c) of the second paragraph of Article 82 EC. The particular prohibition on discrimination laid down in this provision is part of the system whereby in accordance with Article 3(1)(g) EC competition in the internal market is to be protected from distortions. The business practice of the dominant undertaking should not distort competition on an upstream or downstream market, i.e. between suppliers or customers of that undertaking. The business partners of the dominant undertaking should not be advantaged or disadvantaged in competition amongst one another.
124. Accordingly, the second half of the sentence in subparagraph (c) of the second paragraph of Article 82 EC is more than just an explanatory addition with declaratory effect. For that provision to apply, it must first be determined that a relationship of competition exists between the relevant trading partners of the dominant undertaking, (113) and secondly be demonstrated that the conduct of the dominant undertaking is likely in the particular case to distort that competition, i.e. to prejudice the competitive position of some of the dominant undertaking’s trading partners in relation to the others.
125. However, proof that quantifiable damage or an actual, quantifiable worsening of the competitive position of individual trading partners of the dominant undertaking actually took place cannot be demanded. That is because, as already stated, Article 82 EC serves primarily to protect competition as an institution, (114) Therefore, under subparagraph (c) of the second paragraph also, discrimination amongst trading partners in competition with each other can be regarded as abusive once the conduct of the dominant undertaking is likely in the circumstances of the individual case to cause competition amongst those trading partners to be distorted.
126. If one applies these criteria, the argument of the Court of First Instance in the contested judgment is revealed as extraordinarily scanty.
127. The Court did nevertheless determine that travel agents in the United Kingdom compete intensively with each other. (115) It further determined that the capacity of travel agents to compete with each other depended on two factors: first their capacity to ‘provide seats on flights suited to travellers’ wishes, at a reasonable cost’, and secondly their financial means. (116)
128. The fact that BA’s commission schemes could lead to sharp and significant variations in the income of the individual travel agents is stated by the Court of First Instance right at the beginning of its judgment and also subsequently in the context of the ‘fidelity-building effect’. (117)
129. Against the factual background of this case, it was legitimate for the Court, in the context of the examination of subparagraph (c) of the second paragraph of Article 82 EC, to proceed immediately, and without any more detailed intermediate steps, to its conclusion that the travel agents’ opportunities for competition amongst each other were prejudiced by BA’s discriminatory reward conditions (118) (even if it would have been better to refer to the likelihood in the particular case of that competition being prejudiced). Whether it was appropriate to use the expression ‘naturally’ in that context need not be determined here.
130. Since the Court of Justice itself has, hitherto, examined only in a very summary way whether and how discriminatory trading conditions have had an impact of the competitive position of trading partners, (119) I conclude that, in this case also, the Court of First Instance can likewise not be accused of an error of law in that respect and that, in the circumstances, it was entitled to assume that BA’s commission schemes had a discriminatory effect for the purposes of subparagraph (c) of the second paragraph of Article 82 EC.
131. The fifth ground of appeal should therefore be dismissed in its entirety as unfounded.
132. I would mention purely for the sake of completeness that, under the case-law of the Court of Justice, a finding that rebates or bonuses granted by a dominant undertaking had discriminatory effect is not a mandatory precondition for finding that there has been an abuse of a dominant market position. In its judgment in Michelin I, for example, the Court found the rebates in question abusive, even though it found their discriminatory effect unproven. (120)
G – Intermediate result
133. Since none of BA’s grounds of appeal have any prospect of success, I conclude that its appeal should be dismissed in its entirety.
V – Costs
134. Under Article 69(2) in conjunction with Articles 118 and 122(1) of the Rules of Procedure, the unsuccessful party is to be ordered to pay the costs if they have been applied for in the successful party’s pleadings. Since the Commission has applied for costs and BA has been unsuccessful, BA should be ordered to pay the costs.
