Language of document : ECLI:EU:T:2011:165

JUDGMENT OF THE GENERAL COURT (First Chamber)

13 April 2011 (*)

(Dumping – Imports of polyethylene terephthalate originating in Taiwan – Determination of the dumping margin – Asymmetrical method of calculation – Export pricing pattern differing according to purchasers and time periods – Full degree of dumping margin cannot be reflected by symmetrical methods of calculation – Duty to state reasons)

In Case T‑167/07,

Far Eastern New Century Corp., formerly Far Eastern Textile Ltd, established in Taipei (Taiwan), represented by P. De Baere, lawyer,

applicant,

v

Council of the European Union, represented by J.-P. Hix and B. Driessen, acting as Agents, and by G. Berrisch, lawyer,

defendant,

supported by

European Commission, represented initially by H. van Vliet and K. Talabér-Ritz, and subsequently by H. van Vliet and M. França, acting as Agents,

intervener,

APPLICATION for annulment of Council Regulation (EC) No 192/2007 of 22 February 2007 imposing a definitive anti-dumping duty on imports of certain polyethylene terephthalate originating in India, Indonesia, Malaysia, the Republic of Korea, Thailand and Taiwan following an expiry review and a partial interim review pursuant to Article 11(2) and Article 11(3) of Regulation (EC) No 384/96 (OJ 2007 L 59, p. 1),

THE GENERAL COURT (First Chamber),

composed of I. Wiszniewska-Białecka, President, F. Dehousse and H. Kanninen (Rapporteur), Judges,

Registrar: K. Pocheć, Administrator,

having regard to the written procedure and further to the hearing on 1 June 2010,

gives the following

Judgment

 Legal context

1        Under Article 1(1) and (2) of Council Regulation (EC) No 384/96 of 22 December 1995 on protection against dumped imports from countries not members of the European Community (OJ 1996 L 56, p. 1), as amended (‘the Basic Regulation’) (now Article 1(1) and (2) of Council Regulation (EC) No 1225/2009 of 30 November 2009 on protection against dumped imports from countries not members of the European Community (OJ 2009 L 343, p. 51; corrigendum OJ 2010 L 7, p. 22)):

‘1. An anti-dumping duty may be applied to any dumped product whose release for free circulation in the Community causes injury.

2. A product is to be considered as being dumped if its export price to the Community is less than a comparable price for the like product, in the ordinary course of trade, as established for the exporting country.’

2        Article 2(10) of the Basic Regulation (now Article 2(10) of Regulation No 1225/2009) provides:

‘A fair comparison shall be made between the export price and the normal value. This comparison shall be made at the same level of trade and in respect of sales made at, as closely as possible, the same time and with due account taken of other differences which affect price comparability. ...’

3        Article 2(11) of the Basic Regulation (now Article 2(11) of Regulation No 1225/2009) provides:

‘Subject to the relevant provisions governing fair comparison, the existence of margins of dumping during the investigation period shall normally be established on the basis of a comparison of a weighted average normal value with a weighted average of prices of all export transactions to the Community, or by a comparison of individual normal values and individual export prices to the Community on a transaction-to-transaction basis. However, a normal value established on a weighted average basis may be compared to prices of all individual export transactions to the Community, if there is a pattern of export prices which differs significantly among different purchasers, regions or time periods, and if the methods specified in the first sentence of this paragraph would not reflect the full degree of dumping being practised. This paragraph shall not preclude the use of sampling in accordance with Article 17 [now Article 17 of Regulation No 1225/2009].’

4        Article 2(12) of the Basic Regulation (now Article 2(12) of Regulation No 1225/2009), provides:

‘The dumping margin shall be the amount by which the normal value exceeds the export price. Where dumping margins vary, a weighted average dumping margin may be established.’

5        Article 2(11) of the Basic Regulation constitutes the transposition into Community law of Article 2.4.2 of the Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade 1994 (OJ 1994 L 336, p. 103; ‘the 1994 Anti-dumping Code’), set out in Annex 1A to the Agreement establishing the World Trade Organisation (‘WTO’) (OJ 1994 L 336, p. 3).

6        Article 2.4.2 of the 1994 Anti-dumping Code reads as follows:

‘Subject to the provisions governing fair comparison in paragraph 4, the existence of margins of dumping during the investigation phase shall normally be established on the basis of a comparison of a weighted average normal value with a weighted average of prices of all comparable export transactions or by a comparison of normal value and export prices on a transaction-to-transaction basis. A normal value established on a weighted average basis may be compared to prices of individual export transactions if the authorities find a pattern of export prices which differ significantly among different purchasers, regions or time periods, and if an explanation is provided as to why such differences cannot be taken into account appropriately by the use of a weighted average-to-weighted average or transaction-to-transaction comparison’.

 Facts

7        The applicant – Far Eastern New Century Corp., formerly Far Eastern Textile Ltd – is a company established in Taiwan which exports polyethylene terephthalate (‘PET’) to the European Union.

8        By Council Regulation (EC) No 2604/2000 of 27 November 2000 (OJ 2000 L 301, p. 21), as last amended by Council Regulation (EC) No 1646/2005 of 6 October 2005 (OJ 2005 L 266, p. 10), the Council of the European Union imposed a definitive anti-dumping duty on imports of certain types of PET originating, inter alia, in Taiwan. The Council fixed the applicant’s dumping margin at 7.8% of the CIF (cost, insurance, freight) Community-frontier import price, on the basis of the comparison between the weighted average normal value and the weighted average of the prices of all exports to the Community (‘the first symmetrical method’). Accordingly, it imposed a definitive anti-dumping duty of EUR 50.2 per tonne on the applicant’s imports of PET.

9        By Council Regulation (EC) No 83/2005 of 18 January 2005 amending Regulation No 2604/2000 on imports of PET originating inter alia in the Republic of Korea and Taiwan (OJ 2005 L 19, p. 1), the Council concluded the interim review conducted under Article 11(3) of the Basic Regulation (now Article 11(3) of Regulation No 1225/2009) and amended the anti-dumping measures imposed by Regulation No 2604/2000. The applicant’s dumping margin was set at 0%. The anti-dumping duty which had been imposed on it by Regulation No 2604/2000 was therefore lifted.

10      As the amendments to the anti-dumping measures imposed by Regulation No 2604/2000 had not extended the original period of validity of those measures, the Commission of the European Communities published, on 2 March 2005, a notice of the impending expiry of the measures (OJ 2005 C 52, p. 2).

11      On 1 December 2005, the Commission initiated an expiry review and a partial interim review of the measures imposed by Regulation No 2604/2000 (OJ 2005 C 304, p. 4). The expiry review period selected by the Commission (‘the investigation period’) ran from 1 October 2004 to 30 September 2005.

12      On 24 November 2006, the Commission sent the applicant a disclosure letter setting out the essential facts and elements on the basis of which it envisaged proposing that the Council should impose a definitive anti-dumping duty of EUR 36.3 per tonne on its imports of PET, corresponding in value to a duty of 3.5%. The applicant’s dumping margin had been obtained by comparing a weighted average normal value with the prices of all individual exports to the Community (‘the asymmetrical method’).

13      Following a request to that effect, detailed calculations were communicated to the applicant on 30 November 2006.

14      By letter of 8 December 2006, the applicant submitted its observations on a series of questions connected with the abovementioned disclosure letter.

