EUR-Lex -  61990CC0105 - EN - Opinion of Mr Advocate General Van Gerven delivered on 14 November 1991. - Goldstar Co. Ltd v Council of the European Communities. - Antidumping - Normal value. - Case C-105/90.

Opinion of the Advocate-General

++++

Mr President,

Members of the Court,

1. In the case now before the Court Goldstar Co. Ltd ("Goldstar"), a company incorporated under Korean law, seeks the annulment of Council Regulation (EEC) No 112/90 of 16 January 1990 imposing a definitive anti-dumping duty on imports of certain compact disc players originating in Japan and the Republic of Korea and collecting definitively the provisional duty ("the definitive regulation"), (1) in so far as that regulation relates to the applicant. In support of its application Goldstar puts forward three submissions which are based primarily on the alleged infringement of Article 2(3)(b) of Council Regulation (EEC) No 2423/88 of 11 July 1988 on protection against dumped or subsidized imports from countries not members of the European Economic Community ("the basic regulation"). (2)

Background

2. Goldstar is a member of the Lucky Goldstar group. It produces electrical and electronic appliances which is sells both in Korea and abroad. During the reference period (1 June 1986 to 31 May 1987) Goldstar sold five models of compact disc players in Korea and in the Community, namely models GCD 603, GCD 605, GCD 606, GCD 613 and GCD 616. The first three models - which Goldstar ceased to produce in 1985 - were sold in Korea and in the Community exclusively under its own brand. The last two models were sold in the Community both under its own brand and to original equipment manufacturers (OEMs), (3) but were put on the market in Korea exclusively under its own brand.

3. In June 1987 the Committee of Mechoptronics Producers and Connected Technologies ("Compact") submitted a complaint to the Commission, on behalf of manufacturers accounting for the bulk of the production of compact disc players in the Community, concerning the importation into the Community of compact disc players originating in Japan and the Republic of Korea.

The Commission' s investigation resulted in the imposition by Commission Regulation (EEC) No 2140/89 of 12 July 1989 ("the provisional regulation") (4) of a provisional anti-dumping duty on imports of certain compact disc players originating in Japan and the Republic of Korea. The provisional anti-dumping duty on imports of Goldstar' s products was fixed by the Commission at 32.5% of the net free-at-Community-frontier price. The rate of that provisional duty was connected with the normal value of the products imported by Goldstar, which was determined by the Commission separately for each model. In the provisional regulation the Commission determined the normal value, according to the quantities sold on the domestic market and the manner of sale (under its own brand or to OEMs), by applying the following three methods:

(a) As regards the GCD 603, GCD 605, GCD 606 and GCD 616 models sold both in Korea and in the Community under Goldstar' s own brand

According to the Commission, those models were sold on the domestic market during the reference period in sufficient quantities (that is to say more than 5% of the volume of exports to the Community of the model concerned). As stated in paragraph 23 of the provisional regulation, the normal value of each model was established on the basis of the weighted average domestic prices of all sales of that model to independent customers.

(b) As regards the GCD 613 model sold both in Korea and in the Community under Goldstar' s own brand

The number of units of this model sold on the domestic market during the reference period amounted to less than 5% of the volume of exports of the same model to the Community. For that reason the Commission determined its normal value on the basis of the constructed value calculated in the manner described in paragraph 48 of the provisional regulation, that is to say on the basis of the weighted averages of the costs incurred and profits realized by Goldstar on the GCD 603, GCD 605, GCD 606 and GCD 616 models sold on the Korean market.

(c) As regards the GCD 613 and GCD 616 models sold only to OEMs in the Community

As stated in paragraphs 51 and 52 of the provisional regulation, the Commission considered that the OEMs to which Goldstar sold its products for export to the Community constituted a separate and distinct class of customer. In determining the normal value of the models sold to those OEMs the Commission made an estimate of the differences which could have existed between prices of own-brand products and OEM prices if OEM sales had taken place on the Korean market. This resulted in a profit rate corresponding to 30% of the profit realized on own-brand sales on the domestic market being taken into account.

4. During the procedure which led to the adoption of the definitive regulation, Goldstar raised objections to the methods employed by the Commission in determining the normal value of the models sold by it. Those objections tally to a large extent with those raised by Goldstar in the procedure which led to the adoption of the provisional regulation. The Council, like the Commission, rejected Goldstar' s objections. It determined the normal value of the models concerned by the same methods as those described above and fixed the definitive anti-dumping duty on imports of Goldstar' s products at 26.1% (instead of 32.5%) of the net free-at-Community-frontier price.

The first submission

5. Goldstar claims that the Council incorrectly determined the normal value of models GCD 603, GCD 605, GCD 606 and GCD 616, sold both in Korea and in the Community under its own brand, on the basis of domestic prices. According to Goldstar, those models were not sold on the Korean market during the reference period in the ordinary course of trade and did not permit a proper comparison, with the result that the Council should, in accordance with Article 2(3)(b) of the basic regulation, have determined the normal value on the basis of the constructed value. In support of that contention, Goldstar puts forward a number of arguments which may be divided into two groups.