135. Under the third subparagraph of Article 69(4) in conjunction with Articles 118 and 122(1) of the Rules of Procedure, the Court could order Virgin, as an intervener, to bear its own costs. In this case, however, since Virgin has intervened in support of the successful party, it seems appropriate to order BA to pay Virgin’s costs also, in accordance with Virgin’s application.
VI – Conclusion
136. In the light of the above considerations, I propose that the Court should:
1) dismiss the appeal.
2) order British Airways plc to pay the costs of the proceedings.
1 – Original language: German.
2 – Commission Decision 2000/74/EC of 14 July 1999 relating to a proceeding under Article 82 of the EC Treaty (IV/D-2/34.780 – Virgin/British Airways) (notified under document number C(1999) 1973) (OJ 2000 L 30, p. 1).
3 – British Airways v Commission [2003] ECR II-5917.
4 – A first complaint was made on 9 July 1993 and an additional complaint was made on 9 January 1998 (see paragraphs 12 and 19 of the contested judgment).
5 – Recitals 90 and 91 of the contested decision and paragraph 22 of the contested judgment.
6 – Recital 31 of the contested decision and paragraph 21 of the contested judgment.
7 – Paragraph 4 of the contested judgment.
8 – Paragraph 14 of the contested judgment and paragraph 7 of the notice of appeal.
9 – Paragraph 5 of the contested judgment.
10 – Paragraphs 6 to 11 of the contested judgment.
11 – Paragraphs 14 to 18 of the contested judgment.
12 – Recitals 29 and 30 of the contested decision and paragraph 23 of the contested judgment.
13 – The expression ‘commission schemes’ is used hereafter as a generic term for marketing agreements applied by BA and their new performance reward scheme.
14 – Recital 96 of the contested decision and paragraph 24 of the contested judgment.
15 – The expression ‘its performance reward systems’ (French: ‘ses systemes de primes de résultat’) in paragraph 25 and elsewhere in the contested judgment is, however, uncertain, for it appears to relate solely to the new performance reward scheme. It is, however, clear from recitals 29, 30, 102 and 109 of the contested decision that there reference is made to both commission schemes – both the marketing agreements and the performance reward scheme.
16 – Recital 102 of the contested decision and paragraph 25 of the contested judgment.
17 – Recital 109 of the contested decision and paragraph 25 of the contested decision.
18 – Recitals 103 and 111 of the contested decision and paragraph 26 of the contested judgment.
19 – See, for example, Case 85/76 Hoffmann-La Roche v Commission [1979] ECR 461, paragraph 91; Case 322/81 Michelin v Commission (MichelinI) [1983] ECR 3461, paragraph 70; Case 31/80 L’Oréal v De Nieuwe AMCK [1980] 3775, paragraph 27; Case C‑62/86 AKZO v Commission [1991] ECR I-3359, paragraph 69).
20 – Michelin I (cited in footnote 19), paragraph 57.
21 – Joined Cases C-395/96 P and C-396/96 P Compagnie Maritime Belge Transports and Others v Commission [2000] ECR I-1365, paragraph 131.
22 – Case 27/76 United Brands v Commission [1978] ECR 207, paragraph 189.
23 – See, to that effect Hoffmann-La Roche, paragraphs 91 and 123, Michelin I, paragraph 70, L’Oréal, paragraph 27, and AKZO, paragraphs 69 and 70 (all cited in footnote 19 above).
24 – AKZO, paragraph 70.
25 – Joined Cases 40/73 to 48/73, 50/73, 54/73 to 56/73, 111/73, 113/73 and 114/73 Suiker Unie and Others v Commission [1975] ECR 1663, paragraph 517 et seq.; Hoffmann-La Roche, paragraph 90 et seq.; Michelin I, paragraph 62 et seq.; Case C-163/99 Portugal v Commission [2001] ECR I-2613, paragraph 50 et seq. See also Case T-30/89 Hilti v Commission [1991] ECR II-1439, paragraph 101; Case T-65/89 BPB Industries and British Gypsum v Commission [1993] ECR II‑389, paragraphs 71 and 120; Case T-228/97 IrishSugar v Commission [1999] ECR II-2969, paragraphs 198, 201 and 213; Case T-203/01 Michelin v Commission (Michelin II) [2003] ECR II-4071, paragraph 53 et seq. Also the contested judgment (cited in footnote 3).
