Australia – Anti-Dumping Measures on A4 Copy Paper: Opening a Door to More Anti-Dumping Investigations

Abstract This paper analyzes the Australia – Anti-Dumping Measures on A4 Copy Paper panel report, the second recent WTO dispute to involve a challenge to Indonesia's paper industry. The Indonesian paper industry benefits from reduced-cost inputs because of the Indonesian government's influence and subsidies over the timber and pulp market. The report offers the first interpretation of ‘particular market situation’ under Article 2.2 of the WTO's Anti-Dumping Agreement. At the same time, it raises questions regarding the appropriateness of using anti-dumping measures to address what are fundamentally subsidy issues. While the panel ultimately found that Australia's measure was inconsistent with Article 2.2, the paper shows that the panel's interpretation of ‘particular market situation’ increases the relative attractiveness of using anti-dumping duties instead of countervailing measures. Two key points on the welfare implications of the decision can be made. The first relates to the motivations of the Australian paper industry and the imperfectly competitive market in which Australian Paper operates. The second is the importance of challenging subsidies rather than imposing anti-dumping duties where the subsidies in question have negative environmental effects.


Introduction
The global paper industry has received increased scrutiny in recent years as concerns regarding greenhouse gas emissions and the deleterious effect that the timber and pulp industries have on the environment mount. Paper is made from pulp which is, in turn, made from timber, and pulp constitutes a key input for the paper industry. The tensions over the global paper industry have manifested themselves in land rights struggles across the globe, from Brazil to Chile to Mozambique to Indonesia to Quebec, as land is taken by government and private companies from indigenous and local communities and converted into timber plantations. Where governments subsidize their domestic timber and pulp industries, the cost of the final product, paper, will be reduced.
Australia -Anti-Dumping Measures on A4 Copy Paper (Australia-Copy Paper) is the second WTO dispute involving challenges to Indonesia's paper industry as a result of the Indonesian government's support of its timber and pulp industries. 1 Indonesia's paper industry has increased its market share in recent years, and in 2019 it was the sixth largest producer in the world. 2 The dispute raises questions regarding the merits of approaching what are fundamentally subsidy issues through the use of anti-dumping (AD) duties rather than through countervailing measures (CM), continuing a long-standing debate amongst economists, lawyers, and policymakers. In addition, and most significantly from a legal perspective, the decision marks the first time a WTO panel has ruled on the meaning of 'particular market situation' as per Article 2.2 of the WTO Anti-Dumping Agreement (AD Agreement). 3 The debate on the feasibility and implications of using AD to tackle subsidies is not new to the literature (see, for instance, Crowley and Hillman, 2018;Fischer and Meyer, 2020). In this paper we focus on a novel dimension of this debate and argue that the interpretation of 'particular market situation' in Article 2.2 of the AD Agreement offers an avenue for AD to be considered as an (optimal) trade policy tool for a downstream industry to tackle a subsidy issue in a foreign upstream market. As we demonstrate, the particular panel interpretation in Australia-Copy Paper has the effect of enhancing the downstream industry's expected payoff from choosing AD instead of CM, therefore increasing the relative attractiveness of an AD policy response.
Economically, the market situation in the upstream timber and pulp industries is not only relevant for the choice of which quantities to use in determining the normal value of A4 paper in Indonesia, but also represents the most important factor shaping the pricing behavior of Indonesian A4 paper producers and its implications for competitors in foreign markets. The decision to 'dump', if any, might not be solely dictated by the paper producers' attempt to undercut the market in Australia, but also by the upstream industries, whose prices are impacted by the government's policies.
Building on this observation, we interpret the complaint and the Australian Anti-Dumping Commission (ADC) trade policy decisions with respect to Indonesia as an AD response to a subsidy issue in an upstream market. We argue that the specific interpretation of the legal text with respect to the 'particular market situation' affects the choice between (downstream) AD and other trade policy toolsmainly CMwhen facing a subsidy issue in a foreign market with implications for domestic downstream markets.
