Anti-dumping MODULE. ESTIMATED TIME: 10 hours OBJECTIVES OF MODULE 3. Understand the basic WTO disciplines related to anti-dumping; and

Transcription

1 MODULE 3 Anti-dumping ESTIMATED TIME: 10 hours OBJECTIVES OF MODULE 3 Understand the basic WTO disciplines related to anti-dumping; and Get acquainted with the different anti-dumping procedures and investigations. 1

2 IN BRIEF GATT and WTO rules do not prohibit "dumping" as such. Rather, they set forth the rules that Members must respect when taking action against dumped imports. For such action to be permissible, Members must determine the existence and amount of dumping, and must establish that dumped imports are causing material injury or threat to, or material retardation of the establishment of, the importing Member's domestic industry producing the product that is "like" the dumped imported product. I. WHAT IS DUMPING? If a company exports a product at a price lower than the price it normally charges in its own home market, it is said to be "dumping" the product. Is this unfair competition? Opinions differ, but many governments take action against dumping in order to defend their domestic industries. The Agreement on Implementation of Article VI of the GATT 1994 (from now on "the AD Agreement") does not pass judgment on dumping. Rather, its focus is on the actions that governments can (and cannot) take in response to dumping in their markets. "Dumping" is defined in both Article VI of the GATT 1994, and in the AD Agreement, as the sale of an imported product in the importing market at less than its "normal value ". As indicated above, most commonly this is where the price of the imported good is less than the price at which the exporter sells that good in its own home market. In this sense, dumping is a situation of international price discrimination. Dumping is NOT the sale of an imported product for less than the price charged for the same product produced domestically. This is price undercutting, which is a factor to be examined in the context of injury analysis, but which is not relevant to whether or not there is dumping. IN BRIEF In the simplest of cases, the existence of dumping is identified by comparing prices in two markets. In this case, dumping would exist where: Price of imported good < Home market price in exporting market IN DETAIL However, the situation is rarely, if ever, that simple, and in most cases it is necessary to undertake a series of complex analytical steps in order to determine the appropriate price in the market of the exporting country (known as the "normal value") and the appropriate price in the market of the importing country (known as the "export price") so as to be able to undertake an appropriate comparison. The calculations can get complicated for a variety of reasons. For instance, imagine a situation where the product subject to investigation is not sold at all in the market of the exporting country. Why could that be? One reason could be that there is no domestic market for that product. Imagine that there is production of ski boots in a tropical country whose citizens rarely if ever travel abroad to snowy countries. There likely would be no domestic sales of ski boots in the producing country (i.e., 100 per cent of the production would be exported), and hence there would be no home market price to which the export price could be compared in order to determine whether the exported ski boots were dumped. That is, the usual price comparison 2

3 calculation could not be performed. The question becomes how, in such a situation, an investigating authority could determine the normal value of the ski boots? This and related questions are addressed later on in the section on normal value. I.A. DUMPING AND THE GATT YEARS FROM ARTICLE VI OF THE GATT Under Article VI of GATT 1947, certain disciplines were established for situations where dumping was causing injury to a domestic industry in the importing market. Article VI allowed an "anti-dumping duty" to be imposed at the border upon importation to offset or prevent the dumping. The level of the duty could be equal to, but not higher than, the margin of dumping. As tariff rates were lowered over time following the original GATT Agreement, anti-dumping duties were increasingly imposed, and the inadequacy of Article VI to govern their imposition became ever more apparent. Article VI requires a determination of injury caused by dumping, but does not contain any guidance as to how that determination is to be made. It addresses the methodology for establishing the existence of dumping, but only in general terms....to THE AGREEMENT ON ANTI-DUMPING PRACTICES Consequently, Contracting Parties to GATT 1947 negotiated successively more detailed Codes relating to antidumping. The first such Code, the Agreement on Anti-dumping Practices, entered into force 1967 as a result of the Kennedy Round. However, the United States never signed this "Kennedy Round Code", which as a result had little practical significance. The anti-dumping Agreement that resulted from the Tokyo Round negotiations (the "Tokyo Round Code"), which entered into force in 1980, represented a quantum leap forward. Substantively, it provided far more guidance about the determination of dumping and of injury than did Article VI, including explicitly requiring that such determinations be made on the basis of an investigation conducted by the authorities of the importing country. Equally important, it set out in substantial detail certain procedural and due process requirements that had to be fulfilled in the conduct of such investigations. Nevertheless, the Code still represented only a general framework for countries to follow in conducting investigations and imposing duties. It was also marked by ambiguities on numerous controversial points, and was limited by the fact that only the 27 Parties to the Code were bound by its requirements. 3

4 I.B. DUMPING AND ANTI-DUMPING IN THE WTO IN DETAIL Unlike the Kennedy Round and Tokyo Round Codes, the WTO Anti-dumping Agreement is a multilateral (as opposed to a plurilateral) agreement.. The AD Agreement therefore must be accepted as part of the "single undertaking" by all current Members and by any country joining the WTO. That is, the AD Agreement applies to all Members. In the following pages, we will review the following elements: The legal documents underpinning anti-dumping actions in WTO; How to establish whether imported goods are being dumped; How to establish whether the dumped imports are causing or threatening to cause injury to the domestic industry; How to determine the level of anti-dumping duties; The procedures to be followed in initiating and conducting investigations, including the collection of information; The procedures for review and termination of anti-dumping duties; Judicial review; Dispute settlement; The Committee on Anti-Dumping Practices; and Notification requirements. Before you dive - hopefully without drowning - into specific provisions of the AD Agreement, please remember that you can contact your tutor anytime in case you have a question or need a clarification on the above. I.C. ARTICLE VI OF GATT 1947 AND THE ANTI-DUMPING AGREEMENT WHO'S WHO? Article VI of GATT allows countries to take action against dumping. The AD Agreement clarifies and expands on Article VI, and the two operate together. Under these provisions, countries are allowed to act in a way that would normally break the GATT principles of binding a tariff and not discriminating among trading partners (which we have seen in the previous module). Typically, anti-dumping action means charging an extra import duty on a particular product imported from a particular exporter in order to bring the price of the imported product up to its "normal value" by offsetting the margin of dumping. As we will see below, anti-dumping measures also may take the form of price undertakings. 4

5 Important note The AD Agreement is rather long and complex. Essentially, it represents an effort to balance potentially conflicting interests: on the one hand, the interest of importing countries in imposing anti-dumping measures to prevent or remedy injury to their domestic industries caused by dumped imports; and on the other hand, the interest of exporters (and importers and consumers) for whom anti-dumping measures and procedures should not themselves become obstacles to fair trade. 5

6 II. ANTI-DUMPING IN BRIEF Article VI of GATT and the AD Agreement explicitly authorize a Member to impose specific anti-dumping measures on imports from a particular source, in addition to ordinary customs tariffs, when the importing Member demonstrates through a properly-conducted investigation that dumping is causing or is threatening to cause material injury to a domestic industry or would materially retard the establishment of a domestic industry. A product is to be considered as being dumped when it is introduced into the commerce of another country at less than its "normal value", normally the comparable price at which the product is sold in the domestic market of the exporting country, or if there is no such price, a comparable price for sale of the like product to a third country market, or the cost of production of the product plus a reasonable amount for selling costs and profit. Under Article VI of GATT 1994, and the AD Agreement, WTO Members can impose anti-dumping measures if they determine: (a) (b) (c) that dumping is occurring; that the domestic industry producing the like product in the importing country is suffering material injury or threat thereof, or that the establishment of a domestic industry is being materially retarded; and that there is a causal link between the two. In addition to substantive rules governing the determinations of dumping, injury, and causal link, the AD Agreement sets forth detailed procedural rules for the initiation and conduct of investigations, the imposition of measures, and the duration and review of measures. IN DETAIL The text Article VI of the GATT reads in relevant portion: Anti-dumping and Countervailing Duties 1. Members recognize that dumping, by which products of one country are introduced into the commerce of another country at less than the normal value of the products, is to be condemned if it causes or threatens material injury to an established industry in the territory of a Member or materially retards the establishment of a domestic industry [A] product is to be considered as being introduced into the commerce of an importing country at less than its normal value [i.e., as being dumped], if the price of the product exported from one country to another (a) is less than the comparable price, in the ordinary course of trade, for the like product when destined for consumption in the exporting country; or (b) in the absence of such domestic price, is less than either (i) the highest comparable price for the like product for export to any third country in the ordinary course of trade, or 6