26 – In the proceedings before the Court of Justice, BA has repeatedly submitted that the Commission is envisaging a reform of its practice in relation to Article 82 EC and is planning the publication of a discussion paper for that purpose.
27 – See in particular paragraphs 273 (last sentence), 278 and 292 of the contested judgment.
28 – Cited in footnote 19.
29 – It is immaterial on this view whether such a condition is determined by contract or applied unilaterally by the dominant undertaking.
30 – Joined Cases C-267/91 and C-268/91 [1993] ECR I-6097.
31 – Case 6/72 Europemballage and Continental Can v Commission (ContinentalCan) [1973] ECR 215, paragraph 26; Case C-333/94 P Tetra Pak v Commission [1996] ECR I-5951, paragraph 37; and Compagnie Maritime Belge Transports (cited in footnote 12), paragraph 112.
32 – In Hoffmann-La Roche and Michelin I (cited in footnote 19), for example, the Court of Justice based its reasoning on exclusionary effect on Article 86 of the EEC Treaty (now Article 82 EC) generally, and not exclusively on subparagraph (b) of its second paragraph. Only in Suiker Unie (cited in footnote 25), paragraph 526, did it refer expressly to subparagraph (b) of the second paragraph of that provision.
33 – Hoffmann-La Roche (cited in footnote 19), paragraphs 82 to 87. Similarly, in relation to the sugar market, Suiker Unie (cited in footnote 25), particularly paragraphs 499 and 510.
34 – Hoffmann-La Roche (cited in footnote 19), paragraph 89. Similarly Suiker Unie (cited in footnote 25), especially paragraphs 518 and 527.
35 – Hoffmann-La Roche (cited in footnote 19), paragraph 90.
36 – Michelin I (cited in footnote 19), paragraph 86.
37 – Michelin I (cited in footnote 19), paragraph 72.
38 – Michelin I (cited in footnote 19), paragraph 66 et seq.
39 – Michelin I (cited in footnote 19), paragraph 81.
40 – See in that regard Hoffmann-La Roche, paragraph 90, and Michelin I, paragraph 85 (cited in footnote 19).
41 – Paragraphs 244 and 245 of the contested judgment.
42 – Michelin I (cited in footnote 19), paragraph 73, first sentence, in conjunction with paragraph 72, last sentence.
43 – See the case-law cited in footnote 19.
44 – See point 33, second indent, of this Opinion.
45 – Michelin I (cited in footnote 19), paragraph 73, first sentence.
46 – Michelin I (cited in footnote 19), paragraph 73, first sentence.
47 – Hoffmann-La Roche, paragraph 90, and Michelin I, paragraphs 71, 73 – second sentence – and 85 (both cited in footnote 19). Similarly in Suiker Unie (cited in footnote 25), paragraph 526.
48 – See Michelin I (cited in footnote 19), paragraphs 70 to 86.
49 – On the significance of the duration of the reference period, see the observations on the first part of the fourth ground of appeal in paragraphs 94 to 98 of this Opinion.
50 – Suiker Unie (cited in footnote 25), paragraph 527, and Hoffmann-La Roche (cited in footnote 19), paragraph 90, last sentence.
51 – See in particular paragraphs 10 and 15 to 17 of the contested judgment, reproduced in points 9 and 10 of this Opinion.
52 – See also Michelin I (cited in footnote 19), paragraph 81.
53 – See point 11 of this Opinion, where the corresponding findings of the Commission, to which the Court of First Instance also refers, are reproduced.
