This document is an excerpt from the EUR-Lex website
Document 61992CC0053
Opinion of Mr Advocate General Jacobs delivered on 10 November 1993. # Hilti AG v Commission of the European Communities. # Appeal - Competition - Abuse of a dominant position - Concept of a relevant market. # Case C-53/92 P.
Opinion of Mr Advocate General Jacobs delivered on 10 November 1993.
Hilti AG v Commission of the European Communities.
Appeal - Competition - Abuse of a dominant position - Concept of a relevant market.
Case C-53/92 P.
Opinion of Mr Advocate General Jacobs delivered on 10 November 1993.
Hilti AG v Commission of the European Communities.
Appeal - Competition - Abuse of a dominant position - Concept of a relevant market.
Case C-53/92 P.
European Court Reports 1994 I-00667
ECLI identifier: ECLI:EU:C:1993:875
Opinion of Mr Advocate General Jacobs delivered on 10 November 1993. - Hilti AG v Commission of the European Communities. - Appeal - Competition - Abuse of a dominant position - Concept of a relevant market. - Case C-53/92 P.
European Court reports 1994 Page I-00667
++++
My Lords,
1. This case comes to the Court by way of an appeal against the judgment of the Court of First Instance of 12 December 1991 in the case Hilti v Commission (2) (hereafter "the judgment"). The judgment rejected Hilti' s application for the annulment of a decision of the Commission in which Hilti was found to hold a dominant position within the EEC in the market for nail guns and for the nails and cartridge strips for those guns and to have abused that position within the meaning of Article 86 of the Treaty. (3) The decision imposed a fine of ECU 6 000 000 and ordered Hilti to put an end to the abuses which had been found.
2. The complaints to the Commission which led to the opening of proceedings against Hilti were made by Bauco (UK) Ltd and Profix Distribution Ltd (then named "Eurofix"). Both Bauco and Profix intervened in support of the Commission in the proceedings before the Court of First Instance. Accordingly, by virtue of Articles 114 and 115 of the Rules of Procedure of this Court, both Bauco and Profix had the right to lodge a response to Hilti' s appeal. However, only Bauco has taken advantage of that opportunity.
3. The nail guns manufactured by Hilti are a technologically advanced means of making secure fastenings in the construction industry. The guns are used together with cartridge strips, cartridges and nails as a "powder-actuated fastening" (or "PAF") system which fires the nails into different materials as required; such a system is not however suitable for use with all types of materials. The cartridges provide the explosive power of the system, and each cartridge strip holds a number of cartridges, which enables the gun to be repeatedly used without the need to reload cartridges. Nails which are compatible with the guns manufactured by Hilti are manufactured and supplied by other firms, as well as by Hilti itself. Bauco and Profix are two such independent manufacturers of nails which can be used in Hilti nail guns. In what follows, all references to "cartridge strips" and "nails" are to components intended to be used in nail guns manufactured by Hilti. Such components are referred to in the Commission' s decision as "consumables".
4. Article 86 of the Treaty, first paragraph, provides that:
"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."
The present appeal is limited to those parts of the judgment of the Court of First Instance which upheld the Commission' s findings that Hilti occupied a dominant position within the common market. Thus Hilti does not now dispute that, on the assumption that it occupied such a position, the behaviour identified by the Commission in its decision amounted to abuse. Nor does Hilti now contest that the behaviour in question might have affected trade between Member States.
5. The abusive conduct alleged by the Commission consisted in Hilti' s exercise of its market power as a producer of nail guns, cartridge strips and nails in such a way as to hinder the entry into and penetration of the market for nails by independent nail producers, or otherwise to damage their business. Such conduct will however only be contrary to Article 86 if Hilti did indeed occupy a dominant position on at least one of the markets in question.
6. In order to establish the existence of a dominant position, within the meaning of Article 86, it is necessary to identify the relevant market or markets in which that dominance obtains. As this Court stated in Continental Can, (4) at paragraph 32 of the judgment:
"For the appraisal of [the applicant' s] dominant position ..., the definition of the relevant market is of essential significance, for the possibilities of competition can only be judged in relation to those characteristics of the products in question by virtue of which those products are particularly apt to satisfy an inelastic need and are only to a limited extent interchangeable with other products."
For the purposes of such an appraisal, moreover, not only the range of relevant products (the "relevant product market") but also the geographical extent of the market must be considered. In Hilti' s submission, the Commission' s decision erred in its identification of the relevant product markets, and Hilti now appeals against those parts of the judgment of the Court of First Instance which upheld the Commission' s identification of those markets. Hilti does not however now contest the Commission' s identification of the relevant geographical market.
7. The Commission identified, in its decision, three distinct relevant product markets in which Hilti was found to be dominant: the market for nail guns, and the markets for cartridge strips and nails for use in the guns. According to paragraph 74 of the decision, Hilti abused its dominant positions in those markets in order to prevent effective competition from new entrants to the market for nails.
8. The delimitation of a relevant product market, for the purposes of establishing the existence of a dominant position within the meaning of Article 86 of the Treaty, is a complex operation involving both findings of fact and the evaluation of those facts in the light of economic principles and legal criteria. (5) In an appeal from a judgment of the Court of First Instance, pursuant to Article 168a(1) of the Treaty, this Court is of course limited to the consideration of points of law. It will be recalled that Article 51, first paragraph, of the Statute of the Court of Justice provides that:
"An appeal to the Court of Justice shall be limited to points of law. It shall lie on the grounds of lack of competence of the Court of First Instance, a breach of procedure before it which adversely affects the interests of the appellant as well as the infringement of Community law by the Court of First Instance."