7 (ii) the cost of production of the product in the country of origin plus a reasonable addition for selling cost and profit. 2. In order to offset or prevent dumping, a Member may levy on any dumped product an anti-dumping duty not greater in amount than the margin of dumping in respect of such product. For the purposes of this Article, the margin of dumping is the price difference determined in accordance with the provisions of paragraph (a) No Member shall levy an anti-dumping [ ] duty on the importation of any product of the territory of another Member unless it determines that the effect of the dumping [ ] is such as to cause or threaten material injury to an established domestic industry, or is such as to retard materially the establishment of a domestic industry. Further, the AD Agreement states: Article 1 (Principles) An anti-dumping measure shall be applied only under the circumstances provided for in Article VI of GATT 1994 and pursuant to investigations initiated and conducted in accordance with the provisions of this Agreement. The following provisions govern the application of Article VI of GATT 1994 in so far as action is taken under anti-dumping legislation or regulations. The AD Agreement also contains detailed provisions elaborating on all of the above elements contained in Article VI of GATT EXERCISES: What is dumping? What are the possible bases for determining normal value? What must a Member do to have the right to apply an anti-dumping measure? 7

8 For determining NV For injury II.A.1. ILLUSTRATION 1 Let's assume that Member A and Member B are Members of the WTO. Vegi Company is a producer of vegetables in Member B. Because of the excessive production of tomatoes in the world during the current year and the increased availability of high quality tomatoes on the international market, Vegi is unable to sell its production of tomatoes for the year, and is facing great losses if it does not find a market for its production. Vegi decides to sell its tomatoes in Member A by offering them at a price ($1.00 per kg) that is below the prevailing market price for tomatoes of similar quality in Member A. Meanwhile, the selling price in Member B for the same tomatoes is $2.00. The tomato producers in Member A experience a slump in their domestic sales of tomatoes, and request the government of Member A to impose anti-dumping duties on tomatoes imported from Member B. The government of Member A could potentially apply an anti-dumping duty on tomatoes up to $1.00/kg., representing the margin of dumping, i.e. the difference between the selling price of tomatoes in Member B and the export price to Member A. To be able to do so, however, Member A first needs to conduct an investigation, and it only could proceed with an anti-dumping duty if in that investigation it determined that: 1) Vegi was exporting its tomatoes to Member A at a price below the price at which Vegi is selling the same tomatoes in its home market (in Member B); 2) the domestic tomato producers in Member A are suffering injury; and 3) there is a causal link between the injury suffered by the domestic industry and the dumped imports. The Products in an Anti-dumping Investigation The three different products in and AD investigation Before moving on to discuss calculation issues, we have to examine the issue of the product/s in an anti-dumping investigation. Conceptually, in an anti-dumping investigation, we can distinguish three different "products": (1) the product under consideration (or the exported product); (2) the product that is "like" the exported product, sold in the exporter's home market; and (3) the product that is like the exported product, produced by the domestic industry of the importing country. The conceptual relationships of these products are shown below: Product under Consideration (Investigated exported product) LIKE PRODUCT in domestic market of exporter, or in 3rd country export market Dumping analysis LIKE PRODUCT produced by domestic industry in importing country Injury and causation analysis Figure 1: Products in an anti-dumping investigation 8

9 At the top of the diagram, we find the "product under consideration". Depending on the jurisdiction, it is also known inter alia as "subject merchandise", "subject goods", "investigated goods", "investigated product", etc. This is the product exported at allegedly dumped prices and hence the product that is alleged to be causing injury to the domestic industry. Why is the "product under consideration" relevant in an anti-dumping investigation? First, because this is the product for which we need to know the export price one of the two elements in a dumping determination - as we will see below when we discuss how to determine the export price; Second, when conducting the injury analysis, the AD Agreement requires an investigating authority to consider among other things the trends in dumped imports of the "product under consideration", as we will see below when we address issues pertaining to the injury determination. More generally, the "product under consideration" is the product covered by the investigation, as identified in the notice of initiation. As such, this product is the focus of and point of reference for all aspects of the investigation. Starting with the application, and the product it alleges to be dumped, the investigation will examine normal values for the product that is "like" that product, to determine the margin of dumping, and will examine the performance of the industry in the importing country producing the "like" product, to determine injury and causation. Ultimately, any anti-dumping measures will apply only to the "product under consideration". $In other words, the "product under consideration" determines the scope of the investigation and of any eventual anti-dumping measures. Because the scope of the eventual anti-dumping measure depends on the definition of the product under consideration, this determination is generally subject to a lot of discussion during the course of an investigation. For example, the domestic industry in the importing country - as a complainant may be interested in a broad definition of "product under consideration", as this would mean wider coverage of any anti-dumping measure that might be applied. On the other side, exporters, importers and consumers may seek as narrow a definition as possible of the "product under consideration", to constrain the scope of application of an eventual anti-dumping measure. II.A.2. CASE STUDY 1 We'll examine an actual (review) investigation conducted by the EC on imports of colour televisions (CTVs) from China, Korea, Malaysia and Thailand. This review concerned a (previously-imposed) anti-dumping measure in force on these products. During the course of the review investigation, questions as to the inclusion or exclusion of certain variants of the product were addressed. (The public notice concerning this review investigation is Regulation (EC) No 1531/2002. The text below is excerpted from that public notice.) a) Product description The product concerned was CTVs with a diagonal screen size of more than 15,5 cm, whether or not combined in the same housing with a radio broadcast receiver and/or clock. The product was classifiable within CN codes ex , , , , ex and In a previous review of this anti-dumping measure, Regulation (EC) No 710/95 excluded from the definition of the product concerned D2MAC sets and high definition televisions (HDTVs) since these products, which introduced qualitative technical changes, were at that time still in the development stage and were not available to the public except in very limited circumstances. Subsequent Regulation (EC) No 2584/98, which 9

10 amended the anti-dumping duty regulation resulting from that previous review, confirmed that these products should not be included in the scope of the product definition at that time, since the investigation had not brought to light any new factual evidence that would justify their inclusion. I In the review investigation examined in this case study, again no new evidence justifying the inclusion of these products was submitted, and it therefore was concluded that D2MAC sets and high definition televisions were not covered by the investigation. b) Arguments of the parties during the review investigation One importer requested the exclusion from the scope of the investigation of so-called internet CTVs, which integrate an internet modem and computer operating system that allow access to the internet via the TV screen, and which are controlled by a remote control unit that includes a full keyboard. In an internet CTV, all the necessary modem circuitry is integrated into the body of the television set, instead of having a separate set-top box. The importer argued that the exclusion was warranted in view of the differences in basic physical and technical characteristics between standard CTVs and internet CTVs, and in view of the different consumer perceptions of these two products. The differences in the basic physical characteristics of internet CTVs consisted of the additional internet components in the CTV, which represented around 60% of the total costs of the internet CTV, and the keyboard that is integrated into the remote control. Regarding the basic technical characteristics, the importer argued that the internet CTV sends and receives data not via broadcasting technology but via the telephone system using the modem. Furthermore, it employs technology which is different from the basic technology in CTVs; it incorporates a system for secure access to the internet, secure socket layer (SSL), a browser technology for the display of internet graphics on standard resolution CTVs, and a modem that translates digital into analogue signals, which can travel over a standard phone line. The importer alleged that the different consumer perception of an internet-ctv compared to a standard CTV was proven by the fact that internet-ctvs were sold at retail level at more than twice the price of a standard CTV. Furthermore, it was argued that the integrated internet circuit imparts a distinct additional quality to an internet-ctv. In support of its position, the importer cited a case on video cassette recorders from Japan and Korea (VCRs) in which it was concluded that where a VCR and a CTV are combined in a single housing, this combination has to be regarded as a distinct product. The applicant claimed that such exclusion was not warranted. It disagreed that internet CTVs had different basic physical and technical characteristics, and argued that the internet component was comparable to the teletext in a CTV. The applicant argued that internet should be considered as a more modern form of teletext and therefore an additional feature of CTVs, rather than a new product lacking the basic characteristics of CTVs. In its view, since CTVs with teletext were covered by the investigation, internet CTVs also should be. The applicant also questioned the importer's arguments concerning consumer perceptions, arguing that internet CTVs had been launched only recently. The applicant further argued that the high percentage of costs represented by the internet element was due to the recent introduction of internet CTVs. In support of this argument, the applicant submitted evidence showing that at the time of the introduction of the teletext facility in CTVs, the cost of teletext was substantially higher than they were at the time of the review investigation examined in this case study. Finally, the applicant contested the relevance of the example of the combined VCR/CTV, arguing that in such a combination the VCR has an independent function, whereas in an internet CTV, the internet function does 10