Goldstar maintains, in the first place, that the expressions "ordinary course of trade" and "a proper comparison" presuppose sales in sufficient quantities on a representative domestic market. Goldstar does not deny that the number of sales on the Korean market during the reference period of the GCD 603, GCD 605, GCD 606 and GCD 616 models amounted for each model to more than 5% of the volume of exports to the Community. Furthermore, it accepts the practice of the Community institutions not to use sales on the domestic market as a basis for determining the normal value where the number of sales on that market falls below the 5% threshold, and to use the constructed value instead. Goldstar maintains, however, that the institutions are obliged to construct the normal value even where the 5% threshold is exceeded if, as in this case, the volume of sales of the models concerned on the domestic market is minimal in absolute terms (I shall not disclose the exact figures on account of their confidential nature). Goldstar further considers that the total size and characteristics of the domestic market should also be taken into account. In the circumstances the total size of the Korean market is negligible (it estimates the total volume of sales of compact disc players in Korea during the reference period at approximately 5 000 units) which, amongst other things, is attributable to the fact that during that period very few Korean productions had been recorded on compact discs. Goldstar therefore alleges that the Council applied the 5% threshold indiscriminately, that is to say without taking account of either the overall number of units sold by Goldstar on the Korean market or the total size and characteristics of that market. On the same grounds, Goldstar maintains that the Council also failed to discharge its duty to state reasons.

Secondly, Goldstar points out that it discontinued production of the GCD 603, GCD 605 and GCD 606 models in 1985, that is to say before the commencement of the reference period. The price at which those outdated models were sold on the domestic market during that period does not, in its view, provide a proper basis for comparison.

6. I am not swayed by Goldstar' s arguments.

Article 2(3)(a) of the basic regulation defines the normal value primarily as

"the comparable price actually paid or payable in the ordinary course of trade for the like product intended for consumption in the exporting country or country of origin."

Other factors, referred to in subparagraph (b)(i) and (ii), may be used as the normal value only

"where there are no sales of the like product in the ordinary course of trade on the domestic market of the exporting country or country of origin, or when such sales do not permit a proper comparison".

As the Court stated in its judgment in Canon (5) (at paragraph 11), it follows from the wording and scheme of the above provisions that regard must primarily be had to the price actually paid or payable in the ordinary course of trade on the domestic market in order to establish the normal value, the other possibilities being merely subsidiary. Accordingly, the Community institutions must determine the normal value as far as possible on the basis of the actual domestic prices and they may depart from that practice only where at least one of the situations provided for in Article 2(3)(b) of the basic regulation has clearly arisen.

7. In accordance with the latter provision, it is necessary to distinguish between two situations. To begin with, there is the situation where "there are no sales of the like product in the ordinary course of trade" on the domestic market. As is clear from the example set out in the fourth consideration in the preamble to the basic regulation, this refers to cases in which sales viewed in themselves do not represent the ordinary course of trade, in particular where a product is sold at prices which are lower than the costs of production, or where transactions take place between parties which are associated or which have a compensatory arrangement. Goldstar has neither alleged, nor by any means demonstrated, that its sales on the Korean market are not in themselves in the ordinary course of trade. It cannot therefore deduce from that passage in Article 2(3)(b) of the basic regulation any grounds for departing from the rule which takes precedence, namely that the normal value must be determined on the basis of domestic prices.

8. The other situation is where sales on the domestic market "do not permit a proper comparison". As is clear from Article 2(4) of the Anti-dumping Code, (6) it is necessary to ascertain whether transactions which are in themselves in the ordinary course of trade have to be left out of account because of the particular market situation. Transactions carried out on the domestic market must reflect normal consumer behaviour and the normal formation of prices in overall terms as well. Only in those circumstances do domestic prices permit a proper comparison with export prices.

As a minimum guarantee that domestic prices are representative and in order to ensure legal certainty the Community institutions have somewhat curtailed their discretion under the aforesaid provisions by adhering in practice to a guideline, more specifically a threshold below which domestic sales should be disregarded. The institutions have thus laid down, for the first time in connection with the anti-dumping procedure concerning electronic typewriters originating in Japan, (7) and have since then systematically applied, the principle that a volume of domestic sales amounting in the case of the product concerned to less than 5% of the volume of exports of that product to the Community cannot be taken into account as a basis for determining the normal value. In its judgment in Canon (at paragraphs 13 to 15), as in the judgment in Brother (8) (at paragraphs 11 to 13), the Court rejected the applicants' objections to the 5% threshold and acknowledged that the threshold may be applied for each model (with its own characteristics). (9) In its later judgment in Neotype Techmashexport (10) (at paragraph 31), moreover, the Court deduced an argument from the 5% threshold discussed in Canon and Brother which it used to reject the allegation that the market of a reference country for the purposes of Article 2(5) of the basic regulation is unrepresentative. In its recent judgment in Noelle (11) the Court added that the mere fact that the volume of production of the reference country amounts to less than 5% of the volume of exports to the Community does not mean that the choice of that country cannot be regarded as appropriate and reasonable, but that a figure of 1.25% is evidence that the market taken into account is scarcely representative.