9. Although that limitation clearly applies to all appeals from the Court of First Instance to this Court, special factors may be relevant in appeals in competition cases which originate in decisions of the Commission finding infringements of Articles 85 and 86 of the Treaty. (6) In particular, it is to be noted that in such cases the Court of First Instance has itself had the task of reviewing the legality of a Commission decision based, on the one hand, upon findings of fact and, on the other, on a possibly complex economic assessment. Issues such as those which arise in the present case of the delimitation of the relevant market and the existence of a dominant position in that market require the application of criteria about which economists may disagree. It is to be noted that judicial review on such issues is more limited than on pure questions of fact: the correctness of the material facts relied upon by the Commission may always be reviewed by the Court of First Instance, whereas its review of matters of economic assessment is normally limited to verifying that the Commission is not guilty of any manifest error of appraisal or abuse of power. (7) As we shall see in greater detail below, the question then arises to what extent the assessment of material facts by the Court of First Instance may itself be reviewed by this Court. (8)
10. The current jurisdiction of the Court of First Instance in competition matters was originally exercised by this Court under Articles 173 and 175 of the Treaty, and was transferred to the Court of First Instance by the Council Decision of 24 October 1988 establishing a Court of First Instance. (9) It is clear from the preamble to that Decision that the reason for the establishment of a Court of First Instance was to enable the task of examining complex factual issues to be carried out by the Court of First Instance rather than by the Court of Justice. Thus the Decision of 24 October 1988 first recites that Article 168a of the Treaty empowers the Council to attach to the Court of Justice a Court of First Instance "called upon to exercise important judicial functions". It further recites that, "in respect of actions requiring close examination of complex facts, the establishment of a second court will improve the judicial protection of individual interests"; and it states that "it is necessary, in order to maintain the quality and effectiveness of judicial review in the Community legal order, to enable the Court [of Justice] to concentrate its activities on its fundamental task of ensuring uniform interpretation of Community law". Finally, it states that it is necessary to "transfer to the Court of First Instance jurisdiction to hear and determine at first instance certain classes of action or proceeding which frequently require an examination of complex facts", including actions brought by natural or legal persons in competition matters. The Council Decision of 8 June 1993 (10) sets out some of the same considerations in transferring to the Court of First Instance, in most cases with effect from 1 August 1993, (11) other classes of action brought by natural or legal persons.
11. As counsel for the Commission submitted at the hearing, it is clear that the Court of First Instance was not intended merely to be inserted as an intermediate court between the Commission and this Court, but rather to take over a substantial part of this Court' s jurisdiction. Thus it is significant that Article 3 of the Decision of 24 October 1988 speaks of the "exercise" by the Court of First Instance of the jurisdiction conferred on the Court of Justice by the Treaties, and that the preamble refers, as we have just seen, to a "transfer" of jurisdiction to the Court of First Instance.
12. Accordingly, it seems to me that, at least in competition cases, it is appropriate to construe strictly the requirement in Article 51 of the Statute that appeals to the Court of Justice are limited to points of law. On that basis, therefore, I now turn to consider Hilti' s individual grounds of appeal.
Hilti' s grounds of appeal
13. Hilti pleads seven grounds of appeal, which can be briefly summarized as follows:
(1) In paragraphs 66 and 67 of the judgment, the Court of First Instance wrongly inferred the existence of separate markets for cartridge strips and nails from the fact that there have since the 1960s been independent producers of nails for use in nail guns and the fact that cartridge strips and nails are specifically manufactured, and purchased by users, for a single brand of gun. The Court of First Instance failed to apply the correct test for defining a relevant product market in spare parts or components and ignored the question of demand substitutability as between different fastening systems.
(2) In paragraph 69 of the judgment, the Court of First Instance reached the conclusion that PAF systems constitute a relevant product market without quantifying the number of cases in which other fastening systems could be substituted for PAF systems.
(3) In paragraphs 70 and 71 of the judgment, the Court of First Instance wrongly upheld a finding by the Commission that other fastening systems were not readily interchangeable with PAF systems, although that finding was based solely on a description of the different characteristics of the products in question and was hence insufficiently grounded.
(4) Similarly, in paragraph 71 of the judgment the Court of First Instance wrongly upheld a finding of non-interchangeability based on the co-existence of different fastening methods over an extended period of time.
(5) In rejecting, in paragraph 74 of the judgment, evidence produced by Hilti tending to show interchangeability between different fastening systems, the Court of First Instance misconstrued the legal burden of proof.
(6) In paragraphs 73 and 76 of the judgment, the Court of First Instance wrongly concluded that the Commission' s findings are corroborated or at least not refuted by certain evidence produced by Hilti, namely an opinion prepared by a Mr Yarrow, a survey conducted by a firm called Rosslyn Research and an econometric analysis by a Professor Albach.
(7) The Court of First Instance failed to have regard to all the relevant evidence adduced by Hilti.
14. In what follows, I shall consider those seven grounds of appeal in turn. It is to be noted however that in some cases the issues raised by individual grounds overlap, so that the grounds cannot always be considered in isolation.
