Important legal notice

61985C0277

European Court reports 1988 Page 05731

Opinion of the Advocate-General

My Lords,

The legal framework and the procedure

Canon Incorporated of Japan (" Canon Inc .") is a company which makes optical and electronic products. Having started as a camera company, it entered the office automation market and started producing electronic typewriters in 1982-83. In most Member States, Canon products are sold to dealers through wholly owned subsidiaries. Canon's three principal European sales companies are located in the United Kingdom (Canon (UK) Ltd, "Canon UK"), Germany (Canon Rechner Deutschland GmbH, "Canon Germany") and France (Canon France SA, "Canon France") and import directly from Japan.

A provisional anti-dumping duty of 33.3% was imposed on Canon Inc.'s electronic typewriters by the Provisional Duty Regulation. It was collected at that rate and a definitive duty of 35% was imposed by the Definitive Duty Regulation. To these regulations and to the Basic Regulation under which they were made, I refer to my Opinion in Joined Cases 260/85 and 106/86 Tokyo Electric Company v Council (" TEC ").

By an application (Case 277/85) lodged at the Court Registry on 10 September 1985 Canon France, Canon Germany and Canon UK brought an action against the Council claiming the annulment of the Definitive Duty Regulation in so far as it concerned them and asking for their costs. By an application (Case 300/85) in largely similar terms, lodged at the Court Registry on 4 October 1985, Canon Inc. brought an action against the Council claiming similar relief.

By another application (Cases 277/85 R and 300/85 R) lodged on 7 October 1985 all four companies sought the suspension of the Definitive Duty Regulation. The cases were joined and the application was dismissed by an Order of the President of the Court of 18 October 1985, which also reserved the costs of the interim proceedings ((1985)) ECR 3491.

Without making a formal objection to the admissibility of Case 277/85, the Council contended in its defence in that case that (i) it was doubtful whether importers were admissible to bring such actions, and (ii) it was unnecessary for the European subsidiaries to bring a separate action from the Japanese parent company. It accordingly asks for the costs of the application in Case 277/85 in any event, since those costs were unnecessarily and unreasonably incurred.

By an Order of 11 November 1985, the Court joined Cases 277/85 and 300/85. All four applicants will be referred to hereafter collectively as "Canon" unless otherwise indicated.

The Commission and Cetma have intervened in support of the Council.

Canon advances five sets of grounds in support of its claim for annulment. They allege: (1) failure to make a fair comparison between normal value and export price, (2) errors in calculating normal value, (3) errors in calculating export price, (4) defective injury assessment and (5) procedural irregularities.

Admissibility

First it is necessary to deal with the admissibility of Case 277/85. In the current state of the law, exporters of the product covered by anti-dumping legislation may challenge that legislation, especially where they are named in it, as is Canon Inc. Importers associated with such an exporter (as opposed to independent importers) may also challenge such legislation, in particular where - as here - the export price has been constructed from their selling prices rather than that of the exporter: see Case 118/77 ISO v Council ((1979)) ECR 1277; Case 307/81 Alusuisse v Council and Commission ((1982)) ECR 3463; and Joined Cases 239 and 275/82 Allied Corporation v Commission ((1984)) ECR 1005. Since the applicants in Case 277/85 are wholly owned subsidiaries of Canon Inc. and since their prices were used to construct export price in the contested regulation, they are in my opinion able to challenge it.

On the other hand, their action was largely duplicated by the case which Canon Inc. subsequently brought in its own name (300/85), a factor to be taken into account in relation to costs.

The substance

Ground 1: Failure to make a fair comparison between normal value and export price

Under this ground, Canon argues as follows. (1) The Commission was obliged by Article 2 (9) of the Basic Regulation to make a fair comparison between export price and normal value; the Commission consistently violated this overriding requirement by using methods of calculation which respectively inflated normal value and depressed export price. Comparing these inevitably yielded an absurdly high dumping margin, namely 76.5%. A correct appraisal of the facts, in accordance with the Basic Regulation, would have revealed no dumping. Canon maintains that its dumping margin was zero. (2) The Commission's method violated Article 2 (9) in a further respect, in that it did not make a comparison at ex-factory level and the same level of trade. (3) The Commission wrongly interpreted Article 2 (10) in considering which allowances to grant as adjustments to reach normal value. Article 2 (10) cannot override the need for a fair comparison required by Article 2 (9); moreover, the Commission's interpretation of the allowances specifically prescribed by Article 2 (10) is unjustified and inconsistent. As a result, the Commission's comparison of normal value and export price was defective.

For the reasons given in my Opinion in TEC, I reject the argument that Article 2 (9) (either alone or read together with Article 2 (10)) imposes any overriding requirement about comparison. As indicated in the "Mini ball-bearing" Cases, the calculation of normal value and the calculation of export price are separate exercises governed by different rules, the former by Article 2 (3) to (7) and the latter by Article 2 (8) of the Basic Regulation. The rules concerning comparison in Article 2 (9) and (10) come into play only after export price and normal value have been established according to their respective rules. They do not override those rules, but lay down rules concerning adjustments which may be made concerning specified matters (differences in physical characteristics of the product, differences in quantities, differences in conditions and terms of sale and differences in import charges and indirect taxes) in order to place the export price and the normal value on a comparable basis.

It follows that Article 2 (9) cannot have the effect of requiring normal value and export price to be calculated in the same way or by "symmetrical" methods, and the Court so held in the "Mini ball-bearing" Cases. In addition, the argument for "symmetrical" methods cannot be reconciled with the terms of Article 2 (3) to (7) and Article 2 (8) because they provide for different methods of calculation.

In this context Canon mentions American practice, in particular the "export sales price offset" ("ESP offset") allowed by the Department of Commerce where the export price (termed "US price" in American law) is constructed. Where it applies, the ESP offset allows general selling expenses to be deducted from normal value (termed "fair market value" in American law) up to the limit of the general selling expenses deducted from the export price. Apparently this practice has been approved by the courts in the USA and it has been published as an administrative rule in the Code of Federal Regulation (19 CFR paragraph 353.15 (c)), but it is not provided for in the American anti-dumping statute and appears to be granted by the Department of Commerce as a concession outside the terms of the anti-dumping statute. The administrative nature of the rule and the ability to change it must reduce its persuasive force in the present context.

In any event the practice cannot be taken over wholesale into the EEC system: the two systems are too different in the details of their operation and there is no warrant for doing so in the legislation. Although the third recital to the Basic Regulation states: "Whereas in applying these rules it is essential, in order to maintain the balance of rights and obligations which these Agreements ((in particular GATT and the Code)) sought to establish, that the Community take account of their interpretation by the Community's major trading partners, as reflected in legislation or established practice", it does not even impose a substantive obligation on the Community authorities to have regard to the matters mentioned. At most therefore a practice such as the American ESP offset is an element from which the Community authorities may derive guidance in applying the Basic Regulation. A reference to the American practice, even by way of guidance, does not establish that the comparison provisions of the Basic Regulation, particularly Article 2 (9), override the provisions on normal value in Article 2 (3) to (7).