9. Goldstar is right in stating that where the volume of domestic sales is greater than 5% of the volume of exports to the Community, it does not necessarily mean that domestic prices permit a proper comparison (any more than that is the case, as is apparent from the aforesaid judgment in Noelle, where the volume of sales is lower). The Council and the Commission are both in agreement with that view, contrary to what Goldstar seems to suggest. The need to take other factors into account as well is apparent from the fourth recital in the preamble to the basic regulation, according to which a constructed normal value must be applied where domestic prices do not "for any reason" form a proper basis for determining the existence of dumping.

It seems to me, however, that since the Community institutions have of themselves, as a minimum guarantee that prices are representative and in order to ensure legal certainty as such, imposed a guideline - which, however, they may not apply indiscriminately - they can and must make an exception to it only where special circumstances arise as a result of which the domestic price does not constitute a proper basis for comparison with the export price.

10. What constitutes special circumstances of that kind? I agree with Goldstar that the characteristics and, in particular, the total size of the domestic market may constitute such circumstances. The Court has implicitly confirmed that view in its judgments in Brother (at paragraph 10) and in Silver Seiko (12) (at paragraph 11). In those cases the applicants also contended that the number of electronic typewriters sold on the Japanese domestic market, as compared with the Community market, was too small. The Court did not reject that argument as such but considered that there was no factual basis for it since some tens of thousands of typewriters were sold annually on the Japanese market and, furthermore, that market was fairly competitive. In those circumstances the Court considered the prices on the Japanese market to be comparable with those on the Community market.

It seems to me that the argument concerning the characteristics of the Korean market relied upon by Goldstar in this case also lacks any factual basis. To begin with, as far as the total size of the market in compact disc players (for all models taken together) is concerned, Goldstar has been unable to substantiate its assertion that during the reference period no more than 5 000 compact disc players were sold on the Korean market. At the hearing the Commission stated that the figure was approximately 6 500, whilst Compact referred to press reports according to which the size of the market was in fact 30 000. In any event, even a market of 5 000 units strikes me as being sufficiently representative in absolute terms, provided that figure constitutes in percentage terms a substantial share of the total volume of exports to the Community. According to the Council' s contention, which has not been disputed, the volume of exports in 1986 amounted to somewhat less than 34 000 units. In other words, the minimal figure referred to by Goldstar of 5 000 units sold on the Korean market nevertheless amounts to 14% of exports of Korean compact disc players to the Community - which is by no means negligible.

11. Goldstar further relies on the - in its view - minimal share (which I shall not disclose on grounds of confidentiality) of domestic sales per model of the GCD 603, GCD 605, GCD 606 and GCD 616 compact disc players. Let me begin with a preliminary remark. The argument, which is in itself correct, that for domestic prices to be comparable there must be a sufficient volume of sales in absolute terms on the domestic market stems from the fact that domestic prices may differ from one sales outlet to another. Furthermore, prices may fluctuate in the course of a single reference period. For that reason the Community institutions usually apply, as they have done in this case, a price corresponding to the weighted average of all sales of the product concerned to independent customers. In order to be representative that weighted average must be based on a minimum number of sales.

That does not mean, however, that the Community institutions should, in addition to the aforesaid relative threshold below which sales should be disregarded, fixed at 5% and calculated for each model, still be required to apply an absolute threshold as well. It makes little sense to establish an absolute minimum threshold in general since the value of an absolute figure depends too closely on the nature of the product. Accordingly, it is necessary to determine, for each product or model, whether there are sufficient sales on the market at stable prices, so that the average price calculated on that basis is sufficiently representative. Apart from a general assertion to the effect that the number of sales on the domestic market is minimal, however, Goldstar has not advanced any arguments demonstrating that the average price applied is not representative. (13)

On the other hand, relative figures which express for a single product or model or for a category of similar products a ratio between the products sold on the domestic and on the export market can in general be established and can relate the size of the domestic market directly to the size of the export market. Since the Korean market in compact disc players, viewed as a whole, amounts to at least 14% of export sales to the Community, and sales on the Korean market for each model also exceed the overall (relative) threshold below which sales should be disregarded, fixed at 5% of exports to the Community, that is a sufficient indication, in my view, that domestic prices are representative and appropriate.