The first ground of appeal
15. The first issue raised by the appeal is whether separate markets can be said to exist in the cartridge strips and nails used in Hilti guns, or whether the guns and their consumables are to be considered as an indivisible whole. Hilti submits that the Court of First Instance failed to apply a rule for the determination of whether there is a separate market for the supply of spare parts of a product which, Hilti suggests, was laid down by this Court in the Hugin case. (12) It is of course clear that a submission that a legal rule has been ignored or misconstrued raises a point of law which can be considered on an appeal to this Court.
16. The Hugin case concerned a refusal by Hugin Cash Registers Ltd to supply spare parts for its cash registers to another firm, Liptons. Hilti refers in particular to paragraph 5 of the Hugin judgment, where the Court stated :
"To resolve the dispute it is necessary, first, to determine the relevant market. ... The question is, therefore, whether the supply of spare parts constitutes a specific market or whether it forms part of a wider market. To answer that question it is necessary to determine the category of clients who require such parts."
The Court concluded that there existed a separate market for spare parts because there were independent undertakings, specializing in the maintenance and repair of cash registers, and in the sale and reconditioning of old machines, which themselves required spare parts, such undertakings comprising a distinct class from the buyers of the machines: see paragraph 7 of the judgment in Hugin. In Hilti' s view, the Court' s reasoning in the Hugin case demonstrates that nails can be regarded as forming a product market distinct from the nail guns in which they are used only if the purchasers of such nails are found to be different from the purchasers of the guns.
17. It seems to me however that the Hugin case does not warrant such a conclusion. It was necessary there to consider the category of clients requiring spare parts because of the particular circumstances of the case. In the first place, Hugin was the monopoly supplier of the spare parts in question. Furthermore, the parts required the services of a specialized technician in order to be fitted, and their value was of little significance in relation to the cost of maintenance and repairs: the users of the cash registers did not therefore operate on the market as purchasers of spare parts. Thus, if the only customers for such parts had been the purchasers of the cash registers themselves, it is clear that there would have been no market in spare parts distinct from the market in those machines and the market for the provision of services of maintenance and repair of the machines: see paragraph 6 of the Court' s judgment.
18. However, as both Bauco and the Commission point out, Hugin is not to be interpreted as laying down any rule to the effect that, for a separate market to exist in spare parts or other components, the purchasers of the spare parts must be distinct from the purchasers of the equipment for which they are intended. Rather than laying down any such rule, the Court in Hugin was merely deciding whether a relevant product market for spare parts existed in the particular circumstances of that case. The present case is in fact quite different from Hugin. The parts purchased for use with the Hilti nail guns are not spare parts requiring the services of an expert technician, but rather consumables for the guns which are designed to be used by any person competent to operate the gun. Such consumables are moreover supplied not only by Hilti itself, but also by independent producers who manufacture nails for use in the Hilti guns. As the Court of First Instance observes in paragraph 67 of its judgment, those circumstances in themselves amount to good evidence of the existence of a separate market for nails. Thus, in contrast to the situation which obtained in the Hugin case, the users of the guns participate directly in the market as purchasers of nails, and it is immaterial that the purchasers of the nails are not a distinct group from the purchasers of the guns.
19. It is true that, in order to identify the markets for nails and cartridge strips as relevant markets in which Hilti is dominant, it may be necessary to consider Hilti' s position in markets which are closely related to the markets for cartridge strips and nails. As we have seen, in its decision the Commission identified the market in nail guns as a relevant product market in which Hilti also held a dominant position. If Hilti is dominant in such a market, it is clear that that can only serve to reinforce its position in the markets for components such as cartridge strips and nails. Conversely, Hilti' s position in the latter markets will be weaker if PAF systems compete with non-PAF systems in a wider market, comprising both kinds of system, in which Hilti is not dominant.
20. The question therefore arises whether PAF systems do indeed constitute a relevant product market, in which Hilti holds a dominant position, and which is distinct from the market for fastening systems as a whole. That may not be the case if there is a significant degree of substitutability between the different systems. Accordingly, in its first ground of appeal, Hilti also raises the question of substitutability between different fastening systems. Since however a number of Hilti' s other grounds of appeal also raise the same question, in particular the second, third, fourth and fifth grounds, I shall consider the matter under those headings.
The second ground of appeal
21. As we have seen, the Commission found that Hilti occupied a dominant position in the markets for nails and cartridge strips as well as in the market for nail guns. Hilti does not now contest the Commission' s findings as to its large market share in nails, cartridge strips and nail guns, and the other advantages, such as patent protection and a well-organized distribution system, which in the Commission' s view served to maintain and reinforce its position in those markets.
22. However, in order to establish the existence of a dominant position, it must be shown that the undertaking concerned enjoys a position of economic strength "which enables it to prevent effective competition being maintained on the relevant market by giving it the power to behave to an appreciable extent independently of its competitors, customers and ultimately of its consumers". (13) As I have already mentioned, Hilti submits that its ability to act independently on the market for nails was affected by the fact that PAF systems form part of a wider market for fastening systems generally. Thus, even if nails, cartridge strips and nail guns are regarded as distinct markets, they may not be "relevant" markets for the purposes of establishing the existence of a dominant position, since users of Hilti nail guns may have the option of switching to other fastening systems which do not require the purchase of cartridge strips or nails. Hilti suggests that, by failing to take into account the high degree of such "demand substitutability" between the different systems, the Commission erred in its definition of the relevant product markets.