11 not work without the CTV, but instead is an addition to it. c) Findings of the investigation The investigation revealed that an internet CTV was a product that combines in a single housing two technologies producing two sufficiently distinct end uses: (1) sending and receiving electronic mail and accessing the world wide web; and (2) viewing television programmes. Given the addition of the internet function, the CTV part of the unit did not necessarily determine the character of the entire product. To the contrary, it was the internet function that predominated over the CTV. Indeed, this combination contained a specific element that imparted an additional function to the internet CTV compared with a standard CTV, allowing the former to be considered as different for the purpose of the review investigation. The above conclusions were reached on the basis of the information gathered in the course of the investigation and relating to the period of investigation. Given the early stage of development of the internet CTV, and the fact that the product was only available to the public in small quantities, it could not be excluded that the conclusions reached regarding product scope would need to be revisited in light of further developments of the product in the future. Thus, in the event of a future new review of the measures in place, the situation of these products would have to be re-examined, on the basis of the information gathered in that future review, to determine whether such exclusion would still be justified. Furthermore, despite differences in screen sizes, sound systems, broadcast systems, screen types and formats, and picture frequency, it was found that all CTVs shared the same basic physical and technical characteristics and the same use, and that therefore they formed a single product. Like product In the course of the investigations, it was established that CTVs originating in or exported from the countries concerned and destined for the Community, shared the same basic physical and technical characteristics and end-uses as CTVs manufactured and sold by the Community industry on the Community market. It was also found that there were no differences between the CTVs produced and sold in the countries concerned, including Turkey which was used as an analogue country, and those exported to the Community, which were both identical to the CTVs manufactured and sold by the Community industry on the Community market. These products are therefore alike within the meaning of Article 1(4) of the basic Regulation. LIKE PRODUCTS IN AD INVESTIGATION "Like product in the domestic market of the exporter..." In examining whether dumping is taking place, the investigating authority will have to determine the "like product in the domestic market of the exporter". This will be the product sold in the domestic market of the exporting country. Why is the "like product in the domestic market of the exporter" relevant in an antidumping investigation? The like product in the domestic market of the exporter is important because its price normally will form the basis of the normal value to be used in the dumping calculation. That is, the first choice for determining the "normal value" of the investigated product is the price at which the "like product" is sold in the domestic market of the exporting country. 11

12 Are not the "product under consideration" and the "like product in the domestic market of the exporter" identical? In case of commodities (e.g., fertilizers, minerals, raw agricultural products), the "like product in the domestic market of the exporter" will most likely be identical to the "product under consideration". That is, for these goods, generally there will not be important product differentiation as between the domestic and export markets. In other cases (e.g., involving consumer goods), the investigating authority will most often be faced with products in the two markets that are not identical. This might be due to, for instance, differences in consumer taste or technical requirements. An example? For instance, colour televisions sold in the US are different from those sold in Europe, inter alia because of differences in voltage (110V in the US, 220V in Europe). Thus, normally a colour TV made for the US market will not work in Europe. This applies to all other electronic goods. What happens if the "product under consideration" and the "like product in the domestic market of the exporter" are not identical? Can we still compare the price of the "product under consideration" to the price at which the "like product" is sold in the domestic market of the exporting country? The AD Agreement states that "like product" must be interpreted to mean a product which is identical, i.e. alike in all respects to the product under consideration, or in the absence of such a product, another product which, although not alike in all respects, has characteristics closely resembling those of the product under consideration. Continuing the above example, the question would be whether otherwise identical 110V and 220V television sets meet the standard of having "closely resembling characteristics". What characteristics should be examined in order to determine whether the product sold in the domestic market of the exporting Member "closely resemble[es] those of the product under consideration"? The AD Agreement is silent on this issue. However, investigating authorities worldwide tend to consider at least the following factors: Physical characteristics of the product; Raw materials used; Manufacturing process; Tariff classification; Consumer preference and end-use; Quality. Thus, in the case of the colour TVs mentioned above: An investigating authority would have to look at the physical characteristics of the TVs sold in the US and in Europe. It would then probably see that some technical components are different because of the voltage and other physical differences. 12

13 It would then examine the inputs used in the production of both TVs. These would include, for both types of TVs, colour picture tubes, transistors, etc. It would also examine the manufacturing process. It would then examine what both types of TVs are used for. It would finally assess aspects relating to consumer preferences and perhaps quality differences between the two types of colour TVs. Ultimately, based on the outcome of the analysis of each of the above factors, the investigating authority would need to arrive at a conclusion on whether the two kinds of colour TVs are, or are not, "like" in the sense of the AD Agreement. What if the product sold in the domestic market of the exporting country is neither identical to, nor has characteristics closely resembling those of, the product under consideration? Then, the price at which that product is sold in the domestic market of the exporting country cannot be used as the basis for the normal value. 13

14 II.A.3. CASE STUDY 2 The following analysis was taken from the same determination examined in Case Study 1, of the EC investigation targeting imports of colour TVs from China, the Republic of Korea, Malaysia and Thailand. Like product In the course of the investigations, it was established that CTVs originating in or exported from the countries concerned and destined for the Community, shared the same basic physical and technical characteristics and end-uses as CTVs manufactured and sold by the Community industry on the Community market. It was also found that there were no differences between the CTVs produced and sold in the countries concerned, including Turkey which was used as an analogue country, and those exported to the Community, which were both identical to the CTVs manufactured and sold by the Community industry on the Community market. These products are therefore alike within the meaning of Article 1(4) of the basic Regulation. The third product is the "like product in the importing country". This is the product manufactured and sold by the domestic industry in its domestic market and plays a central role in an injury determination. Why is the "like product in the importing country" relevant in an anti-dumping investigation? The "like product in the importing country" is relevant because the AD Agreement requires that an investigating authority determine the impact of the dumped imports on the state of the domestic producers of the like product in the importing country. The Agreement further requires that the investigating authority evaluate the state of that industry by examining a number of factors such as the development of sales of the like product manufactured by the domestic industry. The injury determination cannot therefore be carried out unless an investigating authority determines whether the domestic industry produces a product "like" the "product under consideration". Here again, the same definition of "like product" applies. That is, "product under consideration" and the product in the importing country are like where they are identical or where the latter has "characteristics closely resembling those of the product under consideration". The factors to be examined are those we have seen above. What if the product manufactured and sold by the domestic industry in the importing country is neither identical nor it has characteristics closely resembling those of the product under consideration? Based on the AD Agreement, in this case there is no domestic industry producing the like product, and hence no basis for an anti-dumping investigation or the imposition of anti-dumping measures. That said, if an industry to produce the like product in the importing market is in the process of establishment, but has not yet come on stream, there is a possibility to carry out an investigation, and impose a measure, on the basis of "material retardation of the establishment of an industry." 14

15 II.A.4. ILLUSTRATION 2 Let's explain this with an example: For instance, suppose that the companies in the importing country being injured produce radios and they complain that due to cheap imports of colour TVs, consumption of radios has fallen, causing them to be materially injured. Can the domestic industry of radios request the initiation of an anti-dumping investigation on colour televisions? No, because radios are neither identical to the product imported - i.e. the product under consideration, the colour TV - nor do they have characteristics closely resembling those of colour TVs. Hence, in this case the domestic producers of radios could not request the initiation of an investigation against colour TVs. Nor could any injury determination be supported based on injury suffered by producers of radios due to imports of colour TVs. If, however, investors are actively engaged in trying to establish a new industry to produce colour televisions in the importing country, an investigation potentially could be conducted to determine whether "establishment" of the industry were being "materially retarded", and in case of an affirmative outcome, a measure could be applied. There are, however, no guidelines in the AD Agreement for how to analyse allegations of material retardation. 15

