The World Trade Organization (WTO) Panel Report in Australia – Anti-Dumping Measures on A4 Copy Paper (Australia – A4 Copy Paper) marks a significant development of the multilateral rules on anti-dumping. Under certain circumstances, WTO agreements permit members to impose anti-dumping measures to counteract the injurious effect of dumping on domestic industries, typically through import duties. The Report is the first to examine in detail when an anti-dumping authority may determine that a “particular market situation” exists in the country of exportation under Article 2.2 of the WTO Anti-Dumping Agreement, potentially justifying the imposition of elevated remedial duties. The Report also develops the jurisprudence on how such remedies may be calculated, expounding the use of benchmark costs for the calculation of a constructed normal value (CNV) under Article 188.8.131.52. These doctrinal questions are central to the longstanding debate over how far the Anti-Dumping Agreement allows anti-dumping measures against state intervention and market distortions. On both fronts, the Australia – A4 Copy Paper panel created flexibilities for WTO members to respond to government-induced distortions. In doing so, the Report deviates considerably from the course set by the Appellate Body in the landmark EU – Biodiesel decision, which seemed to confine anti-dumping measures to responding to private action.Footnote 1 At the same time, the panel left open several important issues relating to the adoption of CNVs and the use of benchmarks for their calculation, leaving wide latitude for investigating authorities to inflate dumping margins in practice.
This dispute arose out of Australia's anti-dumping investigation of A4 copy paper exported from Brazil, China, Indonesia, and Thailand, leading to the imposition of anti-dumping duties on a range of exporters from these countries.Footnote 2 Australia subjected two major Indonesian exporters to the highest duties.Footnote 3 These higher rates largely resulted from the finding of Australia's Anti-Dumping Commission (Commission) that a particular market situation existed in Indonesia's A4 copy paper market due to ongoing government support for the pulp industry.Footnote 4 The Commission found that Indonesian intervention had artificially lowered the domestic price of paper, and hence relied on a counterfactual normal value for Indonesian A4 copy paper. In doing so, the Commission disregarded the actual cost of the key input incurred by the Indonesian producers or exporters—paper pulp. In its view, this cost did “not reasonably reflect a competitive market cost.”Footnote 5 Instead, the Commission resorted to external benchmarks based on the prices of pulp exported to China and Korea from several South American countries.Footnote 6 This use of surrogate production costs resulted in a higher CNV, and hence a finding of higher dumping margins, justifying higher anti-dumping duties.
Under Article 2 of the Anti-Dumping Agreement, a normal value should be calculated based on the domestic price of the subject goods in the country of exportation unless special circumstances exist. Article 2.2 identifies several such circumstances, and authorizes use of alternative proxies for constructing normal value:
When there are no sales of the like product in the ordinary course of trade in the domestic market of the exporting country or when, because of the particular market situation or the low volume of the sales in the domestic market of the exporting country, such sales do not permit a proper comparison, the margin of dumping shall be determined by comparison with a comparable price of the like product when exported to an appropriate third country, provided that this price is representative, or with the cost of production in the country of origin plus a reasonable amount for administrative, selling and general costs and for profits (emphasis added).
Thus, when an investigating authority determines that a country is in a particular market situation, it may disregard the domestic price of those goods and instead construct a normal value (under defined limits) for the determination of dumping margins. However, the Anti-Dumping Agreement does not define particular market situation, leaving substantial discretion in the hands of domestic investigators. Prior to this dispute, this concept was raised in only one General Agreement on Tariffs and Trade (GATT) disputeFootnote 7 and has never been interpreted by WTO tribunals.