54 – Paragraphs 272 and 273 of the contested judgment.
55 – Hoffmann-La Roche (cited in footnote 19), paragraph 41, and Compagnie Maritime Belge Transports (cited in footnote 21), paragraph 132.
56 – See also Michelin I (cited in footnote 19), paragraph 82.
57 – Paragraphs 276 and 277 of the contested judgment.
58 – Case C-37/03 P BioID v OHIM [2005] ECR I-0000, paragraphs 43 and 53; Joined Cases C‑204/00 P, C-205/00 P, C-211/00 P, C-213/00 P, C-217/00 P and C-219/00 P Aalborg Portland and Others v Commission (Cement) [2004] ECR I-123, paragraphs 47 to 49.
59 – Hoffmann-La Roche, paragraph 90, and Michelin I, paragraph 85 (cited in footnote 19).
60 – Suiker Unie (cited in footnote 25), paragraph 518, Hoffmann-La Roche (cited in footnote 19), paragraphs 90 and 100, and Michelin I (cited in footnote 19), paragraphs 71 and 72. See also paragraph 244 et seq. of the contested judgment (cited in footnote 3).
61 – Cited in footnote 19, paragraph 90.
62 – Hoffmann-La Roche (cited in footnote 19), paragraph 100.
63 – Similar considerations on the account to be taken of efficiency advantages can be found, in the area of merger control, in recital 29 of Council Regulation (EC) No 139/2004 of 20 January 2004 on the control of concentrations between undertakings (the EC Merger Regulation; OJ 2004 L 24, p. 1) and in paragraphs 76 to 88 of the Commission’s Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings (OJ 2004 C 31, p. 5), with further reference to Article 81(3) EC in paragraphs 135 and 141, last sentence, of the ‘Commission notice – Guidelines on Vertical Restraints’ (OJ 2000 C 291, p. 1).
64 – It need not be determined here whether quantity rebates which are based on objective quantities the same for all purchasers can have abusive effects in individual cases on account of the criteria and circumstances of their being granted. See Michelin II (cited in footnote 25) and Portugal v Commission (cited in footnote 25), paragraph 50 et seq.
65 – See paragraphs 279 to 291 of the contested judgment.
66 – See point 55 of this Opinion and the case-law cited in footnote 58. That obviously applies only where there has been no falsification of facts or evidence, which has, however, not been argued in this case.
67 – Paragraph 293 of the contested judgment.
68 – Paragraph 294 of the contested judgment.
69 – Paragraph 293 of the contested judgment.
70 – Paragraph 295 of the contested judgment.
71 – Continental Can (cited in footnote 31), paragraph 26, and Hoffmann-La Roche, paragraphs 91, 123 and 125, Michelin I, paragraph 70, and L’Oréal, paragraph 27 (all cited in footnote 19).
72 – See to that effect ContinentalCan (cited in footnote 31), paragraph 26, and Hoffmann-La Roche (cited in footnote 19), paragraph 125.
73 – Joined Cases 6/73 and 7/73 Istituto Chemioterapico Italiano and Commercial Solvents v Commission [1974] ECR 223, paragraph 25.
74 – Michelin I (cited in footnote 19), paragraph 57.
75 – Paragraph 85 of the notice of appeal.
76 – Suiker Unie (cited in footnote 25), paragraph 526, Hoffmann-La Roche (cited in footnote 19), paragraph 90, and Michelin I (cited in footnote 19), paragraph 73, second sentence, and paragraph 85, first sentence. On the criterion of likelihood, see also Case C-7/97 Bronner [1998] ECR I-7791, paragraph 38.
77 – As already stated, however, there is no abuse if there is an objective economic justification for the conduct of the dominant undertaking (see points 56 to 60 of this Opinion).
78 – See points 45 and 46 of this Opinion.
79 – See the judgment in AKZO (cited in footnote 19), paragraph 163, in which the Court of Justice saw the absence of any significant influence on the affected parties’ respective market shares as a reason to reduce the amount of the fine imposed by the Commission.