16 III. DETERMINATION OF DUMPING IN BRIEF Dumping is established by comparing the "normal value" and the "export price". Generally, the normal value is the price at which the like product is sold for consumption in the market of the exporting country. The export price is the price at which the exporter sells the product to the importing country. The AD Agreement states that a product is to be considered as being dumped where the export price of the product exported from one country to another is less than the normal value of that product, which normally is the comparable price, in the ordinary course of trade, for the like product when destined for consumption in the exporting country. Any dumping calculation will include the following four steps: The observed or constructed export price; The observed or constructed normal value; The adjusted normal value and the adjusted export price (reflecting adjustments to ensure comparability); and The margin of dumping. These steps will be examined in the following sections. III.A. DETERMINATION OF THE EXPORT PRICE Although the term "export price" is not defined in the AD Agreement, the export price will normally be the price charged by the exporter for the product when exported to the importing Member. III.A.1. EXCEPTIONS The AD Agreement recognizes that in certain circumstances, the price from the exporter to the importer, or to a third party, may not be reliable. One such circumstance is where there is no export price. This could occur, for example, where the product is sold on consignment (i.e., the selling price is not fixed until the product is actually sold to a purchaser in the importing country), or transferred to a related entity for further processing before sale in the importing country. Another situation where the price charged by the exporter may not be a reliable indicator of the export price to be used in the dumping calculation is where the exporter sells the product to a related importer. This happens very often where large companies, such as Sony, Samsung, General Electric, etc. are involved in investigations. Another circumstance that can lead to price unreliability is a "compensatory arrangement" between the exporter and the importer or a third party, such as where the exporter gives discounts, refunds or rebates, after the export transaction has taken place. 16

17 In such cases, the AD Agreement provides for an alternative method of determining the export price to be used in the dumping calculation, namely a "constructed export price". Such a price is to be calculated on the basis of the price at which the imported products are first resold to an independent buyer in the importing country. If the imported product is not resold to an independent buyer, or is not resold as imported, the authorities may determine a reasonable basis on which to construct the export price. III.A.2. EXAMPLE: AN INVESTIGATION OF PARACETAMOL - EXPORT PRICE We will explain each of the steps in the calculation with the help of an example. Member A initiated an investigation concerning imports of paracetamol from Member B. The sole producer of paracetamol in Member B is "Paracetamol PLC". This company provided a response to the questionnaire it received from the investigating authority in Member A. The investigating authority will have to calculate the margin of dumping. The first step will be to calculate the export price. As a general rule, the margin of dumping is to be calculated by comparing weighted average export price to the weighted average normal value. Thus, the first step is to calculate these weighted averages. This process is shown in the table below for the export price: Number of Sale Date Gross Price CIF (USD/Kg) Quantity (Kg) Price times quantity Total Weighted Average Export Price (PxQ) Q 7.33 Table 1: Calculation of the weighted average export price The investigating authority does not have any indication that the export prices for the 12 transactions reported in the above table might be affected by an association or compensatory arrangement between Paracetamol PLC and the importer. The investigating authority will therefore determine the export price on the basis of the price paid by the importer. In this simple example, the weighted average export price (WAEP) will be 7.33 USD/kg. The next step is to calculate the normal value. We now turn to that issue. 17

18 III.B. DETERMINATION OF THE NORMAL VALUE III.B.1. GENERAL RULE The AD Agreement recognizes three possible options to determine normal value: Preferred basis: Domestic prices in the exporting country Exception: Export price to a third country Note: No hierarchy between these two Exception: Constructed value in the exporting country Figure 2: Options for the determination of the normal value in an anti-dumping investigation The first, preferred, basis in the AD Agreement for determining normal value is, as discussed above, the price at which the exporter sells the like product "when destined for consumption in the exporting country", i.e. in the exporter's domestic market. When normal value cannot be determined on that basis, two alternatives are available: the price charged by the exporter in another country (known as third country market price); or a constructed value obtained by adding to the cost of production of the like product in the country of origin a reasonable amount for selling, general and administrative expenses and for profits. (This is referred to as constructed normal value). III.B.2. ILLUSTRATION 3 Let's suppose that the Gambia initiates an anti-dumping investigation regarding imports of wheat flour from the United States. How will the investigating authority in the Gambia obtain information on the prices at which the US exporters sell wheat flour in the US domestic market? The Gambian investigating authority will be asking the US wheat flour exporters to provide that information. The investigating authority also may request information on the prices of US exports of wheat flour to third country markets, as well as on the cost of production, selling costs and profits for US production of wheat flour. In this way, the investigating authority would have all of the information necessary to determine normal value on any of the bases provided for in the AD Agreement. Normal value will be based on sales prices of the like product in the domestic market unless any of the following situations arise: There are no sales of the like product in the domestic market of the exporting country - recall the case of ski boots in Malaysia! There are sales in the home market but in low volume - normally where home market sales represent less than 5% of the volume of export sales to the importing Member. This situation arises frequently in 18

19 countries with small domestic markets. (The AD Agreement, however, encourages the use of a lower ratio where the evidence demonstrates that such lower ratio nevertheless would allow for a proper comparison.) There is a "particular market situation" that justifies not using the prices of sales of the like product in the domestic market The AD Agreement, however, does not define what might constitute such a "particular market situation". Sales in the home market are not "in the ordinary course of trade" - again, this term is not defined by the AD Agreement. The following situations might lead to findings of sales not being in the ordinary course of trade: There are sales between related parties. Similar to the situation that may arise when calculating the export price, there are many instances where the exporter sells in its domestic market through related parties. In such instances, a question may arise whether the price charged to the related purchaser is reliable. If not, that price might be rejected as the basis on which to calculate normal value. There are sales below cost: Under certain circumstances, the AD Agreement allows sales below cost to be treated as not being in the ordinary course of trade, and thus disregarded, i.e. not included, in the normal value calculation normal value. There are liquidation sales: An example would be sales of clothing at the end of a season at very low prices. There are sales to employees Where, because of the occurrence of any of the above situations, normal value cannot be determined on the basis of domestic sales prices, an investigating authority will have to proceed with any of the alternative methods explained below. III.B.3. EXCEPTIONS Third country market price. The AD Agreement contains few rules on how to calculate normal value on the basis of export prices to a third market, other than that the third country should be "appropriate", and that the export price to that country should be "representative". Members determining normal value on this basis have developed understandings about these terms. Thus, for instance they generally test the volume of exports to a given third country against the volume of exports to the importing Member, i.e. the Member conducting the investigation. If the volume sold to the third country is very low in comparison to the volume of exports to the importing Member, they may not select the sales to that third country as a basis for the determination of normal value. (This test, which although not provided for in the AD Agreement is applied by numerous Members, is similar to the so-called "home market sufficiency test" provided for in the AD Agreement, whereby home market sales shall be considered to be of sufficient quantity for the establishment of normal value if they are at least five per cent of the sales of the product to the importing Member.) Constructed normal value (CNV). As far as constructed normal value is concerned, as noted this is not a price charged by the exporter but rather is a value calculated by the investigating authority in the importing Member, exclusively for the purpose of its anti-dumping investigation. The AD Agreement tells us that to obtain the CNV, an investigating authority must sum up the cost of production, and reasonable amounts for selling, general and administrative costs and for profits. The AD Agreement contains rules on how costs should be 19

20 calculated. It also establishes a set of rules for the calculation of the selling, general and administrative costs as well as of profits. III.B.4. OUR INVESTIGATION OF PARACETAMOL - NORMAL VALUE In our investigation, the next step will be to determine the price at which Paracetamol PLC sold paracetamol in its domestic market. For that purpose, the investigating authority of Member A asked Paracetamol PLC to provide information on its selling prices of paracetamol in Member B (i.e., on its home market prices). The information received from Paracetamol PLC is set forth in the table below. Again, the weighted average needs to be calculated, and this process is as follows: Number of Date Gross Price Quantity Price times Sale CIF (Kg) Quantity (USD/Kg) Total Weighted Average Normal Value (P*Q) Q 6 Table 2: Calculation of the weighted average normal value As we see in the above table, Paracetamol PLC made 12 sales in its domestic market, each of 20 kg. Thus, in total it sold 240 kg. of paracetamol in the domestic market. As we have seen in the earlier section, the amount exported to Member A was 120 Kg. The volume sold in the domestic market therefore is twice the volume exported, i.e., far more than the 5 per cent threshold below which home market sales could be considered to be insufficient as a basis for determining the normal value. Therefore, sales in the domestic market cannot be disregarded on account of low volume. In addition, let us assume that there is no "particular market situation", or any reason to believe that any of the domestic market sales transactions was not in the ordinary course of trade. Hence, the normal value in this 20