Finally, reliance is placed on the word "generally" in the sentence "allowances generally will not be made for differences in overheads and general expenses", as well as on the word "guidelines" in the opening lines of Article 2 (10). It is true that Article 2 (10) (c) does not purport to lay down an absolute rule, but a guideline which will admit of exceptions. However, Recitals 24 and 25 both admit that it is only a general rule, and cannot be impugned on that account. Moreover, such exceptions have to be substantiated by the party claiming to benefit from them. The second sentence of Article 2 (10) provides: "Where an interested party claims such an allowance, it must prove that its claim is justified". It was therefore for Canon to show that the expenses in respect of which it was claiming allowances were, exceptionally, of a kind which had a direct relationship to the sales under consideration. In the present proceedings Canon has failed to demonstrate this and therefore its contentions in this respect fall to be rejected (see Case 258/84 Nippon Seiko v Council, cited in my Opinion in TEC, paragraph 45).

Accordingly, in my opinion, Canon's first ground of annulment falls to be dismissed in its entirety.

Ground 2: Errors in calculating normal value

By its second ground of annulment Canon argues as follows . ( 1 ) The Commission was obliged by Article 2 ( 3 ) ( a ) and 2 ( 3 ) ( b ) of the Basic Regulation to calculate normal value on the basis either of the domestic prices charged by Canon Inc . or, if it rejected those prices by reference to Article 2 ( 7 ), on the basis of cost of production at ex-factory level . The Commission, in proceeding to base normal value on the prices charged by Canon Sales, failed to comply with this obligation . ( 2 ) Construction of normal value on the basis of cost of production was in any event appropriate in the light of the small number of sales of Canon electronic typewriters on the Japanese market, and in light of that market' s special features . The domestic sales of the two models for which normal value was based on actual domestic price were equal to only 1.4% of the total volume of Canon' s exports to the Community . ( 3 ) The methods followed by the Commission in calculating normal value were unfair, and were contrary to Article 2 ( 3 ), ( 9 ) and ( 10 ) of the Basic Regulation, in particular for the following reasons : ( a ) As to the two models where normal value was based on domestic price, the Commission refused to allow appropriate deductions from that price in order to yield an ex-factory price . The so-called normal value thus achieved was therefore artificially high . ( b ) As to the four models where normal value was constructed, the Commission wrongly doubled the actual costs of manufacture and overheads of Canon Inc . by adding on additional factors . ( c ) As to the said four models, the profit margin of 47% on costs ( 32.39% on turnover ) added on top of the actual costs of production and the assumed additional costs was grossly excessive . ( d ) As to all six models, the Commission' s method of calculating normal value yielded a figure which was not equivalent to an ex-factory price .

The Community authorities were plainly entitled to treat Canon Sales as being "associated" with Canon Inc . for the purposes of Article 2 ( 7 ) of the Basic Regulation and therefore had a discretion to disregard Canon Inc .' s prices to Canon Sales for the purpose of determining normal value . Canon has failed to show that that discretion was exceeded or improperly used . Therefore its argument that normal value should be based on Canon Inc .' s prices to Canon Sales must be rejected .

Article 2 ( 3 ) ( a ) of the Basic Regulation provides that normal value shall be "the comparable price actually paid or payable in the ordinary course of trade for the like product intended for consumption in the exporting country or country of origin ". This provision in my opinion includes in its scope the price charged by a producer' s sales company to its customers . The prices charged by Canon Sales to its customers are plainly "in the ordinary course of trade" within the meaning of this provision, and the Community authorities were entitled to use them for the purpose of establishing normal value as they did ( Recital 8 to the Definitive Duty Regulation ). Canon' s argument that, if Canon Inc .' s prices are not used, normal value must either be constructed or based on export price to third countries under Article 2 ( 3 ) ( b ), must, I consider, be rejected .

The 5% threshold adopted in Recital 4 to the Definitive Duty Regulation as regards the volume of sales needed for use as a basis for normal value is not merely lawful but it enhances legal certainty .

Whereas this threshold is 5% by volume "of exports to the Community", Canon states that the US practice is to disregard domestic sales if less than 5% of exports are to countries other than the USA . The weight to be given to US practice is limited for the reasons previously outlined . Within those limitations, on the one hand, it is to be observed that the American practice confirms the figure of 5% for the threshold and, in any event, I do not accept that exports to countries other than the one concerned by the alleged dumping are necessarily the appropriate ones to take . As at present advised, I think that exports to the country ( or economic community ) concerned by the dumping constitute a more appropriate standard of comparison . I therefore do not find any grounds for contesting the 5% threshold in Recital 4 .

Two of Canon' s models are said to have had sales in Japan exceeding the threshold during the reference period : the AP 400 and AP 500 . Canon says that domestic sales of those two models amounted to only 1.4% of the total volume of its exports of electronic typewriters to the Community . That figure is irrelevant because it compares sales of only those two models in Japan with sales of six models in the EEC . The percentage has to be worked out model by model . The Council has stated that domestic sales of model AP 400 represented 7.6% in volume of exports to the Community, while the figure for the model AP 500 was 8.7 %. In answer to questions by the Court, Canon has supplied confidential sales figures which seem to me substantially to confirm those percentages . Canon' s argument must therefore be rejected .

Whilst it is true that alpha-numeric electronic typewriters are in an unusual situation in Japan because the language is not written in Roman characters, that can provide no reason - contrary to what Canon alleges - to depart from a normal application of Article 2 ( 3 ) ( a ) of the Basic Regulation . If the domestic sales reach a sufficient volume I do not see how such circumstances can alter the conditions laid down in the legislation under which actual domestic price can be used as a basis for normal value .

Canon argues that the price of exports to the US market should have been used to establish normal value . That argument cannot apply to the two models for which normal value was - in my view properly - based on actual domestic price . As regards the remaining models, I read Article 2 ( 3 ) ( b ) as giving the Community authorities a discretion to choose between either constructing normal value or using third-country export prices . Canon has failed to show that that discretion was wrongly used, whereas the Council has explained the reasons for the Community authorities' choice in the second paragraph of Recital 4 to the Definitive Duty Regulation . Canon' s argument on this point therefore falls to be rejected .

Regarding Canon' s third set of arguments, certain of these arguments confuse points concerning allowances for purposes of comparison with points concerning the calculation of normal value . As already said, they fall to be dealt with separately, and those concerning comparison have in substance been dealt with above and in my Opinion in TEC .