80 – Paragraph 293 of the contested judgment. The full sentence reads: ‘It is sufficient in that respect to demonstrate that the abusive conduct of the undertaking in a dominant position tends to restrict competition, or, in other words, that the conduct is capable of having, or likely to have, such an effect.’ The French version reads: ‘Il suffit à cet égard de démontrer que le comportement abusif de l'entreprise en position dominante tend à restreindre la concurrence ou, en d’autres termes, que le comportement est de nature ou susceptible d'avoir un tel effet.’
81 – For example, in Michelin I (cited in footnote 19), paragraph 73, second sentence, the criterion in English was ‘… whether the discount tends to remove or restrict …’ and in the authentic language of the case, French ‘… si le rabais tend … à enlever … ou à restreindre …’; the French expression ‘tend à …’ was also used in Hoffmann-La Roche (cited in footnote 19), paragraph 90; emphasis added.
82 – Paragraphs 294 to 298 of the contested judgment.
83 – See point 55 of this Opinion and the case-law cited in footnote 58.
84 – Case C-499/03 P Biegi Nahrungsmittel and Commonfood v Commission [2005] ECR I-1751, paragraph 41; Joined Cases C-2/01 P and C-3/01 P BAI and Commission v Bayer [2004] ECR I-23, paragraph 47.
85 – Paragraph 298 of the contested judgment.
86 – See point 68 of this Opinion.
87 – Continental Can (cited in footnote 31), paragraph 26, and Hoffmann-La Roche (cited in footnote 19), paragraph 125. Contrary to what BA argues, I do not think that Advocate General Jacobs expresses a view that is any different in his Opinion in Case C-7/97 Bronner [1998] ECR I‑7791, point 58. He merely states there that ‘the primary purpose of Article [82] is to prevent distortion of competition – and in particular to safeguard the interests of consumers –’, and thus also appears to take the view that Article 82 EC protects competition as an institution, thereby also indirectly safeguarding the interests of consumers.
88 – See point 35 of this Opinion and the case-law cited in footnote 31.
89 – See the case-law cited in footnote 32.
90 – See to that effect Suiker Unie (cited in footnote 25), paragraph 526, where the Court of Justice stated that a rebate scheme was ‘likely to limit markets to the prejudice of consumers within the meaning of Article [82](b), because it gave other producers and especially those having their places of business in other Member States no chance or restricted their opportunities of competing with [the product] sold by [the dominant undertaking]’.
91 – That is made particularly clear in paragraphs 296 and 311 of the contested judgment. (The Court of First Instance was also right to hold that BA’s commission schemes were likely in any case to have an exclusionary effect and thus adversely affect competition; see my observations on the first and second grounds of appeal in paragraphs 35 et seq. and 67 et seq. of this Opinion.)
92 – To that effect see also Michelin I (cited in footnote 19), paragraph 81, in which the Court draws attention to the ‘relatively long reference period’ of the target rebates (one year).
93 – That uncertainty can be increased further by the lack of transparency of the rebate or bonus system (Michelin I (cited in footnote 19), paragraph 83).
94 – See in particular point 45 of this Opinion.
95 – See paragraph 271 et seq. of the contested judgment.
96 – See the presentation of the facts in paragraphs 8 to 11 of the contested judgment on the one hand and in paragraph 15 of that judgment on the other.
97 – Paragraphs 272 and 273 of the contested judgment.
98 – Paragraphs 276 to 278 of the contested judgment.
99 – See point 55 of this Opinion and the case-law cited in footnote 58. That naturally applies only in the absence of any distortion of the facts or evidence, which has, however, not been alleged in this case.
100 – Paragraph 272 of the contested judgment.
101 – Paragraph 240 of the contested judgment.
102 – Paragraph 238 of the contested judgment.
