4.646 Firstly, due account should be taken, as stressed throughout the investigation, of the large number of exporters involved. Indeed, Assopiastrelle listed 205 producers, of which 78 were apparently exporters of the product investigated. 4.647 This justifies, under Article 6.10 of the Agreement, the decision to make a determination on the basis of the sample procedure. Moreover, the sample used was put forward by Assopiastrelle, acting as representative of the Italian manufacturing firms, in view of the large number of manufacturing enterprises involved. In this connection, the presentation by Assopiastrelle of the �methodology for the selection of samples� states, inter alia, that �On behalf of the Italian manufacturers of porcellanato, Assopiastrelle suggested limiting the supply of information on dumping and injury to a sample considered representative of all of the companies�. It goes on to state that �The purpose of Assopiastrelle�s proposal was to ensure greater cooperation by the Italian industry concerned and at the same time to simplify the task of the authorities investigating the Italian producers�.

4.648 This statement by the Association clearly shows that for the purposes of calculating the margin of dumping, the sample is claimed to be a valid methodology.

4.649 In reply to the proposal submitted by Assopiastrelle, the DCD, by Note No. 273-000471/98, wrote to the Economic and Trade Advisor of the Italian Embassy in Argentina as follows:

However, according to the information received, we are faced with an investigation involving a large number of Italian exporters, and would therefore like to limit the procedure to the few Italian companies representing most of Italian exports to Argentina.

4.650 In this connection it was pointed out that:

� the implementing authority, pursuant to Article 6.10 � may resort to statistically representative samples for the purposes of determining the margin of dumping in the investigation. Similarly, in order to apply this procedure, the consent of the exporters concerned is needed as long as the conclusions reached would have to be extended to the other firms that were not included in the selection.

While the investigation should extend to all firms which have exported to Argentina from the investigated origin during the period concerned, for purely methodological purposes, the analysis should be confined to a representative number of exporters of the goods involved.

4.651 The note concluded by stating that �having received the express consent of the firms, we shall proceed as stated, subject to the relevant legislation in force�.

4.652 Thus, there is on the one hand a recognition by the EC of the validity of a selection or sample as a basis for determining the margin of dumping. It cannot be inferred from the acceptance of the sample that if the sample did not serve the proposed purpose, the authority had no alternative. In other words, either the authority calculates a margin of dumping for all of the companies exporting to Argentina beyond the sample, which would clearly be unreasonable since the very purpose of the sample was to avoid such a situation, or it uses another alternative to make up for the shortcoming.

4.653 The sample did not, as such, serve the purpose, and the authority therefore warned of the possibility of �using samples which are statistically valid on the basis of information available to the authorities at the time of the selection� (Article 6.10). This became necessary in order to complete the missing information. The exporting firms were certainly not left defenceless and could have protested against the note or asked the authority to reconsider its approach; or they could have provided the missing information, thereby enabling the authority to determine a margin of dumping for each firm on the basis of the sample. The exporters did none of this, and legally consented to the approach the authority said it would use in calculating the margin of dumping. It is difficult to see how the situation that the exporters themselves contributed to creating can now be invoked against the DCD, proposing a sample that could not be used for the purpose put forward.

4.654 In any case, even if despite the above considerations Argentina is deemed to have deviated from the requirements of Article 6.10 of the AD Agreement, quod non, this would be a case of harmless error, since Argentina could not apply Article 6.10 of the AD Agreement with the supporting information provided by the sample. The error would have been harmless, since the EC did not demonstrate how this situation injured the companies by producing a higher margin of dumping for its exports. In fact, the EC did not even claim that this was the case, but confined itself to arguing that Article 6.10 of the AD Agreement had been violated, thereby tacitly acknowledging that it did not happen.

(b) Arguments of Argentina in its first oral statement relating to the EC�s claim under Article 6.10 of the AD Agreement

4.655 In its first oral statement, the EC made the following arguments relating to the EC�s claim under Article 6.10 of the AD Agreement.

4.656 Pursuant to Article 6.10, in particular paragraph 1 thereof, the DCD selected the enterprises that would be the subject of the investigation. To that end, it accepted Assopiastrelle�s proposal, as explained in paragraphs 115 and 116 of Argentina�s first written submission.