As to the two models for which normal value was based on actual domestic price, Canon says : "(( The Commission )) refused to take the Canon Inc . to Canon Sales price as its point of departure on the domestic side . It insisted on going further down the commercial chain to Canon Sales' prices to customers . These prices were much higher due to the special nature of the market and the high costs of selling electronic typewriters ". This statement seems to me to confirm that the Community authorities were entitled to use the prices of Canon Sales because it indicates that the price from Canon Inc . to Canon Sales was far below the prices actually prevailing on the market and was therefore more like a transfer price than a price in the ordinary course of trade . I have also already indicated why the "special nature of the market" is of no assistance to Canon . Finally as to the causes of the market prices being higher, Cetma alleges inter alia that profit margins on the protected and partly cartelized Japanese market are well known to be high . I do not think it necessary to express a concluded view on the causes . What is important here is that Canon admits that the Japanese market prices of the products in question were high . There is no suggestion that the finding made in that respect was mistaken . Far from being "a recital of unfairness" as alleged, the procedure up to that point was lawful and proper .

Then mention is made of the deduction of Canon Sales' selling costs, in particular the costs borne by Canon Sales in advertising typewriters in Japan . This submission is difficult to understand . Article 2 ( 3 ) ( a ) on normal value makes no provision for any deductions . It provides that the normal value shall be "the comparable price actually paid or payable in the ordinary course of trade ". The price "in the ordinary course of trade" is the first arm' s length sale, and since export price was equally taken from the first arm' s length sale it is in that respect "comparable ". I do not think it possible to read any requirement for further deductions into the word "comparable" here . Under Article 2 ( 3 ) ( a ) the price is to be taken as found . Any further adjustments or deductions can only be such as come within the terms of Article 2 ( 9 ) and ( 10 ). Those provisions go to comparison, not the establishment of normal value . There is no way in which the selling costs of Canon Sales could be deducted under Article 2 ( 3 ) ( a ). The treatment of selling costs by the Community authorities cannot invalidate their finding of normal value based on actual domestic prices for the two models concerned .

Even when this complaint is examined under the provisions to which it properly relates, it fails . Canon says : "The Commission refused to deduct all of Canon Sales' costs, despite the fact that Canon Sales' only function was selling ". The Community authorities, however, were under no obligation - as implied by Canon - to deduct all of the selling costs of Canon Sales . In my Opinion in TEC I have explained why, in my view, the exporters cannot claim such deductions simply because in the same corporate group they happen to have split up the manufacturing and the selling functions between formally separate bodies .

Next, Canon complains that the Commission "refused to deduct a reasonable distributor' s profit margin to parallel the deductions made from the prices charged by Canon' s European subsidiaries ". As I see it, this deduction cannot come within the terms of Article 2 ( 3 ) ( a ), or, indeed, Article 2 ( 3 ) ( b ) ( i ). A "reasonable margin of profit" is mentioned in relation to normal value only in Article 2 ( 3 ) ( b ) ( ii ), concerning constructed normal value . By definition that provision does not apply to the case - which is under consideration here - where normal value is based on actual domestic price . Again, the argument is irrelevant and cannot invalidate the Community authorities' finding of normal value on the basis of actual domestic price for the two models concerned . Secondly, even in relation to constructed value, the argument that parallel methods have to be applied in constructing normal value and in constructing export price is unfounded for the reasons given above and in my Opinion in TEC .

Canon has therefore failed to demonstrate any defect in the way the Community authorities established the normal value of the two models for which it was based on actual domestic price .

As regards the other four models, I reject for the reasons I gave in my Opinion in TEC the argument that the Community authorities were not entitled under the Basic Regulation to construct a "surrogate" normal value using elements of actual domestic price for guidance .

The argument that the procedure followed did not achieve a fair comparison at the ex-factory level goes to comparison not to the construction of normal value and, even when considered in relation to comparison, falls to be rejected for the reasons given above and in my Opinion in TEC .

The argument that the expenses included in constructed normal value were unlawful because they were not treated symmetrically to those on the export price side falls to be rejected for the reasons given above .

Canon insists in particular on the Commission' s refusal to deduct advertising expenses borne by Canon Sales . However, advertising costs are relevant not to establishing normal value but to achieving price comparability under Article 2 ( 10 ), and Article 2 ( 10 ) ( c ) provides that allowances for differences in overheads and general expenses, such as advertising, will not generally be made . Canon has failed to establish that this general rule should not apply to the present point . Therefore its argument falls to be rejected not only in relation to normal value but also in relation to comparison .

Canon also attempts to bring its alleged advertising costs into a discussion of the profit margin used for the purpose of constructing normal value . Canon argues as follows : "The profit margin of 47% was obtained by subtracting all costs of Canon Sales and Canon Inc . from Canon Sales' price to dealers . However, in the case of Canon Sales, the Commission failed to take into account the actual costs borne by Canon Sales in selling electronic typewriters . The Commission chose to calculate Canon Sales' profit on electronic typewriters as if it had not incurred ( to give one example ) especially heavy advertising costs . This made an enormous difference . Advertising amounted to 7% of overall turnover of Canon Sales . However, for electronic typewriters taken alone, actual provable advertising costs amounted to 26% of turnover . Thus, using the wrong figure had the effect of substantially overstating the profit ". To work out the profit on actual domestic sales, the only way of proceeding is to deduct the cost from the selling price . That is not challenged . Canon' s case is that the amount of the cost deducted should have been larger . The point is developed in Canon' s reply as follows : "(( The Commission )) disregarded the actual costs incurred by Canon Sales in advertising and promoting electronic typewriters, basing its conclusions on all the products handled by Canon Sales, regardless of the proven and heavy costs of marketing electronic typewriters . In fact, as the evidence given to the Commission showed, the costs of advertising electronic typewriters were nearly three times the costs of advertising all products when stated as a percentage of turnover . Had the Commission accepted the self-evident fact that these costs reduced the profits of Canon Sales on sales of electronic typewriters, the profit margin attributed to Canon ( and to other exporters who had no domestic sales ) would have been considerably lower .

I deal with this point in particular detail because, as indicated in this quotation, the 47% profit margin derived from the Canon products was used in every case ( except Brother and Silver Seiko ) where normal value was constructed in the present proceeding . Therefore the point is of importance to all the other exporters ( except Brother and Silver Seiko ) for whose products normal value was constructed, not only to Canon .

Although advertising costs are cited as "one example", the costs mentioned by Canon are in fact confined to advertising and promotion . No other costs have been urged upon the Court . Therefore, in my opinion, Canon has not even raised an issue regarding costs other than advertising and promotion costs .

