SECOND CONTRIBUTION TO DISCUSSION OF THE NEGOTIATING

GROUP ON RULES ON ANTI-DUMPING MEASURES

Paper by Brazil;  Chile;  Colombia;  Costa Rica;  Hong Kong, China;  Israel;

Japan;  Korea;  Norway;  Separate Customs Territory of Taiwan, Penghu,

Kinmen and Matsu;  Singapore;  Switzerland;  and Thailand

            The following communication has been received on 27 June 2002 from the Permanent Missions of Brazil;  Chile;  Colombia;  Costa Rica;  Hong Kong, China;  Israel;  Japan;  Korea;  Norway;  Separate Customs Territory of Taiwan, Penghu, Kinmen and Matsu;  Singapore;  Switzerland;  and Thailand.

_______________

            This is a second contribution to discussion of the Negotiating Group on Rules on anti-dumping measures.  As we recall, the first paper is contained in document TN/RL/W/6.

            Further contributions may be made for future sessions of the Negotiating Group.  As such, this paper does not necessarily represent the full view of every co-sponsor.  Likewise, we continue to encourage other Members to contribute with their points of view.

            This paper indicates some additional issues of relevant provisions that we seek to clarify and improve, including the ruling out of abusive interpretation of the current AD Agreement.

            The order of the issues in this paper is not an indication of priority among them.

            In order to facilitate the understanding of the issues involved, some examples are provided.  By definition, the examples are limited.  They cannot reflect the full scope of the problems and provisions in question.

            In indicating the issues, some questions are raised with the purpose of promoting the discussion while giving a sense of the direction the negotiations should have.

1.         Definition of Product Under Investigation/Consideration and Like Product

The AD Agreement does not currently contain any provision defining the product under investigation/consideration.  This ambiguity allows the authorities to define a group of products destined for very different market segments to be a single “product under investigation” based on the petition of domestic industry, or based on the authorities’ own discretion.  The purpose of AD measure is to remove injury caused by dumped imports.  In light of this purpose, is it appropriate to exploit the fact that a certain specific product has been found to be dumped, and then to “use” that finding for one product to apply AD measures to other product, which is a superficially similar but substantially different product, for example, a product that is destined to a different market segment? What rules would provide a more rational and disciplined framework to determine the scope of “product under investigation,” so that AD measures would only be applied to those products found to be “dumped” and causing injury?  Shouldn’t there be appropriate criteria for determining the “product under investigation” to limit arbitrary expansions of product scope?

Determining the scope of “product under investigation” at initial stage of investigation could have significant effect on the whole process of the AD proceedings.  For example, the scope of the product under investigation/consideration defines the scope of "like product," which in turn determines domestic producers that constitute the "domestic industry" and also determines the data that will be necessary for the injury analysis. 

Allowing an undisciplined definition of the scope of "product under investigation/ consideration" could also permit the authorities to include products developed at a later time in the scope of the "product under consideration", even if such later-developed products would be radically different.  Should the authorities be given so much discretion to effectively redefine the scope after affirmative determination?

(An Illustrative Example)

Country A imports tyres used for buses and tractors.  Based on a petition filed by the domestic producers of bus and tractor tyres, an anti-dumping investigation on both tyres (“product under investigation”) is initiated.  Even though tractor tyres are for a market segment, which is very different from the bus tyre market segment, ultimately the authorities treat both kinds of tyres as a single product under consideration.  But, in reality, the dumping margin on tractor tyre is 10 per cent and that on bus tyre is de minimis.  In spite of the de minimis margin on imports of bus tyres, the average margin of dumping for the “product under investigation” is positive.  As a result, and considering a positive determination of injury on the single product under investigation, an anti-dumping duty on bus and tractor tyres is applied because the authorities treated them as a single product group.

Is such a scope of “product under investigation” justified?  Is it acceptable to treat these different types of tyres, belonging to different market segments, within the same scope?  Does it make sense to leave the authorities so much discretion to define the scope of “product under investigation?”

2.         Definition of Domestic Industry

            Article 4.1 of the AD Agreement defines the term “domestic industry” as referring to (a) the domestic producers as a whole of the like products or to (b) those of them whose collective output of the products constitutes a major proportion of the total domestic production of those products.

            Shouldn’t the AD Agreement have clearer criteria for the definition of the term “major proportion”?  Shouldn’t there be criteria to determine when the authorities are allowed, in exceptional cases, not to use the “domestic producers as a whole of the like products”?