12. Apart from the size of the Korean market as a whole, on the one hand, and its own sales figures for each model, on the other, Goldstar does not rely on any other special circumstances in challenging the view that domestic prices are representative. In particular, Goldstar does not deny that the Korean market is "fairly competitive", a feature which the Court took into account in its judgments in Brother and Silver Seiko in considering whether prices were representative. More particularly, in that connection the applicant has not alleged, still less has it demonstrated, that domestic prices would have been lower if the volume of sales and competition on the domestic market had been greater. Instead it is apparent from the documents before the Court (14) that Goldstar consciously decided as part of its strategy to sell compact disc players in Korea as luxury goods for relatively high domestic prices (and, consequently, profit margins).

I attach considerable importance to that factor. It seems to me that a producer cannot rely on his own deliberate decision to place his product in a given sector of the market and in a given price category - which may also be the reason for which, if that strategy proves to be mistaken, he sells fewer units than his competitors - as a special circumstance when calling in question the view that the domestic prices charged by him are representative. Only external circumstances to which producers must submit are to be taken into account. If it were otherwise, producers themselves could create circumstances which make it necessary to depart from the principle that the normal value is determined primarily on the basis of the domestic prices actually charged.

13. On the basis of the foregoing considerations, therefore, I have come to the conclusion that the Council was right to consider that the number of units sold by Goldstar on the Korean market and the total size of that market do not constitute grounds for rejecting domestic prices as an unsuitable basis for comparison.

It also follows that the Community institutions were entitled to rely on the "5% threshold" guideline and that the argument based on Article 190 of the EEC Treaty concerning the absence of a statement of reasons must therefore be rejected. Paragraph 21 of the definitive regulation refers to paragraph 27 of the provisional regulation, which states as follows:

"The Commission considers that a low absolute amount of domestic sales cannot by itself be a sufficient reason to consider that these sales are not in the ordinary course of trade and justify a deviation from the normal practice of establishing normal values on the basis of the domestic sales of models which in quantity terms exceed 5% of export sales to the Community."

That statement reflects the reasoning of the Community institutions sufficiently clearly.

14. Finally, I consider that the argument deduced by Goldstar from the fact that production of the GCD 603, GCD 605 and GCD 606 models was discontinued before the commencement of the reference period is also unfounded. As is clear from the wording of Article 2(3)(b) of the basic regulation it is necessary to take account of "sales" on the domestic market. That provision draws no distinction between products according to whether or not they are still being manufactured. The fact that the models sold are no longer produced does not, therefore, in itself constitute a ground for rejecting the domestic prices of those models as unsuitable. A different decision could lead to a situation in which the normal value of comparatively novel products such as compact disc players, new models of which are constantly being brought onto the market, can be determined on the basis of domestic prices only by way of exception.

15. In the light of the foregoing considerations, therefore, I am of the opinion that Goldstar' s first submission must be rejected.

The second submission

16. Goldstar alleges in its second submission that the Council infringed Article 2(3)(b)(ii) of the basic regulation by determining the constructed value of model GCD 613 sold under its own brand on the domestic market and in the Community, and more particularly the amount for profit to be added to the production costs, by reference to the average profit realized by Goldstar during the reference period on its domestic sales of other models, namely the GCD 603, GCD 605, GCD 606 and GCD 616 units (for reasons of confidentiality I shall not disclose that average, which is expressed as a percentage of the cost of production). As stated earlier (paragraph 5 above) Goldstar accepts that in the case of a product such as the GCD 613 model, sales of which were beneath the relative threshold below which domestic sales are to be disregarded, fixed at 5% of the volume of exports, the Community institutions were entitled (and obliged) to take a constructed value into account. (15) According to Goldstar, however, the average profit on sales of those other models is unsuitable for two reasons. To begin with, it is calculated on the basis of sales on the domestic market which were not made in the ordinary course of trade and/or which did not permit a proper comparison. Furthermore, Article 2(4) of the Anti-dumping Code describes the profit made by an individual exporter on the domestic market as not suitable for use where, as in this case, it exceeds the profit normally realized on sales of like products on the domestic market.

17. I do not need to consider the first argument, having come to the conclusion in connection with my examination of the first submission that own-brand sales of the GCD 603, GCD 605, GCD 606 and GCD 616 models on the domestic market were made in the ordinary course of trade and permitted a proper comparison.

The second argument, on the other hand, deserves closer attention. Since the Court' s judgment in Nakajima (16) (in particular at paragraphs 26 to 32), it has been clear that producers may challenge the validity for legal purposes of an anti-dumping regulation on the basis of the provisions of the Anti-dumping Code.