As regards advertising and promotion costs, Canon has submitted no evidence to substantiate its allegations . On the other hand, the Council in its rejoinder states as follows : "The advertising costs as submitted by Canon included 198 separate cost items . When these items were scrutinized it was found that : only 59 of the 198 different cost items were accompanied by documentation; for an overwhelming number of the 59 items it was not clear whether the documentation ( simply photocopies of what appear to be advertisements ) did in reality relate to the items listed by Canon; 30 of the 59 items referred exclusively to those electronic typewriters which were not the subject of the investigation and which were excluded from the duty ( Article 1 ( 3 ) of Regulation No 1698/85 ) and for which the profitability had not to be established; several of the items exclusively referred to products other than electronic typewriters, namely so-called DWs, AP 89 ( floppy disk units ) and screens; several of the items referred to both electronic typewriters as investigated and to other products such as screens; the sales material ( price lists, technical brochures ), to which apparently some of Canon' s other items refer, covers not only the electronic typewriters under investigation but also a floppy disk unit, screens, small typewriters of the kind not included in the investigation, accessories and other products; none of the 198 cost items was accompanied by invoices which would have revealed the amount of costs actually paid to outside parties . It is furthermore not clear whether any of the alleged costs resulted from one part of the Canon group charging another part of the Canon group . The vouchers added by Canon appeared to be merely internal manuscript records, and did not provide any corroborative evidence ". It follows that Canon fails to substantiate its allegation that the Community authorities did not take due account of advertising and promotion costs in calculating the 47% profit margin . The allegation that the profit margin was wrongly calculated therefore falls to be rejected .

An additional reason for rejecting it - but a reason which applies only to Canon and not to the other exporters - is that Canon apparently accepted the same cost figures ( including advertising costs ) when they were used for the SGA expenses element of constructed normal value . Lower SGA expenses result in a lower constructed normal value, which is favourable to Canon . However, Canon can hardly be heard to contest figures in one part of a calculation which is unfavourable to it if it accepts them in another part of the calculation which it considers favourable .

Further in relation to the element for SGA expenses included in constructed normal value, Canon submits that the Community authorities should have had regard to overhead expenses relating to export sales not overhead expenses relating to the Japanese market, and it seeks to assert that Commission Regulation No 3453/81 imposing a provisional anti-dumping duty on imports of certain cotton yarns originating in Turkey ( Official Journal 1981, L 347, p . 19 ) lays down a rule to that effect . In my view it does not, if only because it was superseded by Council Regulation No 789/82 imposing a definitive anti-dumping duty on the same products ( Official Journal 1982, L 90, p . 1 ); but even giving due weight to both of these regulations it is clear that they concern a special case where the Community authorities had so little information about sales on the domestic market that they had to fall back on data concerning export sales to provide them with guidance in calculating the normal value . They do not lay down any such rule as Canon contends for, and where - as in this case - ample information is available about overhead costs on the domestic market it is impossible to accept a contention that the Community authorities were obliged to ignore it and look instead at overhead costs on exports .

Finally, as regards the "reasonable margin of profit" to be included in constructed normal value, Article 2 ( 3 ) ( b ) ( ii ) of the Basic Regulation specifies : "As a general rule, and provided that a profit is normally realized on sales of products of the same general category on the domestic market of the country of origin, the addition for profit shall not exceed such normal profit ". Canon submits that "it would therefore have been more appropriate for the Commission to examine profit margins in the office equipment sector in Japan as a whole ". First, this submission is not supported by the text, because the provision does not state that the profit margin must be that in the same general category of goods but only that it must not exceed that in the said category . Secondly, as I said in my Opinion in TEC, "the same general category" of goods must, on a reasonable reading and in the light of the concept of "like product" in Article 2 ( 2 ) and ( 12 ), be taken to mean electronic typewriters, and I would reject any wider category of office equipment for this purpose . Therefore this submission must be rejected and the various profit figures put forward by Canon in connection with it must also be disregarded .

What is required by Article 2 ( 3 ) ( b ) ( ii ) is that the profit margin should be "reasonable" and, as I said in my Opinion in TEC, it is reasonable to use profit margins actually obtained on the domestic market . According to Recital 16 to the Definitive Duty Regulation, it was considered reasonable to compute the constructed value for the four Canon models concerned using Canon' s profit margin on its sales of the other two models on the domestic market . In my opinion, that was a sound guide to the profit normally realized, and that approach was in conformity with the Basic Regulation . Canon alleges that "its management accounts showed a profit on electronic typewriters of 7.2% during the period ". These accounts have not been produced to the Court . It is not clear what models they covered or whether they include export sales . No indication is given as to how the figure was worked out, and whether the calculation used elements which might be irrelevant for present purposes . The words "management accounts", although vague, suggest that the accounts are only those of Canon Inc . not of the Canon group as a whole, which for the reasons already given do not provide an appropriate basis . In particular if the 7.2% figure is based on a transfer price between Canon Inc . and Canon Sales and if it represents only the profit of Canon Inc . and not also that of Canon Sales, it cannot be relied on for the construction of normal value . In the absence of evidence submitted to the Court, it is a bare assertion of doubtful value and provides no reason to overturn the profit figure used by the Community authorities .

Accordingly, both as regards the two models for which normal value was based on actual domestic price and as regards the models for which normal value was constructed, Canon' s second ground of annulment alleging errors in the calculation of normal value falls to be rejected in its entirety .

Ground 3 : Errors in calculating export price

By its third ground of annulment Canon alleges that the methods used by the Commission in calculating export price were unfair and inaccurate and were therefore contrary to Articles 2 ( 8 ) and 2 ( 9 ): the Commission arrived at the export price by making deductions corresponding to the overhead expenses of Canon Inc .' s European subsidiaries, as well as deducting an assumed profit; the Commission should have calculated export price in parallel to its calculation of normal value .

Since Canon France, Canon Germany and Canon UK were wholly owned subsidiaries there was plainly an association between them and Canon Inc . within the meaning of Article 2 ( 8 ) ( b ) of the Basic Regulation, and in my view the Community authorities were therefore entitled pursuant to that provision to construct the export price "on the basis of the price at which the imported product is first resold to an independent buyer ". I reject the contention that the Community authorities were required to base export price on the price between Canon Inc . and its European subsidiaries .

Where, as here, export price is based on the first arm' s length sale, Article 2 ( 8 ) ( b ) stipulates that "allowances shall be made for all costs incurred between importation and resale, including all duties and taxes, and for a reasonable margin of profit ". Canon does not challenge that rule but alleges a number of errors in the way it was applied . First, it says that it was erroneous to take 5% as a reasonable profit margin and not the 3% which Canon proposed . Canon advances no evidence or argument to substantiate that submission . The Council on the other hand explains that the 5% was derived from independent importers' profit margins and it was appropriate to use them as being the most objective basis available for the purpose of arriving at a satisfactory estimate of the arm' s length export price . That approach appears to be within the discretion of the Community authorities in establishing a "reasonable margin of profit ". Since Canon advances no reason to rebut it, Canon' s allegation of error must be dismissed .