23. Thus Hilti submits that the Court of First Instance erred in law in upholding the Commission' s finding that PAF and other fastening systems are not readily substitutable for one another. In its various grounds of appeal, Hilti makes a number of submissions on the issue of substitutability.
24. In the first place, in its second ground of appeal Hilti alleges that the Court of First Instance has failed to make any finding which quantifies the number of cases in which PAF systems cannot readily be replaced by other forms of fastening system. Hilti refers in particular to the statement made in paragraph 69 of the judgment that:
"The specific features of PAF systems, set out in paragraph 62 of the Decision, are such as to make them the obvious choice in a number of cases."
Hilti suggests that, in the absence of a finding that the number of such cases is more than a few, or indeed more than trivially small, the Court of First Instance is not entitled to infer that PAF and other fastening systems are not interchangeable.
25. It would appear however that that submission is contradicted by the wording of paragraph 69 itself, which continues:
"... It is evident from the documents before the Court that in many cases there is no realistic alternative either for a qualified operator carrying out a job on site or for a technician instructed to select the fastening methods to be used in a given situation." [my emphasis]
Furthermore, in paragraph 71 of the judgment it is stated that the Commission' s findings:
"... leave no real doubt as to the existence, in practice, of a variety of situations, some of which inherently favour the use of a PAF system whilst others favour one or more other fastening systems." [my emphasis]
It seems to me that those passages indicate that the Court of First Instance made a finding of fact, according to which there are a significant number of cases in which there is only a low degree of demand substitutability between the different systems. The submission in Hilti' s second ground of appeal is accordingly to be rejected.
The third and fourth grounds of appeal
26. The Court of First Instance based its conclusion that PAF and other fastening systems do not form part of the same relevant product market on an examination of the following evidence: (i) the differing qualitative characteristics of the products at issue, as described by the Commission; (ii) the co-existence of different fastening systems on the market over an extended period of time; (iii) the expert opinion prepared by Mr Yarrow; and (iv) the survey by Rosslyn Research. The Court of First Instance moreover considered and rejected the conclusions of the econometric analysis carried out by Professor Albach.
27. In its third ground of appeal Hilti submits that the Court of First Instance erred, in paragraphs 70 and 71 of the judgment, in basing its conclusion concerning the relevant product market solely on the differing characteristics of the fastening systems in question. Similarly, in its fourth ground of appeal Hilti maintains that in paragraph 71 of the judgment the Court of First Instance erred in inferring, from the fact that PAF and non-PAF systems have co-existed over an extended period of time, the conclusion that the different systems were not interchangeable (and hence not part of the same relevant product market). It seems to me however that those submissions misrepresent the reasoning of the Court of First Instance. As we have seen, the Court of First Instance reached its conclusions regarding the relevant product market on the basis of a number of findings which must be considered jointly. It would therefore be wrong to criticize the Court of First Instance for basing its conclusions on any one of the findings considered in isolation.
28. It must none the less be examined whether, in reaching its conclusions, the Court of First Instance took account of all the relevant factors. If relevant factors were not taken into consideration, the Court of First Instance will have erred in law by basing its conclusions on insufficient reasoning. It is clear that a failure to take relevant factors into consideration in reaching a conclusion of law is in itself an error of law which might render the judgment liable to be annulled. It is to be noted that, in such a case, this Court is not reviewing the findings of fact made by the Court of First Instance, but rather examining whether sufficient findings have been made to support the legal conclusions drawn. It seems to me moreover that the identification of the relevant product market, for the purposes of a finding of dominance, must be regarded as a conclusion of law rather than as a pure finding of fact.
29. In its third ground of appeal, Hilti submits that the reasoning in paragraphs 70 and 71 of the judgment is contrary to established principles of Community law concerning the definition of the relevant market. Hilti refers in particular to the Michelin case, (14) where this Court stated, at paragraph 37 of its judgment:
"As the Court has repeatedly emphasized ... for the purposes of investigating the possibly dominant position of an undertaking on a given market, the possibilities of competition must be judged in the context of the market comprising the totality of the products which, with respect to their characteristics, are particularly suitable for satisfying constant needs and are only to a limited extent interchangeable with other products. However, it must be noted that the determination of the relevant market is useful in assessing whether the undertaking concerned is in a position to prevent effective competition from being maintained and behave to an appreciable extent independently of its competitors and customers and consumers. For this purpose, therefore, an examination limited to the objective characteristics only of the relevant products cannot be sufficient: the competitive conditions and the structure of supply and demand on the market must also be taken into consideration."
Hilti contends that the Commission and the Court of First Instance "did not rely on any such necessary findings".
30. The Michelin case concerned the market for new replacement tyres for heavy vehicles, that is to say lorries, buses and similar vehicles. It was common ground between Michelin and the Commission that the original tyres supplied with the vehicles should be left out of account; as the Court observed at paragraph 38 of the judgment:
"Owing to the particular structure of demand for such tyres characterized by direct orders from car manufacturers, competition in this sphere is in fact governed by completely different factors and rules."
That illustrates the principle that even products with identical qualities may be found to belong to different markets once the structure of demand is taken into consideration. As far as the market in replacement tyres was concerned, the Court made three observations (see paragraphs 39 to 41 of the judgment): first, as regards the requirements of the users, heavy-vehicle tyres are not interchangeable with car and van tyres; secondly, as regards the structure of demand, the buyers of heavy-vehicle tyres, who are mostly trade users, can be distinguished from the average buyer of car or van tyres; and finally:
"... there is no elasticity of supply between tyres for heavy vehicles and car tyres owing to significant differences in production techniques and in the plant and tools needed for their manufacture".