3.         Standing Rules

            Article 5.4 of the AD Agreement provides that petitions need support by domestic producers whose collective output constitutes more than 50 per cent of the total production by that portion of the domestic industry expressing either support for or opposition to the petition.  This provision also states that “no investigation shall be initiated when domestic producers expressly supporting the application account for less than 25 per cent of the total production of the like product produced by the domestic industry.”

            Under these current rules, an application can be filed with the support of those representing only a minority portion of the total domestic production of the like product.  Is it acceptable to consider that the application represents the total domestic production when the application is supported by such a small portion of the total domestic production? Shouldn’t the application be supported by at least more than 50 per cent of the total domestic production?

            (An Illustrative Example)

            An application for AD investigation was filed based on the support of domestic producers whose collective output constitutes only 25 per cent of the total domestic production.  Although the collective output of those who expressly opposed the application constituted 20 per cent of the total domestic production and the collective output of those who were silent constituted 55 per cent, the authorities judged that the application satisfied standing requirement.

            Can the authorities determine injury by data from the minority portion of the total domestic production?  Suppose, in the course of the AD investigation, the supporters of the application responded to the questionnaire regarding injury from the authorities, but other domestic producers ignored the questionnaire.  As a result, the authorities determined injury based only on data which represents minor portion, i.e. 25 per cent, of the total domestic production.

            Should such an AD investigation, based on a minority portion of the total domestic production, be considered appropriate?  Is it appropriate to consider a mere 25 per cent as representative of the total domestic production? Moreover, in order to avoid irresponsible expression of support, shouldn’t we require that legally sufficient “ support” should include complete data relevant to assessment on injury, and the causal relationship between that injury and imports?

4.         Initiation Standards

            Article 5.3 of the AD Agreement requires that the authorities shall examine the accuracy and adequacy of the evidence before initiating the investigation.  However, the interpretation of the terms “examine”, “accuracy”, and “adequacy” merits clarification.  In some cases, an investigation begins solely based on the facts alleged in the petition, with little or no effort to corroborate.  Sometimes the facts alleged in the petition turn out to be false, sometimes intentionally false.

            Furthermore, in light of the fact that investigation incurs extremely heavy burden on the respondents, shouldn’t the requirements to initiate an investigation be improved and clarified in order to allow more meaningful “examination” of the basis for beginning the investigation?

            (An illustrative example)

            An anti-dumping investigation was initiated based on a petition.  The application included some facts indicating the existence of “dumping” and “injury.”  However the facts written in the application obviously included some differences from the data recognized and owned by the respondents, because the petitioners adopted the data from non-public data resources such as individual research papers.

            In such a case, shouldn’t the authorities actually examine the accuracy of data by gathering information from public records, intra-governmental records, and other available sources, as appropriate?

5.         Determination of Normal Value – Affiliated Parties and Their Transactions

            The current AD Agreement does not specifically address the issue whether home-market sales to affiliates may be included in, or excluded from, the calculation of normal value.  Nor does it include any provision regulating transactions involving affiliated suppliers.  Moreover there is no appropriate definition of “affiliation” in the AD Agreement.  Footnote 11 provides the definition of affiliation in the narrow context of determining the domestic industry.  It, however, does not constitute a sufficient basis to define affiliation in the context of normal value because it does not stipulate when and how transaction prices between affiliates may be rejected or adjusted.

            The lack of clear provisions with respect to determining affiliation and determining when affiliated party transaction prices should be considered unreliable often forces respondents to unnecessarily submit extensive sales and cost data of affiliates and other market data.  Moreover, the authorities often apply improper adjustment on normal value, based on alleged “affiliation”, even though the affiliation does not distort the transaction prices.

            (An Illustrative Example)

            A petition was filed against Company A, a producer of polyvinyl alcohol, a chemical product.  Before the petition was filed, Company A executed a supply contract of vinyl acetate monomer ("VAM"), the major raw material of polyvinyl alcohol, with Company B.  Although the transaction price of VAM was approximately 10 per cent lower than the commodity market prices of VAM, Company B nevertheless earns a profit on sales of VAM to Company A.

            In the course of an AD investigation, the authorities found that Company A owns 6 per cent of the outstanding stock of Company B, and decide to consider the two companies to be “affiliated.”  The authorities then adjusted the transaction price from Company B to Company A to account for the 10 per cent difference, claiming that such adjustment was to reflect the price in the ordinary course of trade between unaffiliated parties.  As a result, Company A was determined to be “dumping,” although it otherwise would not be dumping.