21 particular case will be based on the prices at which Paracetamol PLC sold the like product in its domestic market. The weighted average normal value (WANV) therefore will be 6 USD/kg. We now have calculated the weighted average (observed) export price (USD 7.33) and the weighted average (observed) normal value (USD Assuming that no adjustments are required to either of these values, the result of comparing them would be that there is no dumping (because the export price is higher than rather than lower than the normal value. It may be, however, that one or both of these values needs to be adjusted to ensure that they are fully comparable before the dumping margin can be calculated. This process is discussed in the next section. III.C. FAIR COMPARISON OF NORMAL VALUE AND EXPORT PRICE IN BRIEF In general terms, the AD Agreement requires that the comparison of the export price and the normal value be "fair". More specifically, the AD Agreement requires that the comparison between export price and normal value must be made at the same level of trade, normally at the ex-factory level, and in respect of sales made at as nearly as possible the same time. Furthermore, based on the results of numerous WTO disputes, we know as well that using "zeroing" in dumping calculations violates inter alia the "fair comparison" rule. ("Zeroing" is assigning a value of zero to any individual negative margins of dumping - i.e., where the export price is above, rather than below, the normal value - in calculating the weighted average margin of dumping for the investigated product.) IN DETAIL Why does the AD Agreement require that a comparison be made at the same level of trade, and normally at the ex-factory level? Concerning level of trade, this is because differences in levels of trade (trader, wholesaler, distributor, retailer, end user, etc.) may affect prices. In particular, a producer typically will vary its prices based on level of trade because the producer's selling costs will vary from one level of trade to another. For example, a producer selling directly to end-users generally would need to make many small sales (incurring selling costs on each one) to reach a given total sales volume, while a producer selling to a trader or wholesaler would need to make fewer, larger sales to reach that same sales volume, and thus would economise on the total cost of selling that volume. Similarly, the requirement normally to compare the export price and the normal value at ex-factory level is aimed at avoiding the distorting effect that factors such as transport, insurance, etc. may have on the comparison of the export price and the normal value. 21

22 III.C.1. ILLUSTRATION 4 Let's look at an example: Suppose that a producer sells urea in its home market at USD 150/ton. The same company exports urea at USD 120/ton to Member A. If we compare these two figures we find that the export price (USD120/ton) is lower than the normal value (USD 150/ton), yielding an apparent dumping margin of USD 30/ton. However, let's suppose that the domestic buyer has asked the producer to deliver the product to the buyer's warehouse, and that the cost of transporting a ton of urea from the factory to the warehouse is USD 50/ton. This is therefore the amount that the producer of urea will have to pay to a transport company to bring the urea from the factory to the warehouse. Thus, what the producer/exporter earns for a ton of urea sold in the domestic market is not USD 150/ton, but USD 100/ton. Let's further suppose that the importer of the good in Member A also has requested that the producer/exporter deliver the urea to the importer's warehouse, and that the cost of transport is USD 20/ton. Again, we deduct the cost of transport and we find that the producer earns USD100 for a ton of urea exported to Member A. In this example, when we deduct the cost of transport from the selling prices in both markets, we find that the export price is equal to the normal value. What does this mean? It simply means that, when we adjust the prices on both sides of the dumping calculation to remove cost elements such as transport, in order to arrive at the "real" prices as required by the AD Agreement, the apparent margin of dumping disappears. That is, there is no dumping. Adjusting - in the circumstances contemplated by the AD Agreement is not optional for the investigating authority; it is the authority's obligation. The obligation exists even if it might be difficult, in a particular case, to make the required adjustments. What sorts of factors require an adjustment? Those affecting price comparability. The AD Agreement names a few: conditions and terms of sale (this includes transport, insurance, credit, guarantees/warranties, packaging, etc.); taxation; levels of trade; quantities; and physical characteristics. But the AD Agreement also requires that any other differences that are also demonstrated to affect price comparability must be adjusted. And it further requires the authorities to indicate to the parties what information is necessary to ensure a fair comparison (i.e. what information is needed to make a given adjustment.) The above example where an adjustment was needed involved differences in transport costs. Another commonly-needed adjustment is for differences in credit costs. A producer may sell "at sight" in its domestic market, while on its export sales it may give 30 days credit. In this case, the investigating authority could adjust the export price by deducting from it the cost relating to the credit provided by the exporter. In this example, no adjustment would be made to the normal value because "at sight" implies no cost to the producer. As another example, imagine that the product is auto CD players. For the export market these units are equipped with an anti-theft device. By contrast, auto CD players sold in the domestic market do not have such a device due to the low rate of theft in that country. The products are therefore different in terms of physical characteristics. An investigating authority could only compare the export price with the normal value after having adjusted for the difference (i.e., the anti-theft device) between the units produced for the two markets. Finally, the AD Agreement states that authorities must indicate to the parties in question what information is necessary to ensure a fair comparison and must not impose an unreasonable burden of proof on those parties. 22

23 III.C.2. OUR INVESTIGATION OF PARACETAMOL - ADJUSTMENTS We have calculated the weighted average (observed) normal value and weighted average (observed) export price in earlier sections. As noted above, if we compare these weighted averages - normal value of USD 6/kg and export price of USD 7.33/kg - no dumping will be apparent. The export price is higher, not lower, than the normal value. Now, let us change the example. exporter has stated the following: In particular, now let's suppose that in its questionnaire response, the "Domestic sales: In the domestic market we give our clients 90 days to pay. The interest rate that banks charge us for shortterm borrowing is 10% per year (0.028% per day). We also deliver the paracetamol to the warehouse of our clients. The cost of transport is 0.38 USD/kg. Export sales: On our export sales, we give our clients 180 days to pay. The interest rate that banks charge us for short-term borrowing is 10% per year. We export on a CIF (cost plus insurance and freight) basis. The total cost of transport and insurance amounts to 2.50 USD/kg." Based on this information, the investigating authority will need to make appropriate adjustments to both the normal value and the export price to take account of the credit, transport and insurance costs that are embedded in the prices charged in the home market and export market. We'll start with normal value. IN DETAIL Adjustments to normal value and export price Based on the information in the exporter's questionnaire response, two adjustments to the normal value are needed. The first relates to the credit costs. The producer has incurred certain credit expenses given that it allows its domestic customers 90 days to pay their bills. How much should we deduct from the weighted average price calculated above (USD 6/kg)? Different investigating authorities calculate credit costs in different ways. One possibility is by applying the following formula: Where: "v" is the value of the sale "i" is the interest rate "d" is the number of days given to pay "365" is the number of days in a year 23

24 The calculation thus would be: ((USD 6/kg) * 0.1 (10% interest rate) * 90 (days given to pay))/365 = USD 0.15/kg Furthermore, an adjustment for the domestic transport and insurance cost is warranted. The amount is provided by the exporter in its questionnaire response: USD 0.38/kg. Having calculated the two adjustments, the final step is to obtain the adjusted normal value, by deducting the credit, transport and insurance costs from the USD 6 weighted average price per kg of each transaction. This gives us a weighted average adjusted home market price of: USD 6.00/kg - USD USD 0.38 = USD 5.47 Thus, the adjusted weighted average home market price, that is, the weighted average normal value, is USD The next step will be to calculate the weighted average export price. The explanation given above for the calculation of the adjustments for credit, transport and insurance costs also applies here. With regard to the credit costs, since the length of credit terms given on export sales is different from that given on sales in the domestic market, we will have to compute the amount again. The formula will be: (USD 7.33/kg) * 0.1 (10% annual interest rate) * 180 days to pay) / 365 = USD 0.36/kg Transport and insurance costs are taken from the questionnaire response of the exporter. They are much higher than those charged in the domestic market because, for exports, the product has to be brought from the factory to the port, loaded into a ship, and shipped to the port in the importing country. The amount to be deducted, as reported in the questionnaire response, is USD 2.50/kg. To arrive at the adjusted weighted average export price, we must deduct the USD 0.35/kg and USD 2.50/kg from the weighted average selling price for exports of USD 7.33/kg, as follows: USD USD USD 2.50 = USD 4.47 Thus, the weighted average export price is USD 4.47/kg. We are now ready to calculate the dumping margin! 24