By "elasticity of supply" is meant the degree of responsiveness of the supply of a product to changes in its price. Thus the elasticity of supply between two products is a measure of the degree of supply substitutability between them: that is to say, of the degree to which it is possible for the manufacturers or suppliers of the first product to switch to producing or supplying the second. Its relevance to the existence of a dominant position on the market for the product is clear: even where a producer has a large share in that market, it will not be able to act independently of other producers in setting its prices if doing so would immediately attract other suppliers into the market who would be able to undercut its price. It is however to be noted that in some cases a very large market share will in itself amount to a barrier to entry, since it may be difficult for alternative suppliers rapidly to meet the demand of those who would like to break away from the undertaking with the largest market share. (15)
31. This Court has expressly considered the structure of supply in a number of other cases, including Hoffmann-La Roche, (16) United Brands, (17) and Continental Can. (18) Cases may of course arise where the point is not in issue, or where the answer is obvious, for instance where the supply of the product or service in question is reserved by national law to a particular undertaking. In such circumstances it is clear that there can be no alternative sources of supply affecting competition on the market for the product. (19) The structure of supply will however normally be relevant in establishing the existence of a dominant position. It is true that it will not always be appropriate to address the question in the context of the delimitation of the relevant market. It appears that the Court has considered the structure of supply in relation to that issue only where it was dealing specifically with the elasticity of supply between different products. In contrast, barriers to market entry or penetration generally have been dealt with under the heading of the undertaking' s "position on the relevant market" or the "structure of the relevant markets". (20) It appears however that the question will usually have to be addressed, either in relation to the definition of the relevant markets or in relation to the strength of the undertaking' s position on those markets.
32. In what follows I shall consider separately the question of the demand for fastening systems and the question of supply. The Court of First Instance reached certain conclusions on the demand side of the market and the question here is whether it erred in reaching those conclusions, as Hilti alleges. In relation to the structure of supply, however, the Court of First Instance does not appear to have made any findings of its own.
(1) The demand for fastening systems
33. The Court of First Instance based its conclusions regarding the demand for different fastening systems on two principal findings of fact: the Commission' s description of the different qualitative features of the products concerned, and the fact that the different fastening systems have co-existed for a substantial period of time. As we have already seen, in its third ground of appeal Hilti submits that a finding that PAF systems have specific features distinguishing them from other systems could not in itself be sufficient to establish the conclusion drawn by the Commission, and upheld by the Court of First Instance, that the systems are not substitutable for one another. Similarly, in its fourth ground of appeal Hilti maintains that a finding that the different systems have co-existed over an extended period of time could not in itself justify the conclusion that the systems are not interchangeable, an inference which Hilti claims is contained in the second sentence of paragraph 71 of the judgment.
34. It is clear however that the second sentence of paragraph 71 must be read in the context of the argument expounded in paragraphs 69 to 72. Paragraph 70 states that the Commission' s description of the specific features of PAF systems, in paragraph 62 of its decision, is sufficiently clear and convincing to justify the conclusions which the Commission draws from them. Paragraphs 71 and 72 of the judgment then continue:
"Those findings leave no real doubt as to the existence, in practice, of a variety of situations, some of which inherently favour the use of a PAF system whilst others favour one or more other fastening systems. As the Commission notes, the fact that several different fastening methods have each continued for long periods to account for an important share of total demand for fastening systems shows that there is only a relatively low degree of substitutability between them.
In such circumstances the Commission was entitled to base its conclusions on arguments which took account of the qualitative characteristics of the products at issue."
It can therefore be seen that the Court of First Instance, as I have already noted, based its conclusion concerning demand substitutability on both of the above two findings considered jointly, rather than on each considered individually. Thus the Court of First Instance stated in paragraph 72 of the judgment that "in such circumstances" the Commission was entitled to base its conclusions on the qualitative characteristics of the products, where the circumstances in question clearly include the co-existence of different fastening systems over long periods of time. Contrary to Hilti' s submission in its fourth ground of appeal, therefore, the Court of First Instance did not rely upon the finding of co-existence taken in isolation.
35. It seems to me moreover that the Court of First Instance was entitled to base its conclusions regarding demand on those two factors considered jointly. Although, as we have seen, Hilti criticizes the reasoning of the Court of First Instance regarding demand substitutability, Hilti does not itself point to any other factor affecting the demand for fastening systems which should have been taken into account. I conclude therefore that the Court of First Instance cannot be criticized for leaving out of account a relevant factor concerning demand substitutability, or any other factor relating to demand.
36. It will be recalled that, in addition to the two factors mentioned above, the Court of First Instance also took into consideration certain expert evidence adduced by Hilti relating to the demand for fastening systems, namely the opinion of Mr Yarrow, the survey by Rosslyn Research and the econometric analysis of Professor Albach. In its fifth and sixth grounds of appeal, Hilti attempts to show that the treatment by the Court of First Instance of that additional evidence was contrary to legal principle or manifestly erroneous in its reasoning. I shall consider those submissions below. It must however first be considered whether the Court of First Instance failed to deal with any relevant issue concerning the supply of fastening systems.