15      At the applicant’s request, a hearing took place at the Commission’s premises on 8 January 2007.

16      Following that hearing, the applicant submitted further observations to the Commission on 10 January 2007.

17      On 22 February 2007, the Council adopted Regulation (EC) No 192/2007 imposing a definitive anti-dumping duty on imports of certain polyethylene terephthalate originating in India, Indonesia, Malaysia, the Republic of Korea, Thailand and Taiwan following an expiry review and a partial interim review pursuant to Article 11(2) and Article 11(3) of Regulation (EC) No 384/96 (OJ 2007 L 59, p. 1; ‘the contested regulation’).

18      As regards Taiwan, the Council stated in recital 66 in the preamble to the contested regulation that, ‘[a]s provided by Article 2(11) and (12) of the Basic Regulation, the weighted average normal values of each type of the product concerned exported to the Community were compared to the weighted average export price of each corresponding type of the product concerned’.

19      In recital 67 to the contested regulation, the Council explained as follows:

‘On the basis of such comparison, the dumping margin found was below the de minimis margin in the case of [the applicant]. … However, for [the applicant], the investigation showed that the [first symmetrical method] did not reflect the full degree of dumping being practised. Indeed, the investigation showed that significant volumes (around 25% of all exports to the Community) were made at significantly low prices and were concentrated on one customer. In addition, exports to all Community destinations were made at significantly decreased prices during the last four months of the [investigation period] in comparison with the first eight months of [that period]. Therefore, another comparison methodology had to be applied. An important difference was found between the dumping margins resulting from [the first symmetrical method] against the [asymmetrical method]. With regard to the transaction-to-transaction comparison, it was not found to be an appropriate alternative comparison method because the process of selecting individual transactions in order to make such a comparison was considered arbitrary in this case. Thus, a comparison [according to the asymmetrical method] was made in accordance with Article 2(11) of the Basic Regulation. Thus a clear pattern of exports differing by customer and by time existed’.

20      Consequently, the Council explained, in recital 68 to the contested regulation, that, in the applicant’s case, it had taken into account the dumping margin yielded by the asymmetrical method. According to recital 69 to that regulation, that margin was 3.5%, which corresponds to the specific duty of EUR 36.3 per tonne. In Article 1(2) of the contested regulation, the Council therefore set an anti-dumping duty in that amount for the applicant.

21      The Commission informed the applicant by letter of 27 February 2007 that the arguments which the applicant had put forward during the procedure leading to the adoption of the contested regulation, concerning the application in its case of the asymmetrical method, had to be rejected. As regards the grounds on which it rejected those arguments, the Commission referred to the contested regulation, published in the Official Journal of the European Union of the same date, and also to the arguments set out in Annex 2 to that letter.

 Procedure and forms of order sought

22      By application lodged at the Court Registry on 18 May 2007, the applicant brought the present action.

23      By document lodged at the Court Registry on 31 August 2007, the Commission applied for leave to intervene in support of the Council. By order of 3 December 2007, the President of the First Chamber of the Court of First Instance (now ‘the General Court’) granted leave to intervene. The Commission has not lodged a statement in intervention.

24      On 17 March 2010, in the context of the measures of organisation of procedure provided for in Article 64 of the Rules of Procedure of the Court, the Court asked the applicant and the Council to reply to certain written questions and to send to it certain documents. The parties complied with those requests within the prescribed period.

25      On the basis of the report of the Judge-Rapporteur, the Court (First Chamber) decided to open the oral procedure.

26      At the hearing on 1 June 2010, the parties presented oral argument and their answers to the questions put by the Court.

27      The applicant claims that the Court should:

–        annul the contested decision in so far as it relates to the applicant;

–        order the Council to pay the costs.

28      The Council contends that the Court should:

–        dismiss the action;

–        order the applicant to pay the costs.

 Law

29      The applicant raises four pleas in law in support of its action: (i) breach of Article 2(11) of the Basic Regulation, on the ground that the Council used the asymmetrical method to calculate the dumping margin; (ii) breach of Article 2(11) of the Basic Regulation and Article 253 EC on the ground that the Council did not fully explain why neither the first symmetrical method nor the method consisting in a comparison of individual normal values with individual export prices to the Community on a transaction-to-transaction basis (‘the second symmetrical method’) would reflect the full extent of the dumping being practised; (iii) breach of Article 2(10) to (12) of the Basic Regulation, on the ground that the Council had applied the zeroing technique (see the definition in paragraph 110 below) in calculating the applicant’s dumping margin; and (iv) breach of the obligation to state reasons, laid down in Article 253 EC, on the ground that the Council failed to state appropriate reasons why, in calculating the dumping margin, the zeroing technique had had to be applied.

30      The first and second pleas may be grouped together, as may the third and fourth pleas.

1.     The first and second pleas: breach of Article 2(11) of the Basic Regulation and breach of Article 253 EC

 Arguments of the parties

31      The applicant refers to the fact that Article 2(11) of the Basic Regulation provides that the asymmetrical method can be used only when two conditions are satisfied: (i) the pattern of export prices differs significantly as between different purchasers, regions or time periods and (ii) the symmetrical methods would not reflect the full degree of dumping being practised. Furthermore, according to the applicant, any comparison carried out pursuant to that provision is subject to the precondition that it be fair. The choice of a method cannot therefore be determined by the fact that it leads to a higher dumping margin.

32      In order to illustrate its argument, the applicant states that if export prices and domestic prices follow the same upward trend during a particular investigation period and if export prices remain above domestic prices, the use of a symmetrical method would not lead to a finding of dumping. However, in the same hypothetical situation, use of the asymmetrical method combined with the zeroing technique would lead to a finding of dumping: all export transactions during the first half of that investigation period would generate positive dumping margins, while those during the second half would generate negative margins. According to the reasoning followed by the Council in the contested regulation, use of the asymmetrical method would therefore be permitted in such a situation, since the pattern of export prices would differ by time period and there would be important differences between the dumping margins yielded through application of the symmetrical methods and the dumping margin yielded through application of the asymmetrical method. However, application of the latter method would be unfair, as no export transaction would be found to have been dumped if it were compared with a domestic transaction carried out on the same date.

33      The applicant claims that the conditions for applying the asymmetrical method were not satisfied in the present case.

34      As regards the first condition, the applicant notes that the Council stated that the pattern of export prices differed significantly by purchaser and by time period.

35      The first of those significant differences in the pattern of export prices was said to arise as a result of the differences between the prices charged to a company established in Austria and the prices charged to other purchasers. The second was said to arise as a result of the differences between export prices during the first eight months of the investigation period and export prices during the last four months of that period.

36      As regards the first of those significant differences, the applicant claims, first, that such a difference was perceived because of a manifest error of assessment consisting in a finding that the prices charged to the company referred to in paragraph 35 above were lower than those charged to other purchasers. However, a month-by-month comparison of the prices charged to that company with those charged to the other purchasers shows that the prices charged to the company in question are only 0.38% lower. The dumping margin on sales to that company was perceived as higher than that of the other purchasers simply because it purchased higher quantities towards the end of the investigation period, when prices generally had declined.

37      Secondly, the applicant claims that the prices on the basis of which the Council calculated the dumping margin of the company referred to in paragraph 35 above were exclusively those of the goods supplied to that company in Austria. In doing so, the Council disregarded – incorrectly – the price of goods supplied to the same company in Italy. The Council’s finding that significant volumes were exported at very low prices and to a single customer is therefore wrong.