Secondly, Canon alleges error by the Commission in "refusing to give credit for interest income received by Canon subsidiaries and in its treatment of the so-called customer cash reduction; and other issues ". These bare allegations are not explained in any further detail or substantiated by evidence . No error on the part of the Community authorities in this connection, in my view, has been demonstrated .

Thirdly, Canon alleges that the advertising costs incurred in connection with the launch of certain models in the United Kingdom, Germany and France should have been allocated over a wider number of models and over a wider geographical area; and should have been amortized over a longer period of time than the particular 12-month period in which they happened to be incurred .

Under Article 2 ( 11 ) of the Basic Regulation cost calculations are in general based on available accounting data, "normally allocated, where necessary, in proportion to the turnover for each product and market under consideration ". The advertising related specifically to three particular models . Although, of course, the advertising might have had the effect of raising public awareness of electronic typewriters generally, that is not a sufficient argument to justify a departure from the general rule, and in my view the Community authorities were entitled to allocate the advertising costs to the models concerned . Although Canon claims to have been developing "a market in Europe", the three subsidiaries had each been given the exclusive distribution rights in their respective Member States, which means that the "market under consideration" was in fact a national market in each case . The Community authorities were therefore not mistaken in allocating the advertising costs to the models, areas and periods of time to which they related, and Canon' s submission on those costs fails .

Finally, Canon' s submission that export price should have been calculated in parallel to normal value is merely repetitive and falls to be rejected for the reasons which I have given above and in my Opinion in TEC .

Accordingly, Canon' s third ground of annulment, concerning export price, in my opinion falls to be rejected .

Ground 4 : Defective injury assessment

Although it was found that Canon had a dumping margin of 76.50%, the exact level of injury attributed to it was only 35.03 %. Since, under Article 13 ( 3 ) of the Basic Regulation, the anti-dumping duty may not exceed the lower of the two figures, it is the injury figure which determined the rate of duty imposed on Canon . Rounded down to the nearest whole number, the rate imposed was 35 %.

Canon' s fourth ground of annulment is that the injury finding in the Definitive Duty Regulation was defective, and Canon advances some 10 arguments in support of that ground . A number of those arguments rest on an expert report compiled for Canon by Dr Jackson . Dr Jackson' s evidence attempts to minimize the European manufacturers' loss of market share at the relevant period and to explain away their diminished profitability by reference to the "life cycle" of products, whilst on the other hand emphasizing difficulties encountered by Triumph-Adler and Olympia in changing over from the production of mechanical and electromechanical typewriters to that of electronic typewriters .

The Council, with the assistance of expert evidence from Mr Reis, challenges the validity of that report in a number of fundamental respects . Thus in particular, it is said that Dr Jackson seeks to deduce the movement of prices, profits and other phenomena from the product life cycle concept, but within a current cycle he can only give a subjective assessment, as reliable judgments can be made only after the event . Moreover, he looks exclusively at the life cycle of the class of products, i.e . electronic typewriters as a whole, whereas for precision a distinction has to be made between the life cycle of a class of products, subdivisions of the class ( in so far as they can be identified ) and individual models or makes . Again, contrary to what Dr Jackson suggests, if a market share grows slowly on a rapidly expanding market, the growth in market share does not exclude dumping : without the dumping it might have grown more quickly . Dr Jackson talks about "revenue" ( i.e . turnover ) or prices, when what counts is profit . Sources for his graphs are not cited ( although some sources are cited in the Reply ). The graphs do not purport to cover all Member States of the EEC . Dr Jackson talks about "company performance", but that is irrelevant to the point at issue, i.e ., whether there was dumping in the product concerned . The fact that a company suffers other - perhaps even more serious - problems does not exclude its also suffering from dumping, neither is that fact relevant to proof of dumping . By superimposing different plots on the same graph - especially Graph F - Dr Jackson masks the fact that even on his figures Canon and the other Japanese producers' turnover in electronic typewriters at all material times continued to grow, whereas that of the three European producers fell in 1983 and 1984 . Dr Jackson' s own Charts D, F, G, H, I and J show depressed sales and profitability for the European manufacturers explicable only by price undercutting and not by Dr Jackson' s "product life cycle" theory . In this respect the profitability graphs submitted by the Council at the interim measures hearing show an even worse drop in profitability on the basis of the European manufacturers' own confidential figures . Some of Dr Jackson' s figures are admitted to be estimates, and all of them are from "outside" sources . The Community authorities have confidential information from the Community producers themselves which Dr Jackson could not have . Therefore it is not surprising that he is simply wrong on a number of points, e.g . Triumph-Adler' s market share ( Chart K ), the proportion of OEM machines sold by the three European companies ( never more than 11%, not "about half" as alleged ) and the decline in profitability ( 63.4% over two years, much more than he alleges ) for the Community producers .

Mr Reis considers that the European manufacturers were not overtaken by the new technology of electronic typewriters but on the contrary pioneered those products and Canon does not dispute that electronic typewriters were first introduced in Europe by Olivetti, in 1978 . Mr Reis states that the first electronic typewriters marketed in the Federal Republic of Germany were marketed in 1979 by Olivetti, Triumph-Adler and Olympia; the first Japanese producer to market them there was Brother in 1981; and Canon did not do so until 1982 . Since it was European manufacturers of electronic typewriters who pioneered and developed the market, the Japanese suppliers who subsequently came on the scene benefited from their investment without having incurred the associated costs . Mr Reis ascribes the success of the Japanese companies in penetrating the European market and building up their market share inter alia to strong support from financially powerful parent companies in Japan, deliberate exploitation of the European industry' s investment in developing a market for a new class of product, technically sound, robust, though not superior, products and aggressive price tactics .

In this highly technical and complex area it is not surprising that experts disagree, but overall I find Mr Reis' s reply to Dr Jackson' s report persuasive .

The first argument advanced by Canon in relation to the issue of injury is that the injury findings are one-sided and that the Commission failed to make a complete and impartial review of the market as a whole . In my opinion, however, Article 4 of the Basic Regulation does not require the Community authorities necessarily to undertake a comprehensive analysis of the situation of the Community industry; it requires them to establish whether the dumped imports are causing injury, which they did in this case .

In this connection Canon mentions the Community authorities' treatment of OEM purchases by the Community manufacturers . Such OEM machines were counted by the Community authorities as Japanese exports, which was proper and is not challenged by Canon . Furthermore, Canon agrees that, as stated in Recital 32 to the Provisional Duty Regulation, the Community producers' share of the Community market fell from about 63% in 1982 to about 51% in 1983-84, specifically referring to machines manufactured in the Community . It would not have been correct to treat OEM machines in any other way, and Canon' s complaint remains only that the Community authorities "did not acknowledge" that the market share they found for Japanese-made machines included a percentage of OEM machines . That complaint, in my view, is unfounded . The Community authorities at no time denied that a proportion of the imports were on an OEM basis; they treated them properly in substance ( which Canon does not deny ). It appears that their approach was well known to Canon ( which does not complain of having been misled ). Not every detail can be stated in the recitals to a regulation, and it seems to me that a matter of secondary importance, such as this one, need not necessarily be spelled out in detail in the recitals .