(2) The structure of supply
37. In contrast to the position in respect of demand, the Court of First Instance does not appear to have reached a finding on the question of supply substitutability between different fastening systems, even though that question was in issue before the Court of First Instance. It is true that the judgment does summarize the arguments of the parties on the question of supply substitutability; the Court of First Instance does not however itself express any view on the matter. The only place in which the judgment discusses the structure of supply in relation to any of the markets in question is in paragraphs 66 and 67: but those paragraphs are concerned, not with the question whether PAF systems form a relevant product market, but with the different question whether the markets in cartridge strips and nails are distinct from the market in nail guns.
38. However, Hilti has not raised the issue of supply substitutability in its appeal. It is true that Hilti relies on the sentence from the Michelin judgment which states that "the competitive conditions and the structure of supply and demand on the market must also be taken into consideration". But the issue of supply substitutability is nowhere specifically identified in Hilti' s pleadings. Nor was the issue separately addressed by the Commission in its response, or by Bauco. The absence of any specific submissions on the issue suggests that it was not treated as being specifically raised in the appeal, and that it would not therefore be appropriate for this Court to deal with it.
39. I conclude therefore that Hilti has not identified any relevant issue, relating either to the structure of supply or to the structure of demand or to the competitive conditions on the market, which the Court of First Instance failed to consider in its judgment. Hilti' s submission that the Court of First Instance failed to make all the findings necessary for a definition of the relevant markets must accordingly fail.
40. I would add that, if the judgment had been found defective in any of these respects, that would not have resulted in the annulment of the Commission' s decision but in the case being referred back to the Court of First Instance in accordance with Article 54 of the Statute.
The fifth ground of appeal
41. In its fifth ground of appeal, Hilti takes issue with the assertion made in paragraph 74 of the judgment that "the evidence produced by the applicant is not such as to weaken the findings made by the Commission". Hilti submits that it was required only to put forward evidence suggesting an alternative explanation for the Commission' s findings; it was not required to produce evidence which would "weaken" those findings. Hilti concludes that the Court of First Instance applies in that paragraph "the wrong legal burden of proof".
42. In support of that submission, Hilti refers to paragraph 63 of the Commission' s decision, where the Commission concludes that:
"... it is not conceivable that small changes in the price of a nail gun, nail and/or cartridge would cause an immediate and large shift to or from alternative fixing methods".
Hilti contends that the use of the term "conceivable" indicates that the Commission was basing its finding on mere supposition. It suggests that to refute a mere supposition it suffices to prove circumstances which cast the facts established by the Commission in a different light: see paragraph 16 of the Court' s judgment in CRAM and Rheinzink. (21)
43. In my opinion, however, an examination of the Commission' s reasoning in paragraphs 63 to 65 of its decision does not support the contention that the Commission was basing its conclusions upon mere supposition. In paragraphs 61 and 62 of the decision, the Commission describes in some detail the factors which may affect the choice of a fastening method in particular circumstances, and the technical characteristics which distinguish PAF systems from other methods of fastening. It is on the basis of those considerations that the Commission reaches the conclusion, in paragraph 63, that PAF and other fastening systems cannot be said to be part of the same relevant market. As the Commission explains in that paragraph:
"... The choice of the best fixing method to use is made on the basis of a specific fixing application on a specific site ... . In view of the great many factors that enter into this choice and of the fact that fixing elements are normally a very small part of building costs, the indications are that the prices of the elements of different fixing methods are not the only or crucial element in the choice of which fixing method to employ for a particular job. Therefore it is not conceivable that small changes in the price of a nail gun, nail and/or cartridge would cause an immediate and large shift to or from alternative fixing methods." [my emphasis]
It can be seen that the Commission' s statement that it is "not conceivable" that small changes in prices would lead to an immediate and large shift between fixing methods is presented as a corollary of the conclusion, already reached on the basis of the evidence, that the choice of a fixing method is not critically dependent upon the prices of the elements of the system, that is to say upon the prices of nail guns, cartridge strips or nails. It is therefore wrong to say that the statement is mere supposition or speculation: it is rather a conclusion reached upon the basis of evidence. In order to cast doubt upon such a conclusion it would have been necessary for Hilti, in the proceedings before the Court of First Instance, either to show that the evidence relied upon by the Commission was insufficient to sustain such a conclusion, or to produce solid alternative evidence that the price of the elements of a fastening system is the crucial factor in the choice between different systems. (22)
44. In my view, therefore, the Court of First Instance was correct in its assumption that it was for Hilti to show that the new evidence it had adduced did indeed cast doubt on the findings of the Commission. Hilti' s submission that the Court of First Instance applied the wrong legal burden of proof must therefore be rejected.
The sixth ground of appeal
45. In its sixth ground of appeal Hilti submits, in effect, that in at least three instances the Court of First Instance misinterpreted the evidence put before it. This ground therefore raises the question whether such a submission involves a point of law which can be considered on an appeal to this Court, or whether it must be regarded as relating only to issues of fact on which the decision of the Court of First Instance is final. As we shall see, there is in my view no doubt that Hilti' s submissions in this ground of appeal raise exclusively questions of fact which cannot be considered by this Court.