Even though a formal assessment of the broader economic and welfare implications of using AD as a trade policy response to a downstream foreign subsidy lies outside the scope of the present paper, we argue that further research in this direction is necessary, especially when the interpretation of the 'particular market situation' under Article 2.2 of the AD Agreement increases the attractiveness for downstream industries to petition for AD investigations. This paper makes two key observations regarding the circumstances and welfare implications of the decision, the first relating to Australia's imperfectly competitive market and the motives behind its AD investigation, and the second relating to the environmental consequences of Indonesia's government supports for its timber and pulp industries.
The paper is organized as follows: Section 2 sets the scene by providing a general background of the case. Section 3 discusses in greater detail the legal features of the 'particular market situation' within the legal structure and jurisprudence of the GATT. Section 4 offers a simple economic framework to think about the 'particular market situation' and its interpretation as factors affecting the attractiveness of downstream AD in the presence of upstream subsidies. We apply our legal and economic analytical tools to the Australia-Copy Paper in Section 5. Section 6 concludes by discussing two implications of our analysis. 2 IDN Financials, Indonesia's pulp and paper are in the top 10 in the world, 28 January 2019, www.idnfinancials.com/archive/news/22291/Indonesias-pulp-and-paper-are-in-the-top-10-in-the-world#:∼:text=JAKARTA.,pulp%20was%20produced% 20per%20year. 3 The GATT panel in EC -Imposition of Anti-Dumping Duties on Imports of Cotton Yarn from Brazil (EEC-Cotton Yarn (GATT)], ADP/137, 4 July 1995 addressed 'particular market situation' regarding Article 2:4 of the Tokyo Round Anti-Dumping Code.

Background on the Australian Anti-Dumping Measure
Australia-Copy Paper arose from a complaint by Paper Australia Pty Ltd (Australian Paper) to the Australian ADC that two Indonesian producers, PT Indah Kiat Pulp and Paper Tbk (Indah Kiat) and PT Pindo Deli Pulp and Paper Mills (Pindo Deli), were engaging in dumping activities in the Australian market of A4 copy paper. With respect to Indonesia, the investigations of the ADC lead to the implementation of AD duties of 35.4% for Indah Kiat and 38.6% for Pindo Deli. Crucial to the calculation of the dumping margins as well as for the case brought by Indonesia to the WTO was the nature of the timber and pulp market in Indonesia. Wood pulp is a key raw material input in the production of A4 copy paper, accounting for 60-65% of the total cost (ADC, 2017, 151).
While Australian Paper is the only domestic A4 copy paper manufacturer in Australia (ADC, 2017), Indah Kita and Pindo Deli are part of a more saturated domestic Indonesian paper industry. This includes Asia Pulp and Paper/Sinar Mas Group, one of the largest paper companies in the world, which was the target of the United States' investigation that resulted in the US-Coated Paper (Indonesia) dispute (US-Coated Paper (Indonesia), para. 2.5).
In its petition to the ADC, Australian Paper claimed that a particular market situation existed in the Indonesian paper market as a result of which domestic sales of A4 copy paper in Indonesia were not suitable for determining normal values. In particular, Australian Paper alleged that A4 copy paper prices in Indonesia were kept artificially low due to Indonesian government influence and subsidies on raw material inputs (Australia-Copy Paper, para. 7.9). The ADC agreed with Australian Paper, finding that the Indonesian government had programs and policies in place that exerted a significant influence over the Indonesian timber and pulp industries. These policies included providing land for plantations and an export ban on logs (ADC, 2017, 146).