18. According to Article 2(3)(b)(ii) of the basic regulation, the constructed value is to be determined by adding a reasonable amount for profit to the cost of production. To that end the provision prescribes three methods which, as the Court stated in Nakajima (at paragraph 61), must be applied in succession. (17) The amount for profit must therefore be calculated primarily by reference to the profit made by the producer on profitable sales of like products on the domestic market. Only if the data concerning such profit are unavailable, unreliable or not suitable for use is the amount for profit to be calculated by reference to the profit realized by other producers in the country of origin on profitable sales of the like product. If neither of those two methods can be applied, the amount for profit is to be calculated by reference to the sales made by the producer or other producers in the same business sector in the country of origin or on any other reasonable basis.

19. Goldstar maintains that the first method of calculation provided for in the basic regulation is "not suitable for use" where the profit which the producer concerned realizes on domestic sales of a like product is greater than the profit normally realized on sales of like products on the domestic market. That contention is based on the last sentence of Article 2(4) of the Anti-dumping Code, which reads as follows:

"As a general rule, the addition for profit shall not exceed the profit normally realized on sales of products of the same general category in the domestic market of the country of origin."

Goldstar apparently proceeds on the assumption (i) that the words "profit normally realized" must be understood as referring to the average profit realized collectively by producers of the country of origin on sales of like products on the domestic market and (ii) that the first method of calculation provided for in Article 2(3)(b)(ii) of the basic regulation, which does in fact refer to the profit realized by the producer concerned, is contrary to the last sentence of Article 2(4) of the Anti-dumping Code as understood above.

20. Goldstar' s approach is based, in my view, on a misinterpretation of the Anti-dumping Code.

It is true that the last sentence of Article 2(4) of the Anti-dumping Code does not expressly indicate by whom the profit referred to therein must normally be realized. That does not mean, however, that it necessarily refers to the average profit realized by all producers together on sales of like products on the domestic market. The wording also bears the interpretation that the addition for profit may not be higher than the profit normally realized by the producer concerned on sales of like products on the domestic market. Such an interpretation is more in keeping with the aim of establishing a constructed value. As the Court has repeatedly stated, (18) the purpose of constructing the normal value is to determine the selling price of a product as it would be if that product were sold in its country of origin or in the exporting country. The most accurate method of determining that price consists in ascertaining the constructed value on the basis of sales of like products on the domestic market by the producer concerned himself. Furthermore, the Commission has rightly pointed out that, as is clear from Article 8(2) of the Anti-dumping Code, preference is given in that code as well to an individual rather than an overall approach.

Moreover, the Court has already stated in general terms, in its judgment in Nakajima (at paragraph 37), that the methods of calculation "embodied" in Article 2(3)(b)(ii) of the basic regulation are compatible with Article 2(4) of the Anti-dumping Code:

"Article 2(3)(b)(ii) of the new basic regulation is in conformity with Article 2(4) of the Anti-dumping Code in so far as, without disregarding the rationale behind the latter provision, the first provision merely embodies, as regards the different situations which may arise in practice, the reasonable methods of calculating the constructed normal value." (19)

Finally, I would remind the Court that Goldstar fixed its domestic prices as part of a conscious strategy to sell compact disc players in Korea as luxury goods. Goldstar can no more rely on its deliberate policy of fixing prices as a special circumstance which has the effect of making its domestic prices unrepresentative (see paragraph 12 above), than on the fact that the high profit margins resulting from the fixing of prices may not be taken into account in determining the constructed value of the GCD 613 model.

In the light of the foregoing considerations, Goldstar' s second submission must also be rejected.

The third submission

21. Goldstar' s third submission is connected with the Council' s decision to determine a separate normal value for sales to OEMs. (20) As stated in paragraph 22 of the definitive regulation, the Council accepted

"that these customers had functions which were clearly distinct from those of other categories of unrelated purchasers and that these different functions were clearly reflected, for the markets concerned, in the quantities sold and in the pattern of prices charged."

Goldstar alleges that the Council infringed Article 2(3)(b)(ii) of the basic regulation in so far as it determined the constructed value of the GCD 613 and GCD 616 models sold to OEMs for export to the Community by reference to a profit margin amounting to 30% of the profit which the applicant realized on sales of those models under its own brand on the domestic market. In support of that submission Goldstar relies on the following four grounds:

(a) there is no connection between the costs and profits for OEM sales and those for own-brand sales;

(b) the Council' s approach is contrary to previous practice;

(c) the rate of 30% is arbitrary;

(d) the Council has contravened the principle of equality.

22. Before Goldstar' s arguments are considered, it must be borne in mind that, as stated earlier, the purpose of constructing the normal value is to determine the selling price of a product as it would be if that product were sold in its country of origin or in the exporting country. In a case such as this, in which a separate normal value was established for sales to OEMs, this means that the selling price for the product must be determined as if that product were sold to OEMs operating on the domestic market.