46. It may of course not always be easy to draw the dividing line between issues of fact and questions of law. The distinction is much debated in national legal systems; in its appeal Hilti appears to suggest that this Court should follow the practice of certain national courts and regard manifest errors of fact, in some circumstances, as amounting to errors of law. However, it seems to me that in appeals against decisions of the Court of First Instance this Court can properly take a restrictive view of the notion of "infringement of Community law by the Court of First Instance" in Article 51 of the Statute. This Court must avoid a situation in which it would have to carry out a further review of findings of fact where it was alleged that an error of fact was manifest. That would have the effect of subverting the purposes for which the Court of First Instance was established. For reasons which I have already given, that consideration applies with particular force in competition cases, where the decision of the Court of First Instance is itself a review with very broad scope of a reasoned decision of the Commission. (23)
47. Thus where the Court of First Instance, without itself hearing new evidence or making any other factual investigations of its own, reviews the findings of fact in the Commission' s decision or the inferences which the Commission has drawn from its findings, it seems to me clear that this Court cannot undertake a further review of the facts found by the Commission. This Court could intervene only if it appeared from the judgment of the Court of First Instance that that Court had wrongly applied Community law by, for example, applying the wrong legal test to an issue such as the definition of the relevant market, or by basing its legal conclusions on insufficient reasoning. It is only such an error which would amount to an "infringement of Community law" for the purposes of Article 51 of the Statute. Such an error will be apparent from the terms of the judgment itself, and it will be sufficient for this Court to review the reasoning contained in the judgment.
48. Is the position any different where the Court of First Instance itself takes evidence, otherwise makes investigations into the facts, and possibly makes new findings of fact? It must be emphasized that any such new findings are not made to be substituted for findings of the Commission, but are made only to confirm or to refute the Commission' s findings; the Court of First Instance is not, in these cases, acting as a trial court making findings for the first time, but is making such findings only for the purpose of reviewing the Commission' s findings. In such a case the Court of First Instance will necessarily, if it is to satisfy the requirement of reasoning, set out its own findings in its judgment and draw the relevant inferences from them to support its own conclusions. Even in that case, therefore, review by this Court can properly be limited to the reasoning as it is set out in the judgment. The reasoning must of course respond adequately to the submissions of the parties and, in doing so, will necessarily disclose the Court of First Instance' s own findings of fact and the inferences it has drawn from them. It will therefore not be necessary for this Court to go behind the judgment of the Court of First Instance to look at the evidence before that court.
49. Difficult cases may of course arise in which this Court will have to determine the extent to which it is prepared to review the reasoning used by the Court of First Instance in its assessment of the facts which have been established. As I have already mentioned, it seems to me that this Court would be entitled to take a restrictive view of its appellate jurisdiction in appeals on competition matters, given the purposes of the establishment of the Court of First Instance and the fact that in competition proceedings there will already have been two reasoned decisions. On any view of the scope of this Court' s appellate jurisdiction, however, it cannot have been intended for the Court to go behind the findings of fact made in the judgment of the Court of First Instance, where that judgment is reviewing findings made by the Commission.
50. With those considerations in mind, I now turn to the errors alleged by Hilti in its sixth ground of appeal.
51. The first error alleged by Hilti is in the statement in paragraph 73 of the judgment to the effect that the Commission' s conclusions are:
"corroborated by the opinion prepared by Mr Yarrow and the survey conducted by Rosslyn Research Ltd, ... inasmuch as they disclose the existence of a large number of nail gun users who could see no realistic alternative to the PAF system in circumstances corresponding to most of those in which nail guns have in fact been used".
To that statement, Hilti objects that the interpretation of the survey conducted by Rosslyn Research is "manifestly wrong" and "based upon a simple mistake in arithmetic". According to Hilti, rather than disclosing the existence of large numbers of such nail gun users, the survey shows that only a minority of nail gun users encounter situations in which they can use no alternative fastening system, and only in a fraction of the cases in which they use nail guns.
52. Hilti therefore objects to a finding that, on the evidence of a particular survey, there are a large number of users, in a majority of cases, for whom PAF systems cannot be replaced by other fastening systems. It seems to me that such a finding is a pure finding of fact made by the Court of First Instance. As we have already seen, it is not for this Court to go behind the findings of fact set out in the judgment; it can, at most, examine the reasoning employed by the Court of First Instance in drawing conclusions from such findings. The interpretation of the results of the Rosslyn Research survey is accordingly a matter which remains within the exclusive competence of the Court of First Instance, and cannot be challenged on appeal.
53. Identical considerations apply in my opinion to the second allegation in Hilti' s sixth ground of appeal, which concerns paragraph 75 of the judgment. In that paragraph the Court of First Instance criticises the methodology adopted in the opinion of Mr Yarrow and in the survey by Rosslyn Research, arguing that the questions put to construction undertakings in the survey:
"are not apt to provide an answer to the fundamental question in this case, namely whether slight but significant differences in the price of nails are likely to shift demand to a significant extent".
In my view, the evaluation of the methodology adopted in an empirical survey is a matter which lies within the exclusive competence of the Court of First Instance, as the tribunal of fact. It is clear that the correct manner in which to frame questions put to construction undertakings, when assessing the responsiveness to price of their decision to employ a particular fastening system, is not a question of law which can be considered in this appeal.
54. Finally, Hilti takes issue with the statement in paragraph 76 of the judgment which criticizes the methodology employed in the econometric study by Professor Albach. It appears that the methodological difficulties in question were addressed in an oral statement made by Professor Albach at the hearing, an extract from which is cited by Hilti in its appeal. It seems to me that it is manifestly for the Court of First Instance, and for the Court of First Instance alone, to consider whether the doubts which had been raised were adequately resolved by Professor Albach' s evidence at the hearing.