The Indonesian timber and pulp market is strongly intertwined with the Indonesian government. Almost all of the forest land in Indonesia is owned by the government of Indonesia, making them the predominant supplier of standing timber (US-Coated Paper (Indonesia), para. 7.49). As the panel in US-Coated Paper (Indonesia) noted, where the government's market share is that high, their position in the market 'approaches that of a sole supplier of the goods' (para. 7.50). In its investigation in that case, the US Department of Commerce found that standing timber was provided by the government of Indonesia to the paper industry, as well as to four other industries (para. 7.136). With the dispute in US-Coated Paper (Indonesia) bringing up issues of both antidumping and countervailing duties, the issue of upstream subsidies was discussed by the panel in the context of the financial contribution by the Indonesian government in the form of log sales, which were uniformly subject to the export ban. There, however, the Panel did not directly address the question of whether the log export ban constituted a financial contribution, leaving the subsidies issue unresolved.
In Australia-Copy Paper, unlike in US-Coated Paper (Indonesia), there was no challenge to the subsidy activity in the upstream timber and pulp market specifically, although the ADC found that domestic price for A4 copy paper was artificially low due to the support of the government of Indonesia for the development of timber plantations and its prohibition on the export of timber logs (Australia-Copy Paper, para. 7.12). This finding unsurprisingly mirrors the finding of the US International Trade Commission in US-Coated Paper (Indonesia).
Demonstrating that there is a subsidy requires showing four elements: (1) a financial contribution, (2) by a government or public body, (3) that confers a benefit, and (4) is specific (SCM, Article 1). To impose anti-dumping duties, on the other hand, it is only necessary to show that the export price was lower than the price of the good in the home market. This is generally an easier determination to make. The Subsidies and Countervailing Measures (SCM) Agreement is meant to address government intervention, while the AD Agreement should address behavior by private firms. As Crowley and Hillman have aptly noted, where there is an interaction between market economies and non-market economies, there will be distortions that may implicate private firm behavior even where they are caused by government intervention (Crowley and Hillman, 2018, p. 196).
In its investigation, the ADC found that there was dumping in relation to A4 copy paper from two Indonesian paper plants. The ADC then turned to determining the amount of duties to impose on them. Here, the ADC opted to disregard the Indonesian domestic sales prices, finding that there was a particular market situation that made Indonesian domestic A4 copy paper prices unsuitable for determining normal values, and calculated a constructed normal value using the export prices of third-party countries. Indonesia challenged this methodology, arguing that Australia incorrectly interpreted the phrase 'particular market situation' (Australia-Copy Paper, para. 3.1).
The panel ultimately found that Australia's measure was not in conformity with its obligations under the AD Agreement (paras. 8.3-8.4). Specifically, while the panel found that the ADC did not act inconsistently with Australia's obligations under Article 2.2 in finding that there was a 'particular market situation' in the Indonesian domestic market for A4 copy paper, it did find that Australia's measure was inconsistent with Article 2.2, first sentence, because ADC disregarded domestic sales of paper by the two companies in question without properly determining that such sales 'did not permit a proper comparison.' It is this finding that is at the heart of our analysis.
3. Article 2.2 of the AD and the Interpretation of 'Particular Market Situation' under the GATT Article 2 of the AD Agreement provides the key legal provisions on when a determination of dumping can be found. Article 2.1 defines dumping. Article 2.2 crucially provides for using constructed values in determining the margin of dumping in certain circumstances where the normal value cannot be calculated. The text of Article 2.2 reads: 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.
Numerous parties have invoked Article 2.2 to justify alternative methodologies in a variety of situations, but surprisingly, only the EC in the GATT dispute EEC-Cotton Yarn and Australia in the present dispute have constructed values based on the perceived existence of a 'particular market situation'. Before we look at how the panel interpreted 'particular market situation' in the Australia-Copy Paper dispute, it is worth briefly mentioning the interpretation provided by the GATT panel in EEC-Cotton Yarn.
The panel in EEC-Cotton Yarn interpreted Article 2:4 of the Tokyo Round Anti-Dumping Code, which Article 2.2 of the AD Agreement largely mirrors, particularly with respect to the language regarding 'particular market situation'. 4 Crucially, the GATT Panel found that the test for having recourse to use of constructed value or third country sales 'was not whether or not a "particular market situation" existed per se', but rather that a '"particular market situation" was only relevant insofar as it had the effect of rendering the sales themselves unfit to permit a proper comparison' (EEC-Cotton Yarn (GATT), para. 478). This requires that there be something intrinsic in the very nature of the sales that dictates that they cannot allow for a proper comparison (para. 478). Further, the GATT panel found that even if something were relevant under Article 2.4, it would be necessary to establish that domestic sales were affected by it in such a way that a proper comparison could not be made (para. 479).