Canon also attempts to suggest that "capacity problems" were in part responsible for holding back growth in the sales of Community producers . That assertion is not made out . The Council has convincingly shown that the Community producers were not operating at full capacity at any time between 1980 and the end of 1983 .

Canon' s second argument is that the Commission relied on misleading or irrelevant factors, ignoring other factors suggesting there was no injury caused by the alleged dumping . It avers that the finding of injury was based almost exclusively on price, on the market share of the imports, and on financial factors which are valueless since the Commission seems to have made no attempt to distinguish the losses caused by structural problems of Community manufacturers from the adverse impact of Japanese competition; that no credit was given for increased production, increased sales, improved turnover, capacity utilization or lower stocks, and that, contrary to Article 3 ( 4 ) of the Code, the Commission uncritically attributed to Japanese imports the adverse consequences for European industry of developments in technology and the inadequate productivity of the domestic industry which at the same time were injuring the industry . It was a technological revolution which almost overwhelmed Olympia and Triumph-Adler, coupled with their own poor productivity because of the change in technology .

Article 3 of the Code makes provision for the determination of injury . Article 4 of the Basic Regulation purports to transpose it into Community law . It is not suggested by Canon ( or any other applicant in the present cases ) that Article 4 of the Basic Regulation is in any respect contrary to Article 3 of the Code . The legality of the contested injury finding falls to be decided by reference to Article 4 of the Basic Regulation . Reference to the Code might become necessary if a need were shown to clarify some point in the Basic Regulation, but none such has been shown in relation to the present issue . Therefore it is not appropriate to resolve the present issue by reference to Article 3 of the Code, as Canon implies, but only by reference to Article 4 of the Basic Regulation .

Article 4 ( 1 ) of the Basic Regulation is particularly important with respect to the question of causation . It provides that a determination of injury shall be made only if the dumped imports are causing ( or threatening to cause ) material injury to the Community industry "through the effects of dumping ". It specifies that injuries caused by other factors which also adversely affect the Community industry "must not be attributed to the dumped imports ".

There clearly have been dumped imports . In my view Canon has failed to show any defect in the finding that it was dumping its products in the Community at a dumping margin of 76 %. Under Article 4 ( 1 ) the question is then whether those dumped imports caused any injury to the Community industry "through the effects of dumping" and if so, how much injury .

The evidence is strongly against Canon' s assertion that Olympia and Triumph-Adler were damaged by a technological revolution rather than by dumped Japanese imports . Dr Jackson' s assertions on this point have, in my view, been disproved by other evidence . The European manufacturers were not surprised by the new technology : they invented it . Olivetti, as well as Olympia and Triumph-Adler, initiated the change in technology alluded to; the Japanese manufacturers followed .

The evidence is also against Canon' s assertion that the Community industry was damaged by its own poor productivity rather than by dumped Japanese imports . The evidence shows that, despite being first in the market, the Community manufacturers were prevented from building up productivity by large-scale Japanese dumping .

As regards what Canon describes as the "structural problems" of the Community industry, the evidence suggests that Olympia and Triumph-Adler changed over from the production of mechanical and electromechanical to the production of electronic typewriters less quickly than Olivetti and invested large amounts of money in the process . It does not, however, support the innuendo by Canon that their fall in market share and in profitability were due to their own conduct .

It is wrong to assert that the Commission made no attempt to distinguish the effects of dumped Japanese imports from any effects of what Canon calls the "structural problems" of the Community industry . The injury was measured particularly by reference to price undercutting, which is normally due to outside competition rather than a manufacturer' s own internal problems . The method of measurement used was, contrary to Canon' s suggestions, well suited to evaluate the injury caused by dumping and to exclude any injury which might be due to other causes, particularly those relating to Community manufacturers' internal arrangements . The wording of the injury finding in the Definitive Duty Regulation clearly shows a separation of injury caused by the dumped imports from injury due to other factors . Thus, Recital 38 states that, "the facts as finally determined show that the injury being caused by dumped imports of electronic typewriters originating in Japan, taken in isolation from that caused by other factors, has to be considered as material . No other factors, such as volume and prices of other imports which were not dumped, or contraction in demand, were found to have been contributory to the injury established ". The words "the injury established" in this recital do not mean "all the problems experienced by Community industry" as Canon implies but plainly refer back to the preceding recitals and mean "the injury described and analysed above", i.e . the injury resulting from dumped Japanese imports . The recital plainly does not deny the existence of "structural problems", as Canon alleges, but does demonstrate that the Community authorities identified the injury specifically caused by dumped Japanese imports and established that no other factors contributed to that particular injury . That approach is entirely in accordance with Article 4 ( 1 ) of the Basic Regulation .

Article 4 ( 2 ) of the Basic Regulation is important as laying down the factors which have to be taken into account in carrying out an examination of injury . It is noteworthy that it provides that no one or several factors can necessarily give decisive guidance . Those factors are :

( a ) volume of dumped ... imports, in particular whether there has been a significant increase, either in absolute terms or relative to production or consumption in the Community;

( b ) the prices of dumped ... imports, in particular whether there has been a significant price undercutting as compared with the price of a like product in the Community;

( c ) the consequent impact on the industry concerned as indicated by actual or potential trends in the relevant economic factors such as :

-production,

-utilization of capacity,

-stocks,

-sales,

-market share,

-prices ( i.e . depression of prices or prevention of price increases which otherwise would have occurred ),

-profits,

-return on investment,

-cash flow,

-employment .

It is unjustified to criticize the Community authorities for having regard to price and market share as these are specifically listed in Article 4 ( 2 ) as relevant factors . It is wrong to describe them as "misleading or irrelevant factors ": they are facts . There could in some cases be error or misinterpretation of the facts, but that is not alleged here . It is not shown how market share or price could be "misleading" in themselves, and by definition they are not irrelevant .