25 III.D. CALCULATION OF THE DUMPING MARGIN IN BRIEF The AD Agreement contains rules governing the calculation of dumping margins. In the usual case, the AD Agreement requires either the comparison of the weighted average normal value to the weighted average of all comparable export prices, or a transaction-to-transaction comparison of normal value and export price. IN DETAIL The following slide shows the basic formula for calculating the margin of dumping in percentage terms: ( ) Figure 3: Basic formula for calculating the margin of dumping Important note Normally, the margin of dumping is expressed as a percentage of the adjusted export price; this is why the difference between the adjusted normal value and the adjusted export price is divided by the adjusted export price. III.D.1. OUR INVESTIGATION OF PARACETAMOL - THE DUMPING MARGIN We have determined that the adjusted weighted average normal value is USD 5.50/kg and that the adjusted weighted average export price is USD 4.48/kg. Let's apply the above formula for calculating the margin of dumping in percentage terms: ( ) / 4.48 = 22.8% Given that the result is a positive number (the normal value is higher than the export price), we have found that the product under investigation is dumped, and that the margin of dumping is 22.8 per cent. Thus, one of the necessary elements for application of a measure - dumping - has been found. A word on "zeroing" Our example above is very simple, as only a single, uniform product is involved. Sometimes, however, the product covered by an anti-dumping investigation may have a number of different presentations, or "models", which are sufficiently distinct from one another that each requires its own margin calculation. (That is, the differences are too significant to be addressed via adjustments.) An example might be fresh, chilled and 25

26 frozen salmon in all presentations (whole, gutted, head-on, head-off, filleted or otherwise cut in pieces, etc.). It could be that the production processes (i.e., cost structure) and pricing of these different presentations or models are sufficiently diverse that each should have its own dumping margin calculation, while ultimately a single, weighted average dumping margin should be calculated for each producer (covering all of the models that it produces). In such a case, it may be that for some models, no dumping (in fact, negative dumping) is apparent, because for that model the weighted average export price exceeds the weighted average normal value. For other models, there may be dumping. The question becomes how to calculate the overall weighted average dumping margin for all of the models combined, and in particular, how in the weight-averaging across models to treat the models where no dumping/negative dumping is apparent. There has been a lengthy series of WTO disputes about this issue. Some Members have applied the practice of "zeroing" the negative margins (i.e., assigning them a value of zero in the weight averaging), on the grounds that a negative margin signifies an absence of dumping ("zero" dumping). Other Members instead factor the full negative margin into their weighted average calculations. The Appellate Body ultimately has ruled in a series of cases that "zeroing" violates various provisions of the AD Agreement, including the "fair comparison" provisions. Thus, the full value of the negative margins of dumping needs to be included in calculating the overall weighted average margin of dumping. The example below illustrates this: Weighted average dumping margin Quantity Margin * Quantity Model 1 8.5% Model 2-5.3% Model 3 9.1% Weighted Average overall margin of dumping, all models: With zeroing (NOT PERMITTED) Without zeroing INDIVIDUAL DUMPING MARGINS AND SAMPLING The AD Agreement requires that in principle, when anti-dumping duties are imposed, a dumping margin be calculated for each exporter known to the investigating authority. Is an investigating authority required to calculate an individual margin of dumping for each exporter even where there are many of them (e.g., hundreds or even thousands)? No. In those cases, the AD Agreement allows an investigating authority to select a sample, and to calculate individual margins of dumping only for the exporters included in the sample. However, the AD Agreement also sets forth rules to determine how the margin of dumping for the nonsampled exporters must be calculated. 26

27 NEXT STEPS Calculation of the dumping margin is not the "end of the road", but only the first step, for the application of an anti-dumping measure. As discussed in the first section, for an anti-dumping measure to be applied, the authorities also must make a determination that the domestic industry producing the like product in the importing country is experiencing injury due to the dumping. We therefore now turn to examining the state of the domestic industry. EXERCISES: 3. How is the dumping margin calculated? 27

28 IV. DETERMINATION OF INJURY AND CAUSAL LINK After calculating the margin of dumping, the next step is to determine whether the domestic industry producing the like product in the importing country is suffering injury as a result of the dumped imports. IN BRIEF The determination of injury and causal link will require us to go through a number of steps: Determination of what is the "like product" produced by the "domestic industry"; Identification of which producers constitute the domestic industry; Determination of whether the domestic industry is suffering injury (including data collection and analysis); Determination of whether the injury suffered by the domestic industry is caused by the dumped imports. These are the basic questions on the injury side of an anti-dumping investigation. IV.A. LIKE PRODUCT We have examined at some length above the concept of "like product". It is important to recall that a product identical to, or with characteristics closely resembling those of, the imported product must be produced in the importing Member. Otherwise, it would not be possible to proceed further with the examination of injury. Let's imagine for instance that apples are the product imported at dumped prices. There are, however, no producers of apples in the importing country. Indeed, bananas are the only fruit produced in the importing country. Before being able to consider whether domestic banana producers are injured by imports of dumped apples, an investigating authority first would have to ask whether bananas, not being identical to apples, nonetheless are "like" apples in the sense of having characteristics closely resembling those of apples. To answer to this question, the investigating authority would have to consider all relevant factors, as mentioned in the previous section. These would need to include in the first place the products' physical characteristics, and also might include raw materials, production process, end use, and users' perceptions,. Let's suppose that the result of the examination is the (perhaps obvious) conclusion that bananas do not have characteristics closely resembling those of apples, i.e., that bananas are not a "like product" to apples. What would this mean? It would mean that no like product is produced in the market of the importing country, and hence that there is no domestic industry to examine. As a consequence, the investigating authority would not have a basis for an injury investigation or determination. Thus, because it would be impossible to make two of the three determinations required for the imposition of an anti-dumping measure (i.e., the determinations of injury and causation), the investigation on imports of apples would have to be terminated immediately, without the imposition of measures! 28

29 IV.B. DOMESTIC INDUSTRY IN BRIEF The AD Agreement defines the term "domestic industry" to mean "the domestic producers as a whole of the like products or those of them whose collective output of the products constitutes a major proportion of the total domestic production of those products". Importers that do not themselves produce the like product therefore cannot be considered "domestic producers". IN DETAIL Why do we need to identify a "domestic industry"? Because the investigating authority will have to determine the existence of injury in respect of a particular domestic producer or group thereof producing the product that is "like" the investigated product. For that purpose, information will have to be sought from that or those domestic producers. While conceptually the domestic industry is composed of all the producers of the like product, the AD Agreement opens the door to considering less than 100 per cent of the domestic producers as the "domestic industry". In this case, the determination of injury will be based on information relating to these producers. DOMESTIC PRODUCERS "AS A WHOLE" OR THOSE ACCOUNTING FOR "A MAJOR PROPORTION OF TOTAL DOMESTIC PRODUCTION" The AD Agreement provides for two bases on which the domestic industry can be defined. First, and clearly the most accurate, is "domestic producers as a whole", that is producers accounting for 100 per cent of domestic production of the like product. There may be situations in which it is difficult or impossible to gather data from, or even to identify, all domestic producers of the product, however. For example, it could be a highly fragmented industry with hundreds or thousands of small producers. The AD Agreement thus also allows the domestic industry to be defined as those domestic producers whose collective output of the like product accounts for "a major proportion" of total domestic production of that product. In fact, the AD Agreement does not express a hierarchy between these two bases for identifying (and thus gathering and analysing data on) the domestic industry. That said, panels and the Appellate Body have ruled that where the domestic industry is defined for purposes of an injury investigation as less than all domestic producers, care must be exercised that the industry is not defined in a way that would introduce a bias into the results of the investigation. In particular, it is not allowed to exclude entire categories of producers of the like product from the domestic industry definition. For example, it is not permissible to exclude producers of one presentation of the like product, where more than one presentation is produced). Nor is it permissible to define the industry as only including "cooperative" domestic producers, or those willing to provide data, or those supporting the application. EXCLUSION OF CERTAIN DOMESTIC PRODUCERS FROM THE DOMESTIC INDUSTRY The AD Agreement recognizes that in certain circumstances, it may not be appropriate to include all producers of the like product in the domestic industry (even where the industry is defined, in principle, as the domestic 29