4.657 At the same time, when the implementinng authority, upon opening the investigation, chose as a �homogenization� criterion the segmentation of the tiles on the basis of their size, it distributed the questionnaires prepared accordingly. The firms involved replied to these questionnaires without raising any objections, showing that they accepted the validity of segmentation by size. Nevertheless, a vast majority of the sample documentation provided by the exporting firms did not correspond to the segmentation used by the implementing authority.

4.658 Thus, when the implementing authority had to decide on the elements necessary for the determination of dumping, it found, as explained below, that it did not have the required documentation from the selected companies corresponding to the segmentation of the tiles on the basis of their size.

(i) Characteristics of the sample

4.659 The fact that the sample companies represented a majority of exports to the Argentine market does not in itself imply that the supporting documentation was representative for each one of the segments selected (20 cm x 20 cm, 30 cm x 30 cm and 40 cm x 40 cm) for the purposes of determining their normal value and export price.

4.660 How could an individual margin of dumping be calculated when the domestic market sales invoices represented only some 1.35 per cent of the value and 1.92 per cent of the physical volume (m2), as stated earlier on?

4.661 How, also, could the DCD have calculated the margin of dumping for 30 cm x 30 cm tiles from Caesar and Marazzi when these companies failed to submit any information on average prices in the Italian domestic market for that category?

4.662 At the same time, as stated in paragraph 112 of Argentina�s first written submission, in the case of Bismantova, 56 per cent of sales on the domestic market were to its controlling company, Rondine; so that without the remaining 44 per cent of the invoices, the calculation would have been somewhat dubious.

4.663 The same doubts arise in the case of Caesar and Marazzi for size 20 cm x 20 cm tiles, for which the firms also failed to provide any information.

4.664 Similarly, Marazzi failed to provide any information concerning the 40 cm x 40 cm. category, while 91 per cent of Bismantova�s domestic market sales in that category were made to its controlling company, Rondine. How, then, could the EC have expected the DCD to establish an individual margin of dumping in these cases?

4.665 All this shows that although the DCD had hoped, through the sample methodology, to be able to determine a margin of dumping for each exporter, this proved impossible because the information from the exporters that was available to the DCD when it issued its final determination did not enable it to do so.

4.666 In short, Argentina considers that the EC proceeds from an erroneous hypothesis when it states that the authority was in a position to make the said calculation on the basis of the elements at its disposal, since the only elements that were in fact available were those which the companies had asked to be considered as a sample, and they turned out to be totally insufficient.

4.667 Argentina submits therefore that it did not deviate from the requirements set forth in Article 6.10 of the AD Agreement.

(c) Arguments of Argentina in its oral statement at the third-party session of the first meeting of the Panel with the parties, relating to the EC�s claim under Article 6.10 of the AD Agreement

4.668 In its oral statement at the third-party session of the first meeting of the Panel with the parties, and in connection to the EC�s claim under Article 6.10 of the AD Agreement, Argentina recalled Japan�s statement that the DCD was required to determine individual dumping margins for each of the four selected exporters, since it had accepted the selection of exporters. Argentina argued that this statement failed to recognize the fact that the information in question was insufficient for such a calculation.

(d) Replies of Argentina to the first set of questions by the Panel relating to the EC�s claim under Article 6.10 of the AD Agreement

4.669 Argentina replied to the first set of questions by the Panel relating to the EC�s claim under Article 6.10 of the AD Agreement as follows:

4.670 The Panel recalled that Argentina argued that the questionnaire responses submitted by the exporters were incomplete since in many cases the exporters failed to report home sales of certain size categories. The Panel asked Argentina to clarify which exporters failed to provide what kinds of home sales? The Panel further asked Argentina to explain whether the exporters concerned made home sales in every size category during the period of investigation. If they did not, the Panel asked whether Argentina contended that the calculation of dumping margins for the product under investigation required the calculation of dumping margins for every size category devised by the DCD.

4.671 Argentina argued that the questionnaire responses were incomplete since they did not contain sufficient information, as stated in the report on the final determination of the margin of dumping. One thing that was missing was sufficiently detailed supporting documentation on domestic market sales.

4.672 To this question, Argentina provided the following reply.

4.673 To help the Panel understand what domestic market sales information according to size was presented by each exporter, Argentina refers the Panel to the Annex to the Final Determination Report.

4.674 At the same time, Argentina reiterates what was explained in paragraphs 112 and 113 of Argentina�s first written submission:

(a) 30 x 30 ceramic tiles:

Two companies in the sample (Caesar and Marazzi) failed to provide any information on average prices for this category.