Among WTO members, Australia was for a time the only major user of the particular market situation method, especially in anti-dumping actions against China. Instead, in relation to China, other members mostly relied on Section 15 of China's WTO Accession Protocol to achieve similar results—by allowing WTO members to treat China as a non-market economy in anti-dumping investigations and to use benchmark prices (as opposed to Chinese prices) in determining normal values.Footnote 8 While the continued invocation of Section 15 for this purpose is controversial,Footnote 9 frequent users like the United States and the European Union have increasingly turned to the particular market situation approach.Footnote 10 In almost all anti-dumping investigations over the past decade, the Australian authority has found China to be in a particular market situation in a variety of sectors.Footnote 11 The Chinese government has consistently challenged these conclusions, both within these investigations and through diplomatic channels. But it has never resorted to WTO proceedings.Footnote 12 Australia's A4 copy paper investigation was one of the few cases where the Australian authority found that a particular market situation did not exist in China.Footnote 13 While China had no incentive to challenge Australia's methodology in this investigation, Indonesia's WTO proceedings addressed China's longstanding concern.
In considering Indonesia's claim that the Australian authority had erred in finding the existence of a particular market situation in Indonesia's A4 copy paper market, the panel started by observing that the term was deliberately left undefined by the drafters and must thus be analyzed case by case (para. 7.21). The panel found that the term “cannot be interpreted in a way that comprehensively identifies the circumstances or affairs constituting the situation that an investigating authority may have to consider” (id.). Furthermore, the panel ruled that the phrase “do not permit a proper comparison” in Article 2.2 sets a separate condition that should not be considered in the determination of what may constitute a particular market situation (para. 7.27). Finally, the panel rejected Indonesia's argument that government action is, in principle, excluded from the coverage of anti-dumping remedies and should be addressed under the WTO Agreement on Subsidies and Countervailing Measures (SCM Agreement) (para. 7.53). In its view, “a situation arising from government action in whole or in part” may constitute a particular market situation (para. 7.55). Based on the above interpretation, the panel found that Australia's general approach to determining that Indonesia had a particular market situation was permissible (para. 7.57).
However, the panel found that the Australian authority did not assess whether the particular market situation identified had precluded a “proper comparison” between domestic and export prices. This assessment is necessary to “determine whether the domestic price can or cannot be used as a basis for comparison with the export price to identify the existence of dumping” (para. 7.74). The panel elaborated that artificially lowered input costs of the sort at issue can affect domestic and export prices to varying degrees, and a determination of the extent may involve consideration of various factors including “the prevailing conditions of competition in each market,” “the existing relationship between price and cost,” and the private commercial decisions of producers and exporters (paras. 7.76, 7.80). Australia did not go beyond determining the existence of a particular market situation to assess whether the distorted pulp cost would in fact make any comparison between domestic paper price and export price misleading (paras. 7.86–7.89). Therefore, Australia's substitution of a CNV for the domestic prices of the two Indonesian exporters failed to satisfy the requirements of Article 2.2 of the Anti-Dumping Agreement (para. 7.90).
The other major issue in this dispute concerned the construction of a normal value. A finding that a particular market situation has precluded a “proper comparison” between domestic and export prices merely justifies the adoption of a CNV. However, it is the calculation of the CNV that determines the magnitude of dumping margins. The level of a CNV typically hinges on the cost of production used for the calculation. Here too, the panel found that Australia's use of benchmark costs for the determination of a CNV fell short of WTO standards. The relevant standards are set out in Article 184.108.40.206 of the Anti-Dumping Agreement which provides:
For the purpose of paragraph 2, costs shall normally be calculated on the basis of records kept by the exporter or producer under investigation, provided that such records are in accordance with the generally accepted accounting principles of the exporting country and reasonably reflect the costs associated with the production and sale of the product under consideration (emphasis added).