            Is it appropriate that the authorities find the affiliation between Companies A and B solely on a basis of small stockholding?  Shouldn’t the AD Agreement at least provide that affiliation may not be found when one company owns a small portion of outstanding voting stocks of the other company and is otherwise unrelated? Is it rational to consider that finding of affiliation between companies without actual control constitute a sufficient evidence that the transaction price is unreliable? Further, shouldn’t the AD Agreement require the authorities to find positive evidence showing that the transfer price is unreliable to revise or reject the transfer price between affiliated parties?  

6.         Injury Determination

            Article 3.4 of the current AD Agreement lists factors that must be considered when injury is determined, but does not provide adequate guidance to evaluate those factors.  Shouldn’t we clarify this provision and its relationship with other provisions of Article 3?

7.         Price Undertakings

            Price undertakings are a useful tool to provide concrete guidelines for exporters to manage their businesses, while protecting the domestic industry from any further injury through the effects of dumping.  Article 8.1 of the AD Agreement, however, does not state what should constitute “satisfactory voluntary undertakings.”  Article 8.3 also provides the authorities with wide discretion to refuse proposals for price undertakings, including undefined “reasons of general policy.”  Because of these ambiguities, the current AD Agreement provides few meaningful procedures for exporters to return to its ordinary business with the price undertakings.

            (An Illustrative Example)

            The authorities preliminarily determined 25 per cent dumping margins on exports of computer displays by Company A.  The margin of underselling, however, was determined to be only 10 per cent.  Exports of Company A represent 40 per cent of total exports from the exporting country X.  Company A proposed a price undertaking.  The proposal states the exporter must raise its price by 15 per cent, and thus completely eliminate the margin of underselling and the injurious effects of any dumping.  Company A also must report its export prices and home market prices to the authorities on a quarterly basis.  Yet the authorities refused to accept the proposal, stating that it has a general policy not to accept price undertakings from a single exporter, which does not represent the majority of all imports from a country.

            Is it appropriate that Company A is deprived of its chance for the price undertaking, simply because other exporters in the country cannot or will not make such a proposal?  Are there circumstances under which the AD Agreement should require the authorities to accept price undertakings?  Shouldn’t we explicitly confirm that such undertakings need only to eliminate injurious effects of dumping? How can we take into account the needs of developing countries? 

8.         Reviews

            The current AD Agreement does not clearly articulate the concepts, procedures and methodologies applicable to reviews under Article 9.3 (anti-dumping duty assessment), Article 9.5 (new shipper reviews), Article 11.2 (revocation reviews) and Article 11.3 (sunset reviews).  The lack of explicit rules makes it possible for the authorities to introduce rules, procedures, and methodologies arbitrarily into these reviews that differ substantially from those in initial investigations and thereby placing undue burden on the respondent.  Is this fair? Shouldn’t the same rules as those used in the initial investigation be applied for reviews?

            Furthermore, shouldn’t the duration of reviews stipulated in Article 11.4 be limited to maximum of 12 months?

            (An Illustrative Example)

            An importing country imposed an antidumping duty on exports of roses by Company A. Company A then adjusted its export price, based on the dumping margin methodologies used in the initial investigation, so as not to commit any further dumping.  Company A believed there was no dumping after the imposition of antidumping duties, and requested a review for anti-dumping duty assessment.  Company A was surprised by results of the review because the dumping margin was determined on each export sale in comparison with monthly weighted-average home market sales, not on the comparison of annual weighted-average prices in both markets that had been used in the initial investigation.

9.         Constructed Export Price: methodology for construction

            Article 2.3 of the AD Agreement stipulates that “export price may be constructed” in certain circumstances, for example, in a situation where an importer of a product is affiliated with its exporter.  In such cases, the provision permits the authorities to construct export price from the price by the importer to unaffiliated buyers.  Such export price is so called “constructed export price” or “CEP.”  Article 2.4 provides certain guidelines applicable to CEP for its fair comparison with normal value.  However, these guidelines are not sufficiently clear, which leads to different practices in different WTO Members and often results in abusive asymmetry deduction of costs and profits from CEP and normal value.  Such an asymmetry in the calculation of CEP and normal value cannot ensure a fair comparison between the two prices.

            Shouldn’t these Articles be clarified so as to explicitly rule out such asymmetry comparison?

            (An Illustrative Example)

            Company X is a producer of DVD players.  The Company X sells its DVD players in its domestic market directly to retailers.  Company X sells DVD players to unaffiliated distributors of DVD players in an importing country through its affiliated importer. 