In the circumstances, however, the Community institutions did not have at their disposal data concerning OEM sales to customers operating on the Korean market since neither Goldstar nor other Korean producers involved in the procedure made sales of that kind. Furthermore, there is no evidence in the documents before the Court that either Goldstar or other producers in the same business sector sold products to OEMs in Korea. From that I draw the inference that the constructed value of the GCD 613 and GCD 616 units sold to OEMs for export to the Community had to be established on the basis of the last method of calculation referred to in Article 2(3)(b)(ii) of the basic regulation, namely "on any other reasonable basis".

As the Court has repeatedly stated (see, most recently, paragraph 63 of the judgment in Nakajima), Article 2(3)(b)(ii) of the basic regulation confers on the Community institutions a broad discretion in determining the "reasonable amount" for costs and profits which must be included in the constructed normal value.

(a) The connection between OEM sales and own-brand sales

23. In general the argument that there is no connection between the costs and profits for OEM sales on the domestic market and those for own-brand sales on that market is untenable. OEMs which sell on the domestic market are in competition with a producer selling on that market under his own brand. In a situation of that kind, it is obvious that the producer concerned will sell to OEMs only if, in so doing, he can clearly raise his profits. Therefore, in establishing the OEM selling price, he must, in addition to the production costs which are the same for both categories, also take account of a number of other factors such as: the higher turnover which he can achieve as a result of OEM sales, the lower costs connected with OEM sales inter alia as regards advertising, but also the increased competition from OEMs on the domestic market and the resultant pressure on the level of prices for the products sold by the producer on the same market under his own brand. In brief, a producer will reasonably charge an OEM customer operating on the same market as himself a price which, taken as a whole, will enable him to realize a higher profit than if he were to sell on that market exclusively under his own brand. Hence there is undoubtedly a connection between the determination of prices and profits for sales to OEMs operating on the domestic market and the determination of prices and profits for own-brand sales on that market. However, it is difficult to express that connection in figures on an abstract plane where there have been no OEM sales on the domestic market, and the Community institutions must therefore be allowed to exercise a certain discretion in that respect.

24. In this case the Community institutions considered that the said connection enabled them to establish the constructed value of (non-existent) OEM sales on the Korean market on the basis of the weighted average of the costs incurred and the profits realized by the same producer on his (actual) own-brand sales on that market, on the understanding that only a portion (30%) of the profit realized on those sales by the producer concerned would be taken into account. In other words, by taking a proportionately smaller profit margin into consideration, the institutions believed they were taking account of all possible differences in costs and profits between OEM sales and own-brand sales by the producer concerned.

It seems to me that, in so doing, the Community institutions did not overstep the limits of their discretion, all the more so as that modus operandi is in keeping with an established practice, as is apparent from three other anti-dumping procedures relied upon by Goldstar (plain paper photocopiers, (21) video cassette recorders (22) and small screen colour television receivers (23)). In those procedures the institutions also determined the constructed value of (similarly non-existent) OEM sales on the domestic market on the basis of the data concerning (actual) own-brand sales on that market, although the amount for profit was established on the basis of a uniform rate of 5% for all producers. Apart from the differences discussed below concerning the method of calculation of the amount for profit in this case (paragraph 25 below), it is clear that in all those procedures the premise was the same, namely that by taking into consideration a smaller profit margin for OEM sales than that for domestic sales all possible differences in costs and profits between OEM sales and other sales had been taken into account.

I therefore consider that it was wrong to reproach the Council for establishing a connection in this case between OEM sales and own-brand sales. Moreover, in its judgment in Nashua (at paragraph 33; see paragraph 28 below) the Court implicitly recognized the establishment of such a connection as valid.

(b) Was the Council' s approach contrary to previous practice?

25. Goldstar points out that in the three other anti-dumping procedures referred to earlier, the Community institutions applied a uniform profit rate for all producers in establishing the constructed value of (non-existent) OEM sales on the domestic market in relation to (actual) own-brand sales on that market, whilst in this case they have applied a profit rate for OEM sales which is also related to own-brand sales but is based on the individual data of each individual producer. (24) In this case, that anomalous method led, in Goldstar' s view, to an unreasonably large profit rate, namely y% (see footnote 24) of Goldstar' s production costs, being taken into account.

I am not swayed by Goldstar' s argument that here too the Council should, in keeping with its previous practice, have taken into consideration a uniform profit rate for all producers of compact disc players operating on the Korean market rather than an individual rate for each producer, in this case Goldstar. I have already pointed out (paragraph 20, above) that the most accurate method of establishing a constructed value is to do so with the aid of the sales made by the producer concerned on the domestic market. Since in the circumstances there were no OEM sales on the domestic market, it was for the Community institutions to establish the constructed value of (non-existent) OEM sales by reference to (actual) own-brand sales, and in so doing they were entitled to give preference to individual data relating to (actual) own-brand sales made by each of the producers concerned. The institutions were thus under no obligation to use a uniform profit rate.