Thus, there are two conditions for the determination of whether costs recorded by producers and exporters should be used for the calculation of a CNV. The first condition merely requires that costs be recorded in accordance with generally accepted accounting principles and hence provides no flexibility for consideration of whether recorded costs are distorted by state intervention. The second condition, according to the Appellate Body in EU – Biodiesel, requires that the cost records suitably and sufficiently reflect the actual costs incurred and does not allow for consideration of the reasonableness of the costs themselves.Footnote 14 Therefore, the “reasonably reflecting test” also provides no flexibility for considering whether state intervention distorts recorded costs. In Australia – A4 Copy Paper, however, the panel distinguished the EU – Biodiesel decision, finding that Australia's application of surrogate production costs was not based on the reasonably reflecting test, but was rather based on an assessment of the reasonableness of the recorded costs, that is, whether the costs were “competitive market costs associated with the production” (para. 7.102) (emphasis added). In other words, as the reasonably reflecting test does not provide room for consideration of the reasonableness of recorded costs, the panel believed that Australia was not applying that test so that the EU – Biodiesel decision does not apply (paras. 7.103–7.107). Instead, the panel held that the term “normally” may provide some flexibility for investigating authorities to consider the reasonableness of costs and to replace distorted production costs with a competitive benchmark. It ruled that there may be circumstances in which benchmark costs may be employed even though the two conditions of Article 220.127.116.11 are fulfilled (paras. 7.110–7.115). However, the panel found it unnecessary to determine the exact scope of these circumstances in this dispute (para. 7.116). The panel merely emphasized that authorities cannot utilize the flexibility provided by “normally” unless they have found that the two conditions under Article 18.104.22.168 are satisfied (para. 7.117). The panel held that Australia did not conduct a sufficient assessment of the two conditions and hence was not entitled to disregard the recorded costs based on the term “normally” (paras. 7.119–7.124).
* * * *
The panel addressed two important issues relating to the use of the particular market situation approach to justify the application of CNVs. First and foremost, the panel's ruling suggests that government intervention in a market may constitute a particular market situation. This is significant because it arguably expands the scope of the WTO anti-dumping rules for addressing market distortions caused by government actions. Prior to this decision, there was some controversy as to whether the use of CNVs to deal with non-market economies was confined to the extreme situations involving complete or substantial state monopoly of trade and price controlsFootnote 15 and to circumstances permitted under the WTO accession instruments of certain new members. For example, as discussed above, Section 15 of China's WTO Accession Protocol sets out China-specific rules that go beyond the Anti-Dumping Agreement allowing WTO members to treat China as a non-market economy in anti-dumping investigations. The panel's decision reads the Anti-Dumping Agreement more broadly to allow targeting all types and levels of government-caused market distortions. This scope would thus legitimate the use of the particular market situation approach by all WTO members against any economies. Lately, China itself used this approach to find that the U.S. energy and petrochemical sector had a particular market situation,Footnote 16 and is investigating whether a particular market situation exists in Australia's wine industry in an ongoing anti-dumping action.Footnote 17
Second, the panel's decision confirms that the existence of a particular market situation does not automatically justify the use of CNVs (in lieu of actual domestic price) to calculate dumping margins. Rather, authorities must take the further step of assessing whether the market situation has precluded a “proper comparison” between domestic and export prices. The panel's decision suggests a test of even-handedness. That is, if the situation concerned has lowered the two prices to the same degree, then the comparability of the domestic price would not be affected, which therefore must be used for the determination of dumping margins. However, if the situation has had a larger or exclusive impact on the domestic price, then a “proper comparison” between the two prices would be precluded, and the use of a CNV would be justified. This even-handedness approach should be welcomed. It would be unnecessary to resort to CNVs if a market situation causes the same level of distortions in domestic and export prices such that the two prices remain comparable. However, the panel has left uncertainties as to how authorities should apply the even-handedness test in determining whether a market situation has had an asymmetric impact on domestic and export prices. WTO tribunals will need to develop more specific rules for the “proper comparison” test to restrain the discretion of domestic authorities so as to avoid abuse of the particular market situation method.
In dealing with how a CNV should be calculated under Article 22.214.171.124, the panel's interpretation of the term “normally” has further developed the jurisprudence established by the Appellate Body in EU – Biodiesel. As noted above, the EU – Biodiesel decision has arguably sought to limit the application of anti-dumping measures to private activities as opposed to government actions, while pushing the latter to be addressed under the SCM Agreement instead.Footnote 18 In Australia – A4 Copy Paper, the panel's interpretation of “normally” has created the flexibility for the use of surrogate costs in the presence of government-induced price distortions and consequently for the use of anti-dumping measures in a way that deviates from the spirit of the EU – Biodiesel decision. While the panel did not clarify exactly what circumstances may fall within the scope of “normally,” it does not impose any substantive limitation either. The satisfaction of the two conditions under Article 126.96.36.199 as a precondition for utilizing the flexibility of “normally” is not a hurdle. In practice, authorities may simply accept the two conditions, i.e., that producers’ records are in compliance with general accounting standards and reasonably reflect the costs actually incurred, to trigger an assessment of whether the costs are competitive prices.