38      The Council should have realised, in the light of the documents in its possession, that the goods being supplied in Austria and in Italy to addressees with the same numerical code were being sold to the same company. The Council’s error cannot be attributed to the applicant, which used in its communications the same customer code for goods exported to both Member States concerned for the company referred to in paragraph 35 above.

39      Thirdly, the applicant states that the company referred to in paragraph 35 above was its largest customer and that there were differences between the quantities purchased by that customer and those purchased by the other customers.

40      As for the existence of significant differences in the pattern of export prices according to time periods, the applicant acknowledges that its export prices were lower during the last four months of the investigation period than during the first eight months of that period. However, they fluctuated without following a specific pattern. Prices first dropped and then rose again. Likewise, as PET is produced from derivatives of crude oil, its price is affected by the volatility of crude oil prices. Lastly, price fluctuations during the investigation period were less pronounced than the normal annual price variation during the previous 16 years and, in consequence, are neither significant nor indicative of ‘targeted dumping’, since export prices, domestic prices and production costs moved in parallel.

41      According to the applicant, the Council thus made a manifest error of assessment in considering that the export prices found during the last four months of the investigation period were evidence of ‘targeted dumping’. Small positive and negative dumping margins were evenly spread out over the entire investigation period and no negative dumping margins generated during the first eight months can be used to disguise positive dumping margins created during the alleged period of ‘targeted dumping’. On the contrary, a dumping margin calculation based on the last four months of the investigation period actually reveals a negative margin of 0.33%.

42      The Council rejected comparison on a monthly basis on the ground that it would be at odds with the principle of the universality of the investigation period. However, under Article 2(11) of the Basic Regulation, for a finding of ‘targeted dumping’ to be made, a comparison must have been undertaken of prices during sub-periods of the investigation period. A comparison of monthly export prices with monthly normal values is necessary for the purposes of assessing whether the existence of differing export prices means that ‘targeted dumping’ exists. In reality, the Council considered that the presence of a pattern of export prices differing significantly by time period was sufficient evidence of the existence of ‘targeted dumping’. That approach would lead to a finding of ‘targeted dumping’ in all cases where prices vary over time.

43      According to the applicant, the unreasonable nature of such a finding is illustrated by a specific example (‘the crude oil example’). The applicant relates that the export price per barrel of crude oil rose steadily between 1 November 2006 and 1 November 2007, from USD 58.98 to USD 89.72. The average export price was USD 63.16 during the first eight months of that period and USD 77.68 during the last four months of that period. If the Council’s argument were accepted, it would lead to a finding that the pattern of prices differed significantly according to time periods.

44      In addition, the applicant argues, the Council’s selection of time periods in this case is arbitrary and the price difference between those time periods – 11% – is not significant. Even if other periods had been selected, the results would have been similar. The difference between the weighted averages in the first four months of the investigation period as compared with the last eight months is 8.05%; the difference between the weighted averages in the first seven months and in the last five months is 9.22%; and the difference between the weighted averages in the first nine months and in the last three months is 8.84%. It is impossible to conclude that a different price pattern exists between the first eight months of the investigation period and the last four months when, if other periods are selected, almost identical price differences are found. What the price data show is not the existence of a different pricing pattern between two different time periods but a gradual decrease in prices.

45      Lastly, the Court did not state in Case T‑274/02 Ritek and Prodisc Technology v Council [2006] ECR II‑4305, paragraph 59, that for the purposes of establishing that the pattern of export prices differs significantly as between different purchasers, regions or time periods, the Council is not required to examine the reasons for the price differences found. That statement relates only to the calculation of the dumping margin. In addition, the Court considered, in paragraphs 61 to 63 of that judgment, whether the cyclical nature of the market and the evolution of worldwide prices were acceptable reasons for the price differences in question.

46      As regards the second condition for applying the asymmetrical method, the applicant observes that the Council failed to state why the symmetrical methods would not reflect the full degree of dumping being practised, as required under Article 2.4.2 of the 1994 Anti-dumping Code and the case-law.

47      In the present case, the first symmetrical method was rejected by the Council because an important difference was found between the dumping margin yielded through application of that method and the dumping margin yielded through application of the asymmetrical method. However, the Council did not explain why a difference of 1.54 percentage points – that is to say, less than the de minimis dumping margin – is significant or why the pricing differences could not be taken into account by the first symmetrical method. Such an explanation would have been appropriate, since the same price evolution took place both on the export market and on the domestic market and since a comparison of monthly weighted averages did not reveal ‘targeted dumping’.

48      Application of the second symmetrical method of calculating the dumping margin was rejected on the ground that the selection of individual transactions for the purposes of that comparison was arbitrary. However, the Council does not explain why that selection was arbitrary, when several countries regularly apply that method.

49      The applicant argues that the second symmetrical method, which is particularly appropriate for situations of price volatility, could easily have been applied in its case. First, there is only one product type, which is identical on the domestic market and on the export market. Secondly, the Council confirmed in the contested regulation that normal value could be established on the basis of the prices paid or payable in the ordinary course of trade by independent purchasers on the domestic market. Thirdly, the Council is in possession of detailed information on all transactions during the investigation period, such as the ‘DMSALES’ file, containing 1 640 transactions, while export sales to the Community can always be compared with domestic sales during the same period. Fourthly, where several sales on the domestic market are made on the same date as an export sale, it is possible to select the most comparable sale by using subsidiary criteria, such as quantity.

50      In order to prove that application of the second symmetrical method is possible, the applicant appends to the reply a calculation of its dumping margin based on that method, which yields a margin of ‑2.76%. Moreover, no ‘targeted dumping’ is found, since the margin for the first eight months of the investigation period is ‑2% and the margin for the last four months is ‑5.45%.

51      In reality, the applicant claims, the Council rejected the symmetrical methods because only the asymmetrical method enabled it to find that dumping had taken place.

52      The crude oil example again demonstrates that that approach is unreasonable. Since crude oil prices are fixed on a global basis, domestic prices and export prices are identical and it cannot be claimed that crude oil is being dumped. However, the margin yielded through application of the asymmetrical method combined with the zeroing technique is 5.15% in the crude oil example. If the Council’s approach were accepted, the symmetrical methods would become the exception and could be used only where prices remain stable throughout the investigation period.

53      According to the applicant, the situation in the present case – in which export prices fell because of a corresponding fall in normal values – is different from that examined in Ritek and Prodisc Technology v Council, paragraph 45 above. In that judgment, which should be read against its factual background, the Court held that the use of the asymmetrical method led to dumping margins which were six to seven percentage points higher than those obtained using the first symmetrical method. Moreover, during the second half of the investigation period, export prices and domestic prices were lower than the production costs of the relevant product. That situation constitutes a particularly serious form of dumping.

54      The Council refers to the fact that a finding of just one of the significant differences in the export price patterns which it identified in the contested regulation, the first by reference to purchasers and the second by reference to time periods, is sufficient to support the conclusion that the first condition for applying the asymmetrical method is satisfied. The applicant must therefore demonstrate the existence of a manifest error with respect to both those differences.

55      As regards significant differences in the pattern of export prices depending on the purchaser, the Council states that a significant share of the applicant’s exports were sold at very low prices to the company referred to in paragraph 35 above. According to the Council, that observation, which is based on a comparison of the average prices of sales made to that company with those made to the other purchasers, is objective and there is no provision requiring the Council to compare prices on a monthly basis.