Recitals 30 to 38 of the Definitive Duty Regulation and Recitals 30 to 33 of the Provisional Duty Regulation ( confirmed in the Definitive Duty Regulation by Recital 32 of the latter ) clearly show that the Community authorities carefully examined the three matters which they were required to under Article 4 ( 2 ) ( a ), ( b ) and ( c ). As regards the factors mentioned in Article 4 ( 2 ) ( c ), the Community authorities were in my opinion under no duty to work through every single one, since they are only cited as examples ( see the words "such as ") of economic factors which might indicate the "consequent impact on the industry concerned ". Although utilization of capacity and stocks do not appear to be expressly mentioned in the recitals of either regulation, I consider that it was within the discretion of the Community authorities to omit mention of them, if they considered that other factors gave them sufficient guidance . As regards increased production, increased sales and increased turnover, Canon' s allegation is simply wrong, because they were dealt with in the recitals . In so far as Canon intends to argue that they were incorrectly dealt with, increase of sales ( or turnover ) has to be set against the fact that the sales did not increase as fast as demand on a market which was expanding rapidly . Furthermore a decrease in profitability due to price undercutting must be given due weight as against rising sales . In my view, the Community authorities were entitled to make the assessment they made in both respects and Canon' s argument falls to be rejected .

Canon' s third argument merely repeats the last of these points, i.e . that the sales of the Community manufacturers grew during the relevant period, and it falls to be rejected for the same reasons .

Canon' s fourth argument is that reduction in profitability is a normal market phenomenon, so that it was wrong to assume that a fall in profitability reflected injury caused by dumped Japanese imports .

This submission rests on a general theory as to the life cycle of products when at best what would be relevant for present purposes is the life cycle of specific models . Secondly, it seems far more likely that profits for electronic products are greatest not during the early stages, as alleged, but that such products are most profitable when they are well established and the initial developing and marketing costs have been absorbed . I do not accept the assertion that a reduction in profitability was a normal market phenomenon for the electronic typewriters in question during the period in question .

I would also reject the suggestion that the Community authorities "assumed" that a fall in profitability was caused by Japanese dumping . On the contrary, the Community authorities have set out in the regulations substantial reasons for their conclusion that "the impact of the low-priced imports has been to decrease significantly the profitability of Community producers" ( Recital 31 of the Provisional Duty Regulation ).

It is important to point out the scale of the reduction in profitability caused by the dumping . According to Recital 31 of the Definitive Duty Regation : "It was established that if the profitability of the total Community industry in 1982, i.e . the year when large-scale imports of Japanese typewriters were starting, was 100 ( index ) the profitability ( expressed as a percentage of turnover before tax ) has declined to 36.6 ( index ) during the reference period . ... Even allowing for seasonal distortions, the quarterly consolidated profitability of sales in the Community during the period of investigation has been regularly below a level at which the existence of this industry could be guaranteed ".

Thus it was found that the profits of the Community industry had been driven down so low that it was facing extinction . It is, I think, impossible to explain that away as a normal market phenomenon . Moreover, the evidence of both sides concurs in asserting the vital necessity for a company which has launched a technically innovative product ( such as an electronic typewriter ) to recoup quickly the usually considerable costs it has incurred on research, development and marketing the new product and to get in more profit to finance the next cycle of technical research . It was at this sensitive stage that the Community producers were deprived by dumping of the profits which they vitally needed . Thus if the evidence about product life cycles has any validity, it is in my view to demonstrate that the decline in profitability caused by dumping in this case was particularly damaging .

Canon' s fifth argument is that it is obscure what evidence the Commission had as to price competition . That is not borne out by the text of the two regulations concerned . According to Recital 31 of the Provisional Duty Regulation : "The resale prices of the dumped imports generally undercut the prices of the Community producers during the period under investigation by varying degrees depending on models and markets . Although there were instances of no undercutting, the latter generally ranged between 11.4 and 30% and in some cases reached 48.5 %". That finding was confirmed in the Definitive Duty Regulation ( Recital 32 ), but at that stage the Community authorities considered it unnecessary to examine the price undercutting in any further detail . The reason for that decision was stated in Recital 33 in the following terms : "It was not considered necessary to undertake a detailed examination of price undercutting by Japanese imports since the prices realized by the Community producers had been depressed by the prices of the Japanese products ". Both the decision and the reasons for it are in my opinion sound . It would indeed have been futile for the Community authorities to examine undercutting of actual prices in any further detail because the prices of Community producers had been depressed by large-scale dumping of Japanese products sustained over a considerable period and the comparison could not have yielded a meaningful result . ( The Community authorities instead constructed the prices within the Community as they would have been if not depressed by dumped imports, and as I said in my Opinion in TEC that was a valid alternative means of comparison ). These full and justified statements in the two regulations in my view demonstrate that Canon' s fifth argument is not made out .

Canon' s sixth argument is that the situation of efficient and inefficient Community producers was incorrectly treated . Canon argues that it should not be blamed for causing injury to two companies ( namely Triumph-Adler and Olympia ) "whose troubles were due to other long-standing factors ". As I have stated above in relation to the second argument on injury, the Council in the regulations did not deal with "the troubles" of the Community manufacturers but ascertained such injury, and only such injury, as was caused to the Community industry by the dumped Japanese imports and imposed only such duty as was adequate to remove that injury . The argument falls to be dismissed for that reason .

Moreover, Canon' s argument about the troubles of the "inefficient" Community producers arising from factors other than dumping is one of causation, but Article 4 of the Basic Regulation does not require that dumping should be the only or even the principal cause of injury . Indeed, one of the major features which distinguishes the present code from its predecessor, the First Anti-Dumping Code ( of 1967 ), is its abandonment of the criterion that the dumping must be "demonstrably the principal cause" of the injury in favour of the test that it is merely "causing injury ". Article 3 ( 4 ) of the present code expressly provides that "there may be other factors" such as technological developments and the productivity of the domestic industry, but all that it requires in that regard is that "the injuries caused by other factors must not be attributed to the dumped imports ". That provision is faithfully reflected in Article 4 ( 1 ) of the Basic Regulation . It follows that in the law as it now stands it is not open to an exporter to plead inefficiency such as that alleged by Canon as a break in a single line of causation which would definitively exclude the exporter' s liability .

Canon' s seventh argument is that the Commission was wrong to base its injury findings on both efficient and inefficient producers . It is clear, however, that Article 4 of the Basic Regulation requires the Community authorities to take the Community industry as they find it for the purpose of determining injury .

It appears by this argument that Canon is seeking to challenge the position taken by the Council in Recital 41 of the Definitive Duty Regulation which states inter alia as follows :

"Moreover, the Council is not convinced that Community interest necessarily requires that the specific situation of an allegedly less efficient producer when confronted with unfair trade practices be disregarded; the Council considers that by setting the injury elimination level by including all three Community producers rather than having regard to the allegedly less efficient one alone, Community interest is appropriately reflected ."

That recital appears under the heading "Community interest" and refers to it explicitly . Article 12 ( 1 ) of the Basic Regulation provides that where dumping and consequent injury have been proved, anti-dumping duty shall be imposed if a third condition is met, namely that "the interests of the Community call for Community intervention ". There is no definition of those interests, and the provision plainly gives the Community authorities wide discretion . Canon has not shown any reason for finding that that discretion has been improperly used .