The final major issue arising from the panel's decision concerns the adjustments that would need to be made to any competitive benchmark to ensure that it reflects “the cost of production in the country of origin” under Article 2.2 of the Anti-Dumping Agreement. In EU – Biodiesel, the Appellate Body established that benchmarks that do not incorporate distortions in the market of the exporting country cannot reflect “the cost of production in the country of origin” and therefore must be adjusted.Footnote 19 Here too, the panel seems to deviate from this approach. Rather, it ruled that Article 2.2 “requires the investigating authority to consider available alternatives for replacing recorded costs so as to use the costs that are unaffected by the distortion to the extent possible” (para. 7.162, emphasis added). This suggests that the required adjustments to benchmarks are limited to the components of producers’ costs unaffected by government-caused distortions (para. 7.164), whereas the EU – Biodiesel decision has arguably required such adjustments to include all conditions in the relevant market including distortions caused by state intervention.Footnote 20 The panel's approach invites an improper comparison between CNV and export prices. Where a particular market situation exists in an upstream market, it typically affects the cost of production for domestic and export sales in the same way. In contrast, a benchmark cost used for the purpose of addressing the market situation would not include the distortion concerned. Therefore, an adjustment to the benchmark that also excludes the distortion would lead to a comparison between an undistorted CNV (which is typically a higher normal value) and a distorted export price. This would in turn result in a higher dumping margin. To avoid such inflation of dumping margins, a better approach would require an adjustment to the benchmark to ensure the cost used for the construction of normal value permits a “proper comparison” between the CNV and the export price. This approach would be needed not only in the typical cases where the distortion concerned has had the same effect on the production cost for both domestic and export sales, as illustrated above. It would also be needed in the unusual cases where a particular market situation does affect the cost of production of domestic goods to a larger extent than the cost of production of export goods. In such circumstances, the different degrees of impact on cost may respectively flow through to domestic and export prices of the end goods. When a benchmark is employed, an adjustment to the benchmark is needed to ensure the cost used for the construction of normal value accurately reflects the difference in the degree of absorption of the cost distortion into domestic and export prices. This would in turn require authorities to determine the exact difference in the degree of impact of the distortion on the two prices.
In sum, the Australia – A4 Copy Paper decision has developed the WTO's anti-dumping jurisprudence in significant ways while also leaving ambiguities around certain highly complex and controversial issues that will be ripe for consideration in future disputes. More specifically, the decision has created flexibilities for WTO members to tackle state intervention and market distortions by recourse to the particular market situation method as a pathway to the adoption of a CNV and subsequently competitive benchmark costs for the calculation of the CNV. In this sense, the decision has extended the scope of the Anti-Dumping Agreement in a way that the EU – Biodiesel decision sought to restrain. Notably, the panel's ruling that an adjustment to selected benchmarks (to reflect the cost of production in the country of origin) should exclude the market distortion concerned cannot be reconciled with the EU – Biodiesel decision. Having created the flexibilities above, the panel's decision leaves uncertainties on issues critical to constraining the flexibilities including how to determine whether a particular market situation has actually precluded a “proper comparison” between domestic and export prices (especially where the situation is in an upstream market), and the circumstances covered by the term “normally.” In the absence of a functional Appellate Body, it is unclear how future panels will address these outstanding issues. In the meantime, these flexibilities and uncertainties will leave considerable discretion to investigating authorities, leading to increasing application of the particular market situation method among WTO members. This will in turn lead to tit-for-tat anti-dumping measures, adding fuel to the current crisis in globalization and international cooperation on trade regulation.