56      The Council contends that the applicant’s argument that the differences in prices charged to the company referred to in paragraph 35 above can be explained by the fact that it was the applicant’s largest customer is redundant, since the reasons for the price differences are irrelevant. Furthermore, the impact on prices of the differences in discounts and rebates and differences in quantities is governed by Article 2(10)(c) of the Basic Regulation. Those adjustments to the export price or to the normal value are made before the analysis provided for in Article 2(11) of the Basic Regulation is undertaken. The Council also stated, at the hearing, that there is nothing in the file to show any correlation between the quantities bought by each purchaser and the prices charged to each of them.

57      Lastly, although the Council acknowledged at the hearing that the company referred to in paragraph 35 above is a single purchaser which is engaged in business both in Italy and in Austria, it contends that the applicant did not rely on that fact at any point in the investigation, even though it had received an electronic copy of the calculation of the dumping margin, one sheet of which revealed that the existence of that difference in the price pattern had been established on the basis of sales made exclusively in Austria. Even if the applicant had raised that argument during the investigation, its position would not have been any stronger. First, if the institutions had accepted the argument, they would have found that the price pattern differed significantly by region. Secondly, the dumping margin of the company referred to in paragraph 35 above is – according to the applicant itself – 4.76% as compared with an overall margin of 1.96%.

58      In conclusion, the Council did not, it contends, make a manifest error of assessment in finding that the export price pattern differed significantly by purchaser.

59      As regards the significant differences in the export price pattern by reference to time periods, the Council states that the applicant’s arguments are unfounded.

60      Lastly, the Council maintains that the applicant’s arguments relating to the second condition for applying the asymmetrical method – that the symmetrical methods would not reflect the full degree of dumping being practised – are also unfounded, and contends that the reasons set out in the contested regulation are adequate in that respect.

61      The Commission contended in essence at the hearing that the various arguments put forward by the applicant should be rejected for the reasons stated by the Council.

 Findings of the Court

62      It is clear from the wording of Article 2(11) of the Basic Regulation that the existence of a dumping margin is normally to be established using one of the two symmetrical methods and that recourse to the asymmetrical method, by way of an exception to that rule, may be had only on the twofold condition that, on the one hand, the pattern of export prices differs significantly as between different purchasers, regions or time periods and, on the other hand, the symmetrical methods do not reflect the full degree of dumping being practised (Case C‑76/00 P Petrotub and Republica v Council [2003] ECR I‑79, paragraph 49).

63      The applicant claims that the two conditions for applying the asymmetrical method were not satisfied in the present case. It also criticises the reasons set out in the contested regulation concerning the second condition for applying that method.

64      It should be noted as a preliminary point that, according to established case-law, the choice between different methods of calculating the dumping margin, such as those referred to in Article 2(11) of the Basic Regulation, require an appraisal of complex economic situations, and review by the Courts of such an appraisal must therefore be limited to verifying that relevant procedural rules have been complied with, that the facts on which the contested choice is based have been accurately stated, and that there has been no manifest error in the appraisal of those facts or a misuse of powers (see Case C‑351/04 Ikea Wholesale [2007] ECR I‑7723, paragraph 41 and the case-law cited).

65      It is therefore necessary to examine whether the Council made a manifest error of assessment in using the asymmetrical method to determine the applicant’s dumping margin. It is also necessary to examine whether the reasons set out in the contested regulation are adequate as regards the second condition for applying that method.

 The first condition for applying the asymmetrical method: there must be a pattern of export prices which differs significantly as between different purchasers, regions or time periods

66      It can be seen from recital 67 to the contested regulation that the Council found that the pattern of export prices charged by the applicant during the investigation period differed significantly as between the different purchasers and the different sub-periods of that period.

67      The applicant denies the existence of those two significant differences in the pattern of its export prices. It is appropriate to examine first the arguments which the applicant puts forward concerning the differences as between the sub-periods of the investigation period and then the arguments concerning the differences as between the different purchasers.

–       The existence of significant differences in the pattern of export prices as between the different sub-periods of the investigation period

68      The Council stated, in recital 67 to the contested regulation, that the pattern of export prices charged by the applicant during the investigation period differed significantly as between the different sub-periods on the ground that, ‘exports to all Community destinations [had been] made at significantly decreased prices during the last four months of the [investigation period] in comparison with the first eight months of [that period]’.

69      The applicant puts forward two arguments to demonstrate that that finding is manifestly erroneous. By the first argument, it claims, in essence, that export prices and domestic prices moved in parallel. By the second, it claims that there was no consistent pattern in export prices.

70      By its first argument, the applicant claims that, in order to establish that the pattern of export prices differs significantly by reference to time periods, the Council cannot merely compare the export prices charged during the different sub-periods of the investigation period. According to the applicant, the Council should also compare export prices and domestic prices. The pattern of export prices differs significantly, for the purposes of Article 2(11) of the Basic Regulation, only where domestic prices do not develop in the same way as export prices.

71      In that regard, it should be noted that Article 2(11) of the Basic Regulation is silent as to how the Council must arrive at the finding that there is a pattern of export prices which differs significantly as between the different sub-periods of an investigation period. What is certain, however, is that that provision does not state that the prices of domestic sales must be taken into account for that exercise. Similarly, the provision in question does not require a dumping margin to be established for the period during which export prices differ significantly as compared with other time periods.

72      Moreover, the Court has held that the purpose of the asymmetrical method is to enable the full degree of the dumping practised to be reflected where, in the event that a difference in the export price pattern has been found, irrespective of its cause, this would not be possible using the two other methods (Ritek and Prodisc Technology v Council, paragraph 45 above, paragraph 54).

73      It must therefore be held that the question whether there is a significant difference in the pattern of export prices as between the different purchasers, regions or time periods is an autonomous question which does not require an examination of factors other than the structure of the export prices charged by the undertaking concerned.

74      The existence of a pattern of export prices which differs significantly by time period must therefore be assessed solely on the basis of an objective analysis of the development of those prices over time, no account being taken of how they compare with domestic prices or of the reasons for the different export prices.

75      That conclusion – contrary to the assertions made by the applicant – does not lead to unreasonable results in a situation where there is a gradual reduction in the price of the product concerned on the domestic market and on the export market, such as that described in the crude oil example. The finding that there is a pattern of export prices which differs significantly as between different time periods is only an interim stage in the calculation of the dumping margin and, accordingly, does not automatically mean that the existence of such a margin can properly be regarded as established.

76      Lastly, that conclusion is not contradicted by the fact that, in Ritek and Prodisc Technology v Council, paragraph 45 above, the Court examined the impact on export prices of certain characteristics of the market concerned (paragraphs 62 and 63 of that judgment).

77      It is sufficient to find in that regard that the Court held, first, that the argument that a substantial number of exports had been dumped during the first half of the investigation period did not alter the finding that a pattern of export prices existed which differed according to time periods (Ritek and Prodisc Technology v Council, paragraph 45 above, paragraph 62).

78      In relation to the arguments concerning the cyclical nature of the price development of the product concerned and the impact of the variation in world prices, the Court made it clear that its findings that the applicants’ assertions in that regard were at odds with the facts noted by the Council in the regulation concerned were findings made ‘in any event’ (Ritek and Prodisc Technology v Council, paragraph 45 above, paragraphs 62 and 63).