In my view, the fact that part of the Community industry is experiencing difficulties due to causes other than dumping makes it more - not less - necessary to prevent additional injury due to dumping . It cannot therefore be argued that the level of protection against dumping must be set at the level needed to protect only the most efficient Community producer . As to whether it should be set at a level to protect the least efficient Community producer, it is clear that the Community authorities did not set it at such a level . Recital 41 states that the level was calculated by reference to all three Community producers; and Recital 36 shows that the calculation was based on an average of the costs of production of all Community producers . Where - as here - the costs of production of the Community producers differ from one to another, it is not possible to give each of them the same degree of protection against dumping; and it seems to me that applying an average degree of protection ( as was done here ) is entirely consistent with the requirements of the Basic Regulation .

Canon also alleges that the approach adopted here diverged without justification from previous practice . However, given the margin of discretion that the Community authorities enjoy in this area, I am not satisfied in this respect that they were bound to follow the same practice in every case . In any event, of the two regulations cited by Canon as establishing a contrary practice, one ( Council Regulation No 1826/84 imposing a definitive anti-dumping duty on imports of vinyl acetate monomer originating in Canada, Official Journal 1984, L 170, p . 70, Recital 14 ) shows in fact that the Council in that case acted in the same way as it did here . Accordingly Canon' s seventh argument falls to be rejected .

Canon' s eighth argument is that the level of the profit margin used in calculating the target price was excessive . Recital 35 to the Definitive Duty Regulation, giving reasons, states that 10% was considered a reasonable profit margin to include in the target sales price of the Community product . It also states that the Community authorities reached that figure despite a demand from the Community industry that it should be much higher ( 20% on turnover or 30% on capital ). Canon suggests, although without any supporting evidence, that the figure of 10% greatly exceeded the profit levels ever achieved on electronic typewriters by at least two of the three complainant companies . The Council has stated that that suggestion is incorrect, and that in fact, although the precise figures are confidential, two of the complainant companies had profits on sales of electronic typewriters in the Community substantially exceeding 10% before dumping began . Canon' s argument on this point is not substantiated .

Canon' s ninth argument concerns the way in which the extent of the injury was measured . The way the injury was measured is set out in detail in Recitals 34 to 38 of the Definitive Duty Regulation . Basically the price of the Community-made product was compared to the price of the dumped import on the Community market . A duty equivalent to the difference between the two prices would bring the price of the dumped imports up to a level where they would no longer unlawfully injure the Community industry . In my view that basic approach - price comparison - is a straightforward way of achieving the aim of the Basic Regulation and is in accordance with its provisions .

For the purpose of that comparison, the prices of the dumped imports were adjusted before being compared with the target price of the Community-made electronic typewriters . According to Recital 34 the reason for that adjustment was that, "unlike many other products, an immediate comparison between imported and Community-produced models was impossible because of their variety and their differing technical specifications ". Canon does not deny that a direct model-for-model comparison was impossible and it does not deny that some adjustment was necessary in order to effect a comparison .

Recital 34 states that the adjustment was made in the following way : "Given that the exporters and the Community industry have each made bona fide evaluations of the percentage differences in the value of the various models, it was concluded that the most reasonable solution was generally to use a figure at the mid-point between these evaluations ". Canon complains that this method is misconceived and that splitting the difference between two estimates of value cannot be relied on to produce a meaningful figure . Canon submits that the Commission should instead have used the cost of production of the differing features, as being the only objectively verifiable indicator .

The adjustments made by the Community authorities do not purport to be precise or to be statistically exact, but only to be a reasonable approximation . In my view, for the particular adjustments concerned, a reasonable approximation was sufficient . Recital 34 states : "When it came to evaluating the technical differences between the most similar models, it became apparent that any evaluation would, to a substantial extent, be influenced by subjective appreciations of the anticipated reactions of prospective purchasers . Furthermore the exporters and the Community industry expressed the view that no objective yardstick for a comprehensive comparison was available ". The subjective element in estimating the value of different features in the eyes of potential customers necessarily excludes any precise result; the evaluation can only be approximate . I do not accept Canon' s suggestion that the figure is meaningless, nor can it be said to be arbitrary . Evaluations were made by the Japanese exporters and by the Community producers, who both have extensive knowledge and experience of the market to inform their judgments on this matter . The estimates were made in good faith, and the Council has stated that for many of the models compared the Japanese exporter and the Community producer in fact agreed on the figure to be used . In these circumstances, "splitting the difference", when there was one, should not be seen as a statistical exercise but a matter of judgment, and was a reasonable course to take . In my view, it has been shown that the method adopted by the Community authorities produced a sound approximation .

I do not consider that the Community authorities were obliged to use the cost of production instead . First, they were engaged on a comparison of prices for the purpose of determining injury, not a comparison of costs . Secondly, it is far from clear that it is practicable to ascertain the cost of production of each of the separate features concerned .

Accordingly I consider that no illegality has been demonstrated in the way the Community authorities calculated the adjustments to be made in respect of the different features of the products compared, and that Canon' s ninth argument falls to be dismissed .

Canon' s 10th argument on the subject of injury is that Canon was given insufficient information and that the statement of reasons in the regulation is inadequate. The details which have been given to the Court of a meeting between the Commission and Canon and of correspondence between the Commission and Canon amply demonstrate that the Community authorities gave Canon all the information requested by Canon which they could give consistently with their duty of confidentiality under Article 8 of the Basic Regulation.

The statement of reasons in the regulation is alleged to be inadequate because it fails to deal with the difficulties caused to the Community industry by its own "structural problems". This is merely a reflection of Canon' s substantive argument about "structural problems" which I consider unfounded for the reasons which I have just given. This argument must be rejected for the same reasons.

I therefore take the view that Canon' s fourth ground of annulment, concerning the injury finding, fails.

Ground 5 : Procedural matters

By its fifth ground of annulment, Canon avers that, since no prudently run enterprise could have anticipated the Commission' s unprecedented methods of calculation, the contested regulation constituted a retroactive penalty. The revolutionary nature of the Commission' s policies imposed on the Commission, and on the Council, especially high procedural standards, which the institutions failed to respect; they did not examine all relevant factors, did not give adequate consideration to arguments submitted, and did not give adequate reasons to explain their actions.

In my opinion, it has not been shown that the methods of calculation were improper. The high rate of the duty imposed was simply the consequence of the very heavy dumping which had caused the Community industry extremely severe injury. No relevant factor has been shown which the Community authorities failed duly to consider. The actions of the Community authorities have not been shown to be unlawful in any respect and are fully explained in the recitals to both regulations concerned. Therefore, in my view, Canon' s fifth ground of annulment should be rejected.

Accordingly, in my opinion, Cases 277 and 300/85 should be dismissed and the applicants ordered to pay the costs of the Council, the Commission and Cetma, including the costs of the proceedings for interim measures.