            An antidumping investigation against DVD players was initiated in the importing country.  The authorities used the CEP method for Company X’s dumping margin calculation, i.e., Company X’s importer’s sales price to unaffiliated distributors.  Company X submitted to the authorities all individual profits gained in its domestic sales department, claiming that these profits must be deducted from the normal value (i.e., sales price to home market retailers) to make due allowance due to the difference in the levels of trade.  The authorities refused to make such deductions from normal value, alleging that submission of these profits were not directly related to the domestic sales and were not sufficient to quantify the difference in levels of trade between CEP and normal value.  The authorities, however, fully deduct all profits -- including identical profits gained at Company X’s importer-- from the CEP, claiming that the importer’s activities is limited to sales to distributors in the importing country.  Because of this asymmetric comparison, dumping margin was created. 

            Shouldn’t the authorities be obliged to deduct these home market profits to observe symmetric deduction of costs and expenses between CEP and normal value?

10.       “All others” Rate

            When the authorities limit their examinations of dumping from an exporting country to be a sample of exporters/producers, Article 9.4 applies in calculation of an “all others” dumping margin rate for exporters/producers, which are not sampled.  The Article obliges the authorities to calculate the “all others” rate based only on actual sales and cost information, which were supplied by exporters/producers and used to calculate their individual margins, but also requires the authorities to disregard such actual information if the resulted margins were zero or de minimis.  Is it logical to require the authorities to ignore zero/de minimis margins, regardless of the fact that zero and de minimis margins also represent actual performance of exporters/producers from the exporting country?

            (An Illustrative Example)

            There were 20 exporters/producers of product X from an exporting country in an antidumping investigation.  Among 20, 3 producers shared 60 per cent of all exports from the exporting country.  The authorities found that there were too many producers/exporters to investigate, and decided to investigate only the top 3 producers for the dumping margin calculation purpose.  The export volume by each of these 3 companies was approximately same.  The authorities finally found the following dumping margins:

                        Company A:     3 per cent

                        Company B:     1 per cent

                        Company C:     0 per cent

                        All Others:       3 per cent

            The authorities applied the Company A’s dumping margin to the all others rate ignoring margins of Companies B and C because their margins were zero or de minimis.  Not-sampled 17 producers are thus subject to 3 per cent antidumping duty for their future export.  These 17 companies, however, would not be subject to any antidumping duty if companies A, B and C were averaged because their weighted-average margin would be de minimis (1.3 per cent (=(3+1+0)/3)).

11.       The Authorities’ Discretion on the Use of Cost Data

            Article 2.2.1.1 of the AD Agreement states that costs shall normally be calculated on the basis of producers’ own accounting records, provided that these records are in accordance with the generally accepted accounting principles (“GAAP”) of the exporting country and reasonably reflect production and sales costs of the product under consideration.  This provision, however, is not specific enough to clarify the circumstances in which the authorities may reject or adjust cost data as maintained in the producers' own cost accounting records.  This ambiguity allows the authorities to reconstruct a producer’s costs arbitrarily, claiming that the particular cost accounting method does not “reasonably” reflect costs for anti-dumping purposes.  Such a practice places undue burden on respondents, even if the respondent keeps its accounting records in accordance with appropriate GAAP and accounting method.  This reconstruction of costs could also create artificial dumping margins.

            Under what circumstances should we require the authorities to accept cost data as recorded in the producer’s accounting book? For example, if the accounting records are audited by duly qualified person or agency, shouldn’t the authorities respect those records?

            (An Illustrative Example)

            A yarn producer, Company X, has three production lines, A, B, C to produce three different kinds of yarns, L, M, and N.  Every production line is able to produce all three kinds of yarns.  Because these three lines have different production capacities, Company X chooses an appropriate production line to produce a specific kind of yarn in accordance with changes of market demands.  Because of this flexibility, Company X calculates and keeps its manufacturing costs of each type of yarn on a plant-wide basis in its ordinary course of business.  This accounting method fully complies with the GAAP in the exporting country.

            In an AD investigation of yarn L, the authorities demand that the per-unit manufacturing cost must be submitted based only on production line A, because yarn L was eventually manufactured only at the production line A during the particular period of investigation.  Company X had to recalculate its per-unit cost of yarn L with enormous resources and expenses to meet the requirements from the authorities.  The recalculated per-unit cost was higher than the cost in the producer's accounting records.  Since Company X has set its yarn L price based on its normal accounting records, home market yarn L sales were considered below the recalculated costs, and resulted in unexpected dumping margins. 

__________