79      Moreover, it should be recalled that in Ritek and Prodisc Technology v Council, paragraph 45 above (paragraph 60), the Court did not call into question the Council’s assessment, set out in recital 30 to the regulation concerned, that the applicants’ assertion that ‘the differences in export prices were because of worldwide trends in prices including normal values … was … irrelevant’ because ‘the appropriate analysis has to be made on export prices to the Community’ and that ‘Article 2(11) of the Basic Regulation requires a demonstration of a pattern of export prices and not an explanation of why such a pattern existed’.

80      The applicant’s first argument must therefore be rejected.

81      By its second argument, the applicant claims in essence that its export prices fluctuated without following a consistent pattern; that the Council’s selection of the sub-periods of the investigation period is arbitrary; and that the price differences during those sub-periods are not significant.

82      It can be seen from the information which the applicant itself provided in the application that its monthly average export prices fluctuated during the investigation period to the extent that they differed significantly. Thus, the average price for December 2004 was 25.7% higher than that for July 2005.

83      It does not emerge from that information, however, that prices fluctuated without following a consistent pattern. The applicant’s monthly average export prices fell slightly in January 2005 and then remained relatively stable until May 2005. In June 2005, they experienced a significant fall and did not, until September 2005, rise to the same level as before that fall. The average prices for June, July and August 2005 are thus considerably below the average for the investigation period (13%, 14.1% and 6.7%, respectively) and, as a consequence, the pattern of export prices differed significantly.

84      As regards the criticism that the investigation period could have been divided in several different ways with very similar results to those emerging from the selection made in the contested regulation, it should be noted that all the combinations of monthly average prices given by the applicant in the reply produce similar results, since they include within the same time segment the prices charged during June, July and August 2005, which are the lowest prices charged during the investigation period. Thus, the applicant’s illustration tends instead to show that the Council’s conclusion is well founded.

85      Accordingly, the applicant’s second argument must also be rejected.

86      In the light of the foregoing, it must be held that the Council did not make a manifest error of assessment in finding that the pattern of the applicant’s export prices differed significantly as between the different sub-periods of the investigation period.

–       The existence of significant differences in the pattern of export prices as between the different purchasers

87      According to recital 67 to the contested regulation, the Council found that the pattern of export prices charged by the applicant during the investigation period differed significantly as between the different purchasers, on the ground that ‘significant volumes (around 25% of all exports to the Community) were made at significantly low prices and were concentrated on one customer’.

88      It is common ground that the customer referred to in the preceding paragraph is the company referred to in paragraph 35 above and that, in calculating the prices charged to that customer, the Council took into account only transactions relating to goods that were delivered to that customer in Austria.

89      Nevertheless, as the applicant points out, during the investigation period the applicant had also sold that company goods that were supplied in Italy. That can be seen inter alia from the copies of the two invoices which the applicant sent to the Court with its reply of 9 April 2010 to the written questions. According to the applicant, it can also be seen from the copy of an extract from its accounts, also sent with that reply, a point which the Council has not disputed.

90      The applicant submits that, throughout the administrative procedure which led to the adoption of the contested regulation, it referred both to the goods supplied in Austria to the company referred to in paragraph 35 above and to the goods supplied to that company in Italy, using the same customer code in both cases.

91      At the hearing, the Council acknowledged that it was apparent from the documents produced by the applicant that, during the investigation period, the applicant had also made sales in Italy to the company referred to in paragraph 35 above. The Council acknowledged, moreover, that those sales should normally have been taken into account in the calculation of the average price of sales to that company.

92      It is apparent from the foregoing that the Council’s assessment with regard to the existence of a price pattern by reference to purchasers is flawed by an error going to the facts.

93      However, and without it being necessary to rule on the Council’s argument that, in essence, the applicant had shown a lack of vigilance during the administrative procedure, it must be held that the Council’s error, although regrettable, does not affect the legality of the contested regulation. Indeed, it is apparent from the documents which the applicant submitted to the Court, which are based on evidence provided during the administrative procedure and, in particular, from the table set out in paragraph 15 of the reply that, if all the sales made to the company referred to in paragraph 35 above were taken into account, those sales would represent 32% of the total sales made during the investigation period, which is a higher figure than that given in the contested regulation (25%). It is also apparent from those documents that the sales to the company referred to in paragraph 35 above were made at very low prices, even if the Council’s error is taken into account. In particular, sales to that company were made at prices below those charged to other customers, thereby producing a dumping margin of 4.76%, instead of 0.73%, according to the applicant’s calculations. Accordingly, the conclusion set out in recital 67 to the contested decision that ‘significant volumes … were [exported] at significantly low prices and were concentrated on one customer’ is not affected by the error detected by the applicant.

94      The other arguments raised by the applicant do not invalidate that conclusion. In particular, with regard to its claim that sales to the company referred to in paragraph 35 above were made during a period when prices were generally low, it is sufficient to note that the first condition for applying the asymmetrical method is the existence of a difference in the pattern of export prices, irrespective of its cause (see paragraph 72 above).

95      Moreover, even if the applicant’s arguments could be interpreted as meaning that the company referred to in paragraph 35 above received discounts linked to the quantities sold and even if the Council erred in failing to take that fact into account, it is sufficient to note that no specific evidence has been produced to that effect.

96      In the light of the foregoing, the applicant’s arguments relating to the Council’s conclusion that the pattern of the applicant’s export prices differed significantly as between the different purchasers must be rejected.

97      In any event, as the Council states, the existence of any one of the three differences in the pattern of export prices referred to in Article 2(11) of the Basic Regulation is sufficient for the first condition laid down in that provision to be satisfied. As was stated in paragraphs 68 to 86 above, the Council was correct in finding in the contested regulation that there was a price pattern which differed significantly as between the different time periods. Moreover, as the applicant and the Council observed at the hearing, the error concerning the pattern of export prices by reference to purchasers did not undermine the Council’s findings concerning the pattern of prices by reference to time periods.

98      In the light of all those considerations, it must therefore be held that, in the present case, the first condition laid down in Article 2(11) of the Basic Regulation is satisfied.

 The second condition: the asymmetrical method is to be used only where the symmetrical methods would not reflect the full degree of dumping practised

99      The applicant claims that the Council did not state the reasons why the symmetrical methods ‘would not reflect the full degree of dumping being practised’ for the purposes of Article 2(11) of Basic Regulation. The applicant’s arguments must be interpreted as referring both to the reasons set out in the contested regulation and to the soundness of the Council’s assessment.

100    It should be noted in that regard – as the applicant points out – that, in Petrotub and Republica v Council, paragraph 62 above (paragraphs 58 and 60), the Court of Justice held that ‘a Council regulation imposing definitive anti-dumping duties and having recourse to the asymmetrical method for the purposes of calculating the dumping margin must in particular contain, as part of the statement of reasons required by Article [253 EC], the specific explanation provided for in Article 2.4.2 of the 1994 Anti-dumping Code’. According to the latter provision, an explanation is to be provided as to why significant differences in the pattern of export prices as between purchasers, regions or time periods cannot be taken into account through application of the symmetrical methods. It should be stated in that regard that the requirement stemming from Article 2.4.2 of the 1994 Anti-dumping Code and the requirement stemming from the second condition in Article 2(11) of the Basic Regulation are, in practice, very similar (see, in that regard, the Opinion of Advocate General Jacobs in Petrotub and Republica v Council, paragraph 62 above, paragraph 82).