26. In that connection it must be pointed out that in Case C-172/87 Mita Industrial v Council, which is still pending before the Court, Mita' s standpoint is precisely the reverse of that adopted by Goldstar in this case. Mita alleges that the Council discriminated against it by establishing in the anti-dumping regulation concerning plain paper photocopiers the constructed value of (non-existent) OEM sales on the domestic market with the aid of a uniform profit rate, with the result that exporters with large profits on their domestic sales were favoured. In paragraph 12 of his Opinion of 13 December 1990 in that case, Advocate General Mischo proposes that the Court should reject that allegation inter alia having regard to the Council' s discretion in that regard.

That position is not incompatible with the view taken in this Opinion. The Community institutions must investigate each case separately and take account of any differences. As the Council has rightly pointed out, there are clear differences between the three other procedures referred to by Goldstar and the present case, in particular as regards the producers' profit margins on their (actual) domestic sales under their own brand. In the procedure concerning plain paper photocopiers, the differences in the producers' profits are unknown; only the average profit margin of 14.6% is mentioned. In the other two procedures, however, reference is made to the differences in the profit margins of the various producers on own-brand sales. Those margins oscillated between 8% and 12% of production costs in the procedure concerning video cassette recorders, and between 7% and 14% in the procedure concerning small screen colour television receivers. Those are relatively small differences and it is understandable, in my view, that the Community institutions, which have a certain discretion, did not opt for an individual approach in those circumstances.

It is otherwise in the case now under consideration. Goldstar' s profit on its own-brand sales on the domestic market is far greater than the abovementioned profit rates in the other procedures. Furthermore, Goldstar itself concedes that its profit (as far as I can see, greatly) exceeds the average profit of the other producers. In those circumstances there were no grounds, in my view, for departing from the practice whereby preference may, as a rule, be accorded to the figures of the producer concerned in establishing the normal constructed value. Moreover, the establishment of a uniform profit rate for all producers may result in placing at an unjustified disadvantage the other Korean producer of compact disc players, referred to in the contested definitive regulation, who also made sales to OEMs in the Community during the reference period.

(c) Is the 30% rate arbitrary?

27. In the light of the Community institutions' preference for establishing in this case the constructed value for OEM sales on the basis of (Goldstar' s) individual data, and given the absence of genuine profit figures in view of the fact that there were no sales of compact disc players to OEMs on the domestic market and also that at the time any experience with regard to domestic OEM sales in other procedures was lacking, the institutions had no choice but to estimate as far as they could the ratio between the profit on OEM sales and the profit on own-brand sales on the Korean market. By fixing that ratio at 30% the Community institutions did not, in my view, exceed the limits of their discretion. It is clear from the following three factors that 30% is not an arbitrary estimate.

In the procedure concerning plain paper photocopiers a constructed value for OEM sales was calculated, as stated earlier, with the aid of a uniform profit rate for all producers fixed at 5% of production costs. That rate corresponded to approximately one-third of the average profit on the domestic sales of those producers, which amounted to 14.6% of production costs. From that it may be inferred that on the assumption that the amount for profit in that procedure was established not on a uniform but on an individual basis, the institutions should also have taken a ratio of one to three between the amount for profit on OEM sales and the profit realized by each producer on sales under his own brand.

Furthermore, the Community institutions now have at their disposal the data relating to the anti-dumping procedure concerning imports of "serial-impact dot -matrix printers" originating in Japan (25) and inter alia the data concerning sales of those products by Japanese producers to OEMs operating on the Japanese market as well. It is apparent from that data that the profit on the OEM sales in question amounts to approximately one-third of the profit on own-brand sales.

Finally, reference may be made to the fact that during the administrative procedure Goldstar itself suggested a proportion of 30%, albeit of the average profit for all producers taken together. (26)

28. For the sake of completeness, allow me briefly to mention that in Nashua as well the applicant contended that the Community institutions had taken an arbitrary profit rate (5%) into account in establishing the constructed value of the products sold to OEMs (see paragraphs 32 to 34 of the judgment). The applicant considered that that profit rate took insufficient account of the differences between OEM sales and own-brand sales. The Court rejected that submission not by reference to the discretion of the institutions but on the ground that the applicant had been unable to substantiate its contention. Goldstar in these proceedings has been no more successful than Nashua in demonstrating that if it had made sales to OEMs on the domestic market its prices would have been lower than the constructed value established by the institutions.

(d) Has the Council contravened the principle of equality?

29. Finally, Goldstar claims that all Korean producers are in a similar situation since none of them has sold its products to OEM purchasers on the domestic market. In the light of the principle of equality, therefore, the Council should have taken the same amount for profit into consideration for each Korean producer.