55. I therefore reach the conclusion that none of the submissions made by Hilti in its sixth ground of appeal can be regarded as raising a point of law which can be considered on an appeal.
The seventh ground of appeal
56. In its final ground of appeal, Hilti submits that the Court of First Instance erred in law by failing to consider all the expert evidence put before it which had a bearing on the definition of the relevant product markets. In support of that submission Hilti observes that the evidence put forward by Hilti is discussed in the judgment in only three short paragraphs, which moreover deal, in Hilti' s view, with "only three very minor aspects" of the evidence.
57. As we have already seen, the Court of First Instance held that the Commission' s conclusions regarding the relevant product markets were corroborated rather than refuted by the expert evidence adduced by Hilti. It does not seem to me however that it can be said to have disregarded that evidence; rather, the Court of First Instance considered the evidence but drew from it different conclusions from those put forward by Hilti. In paragraph 4.58 of its appeal, Hilti sets out what it considers to be the most relevant conclusions to be drawn from the evidence analysed by Mr Yarrow and Professor Albach. However, the fact that the Court of First Instance did not draw the same conclusions does not indicate that it failed to have regard to that evidence.
58. In my opinion, therefore, Hilti' s submission in its seventh ground of appeal must equally be rejected. It follows that Hilti' s appeal fails.
59. Finally, it is to be noted that Bauco' s submission that the fine imposed on Hilti should be increased does not call for consideration. It is clear that an intervener before the Court of First Instance may only seek an order in support of or opposing, in whole or in part, an order sought by one of the parties, and that on an appeal it may not seek in its response a different form of order from that sought at first instance. (24) Bauco is accordingly not entitled to seek an order increasing the amount of the fine, whether in its original intervention before the Court of First Instance or in its response to the appeal.
Conclusion
60. I am accordingly of the opinion that the Court should:
1. dismiss the appeal;
2. order Hilti to pay the costs, including the costs incurred by Bauco.
(*) Original language: English.
(2) - Case T-30/89 Hilti v Commission [1991] ECR II-1439.
(3) - Commission Decision 88/138/EEC of 22 December 1987 relating to a proceeding under Article 86 of the EEC Treaty (IV/30.787 and 31.488 - Eurofix-Bauco v Hilti), OJ 1988 L 65, p. 19.
(4) - Case 6/72 Europemballage and Continental Can v Commission [1973] ECR 215. See similarly Case 31/80 L' Oréal v De Nieuwe AMCK [1980] ECR 3775, paragraph 25, and Case C-62/86 AKZO v Commission [1991] ECR I-3359, paragraph 51.
(5) - See in particular the discussion in Whish, Competition Law (2nd edition; London 1989), pp. 278-287.
(6) - On the application of the limitation to appeals in staff cases, see for example: Case C-115/90 P Turner v Commission [1991] ECR I-1423; Case C-283/90 P Vidrányi v Commission [1991] ECR I-4339; Case C-107/90 P Hochbaum v Commission [1992] ECR I-157; Case C-18/91 P V. v Parliament [1992] ECR I-3997.
(7) - Cf. Case 42/84 Remia v Commission [1985] ECR 2545, paragraph 34.
(8) - See below, paragraphs 46 to 49.
(9) - Council Decision 88/591/ECSC, EEC, Euratom, OJ 1988 L 319, p. 1.
(10) - Council Decision 93/350/Euratom, ECSC, EEC, OJ 1993 L 144, p. 21.
(11) - The entry into force of the Decision is deferred, in anti-dumping and anti-subsidy cases, to a date that the Council shall fix by unanimous decision: see Article 3.
(12) - Case 22/78 Hugin v Commission [1979] ECR 1869.
(13) - Case 27/76 United Brands v Commission [1978] ECR 207, paragraph 65.
(14) - Case 322/81 Michelin v Commission [1983] ECR 3461.
(15) - See Case 85/76 Hoffmann-La Roche v Commission [1979] ECR 461, paragraph 41.
(16) - Cited in note 14; see paragraphs 33, 34 and 48 of the judgment.
(17) - Cited above in note 12; see paragraph 122 of the judgment, where the matter is however considered under the heading of the position of United Brands on the relevant market rather than that of the definition of the relevant market.
(18) - Cited above in note 3; see paragraphs 33 to 36 of the judgment.
(19) - See for example Case 26/75 General Motors v Commission [1975] ECR 1367; Case 226/84 British Leyland v Commission [1986] ECR 3263; Case C-41/90 Hoefner and Elser [1991] ECR I-1979.
(20) - See note 16 above, and see also Hoffmann-La Roche, cited in note 14, paragraphs 33 and 34 of the judgment.
(21) - Joined Cases 29 and 30/83 CRAM and Rheinzink v Commission [1984] ECR 1679.
(22) - Cf. the remarks of Advocate General Sir Gordon Slynn in Joined Cases 100-103/80 Musique Diffusion Française v Commission [1983] ECR 1825, pp. 1930-1931.
(23) - See paragraphs 9 to 12 above.
(24) - See Article 116(4) of the Rules of Procedure of the Court of First Instance, and Article 116(1) of the Rules of Procedure of the Court of Justice.