101    It is therefore necessary to determine whether the Council provided an adequate explanation in the contested regulation as regards the exclusion of the two symmetrical methods and, if so, whether that explanation is valid.

–       The explanation with regard to the first symmetrical method

102    Recital 67 to the contested regulation states:

‘On the basis of [a comparison in accordance with the first symmetrical method], the dumping margin found was below the de minimis margin in the case of [the applicant]. … However, for [the applicant], the investigation showed that [the first symmetrical method] did not reflect the full degree of dumping being practised. Indeed, the investigation showed that significant volumes (around 25% of all exports to the Community) were made at significantly low prices and were concentrated on one customer. In addition, exports to all Community destinations were made at significantly decreased prices during the last four months of [the investigation period] in comparison with the first eight months of [the investigation period]. Therefore, another comparison methodology had to be applied. An important difference was found between the dumping margins resulting from [the first symmetrical method] against the [asymmetrical method]. … Thus, a comparison on the basis of the [asymmetrical method] was made in accordance with Article 2(11) of the Basic Regulation. Thus a clear pattern of exports differing by customer and by time existed’.

103    It is settled law that the statement of reasons required under Article 253 EC must show clearly and unequivocally the reasoning of the institution which adopted the contested measure, so as to enable the persons concerned to ascertain the reasons for it in order to defend their rights and to enable the Community Court to exercise its power of review. It is not necessary for the reasoning to go into all the relevant facts and points of law, since the question whether it meets the requirements of Article 253 EC must be assessed with regard not only to its wording but also to its context and to all the legal rules governing the matter in question (see Case C‑76/01 P Eurocoton and Others v Council [2003] ECR I‑10091, paragraph 88 and the case-law cited). In particular, it would be unreasonable to require a detailed description of each of the factors underpinning the contested measure, particularly where, as in the present case, the applicant was closely associated with the administrative procedure (see, to that effect and by analogy, Case C‑413/06 P Bertelsmann and Sony Corporation of America v Impala [2008] ECR I‑4951, paragraph 180).

104    The explanation provided by the Council in recital 67 to the contested regulation makes it possible to understand the reasons why it considered that the first symmetrical method would not reflect the full degree of dumping being practised and, accordingly, the reasons why it was not possible, through the use of that method, to take into account appropriately the significant differences in the pattern of export prices.

105    It can be seen from recital 67 to the contested regulation that the Council did not merely state that use of the first symmetrical method would not reflect the full degree of the dumping. First, the Council also stated that application of the first symmetrical method led to the finding that the dumping margin was below the de minimis level. Secondly, it stated that there was a difference between the dumping margins established in accordance with the first symmetrical method and that established in accordance with the asymmetrical method. Thirdly, the Council stated that that difference was ‘important’. It also emerges from that recital that, as a result of that situation, the Council necessarily formed the view that the significant differences in the pattern of export prices noted in recital 67 to the contested regulation could not be taken into account through use of the first symmetrical method.

106    Thus, the explanation in question enables the applicant – which, moreover, was closely associated with the administrative procedure – to challenge the soundness of the assessment made by the Council and it enables the Court to exercise its power of review, without there being any need to consider whether other documents in the file and, in particular, the letter sent by the Commission to the applicant on 27 February 2007 (see paragraph 21 above) might have supplemented the reasons stated in the contested regulation.

107    It is therefore necessary to examine the ground relied on by the Council with regard to the substance.

108    First of all, it should be noted that the purpose of the asymmetrical method is to enable the full degree of the dumping practised to be reflected where, in the event that a difference in the export price pattern has been found, irrespective of its cause, this would not be possible using the other two methods (see paragraph 72 above).

109    By the same token, a finding of dumping, the first stage in the assessment of whether an anti-dumping duty should be imposed, is a purely objective comparison between the normal value and export prices. That comparison, conducted in accordance with Article 2 of the Basic Regulation, is based on an examination of the economic and accounting data of the undertakings concerned and in no way extends to looking into the reasons for domestic and export price levels (Ritek and Prodisc Technology v Council, paragraph 45 above, paragraph 59).

110    It should further be noted that it is common ground that, although the contested regulation does not state this, the applicant’s dumping margin was calculated, in the context of the asymmetrical method, using ‘zeroing’, the technique by which a dumping margin of a negative amount – a sign that an export sale has been made at a price above the normal value – is set to zero in order to prevent the disguising effect that taking that dumping margin into account would have on the positive dumping found to have taken place elsewhere (see, in that regard, Ritek and Prodisc Technology v Council, paragraph 45 above, paragraph 97).

111    In that regard, it should be borne in mind that, even though the second condition for applying the asymmetrical method is certainly not designed to enable the method applied for calculating the dumping margin to be the method yielding the highest result, but to reflect the full degree of dumping being practised, the asymmetrical method, provided that it includes zeroing, will still always lead, in cases where certain export transactions have been made without dumping, to a higher dumping margin than that yielded through application of the first symmetrical method (see, to that effect, the Opinion of Advocate General Jacobs in Petrotub and Republica v Council, paragraph 62 above, paragraphs 8 to 15). Thus the obtaining, through application of the asymmetrical method, of a dumping margin higher than that yielded through application of the first symmetrical method is bound to reflect the fact that, in parallel with dumped transactions, transactions were carried out without dumping. In those circumstances, the fact of obtaining through use of the asymmetrical method a dumping margin that is higher than that obtained through use of the first symmetrical method is not entirely irrelevant for the purposes of establishing whether the latter method reflects the full degree of dumping practised (Ritek and Prodisc Technology v Council, paragraph 45 above, paragraph 83).

112    In the present case, it should first be noted that application of the asymmetrical method or of the first symmetrical method leads to a finding that a dumping margin exists, even though, in the case of the first symmetrical method, that margin is below the de minimis level, within the meaning of the Basic Regulation.

113    Secondly, as the Council rightly states, the dumping margin established in accordance with the asymmetrical method (3.55%) is over 80% higher than the dumping margin established in accordance with the first symmetrical method (1.96%). That difference cannot be described as insignificant.

114    Thirdly, the dumping margin calculated in accordance with the first symmetrical method is in the present case below the de minimis level, whilst the dumping margin calculated in accordance with the asymmetrical method is above that level.

115    Fourthly, as can be seen from the documents which the applicant appended to the letter of 10 January 2007 by which it sent submissions to the Commission following the hearing held on 8 January 2007, its dumping margin, calculated on a monthly basis, was positive for over half of the investigation period. In particular, the applicant’s dumping margin was positive for half of the period which the Council identified in recital 67 to the contested regulation as reflecting a significant decrease in export prices. Moreover, it should be noted that, as is clear from paragraphs 87 to 95 above, a substantial proportion of export sales was made to one customer at significantly low prices.

116    In the light of the foregoing, it must be held that the Council did not make a manifest error of assessment in deciding not to apply the first symmetrical method but to apply the asymmetrical method instead.

117    That conclusion is not undermined by the observations made by the applicant on the basis of Ritek and Prodisc Technology v Council, paragraph 45 above.

118    It should be noted that the ground on which the Court found in that judgment that the Council had been correct not to apply the first symmetrical method is that the dumping margin calculated in accordance with that method was 50% lower than that calculated in accordance with the asymmetrical method, in the case of one of the undertakings concerned, and varied by nearly six percentage points in the case of the other undertaking concerned (Ritek and Prodisc Technology v Council, paragraph 45 above, paragraph 84).