This argument must also be rejected. As stated earlier, Goldstar' s profit on its domestic sales was greater than that of the other Korean producers. The method adhered to by the Community institutions of calculating the constructed value of the products sold to OEMs takes account of those differences in profits, with the result that no breach of the principle of equality can be alleged.

30. In the light of the foregoing considerations I propose that the Court dismiss the action and order the applicant to pay the costs, including those of Compact' s intervention. Under the first paragraph of Article 69(4) of the Rules of Procedure, as amended on 15 May 1991, the Commission, must bear the costs of its intervention.

(*) Original language: Dutch.

(1) - OJ 1990 L 13, p. 21. In Case C-104/90 Matsushita Electric Industrial also seeks the annulment of Regulation No 112/90 but on grounds other than those relied upon by Goldstar.

(2) - OJ 1988 L 209, p. 1.

(3) - In its judgments of 14 March 1990 in Joined Cases C-133/87 and C-150/87 Nashua Corporation v Commission and Council [1990] ECR I-719, at paragraph 3, and in Case C-156/87 Gestetner Holdings v Council and Commission [1990] ECR I-781, at paragraph 3, the Court described an OEM as a supplier under its own brand of products manufactured by other undertakings.

(4) - OJ 1989 L 205, p. 5.

(5) - Judgment of 5 October 1988 in Joined Cases 277 and 300/85 Canon v Council [1988] ECR 5731.

(6) - Agreement on implementation of Article VI of the General Agreement on Tariffs and Trade, adopted on behalf of the European Economic Community by Council Decision 80/271/EEC of 10 December 1979 concerning the conclusion of the Multilateral Agreements resulting from the 1973 to 1979 trade negotiations (OJ 1980 L 71, p. 1).

(7) - Commission Regulation (EEC) No 3643/84 of 20 December 1984 (OJ 1984 L 335, p. 43), in particular the ninth recital in the preamble, and Council Regulation (EEC) No 1698/85 of 19 July 1985 (OJ 1985 L 163, p. 1), in particular the fifth recital in the preamble.

(8) - Judgment of 5 October 1988 in Case 250/85 Brother v Council [1988] ECR 5683.

(9) - See the judgment in Canon, cited in footnote 5, at paragraph 14.

(10) - Judgment of 11 July 1990 in Joined Cases C-305/86 and C-160/87 Neotype Techmashexport v Commission and Council [1990] ECR I-2945.

(11) - Judgment of 22 October 1991 in Case C-16/90 Noelle v Hauptzollamt Bremen-Freihafen [1991] ECR I-516 at paragraph 22.

(12) - Judgment of 5 October 1988 in Joined Cases 273/85 and 107/86 Silver Seiko v Council [1988] ECR 5927.

(13) - In response to a question from the Court concerning the appropriate level of an absolute figure, Goldstar' s representative at the hearing stated that 50 units was a fair criterion.

(14) - See, in particular, paragraph 21 of Annex VI to Goldstar' s application.

(15) - In its view this should also have been done for the other models, a contention which I have rejected above.

(16) - Judgment of 7 May 1991 in Case C-69/89 Nakajima All Precision v Council [1991] ECR I-2069.

(17) - Since Goldstar does not contest the calculation of the production costs including an amount for expenses, I shall refer henceforth only to the provisions of that article which relate to the determination of the amount to be added for profit. For the sake of brevity, I shall confine myself to producers, and not to exporters.

(18) - See Brother (at paragraph 18), Canon (at paragraph 26), Silver Seiko (at paragraph 16) and Nakajima (at paragraph 64).

(19) - Unofficial translation.

(20) - For a definition of an OEM, see footnote 3.

(21) - Council Regulation (EEC) No 535/87 of 23 February 1987 imposing a definitive anti-dumping duty on imports of plain paper photocopiers originating in Japan (OJ 1987 L 54, p. 12).

(22) - Council Regulation (EEC) No 501/89 of 27 February 1989 imposing a definitive anti-dumping duty on imports of certain video cassette recorders originating in Japan and the Republic of Korea and definitively collecting the provisional duty (OJ 1989 L 57, p. 55).

(23) - Commission Regulation (EEC) No 3232/89 of 24 October 1989 imposing a provisional anti-dumping duty on imports of small screen colour television receivers originating in the Republic of Korea (OJ 1989 L 314, p. 1).

(24) - The uniform rate used in the three procedures mentioned above amounts to 5% of the average production costs for the products concerned. The rate used in this case amounts to 30% of x% (being Goldstar' s profit rate on domestic own-brand sales, which I shall not disclose on grounds of confidentiality) of Goldstar' s production costs, namely (30% of x% =) y%.

(25) - Council Regulation (EEC) No 3651/88 of 23 November 1988 imposing a definitive anti-dumping duty on imports of serial-impact dot-matrix printers originating in Japan (OJ 1988 L 317, p. 33).

(26) - See paragraph 40 of the definitive regulation.

Full & Egal Universal Law Academy