30 producers as a whole). industry": In particular, Members are permitted to exclude from the definition of "domestic Producers related to the exporters or importers under investigation; or Producers who are themselves importers of the allegedly dumped product. The AD Agreement contains its own definition of when a producer is to be considered "related" to an importer or an exporter. IV.C. INJURY INJURY AS A LEGAL CONCEPT IN BRIEF The AD Agreement does not define the term "injury", although it does specify that three types of injury may be found in an anti-dumping investigation: "Material" injury to a domestic industry; Threat of material injury to a domestic industry; or Material retardation of the establishment of a domestic industry. IN DETAIL The AD Agreement does not define any of the above types of injury but it tells an investigating authority what it must examine in order to determine whether a domestic industry has suffered "material" injury or threat thereof. The AD Agreement does not provide any specific rules on how a determination of "material retardation" must be conducted, and "material retardation" cases are rare. We turn now to the rules applicable to injury investigations. General requirements applicable to injury determinations: As an overarching obligation, the AD Agreement requires that a determination of injury must be based on "positive evidence" and involve an "objective examination" of: The volume of the dumped imports; The effect of the dumped imports on prices in the domestic market for like products; and The consequent impact of these imports on domestic producers of such products. The AD Agreement develops in detail what is expected, at a minimum, from an investigating authority with respect each of the above items: With regard to the volume of the dumped imports, an investigating authority must consider whether there has been a significant increase in dumped imports, either in absolute terms or relative to production or consumption in the importing Member. 30

31 Important note It is not required that the volume of dumped imports has increased, but rather that the investigating authority consider whether there has been any such increase. However, a decrease in the volume of dumped imports might be a factor weighing against a finding that the domestic industry has suffered injury due to dumped imports. It also should be noted that the increase might be in "absolute" terms, or "relative" to production or consumption. Thus, there might be situations where there is no absolute increase in dumped imports because, for instance, the total consumption of the like product has decreased, but dumped imports, in relative terms (to production or consumption) might have increased. As far as the effect of the dumped imports on prices is concerned, an investigating authority must consider whether there has been a significant price undercutting by the dumped imports as compared with the price of the like product produced in the importing Member, or whether the effect of such imports is otherwise to depress prices to a significant degree, or to prevent price increases that otherwise would have occurred, to a significant degree, for that like product. EFFECT OF DUMPED IMPORTS ON THE PRICES OF DOMESTIC LIKE PRODUCTS Thus, an investigating authority must examine whether dumped imports have any effect on the prices at which the domestic industry sells the like product. As noted, these "effects" can be in the form of "price undercutting", "price depression" or "price suppression". In some cases, "price undercutting" might be observed together with, for instance, "price suppression". Let's look at an example: Price DLP COP DLP Price Dumped I Price domestic like product Cost of production, domestic like product Price dumped imports Figure 3: Relationship of prices of dumped imports, and costs and prices of domestic like products 31

32 While the AD Agreement does not set forth any particular methodology to assess "price undercutting", generally the term means selling at a lower price. Thus, assessing whether there is price undercutting by dumped imports involves comparing the price of those imports with the price at which the domestic industry sells the like product in the domestic market. In the example in Figure 4, in the most recent period of time, i.e. 2011, the price of dumped imports is lower than the price of the domestic like product (140 vs. 145), i.e., there appears to be price undercutting by the dumped imports during that period. Furthermore, the data presented show that the domestic industry's cost of production has been rising steadily over the entire period for which data are available. Under this circumstance, the domestic industry could have been expected to increase its selling prices in order to pass on the cost increases. However, in the above case we see that this did not happen. Rather, at the same time that unit production costs increased from 125 to 145, the domestic industry's selling price decreased from 150 to 145. In the meantime, the price of dumped imports dropped from 200 to 140. One possible conclusion is that the price pressure from the dumped imports has prevented the domestic industry from raising its prices, which otherwise it would have done. In other words, this could be a case of "price suppression"? Finally, we observe that the domestic industry's selling prices decreased during the 5-year period, from 150 to 145. This could be an indication that the dumped imports have depressed the domestic industry's prices to a significant degree. It should be noted, however, that the AD Agreement does not contain any methodology or guidance as to how to assess whether there is "price suppression" or "price depression" due to dumped imports, or as to when any such suppression or depression is "significant". IMPACT OF THE DUMPED IMPORTS - THE CONDITION OF THE DOMESTIC INDUSTRY Finally, the examination of the impact of the dumped imports on the domestic industry concerned must include an evaluation of all relevant economic factors and indices having a bearing on the state of the industry, including: Actual and potential declines in sales, profits, output, market share, productivity, return on investments, or utilization of capacity; Factors affecting domestic prices; The magnitude of the margin of dumping; Actual and potential negative effects on cash flow, inventories, employment, wages, growth, ability to raise capital or investments. The AD Agreement provides that this list is not exhaustive, and that no one or several of these factors will necessarily be determinative in a given case. Nevertheless, in any analysis of injury, all 15 of the above factors must be considered (i.e., analysed). Investigating authorities should treat injury investigations on a case by case basis, with no single factor or group of factors necessarily being decisive in all cases. Each investigation involves a unique set of facts, and thus all of the above mentioned factors must be evaluated, and considered within the overall context in which the domestic industry operates. This context could include business cycles, product life cycles, conditions of competition among domestic and other competitors, or other circumstances. These considerations should be 32

33 part of the assessment by the investigating authorities as to whether the industry is suffering injury (as opposed, for example, to experiencing typical business upswings and downswings). The ultimate conclusion as to whether or not the domestic industry is being injured may rest on fewer than all of the obligatory, listed factors, but it must be apparent that all were examined. SPECIFIC REQUIREMENTS FOR DETERMINATIONS BASED ON THREAT OF MATERIAL INJURY A finding of current, present material injury is not necessary for an anti-dumping measure to be applied. Rather, a finding that there is a threat of material injury is one of two alternative bases for a measure (material retardation of establishment of an industry being the second). "Threat" implies that if certain changes take place - notably increases in the volume or rate of the dumped imports, reductions in their prices, etc. - material injury will occur in the imminent future. Although the threat of injury analysis is necessarily prospective in nature, the AD Agreement emphasizes that a threat determination cannot be based "merely on allegations, conjecture or remote possibility", but must instead be based on facts. Furthermore, "the change in circumstances which would create a situation in which the dumped imports would cause injury" needs to be "clearly foreseen and imminent". The AD Agreement sets forth a non-exclusive list of specific factors to be considered (in addition to the 15 listed injury factors discussed above) in determining whether the domestic industry faces a threat of material injury, including: The rate of increase in dumped imports A significant rate of increase of dumped imports into the domestic market indicating the likelihood of a substantially increased importation. Capacity Sufficient freely disposable capacity of the exporter, or an imminent, substantial increase of that capacity, indicating the likelihood of substantially increased dumped exports to the importing Member's market, taking into account the availability of other export markets to absorb any additional exports. Price Suppressing or Depressing Effects Whether the dumped imports are entering the market at prices that will have a significant depressing or suppressing effect on domestic prices, and would likely increase the demand for further imports. Inventories Inventories of the imported product being investigated, whether held in the country of export or in the importing investigating country. (A buildup of inventories of the domestically-produced like product also might be relevant.) As noted, a determination of threat of injury cannot be based solely on these threat-specific factors, but also must take into account all 15 of the factors pertaining to the condition of the domestic industry, discussed in the previous section. Furthermore, the AD Agreement requires that, where an affirmative injury determination is based on the threat of material injury, "the application of anti-dumping measures shall be considered and decided with special care". This provision suggests that an even higher level of scrutiny is expected in investigations predicated on 33