119    As the applicant observes, the Court also found in that judgment that, during the second half of the investigation period, the export prices concerned had been lower than production costs and had therefore constituted a particularly serious form of dumping (Ritek and Prodisc Technology v Council, paragraph 45 above, paragraph 84). However, since the existence of exports at a price below the cost of production does not in itself constitute a form of dumping under the system established under the Basic Regulation, it must be held that the Court mentioned that fact only in order to categorise as particularly serious the dumping reflected by the significant difference between the dumping margin yielded through application of the first symmetrical method and that yielded through application of the asymmetrical method. Moreover, there is no reason to consider that the seriousness of the dumping is a condition for applying the asymmetrical method. It should also be held that the Court’s additional point, made in paragraph 84 of Ritek and Prodisc Technology v Council, paragraph 45 above, referred to a factual finding by the Council in the regulation at issue in that case (see, in that regard, paragraph 69 of the judgment). In those circumstances, it cannot be inferred from Ritek and Prodisc Technology v Council, paragraph 45 above, that the fact that the dumping is alleged not to be serious in the present case should necessarily mean that the asymmetrical method should not be applied.

120    Lastly, there is no need to consider whether the Council is also entitled to conclude that the first symmetrical method would not reflect the full degree of the dumping being practised in situations like that described by the applicant in the crude oil example where, despite the existence of an important difference between the dumping margin calculated in accordance with the asymmetrical method and that calculated in accordance with the first symmetrical method, the monthly average domestic prices were, throughout the investigation period, below the corresponding monthly average export prices. As has already been observed, the applicant’s dumping margin, calculated on a monthly basis, was positive for over half of the investigation period. In particular, as was noted above, the applicant’s dumping margin was positive for half of the period identified by the Council in recital 67 to the contested regulation as reflecting a significant decrease in export prices.

–       The explanation with regard to the second symmetrical method

121    The Council stated in recital 67 to the contested regulation that the second symmetrical method was not an appropriate alternative comparison method as regards the applicant, ‘because the process of selecting individual transactions in order to make such a comparison was considered arbitrary in this case’.

122    It must be held that the arbitrariness of the selection of individual transactions in a particular case for the application of the second symmetrical method constitutes in itself a valid ground for not applying that method. Otherwise, calculation of the dumping margin would be influenced by the Council’s subjective choices, whereas a finding of dumping, the first stage in the assessment of whether an anti-dumping duty should be imposed, is to be based on a purely objective comparison of the normal value and the export prices (Ritek and Prodisc Technology v Council, paragraph 45 above, paragraph 59).

123    The applicant claims, however, that the Council did not explain why the selection of individual transactions was arbitrary in the applicant’s case. Moreover, the applicant claims that the second symmetrical method could easily have been applied in this case.

124    Accordingly, it is first necessary to consider whether the reasons set out in the contested regulation are inadequate on the ground that the Council did not explain why it would be arbitrary to select individual transactions for the purposes of the comparison. Secondly, it is necessary to determine whether the Council made a manifest error of assessment in not applying the second symmetrical method on that ground.

125    In that regard, it should be borne in mind that, as was stated in paragraph 103 above, it is not necessary for the reasoning to go into all the relevant facts and points of law, since the question whether the requirement to state reasons is met must be assessed with regard, inter alia, to the context of the measure and to all the legal rules governing the matter in question.

126    Consequently, if the contested measure clearly discloses the essential objective pursued by the institution, it would be unreasonable to require a specific statement of reasons for each of the technical choices made by the institution (see Case T‑171/97 Swedish Match Philippines v Council [1999] ECR II-3241, paragraph 82 and the case-law cited).

127    As the Council points out, the possibility of selecting, in a way that is not arbitrary, individual transactions for the purposes of the comparison required in order to apply the second symmetrical method depends on the applicant’s sales structure, which is well-known to the applicant. It would therefore be unreasonable to require the Council to reproduce, in the contested regulation, the characteristics of that structure which, in its opinion, make the selection of individual transactions an arbitrary exercise.

128    Moreover, it should be noted that the Commission had already indicated to the applicant, no later than 24 November 2006, in the disclosure letter which it sent to it, that the process of selecting individual transactions in order to apply the second symmetrical method was considered arbitrary in the applicant’s case.

129    However, in its reply of 8 December 2006 to that letter, the applicant did not challenge the Commission’s preliminary conclusion or ask for more information about the reasons why application of the second symmetrical method would be arbitrary.

130    Nor did the applicant challenge, in the letter of 10 January 2007, by which it sent the Commission submissions following the hearing held on 8 January 2007, the preliminary conclusion conveyed by the Commission in the disclosure letter, or query the reasons why application of the second symmetrical method was considered to be arbitrary.

131    Accordingly, it must be held that the reasons set out in the contested regulation are adequate in that respect.

132    As regards the soundness of the case made by the Council for not applying the second symmetrical method, the applicant gave an example in the reply which showed that the dumping margin could be calculated using that method.

133    The Council argued in the rejoinder – and was not challenged on that point by the applicant – that, in its example, the applicant compared 31 out of 98 export transactions with loss-making domestic transactions which had not been made in the ordinary course of trade.

134    Similarly, the Council noted – and was not contradicted by the applicant – that, in its example, the applicant had used the same domestic sale several times in a comparison with various export sales, without giving any explanation in that regard, which, as the Council rightly states, does not show that the example provides a selection of individual transactions which is not arbitrary.

135    Lastly, the selection method proposed by the applicant, which consists inter alia in comparing each export sale with the domestic sale made on the closest date and, where several domestic sales are made on the same day, with the sale relating to the nearest quantity, is not appropriate for a domestic sales structure like that of the applicant, as the Council rightly argues.

136    First, that method often leads to the comparison of transactions which relate to totally different quantities, since the quantities involved in domestic sales (between 1 tonne and 31 500 tonnes) are much lower than those involved in export sales (between 4 300 tonnes and 504 000 tonnes).

137    Secondly, the prices of the applicant’s domestic sales on dates that were close – or even on the same day – are very different, although those differences are not clearly related to the volume of sales.

138    That fact is clear from the extract from the table headed ‘DMSALES’, which the applicant appended to its example of a comparison made using the second symmetrical method.

139    Thus, for example, on 18 October 2004, the applicant made eight domestic sales at net unit prices of between 37 and 44 Taiwan dollars (TWD)/kg involving quantities of between 1 000 kg and 25 580 kg. It made five sales involving exactly the same quantity – namely 21 000 kg – at net unit prices of between 37 TWD/kg and 42 TWD/kg.

140    Similar situations may be observed in the case of most of the domestic sales which appear in that extract from the table ‘DMSALES’.

141    This can be seen even from the example given by the applicant to support the method that it proposes for selecting the individual transactions. Thus, on 19 November 2004, the applicant made five domestic sales at prices of between 33.38 and 45.1 TWD/kg involving quantities of between 2 100 kg and 23 060 kg. The prices of the two most similar sales in terms of quantity, the first involving a quantity of 21 000 kg and the second a quantity of 23 060 kg, were 43.65 TWD/kg and 33.38 TWD/kg, respectively. It must therefore be held that the quantity criterion does not enable the most representative transaction to be selected and involves the risk of producing arbitrary results.