In the case of Bismantova, it was found that 56 per cent of sales on the domestic market were to its controlling company, Rondine.

(b) 20 x 20 ceramic tiles:

The same two companies failed to provide any information for this category.

(c) 40 x 40 ceramic tiles:

The company Marazzi failed to provide any information for this category.

In the case of Bismantova, it was found that 91 per cent of sales reported for the domestic market were to its controlling company, Rondine.

4.675 This lack of information in itself makes it impossible to calculate the individual margin of dumping for the company Marazzi in any of the three categories and for Caesar in categories 30 x 30 and 20 x 20. In the case of Bismantova, the high percentage of sales in abnormal trading conditions made it impossible to calculate the margin of dumping for the categories 30 x 30 and 40 x 40.

4.676 The implementing authority decided to analyse the product in accordance with the adopted segmentation, 20 x 20, 30 x 30 and 40 x 40, taking account of the elements it had at its disposal. There was no objection to this. The implementing authority expected the selected sample companies to be representative, in terms of their exports, under Article 6.10 of the AD Agreement, and that they would also provide information that would enable it to make a determination of dumping on the basis of the above categories. It is impossible to establish, from the information they supplied, whether or not they in fact sold all the sizes in question on the domestic market. But based on the information in the record of the case, the implementing authority understood that the criterion which would enable it to make a proper determination of dumping was the segmentation of the product under investigation by size.

4.677 Consequently, the DCD had to calculate the margin of dumping of the product under investigation for each one of the size categories analysed.

(e) Arguments of Argentina in its second written submission relating to the EC�s claim under Article 6.10 of the AD Agreement

4.678 In its second written submission, Argentina made the following arguments relating to the EC�s claim under Article 6.10 of the AD Agreement.

4.679 The EC�s claim concerning the failure to calculate individual margins of dumping for the exporting firms is based on an erroneous analysis of the elements making up the record which inevitably leads to a mistaken conclusion concerning compliance with Article 6.10.

4.680 Firstly, the EC claims that Article 6.10 has been violated, but analysing the content of the obligation that it claims was violated (� � determine an individual margin of dumping for each known exporter or producer concerned ��), it disregards other provisions of the AD Agreement which necessarily condition the interpretation of that provision.

4.681 In other words, the EC speaks of a violation of Article 6.10 that did not occur � since the calculation of the margin of dumping could not take place on the basis of the characteristics of the sample � while at the same time disregarding the determination of the margin of dumping for the product that ultimately served as a basis for adopting the measure, which, for its part, did comply with Article 2 of the AD Agreement. Furthermore, the product was adjusted in order to arrive at a �fair� comparison in conformity with Article 2.4.

4.682 Regarding the first element, i.e. the calculation of the margin of dumping for the product, which is highly relevant to the analysis of the Article 6.10 obligation, in Argentina�s view the individual margin of dumping cannot be analysed for each firm outside the context of the calculation of the margin of dumping for the product under investigation. This calculation of the margin of dumping for the product is related to the definition of dumping in Article 2.1 which matches the definition of the product given by the Appellate Body (Appellate Body Report, European Communities � Anti-Dumping Duties on Imports of Cotton-Type Bed Linen from India, WT/DS141/AB/R, adopted 12 March 2001, at para. 51).

4.683 This finding of dumping must be made with respect to the product under investigation, i.e. the �product� that is introduced into the commerce of another country at less that its normal value on the market of the exporting country.

4.684 In broaching the subject of Article 6.10, the EC disregards Argentina�s argument that the sample proposed by Assopiastrelle being a valid methodology for calculating the margin of dumping, it could not be claimed that when the documentation provided by the sample exporting companies turned out to be insufficient, the Authority was left without any alternative for determining a margin of dumping according to the duly adopted segmentation.

4.685 In this case, the product under investigation was ceramic tiles from Italy, �in all sizes�, which includes, specifically, sizes 20 x 20, 30 x 30 and 40 x 40. This type of porcellanato makes up 99.29 per cent of imports from the investigated origin comparable to the like domestic product.

4.686 Since the Authority had thus defined the like product, the EC�s assertion in paragraph 66 of its oral statement that �the product under investigation was ceramic tiles and not each of the size categories defined by the DCD� is erroneous. It is on the basis of its own definition that the DCD calculated the margin of dumping for each one of the three categories of the product based on the adopted segmentation.