34 threat of, as compared to actual, material injury. Thus, the investigating authorities should provide a particularly thorough discussion of the analysis and rationale underlying a decision to impose anti-dumping duties in any final determination that is based on a determination of threat of material injury, setting forth the factual evidence leading to the conclusion that injury would be imminent if no anti-dumping action were taken. MATERIAL RETARDATION OF THE ESTABLISHMENT OF AN INDUSTRY As noted, the AD Agreement provides for a third possible basis for an injury finding, material retardation of the establishment of an industry. Formally, all of the requirements for analysing the state of the domestic industry (the 15 listed factors, the question of the impact of dumped imports on the condition of the domestic industry, etc.) are applicable in material retardation cases, but there is no specific guidance in the AD Agreement as to how to do this, and there have been very few anti-dumping measures based on material retardation. It is clear that in practice, many of the listed factors may be of limited relevance (for example, level of production may not be able to be analysed if the industry in the process of establishment has not yet started producing). Furthermore, the AD Agreement contains no definition of what constitutes an industry "in establishment". Thus, investigating authorities should proceed very cautiously in considering allegations of material retardation due to dumped imports. IV.D. DEMONSTRATION OF A CAUSAL LINK Let's suppose that the investigating authority of Member B has found that the imported goods from Member A are dumped and that the domestic industry of Member B is suffering injury. This still is not a sufficient basis to apply an anti-dumping measure. Rather, the final requirement under the AD Agreement is a demonstration that the dumped imports have caused injury, and this demonstration must be based on an examination of all relevant evidence. While imports need not be the only cause, or even the predominant cause, of any injury found to exist, there must be a causal relationship. Furthermore, as part of this analysis, the authority must ensure that it does not attribute to the dumped imports any injury caused by known factors other than those imports. In respect of the causation and non-attribution requirements of the AD Agreement, the Appellate Body has ruled that the investigating authority must demonstrate on the basis of the evidence that there is an affirmative genuine and substantial relationship of cause and effect between the dumped imports and the injury. In addition, in conducting the non-attribution analysis, the investigating authority must separate and distinguish the injury caused by other factors from that caused by dumped imports. The AD Agreement does not specify particular factors or give detailed guidance on how the relevant evidence is to be evaluated. Again, recalling the overarching obligations, the injury/causation analysis must be based on positive (i.e., robust, verifiable) evidence, and involve an objective examination of the volume of dumped imports and their effect on domestic prices, and the consequent impact of those imports on the condition of the domestic industry. All of the factors discussed in the preceding section are to be taken into account in this analysis. In addition, the AD Agreement lists some possible factors to examine in the consideration of known factors other than dumped imports that may be causing injury. These factors are: Volumes and prices of imports not sold at dumped prices. Returning to the example referred to above, let us suppose that the dumped imports from Member A represent 15% of the total imports in Member 34

35 B and that the prices of the other, non-investigated imports (from countries C, D and E) are considerably lower than the price of the dumped imports. The question that one could pose is whether injury suffered by the domestic industry is due to the much larger volume of imports from countries C, D and E (which, not being investigated, cannot be treated as "dumped"), rather than to the dumped imports from Member A. (There is no clear-cut right or wrong answer to this example, it is simply meant to point to issues and questions that would need to be explored in the "non-attribution" analysis. Contraction in demand and changes in the pattern of demand. One example might be hula hoops, which were a fad during a particular year, but which went out of fashion the following year. The sharp drop in the domestic industry's production, employment, profits etc. for hula hoops might be solely attributable to this contraction in demand, and not to dumped imports. Developments in technology. An example could be portable digital storage devices, from floppy disks to micro disks to data sticks. A decline in production, employment, profits etc. for one kind of device might well be due to its displacement in the market by a next generation device, which uses different technology. Export performance and productivity of the domestic industry. Let's suppose that a domestic industry historically has exported 90% of its output of a product, because the domestic market is very small. Let's further suppose that the industry's exports cease abruptly, due to new import restrictions in the export markets. At the same time, the industry starts facing dumped imports in its domestic market, and its selling price in that market declines by 3 per cent. An investigating authority finds that the imports are dumped and that the domestic industry is injured. The question, however, is whether the dumped imports have contributed in a real and substantial way to the injury, or whether in fact the injury is exclusively a function of the disappearance of the export markets. The AD Agreement contains no guidance on how either the causal relationship or non-attribution is to be established. In practice, different Members use different methodologies, ranging from detailed qualitative analyses of prices, costs, other conditions of competition to sophisticated econometric models. While there is no one right approach, the most important thing is for the investigating authority to provide in its reports and determinations a detailed explanation and justification, based on the factual evidence on the record, relating to both the finding of affirmative causality and the nonattribution analysis. EXERCISES: How is a determination of injury to the domestic industry reached? How is a causal link between dumped imports and injury established? 35

36 V. PROCEDURAL REQUIREMENTS V.A. INTRODUCTION The AD Agreement contains detailed procedural requirements on a number of issues, which are applicable at different stages of the proceeding. These are contained in among others, Articles 5 (Initiation and Subsequent Investigation), 6 (Evidence), 7 (Provisional Measures), 8 (Price Undertakings), 9 (Imposition and Collection of Anti-dumping Duties), 10 (Retroactivity), 11 (Duration and Review of Anti-dumping Duties and Price Undertakings), 12 (Public Notice and Explanation of Determinations), 13 (Judicial Review), 14 (Anti-dumping Action on Behalf of a Third Country), 15 (Developing Country Members), 16 (Committee on Anti-dumping Practices), 17 (Consultation and Dispute Settlement), 18 (Final Provisions), and Annexes I (Procedures for the On-The-Spot Investigations Pursuant to Paragraph 7 of Article 6) and II (Best Information Available in Terms of Paragraph 8 of Article 6). TIP Before plunging into those provisions, it must be stressed that anti-dumping investigations conducted by WTO Members must, at a minimum, comply with the requirements contained in the Agreement. WTO Members are free, and in some cases even encouraged by the AD Agreement itself, to establish additional procedural requirements. Should a Member fail to respect certain provisions of the AD Agreement, Article 13 allows an interested party to request the judicial review of final determinations and reviews of determinations, before an independent judicial, arbitral or administrative tribunal maintained for that purpose within the jurisdiction of the importing Member. Thus, the investigating authority may be found to have violated provisions pertaining to the use of anti-dumping measures under the national legislation implementing the AD Agreement. In addition, under the AD Agreement, an affected Member can use the WTO dispute settlement mechanism if it considers that the authorities of the importing Member have violated one or more substantive or procedural provisions of the AD Agreement. Whether under national procedures or WTO dispute settlement, findings of violation may require the measure to be removed. It is therefore essential, even for Members that are new users of anti-dumping, to comply strictly with all of the requirements set forth in the AD Agreement. 36

37 STAGES OF AN INVESTIGATION Any investigation may be conceptually and chronologically divided into the following stages: Initiation Information gathering [Preliminary determination] Further investigation Verification of information Final determination DURATION: Normally 12 months; cannot exceed 18 months Figure 4: Stages of an investigation Any anti-dumping measure, either in the form of anti-dumping duty or price undertaking, can only be imposed after an investigation consistent with the AD Agreement has been conducted. In a nutshell, this entails a number of distinct steps or stages, the first of which is that an investigation be initiated. Once this has been done, a number of obligations will apply, including informing the parties (exporters, importers, domestic producers, etc.) of the information that they need to submit. This starts the process of gathering the information to be used in the investigation. The examination, verification and analysis of this information will form the heart of the investigation. Relevant information that is found to be complete and accurate will be taken into account in the preliminary and/or final determinations, which are the next steps in the investigation process. Only if there are affirmative final determinations of dumping, injury and causation, can a measure ultimately be applied. Those determinations in turn must be based on the information before the investigating authority. In particular, the focus of the authority's analysis of the information is to determine whether the conditions for imposing measures are met. Should the relative determinations be affirmative, the government of the Member may then consider whether it is in its interest to impose a measure, whether any measure should be for the full amount of the dumping margin or something less, whether to accept a price undertaking from an exporter in lieu of imposing a duty, and similar questions. Should the Member determine, however, that the dumping margin is zero or "de minimis", that the volume of imports is negligible, that the domestic industry is not injured, or that dumped imports have not caused injury, there will be no legal basis for any measure. In such cases, the Member must terminate the proceeding without the imposition of measures. The entire process of investigation, from initiation to final determinations of dumping, injury and causation, and decisions related to the application of measures, normally lasts for at least 12 months. While some investigations proceed more quickly than this, in many other cases the process will take longer. The AD Agreement makes some allowance for this, providing that in exceptional circumstances an investigation can last as long as, but in no case more than, 18 months following initiation. 37