142    The conclusion must therefore be drawn that the applicant does not demonstrate that the Council made a manifest error of assessment in considering that the selection of individual transactions for the purposes of applying the second symmetrical method was arbitrary in the applicant’s case.

143    In the light of the foregoing, it must be held that, in the present case, the second condition laid down in Article 2(11) of the Basic Regulation is satisfied and that the reasons stated in the contested regulation are adequate in that regard.

144    The first and second pleas must therefore be rejected.

2.     The third and fourth pleas: breach of Article 2(10) to (12) of the Basic Regulation and breach of Article 253 EC

 Arguments of the parties

145    The applicant states that all the negative dumping margins were zeroed when its dumping margin was calculated in accordance with the asymmetrical method.

146    The applicant claims that, in its report of 9 January 2007, ‘United States – Measures relating to Zeroing and Sunset Reviews – AB–2006–5’, (WT/DS322/AB/R, paragraph 115) (‘the “Zeroing Measures” Report’), the WTO Appellate Body concluded that dumping margins can be found to exist only in relation to the investigation as a whole. Thus, when the dumping margin is calculated on the basis of multiple comparisons, the results of those comparisons must be aggregated. According to the applicant, the results obtained using the three comparison methods referred to in Article 2(11) of the Basic Regulation must therefore be aggregated for the purposes of calculating the dumping margin for the product as a whole. The aggregation process is governed by Article 2(12) of the Basic Regulation, which does not impose different aggregation methods depending on the comparison method selected. Since zeroing is prohibited when aggregating the results of applying the symmetrical methods, it is also prohibited in the context of the asymmetrical method.

147    The applicant acknowledges that the above interpretation remains as yet untested and that, as Community law now stands, zeroing is not prohibited in every situation. However, other calculation methods are possible, as no obligation to resort to zeroing where the asymmetrical method is applied can be inferred from the wording of Article 2(11) and (12) of the Basic Regulation. The ‘mathematical equivalence’ argument put forward by the Council, whereby the use of zeroing in combination with the asymmetrical method is necessary so that the dumping margin yielded through application of that method is not always identical to that yielded through application of the first symmetrical method, is therefore incorrect, as the Commission contended in WTO dispute settlement proceedings.

148    The WTO Appellate Body envisaged alternatives to zeroing in the context of the asymmetrical method, thus confirming its earlier rejection of the ‘mathematical equivalence’ argument in the ‘Zeroing Measures’ Report. The Council itself accepts that other methods might be envisaged in order to take the differences in pricing patterns by time period into account. However, it considers that those methods would not be capable of taking into account differences in pricing patterns by reference to the purchaser.

149    The Council has not provided any information which would show that the zeroing technique was necessary in order to address the differences in prices found between the first eight months of the investigation period and the last four months. Multiple comparisons could also have taken those differences into account. Use of such comparisons cannot be rejected on the ground that the investigation period must not be subdivided, as that argument is contrary to Article 2(12) of the Basic Regulation, to the Commission’s practice and to that of the WTO Dispute Settlement Body. By automatically applying the zeroing technique, the Council acted in breach of the fair comparison requirement laid down in Article 2(10) and (11) of the Basic Regulation and infringed the provisions, laid down in Article 2(12) of that regulation, concerning the determination of a weighted average dumping margin for the product as a whole.

150    Furthermore, the applicant claims that, in so far as the contested regulation does not state that the applicant’s dumping margin was calculated using the zeroing technique or the reasons why that was done, the obligation to state reasons referred to in Article 253 EC was infringed.

151    The Council contends that the applicant’s arguments are unfounded.

 Findings of the Court

152    As was stated in paragraph 110 above, it is common ground that, although the contested regulation does not state this, all negative dumping margins were zeroed when the applicant’s dumping margin was calculated in accordance with the asymmetrical method.

153    The zeroing technique, although not mentioned in the 1994 Anti-dumping Code or the Basic Regulation, is commonly used by importing countries and customs unions, including the European Union (Ritek and Prodisc Technology v Council, paragraph 45 above, paragraph 97).

154    As regards the ‘Zeroing Measures’ Report, suffice it to say – without it being necessary to rule on whether a Court of the European Union is bound by reports of the WTO Appellate Body – that the applicant is not claiming that the observations made by the WTO Appellate Body in that report can be interpreted as expressly criticising the combined use of zeroing and the asymmetrical method.

155    In that regard, it should be observed that the applicant was unable to cite any decision of the WTO Appellate Body that could be interpreted as containing a direct criticism of the zeroing technique in the context of the asymmetrical method.

156    Furthermore, the Court has held that it is neither contrary to Article 2.4.2 of the 1994 Anti-dumping Code or Article 2(11) of the Basic Regulation, nor unfair for the purposes of Article 2(10) of that regulation, to employ the zeroing technique in the context of the asymmetrical method, where the two conditions for applying that method are satisfied (Ritek and Prodisc Technology v Council, paragraph 45 above, paragraph 107).

157    In any event, as the Council points out, the Court has held that the zeroing technique has proved – contrary to the assertions made by the applicant – to be mathematically necessary in order to distinguish, in terms of its results, the asymmetrical method from the first symmetrical method. In the absence of that reduction, the asymmetrical method will always yield the same result as the first symmetrical method (Ritek and Prodisc Technology v Council, paragraph 45 above, paragraph 109; see also, to that effect, Opinion of Advocate-General Jacobs in Petrotub and Republica v Council, paragraph 62 above, paragraphs 8 to 15).

158    Accordingly, the need to use the zeroing technique in the context of applying the asymmetrical method results directly from Article 2(11) of the Basic Regulation and from Article 2.4.2 of the 1994 Anti-dumping Code, which defines the asymmetrical method as an alternative to the symmetrical methods.

159    In the light of the foregoing, it must be held that it is unnecessary to state a reason for using the zeroing technique in the context of the contested regulation.

160    It can be seen from the application, moreover, that the applicant fully understood, from the time of the administrative procedure, that the Council had used the zeroing technique with the asymmetrical method.

161    As regards, lastly, the applicant’s argument that the zeroing technique is not the only technique for taking situations of ‘targeted dumping’ into account, it should be observed that, even if this were correct, the fact remains that the applicant is not claiming that the Council made a manifest error of assessment in deciding to apply the zeroing technique together with the asymmetrical method, rather than those alternatives.

162    The last two pleas must therefore be rejected as unfounded.

163    Since all the pleas for annulment have been rejected, the action must be dismissed.

 Costs

164    Under Article 87(2) of the Rules of Procedure, the unsuccessful party is to be ordered to pay the costs if they have been applied for in the successful party’s pleadings. Since the applicant has been unsuccessful, it must be ordered to pay the costs, in accordance with the form of order sought by the Council.

165    Under the first subparagraph of Article 87(4) of the Rules of Procedure, institutions which have intervened in the proceedings are to bear their own costs. The Commission, which has intervened in support of the Council, must therefore bear its own costs.

On those grounds,

THE GENERAL COURT (First Chamber)

hereby:

1.      Dismisses the action;

2.      Orders Far Eastern New Century Corp. to bear its own costs, together with those incurred by the Council of the European Union;

3.      Orders the European Commission to bear its own costs.

Wiszniewska-Białecka

Dehousse

Kanninen

Delivered in open court in Luxembourg on 13 April 2011.

[Signatures]


* Language of the case: English.