4.687 Having made a calculation of the margin of dumping which was in keeping with its definition of the like product, the authority would have been in a position to proceed with the calculation of the margin for each exporter � Article 6.10 � if it had had the necessary information from the sample that the exporters themselves proposed.

4.688 The sample provided be Assopiastrelle was claimed to be a valid methodology for calculating the margin of dumping. However, as already mentioned on many other occasions, the sample presented a number of shortcomings that made it impossible to calculate an individual margin of dumping for each firm, among which the most significant were the following:

(a) 30 x 30 ceramic tiles: Two companies in the sample (Caesar and Marazzi) failed to provide any information on average prices for this category. In the case of Bismantova, it was found that 56 per cent of sales on the domestic market were to its controlling company, Rondine.

(b) 20 x 20 ceramic tiles: The same two companies failed to provide any information for this category.

(c) 40 x 40 ceramic tiles: The company Marazzi failed to provide any information for this category either. In the case of Bismantova, it was found that 91 per cent of sales on the domestic market were to its controlling company, Rondine.

4.689 This lack of information in itself made it impossible to calculate the individual margin of dumping for the company Marazzi in any of the three categories and for Caesar in categories 30 x 30 and 20 x 20. In the case of Bismantova, the high percentage of sales in abnormal trading conditions made it impossible to calculate the margin of dumping for the categories 30 x 30 and 40 x 40.

4.690 Thus, Argentina does not think that the DCD can be blamed for not having calculated the individual margins of dumping for each exporting firm, since the firms in question did not provide the information needed for that purpose.

4.691 Similarly, Argentina would like to point out that in response to the questionnaires of 30 October 1998, the Argentine representative of Assopiastrelle sent a note dated 12 May 1999 to its principal in which it states that �regarding non-confidential invoices, I suggest to select the same invoices of each segment (20 x 20, 30 x 30, etc.) with prices closer to the weight average of the segment � �. This note shows once again that the exporters were aware of, and indeed agreed with, the segmentation in question.

(f) Arguments of Argentina in its second oral statement relating to the EC�s claim under Article 6.10 of the AD Agreement

4.692 In its second oral statement, Argentina made the following arguments relating to the EC�s claim under Article 6.10 of the AD Agreement.

4.693 As explained in paragraph 112 of Argentina�s first written submission, the information on domestic market sales by size submitted by each exporter was as follows:

(a) 30 x 30 ceramic tiles: two companies in the sample (Caesar and Marazzi) failed to provide any information on average prices. In the case of Bismantova, it was found that 56 per cent of sales on the domestic market were to its controlling company, Rondine.

(b) 20 x 20 ceramic tiles: the same two companies failed to provide any information.

(c) 40 x 40 ceramic tiles: the company Marazzi failed to provide any information for this category either, while in the case of Bismantova, it was found that 91 per cent of sales on the domestic market were to its controlling company, Rondine.

4.694 As stated in paragraph 113 of Argentina�s first written submission, this lack of information in itself made it impossible to calculate the individual margin of dumping for the company Marazzi in any of the three categories and for Caesar in categories 30 x 30 and 20 x 20. In the case of Bismantova, the high percentage of sales in abnormal trading conditions made it impossible to calculate the margin of dumping for the categories 30 x 30 and 40 x 40.

4.695 With respect to paragraph 72 of the EC�s second written submission and paragraph 66 of its second oral statement, Argentina reiterates that the product under investigation was ceramic tiles from Italy, �in all sizes� which includes, specifically, sizes 20 x 20, 30 x 30 and 40 x 40. As already mentioned, this type of porcellanato makes up 99.29 per cent of imports from the investigated origin comparable to the like domestic product.

4.696 Argentina therefore repeats that since the authority had thus defined the like product, the EC�s assertion that �the product investigation was ceramic tiles and not each of the size categories defined by the DCD� is erroneous. It is on the basis of its own definition that the DCD calculated the margin of dumping for each one of the three categories of the product based on the adopted segmentation.

4.697 In its reply a question by the Panel following the first meeting, the EC agrees with Argentina�s statement that if the implementing authority has to resort to the facts available, the margin of dumping determined will depend on those facts.

4.698 The Panel did not ask Argentina any questions following the second meeting relating to the EC�s claim under Article 6.10 of the AD Agreement.

Continuation: 3. Third Parties: Japan

Return to Contents