103 – Paragraphs 235 and 236 of the contested judgment.
104 – The need for such an assessment of the circumstances of the individual case is emphasised not least in Michelin I (cited in footnote 19), paragraph 87 et seq.
105 – See point 55 of this Opinion and the case-law cited in footnote 58. That naturally applies only in the absence of any distortion of the facts or evidence, which has, however, not been alleged in this case.
106 – Consistent case-law; see for example Case C-434/02 Arnold André [2004] ECR I-11825, paragraph 68; Case C-210/03 Swedish Match [2004] ECR I-11893, paragraph 70; Case C‑110/03 Belgium v Commission [2005] ECR I-2801, paragraph 71; Case C‑403/03 Schempp [2005] ECR I-0000, paragraph 28; and Joined Cases C-453/03, C‑11/04, C-12/04 and C-194/04 ABNA and Others [2005] ECR I-0000, paragraph 63.
107 – Michelin I (cited in footnote 19), paragraph 90.
108 – A differentiated analysis of quantity rebates can be found, for example, in Portugal v Commission (cited in footnote 25), paragraph 50 et seq.
109 – See my observations on the first ground of appeal in points 44 to 62 of this Opinion.
110 – To that effect see also Hoffmann-La Roche (cited in footnote 19), paragraph 90, where the Court held in relation to fidelity rebates: ‘Furthermore the effect of fidelity rebates is to apply dissimilar conditions to equivalent transactions with other trading parties in that two purchasers pay a different price for the same quantity of the same product depending on whether they obtain their supplies exclusively from the undertaking in a dominant position or have several sources of supply’.
Michelin I (cited in footnote 19), paragraph 87 et seq., does not contradict that view. In that case, the Court of Justice saw discrimination as unproven, because the Commission’s findings on the method of functioning of Michelin’s rebate system were subsequently found to be incomplete, so that the possibility could not be excluded that the Commission thereby overlooked legitimate business considerations of Michelin (see paragraphs 89 and 90 of the judgment).
111 – Hoffmann-La Roche, paragraph 91, and AKZO, paragraph 69 (cited in footnote 19).
112 – In Michelin I (cited in footnote 19), paragraph 87 et seq., no finding of discrimination was made, so that the problem of whether a disadvantage was suffered in competition did not arise. Portugal v Commission (cited in footnote 25), paragraph 50 et seq., and Case C-82/01 P Aéroports de Paris v Commission [2002] ECR I-9297, paragraph 114 et seq., deal only with the question of discrimination; at any rate it is not clear whether the question of disadvantage to trading partners was at issue in the respective proceedings. In Suiker Unie (cited in footnote 25), paragraphs 522 to 525, brief consideration is given to the competitive relationship between the customers discriminated against. United Brands (cited in footnote 22), paragraphs 232 to 234, and Case C-179/90 Merci Convenzionali Porto di Genova v Siderurgica Gabrielli [1991] ECR I-5889, paragraph 19, suggest that the Court of Justice regards a summary examination of the effects of the conduct of dominant undertakings on the competitive position of their trading partners as necessary.
113 – See Suiker Unie (cited in footnote 25), paragraphs 524 and 525.
114 – See my observations on the second ground of appeal in paragraphs 67 to 78 of this Opinion.
115 – Paragraph 237 of the contested judgment.
116 – Paragraphs 237 and 238 of the contested judgment.
117 – See, on the one hand, paragraph 23 of the contested judgment, reproduced in point 11 of this Opinion and, on the other, paragraphs 272 and 273 of the contested judgment.
118 – Paragraph 238 of the contested judgment.
119 – See in particular United Brands (cited in footnote 22), paragraphs 232 to 234, Merci Convenzionali Porto di Genova (cited in footnote 112), paragraph 19, and Portugal v Commission (cited in footnote 25), paragraph 50 et seq.
120 – Michelin I (cited in footnote 19), paragraphs 86 and 91.