Why Might a Country Choose AD Rather than CM?
In order to understand the rationale of Australia in opting to pursue AD rather than CM in this case, we have to look more generally at what motivates countries in making these choices. Applying a simple 2 countries-2 sectors economic framework that models the choice between AD and CM in a setup that generalizes the key features of the Australia-Copy Paper case allows us to assess the interpretation of 'particular market situation' given by the panel report and to discuss its economic and legal implications.
In this generalized framework, an economy consists of two trading jurisdictions, A and B, and two industries I and F producing respectively the intermediate and the final good in both countries. B subsidizes its I industry and, as a consequence, the input price for the final good industry in B decreases. We remain completely agnostic as to the potential existence of a rationale for B's final good industry to engage in dumping activities in A. 5 It is immaterial here why they may or may not choose to engage in dumping activity. Instead, we posit that, because of the subsidy, and because of a positive pass-through from the upstream sector to the output price in F, the export price of F's output produced in B and exported to A decreases. In other words, the upstream subsidy results in a reduction of the output price of the industry in the exporting country, which means that the cost of the exported final good decreases and becomes cheaper in country A.
There are two alternative actions available to industry F in country A: petitioning for an AD investigation or petitioning for an investigation regarding subsidies related to the final good imported from B. Each action can be mapped to an expected payoff, which represents what industry F in country A is expecting to get by choosing to petition for the related trade policy response. In particular, industry F in country A will choose to petition for AD investigations if the expected payoff from that action exceeds the expected payoff from petitioning for a subsidies investigation.
The expected payoff of the two actions will also depend, among other things, on all sorts of relevant deterministic and/or stochastic parameters in the economy as well as from strategic interaction with other agents (e.g., industry I in A, industry F in B, government agencies, the WTO etc.). Keeping things simple, we focus on what we think matters most for examining how the interpretation of 'particular market situation' in Article 2.2 of the AD Agreement may affect the decision to start a petition for an AD investigation.
The crucial parameter for our discussion is given by the dumping margins, which in turn depend on what AD law prescribes in terms of the methodology that can be used to compute them. The requirement is to start by comparing the export price paid by A's consumers for the output of industry F in country B, either with the price charged for the same good in the exporting country B (after removing all costs that materialize after the good leaves the factory), or with the average total costs for industry F in B. In our setup, both these values are affected by the subsidy. More precisely, other things being equal, the presence of B's subsidy to its I industry makes these values smaller than they would have been without it. Whichever way the dumping margin is calculated, the subsidy to upstream inputs decreases that margin. Intuitively, the lower the dumping margins, the lower the expected payoff of petitioning for AD investigations will be.
If this were the end of the story, an upstream subsidy would just make the dumping margins smaller, thus decreasing the expected payoffs of AD. Where things become interesting is where we start looking at other approaches to compute dumping margins. These alternative methodologies generally allow substantial discretion, which can be used to increase dumping margins on the part of the investigators. Article 2.2 discussed above introduces and disciplines the possibility of computing the dumping margins by comparing the export price with quantities other than the output or input prices in country B's market. This possibility is conditional, among other things, on the existence of a 'particular market situation' that 'does not permit a proper comparison' between the export price and the price/cost in country B's market.
An interpretation of Article 2.2 that allows an anti-dumping investigation to consider B's subsidy to its I industry as a particular market situation valid under Article 2.2 would allow country A to use constructed quantities, higher than the output or input prices in the B market, to compute the dumping margins. This possibilityassuming that country A is successful at zero cost in showing that the subsidy prevents a proper comparison as indicated in Article 2.2opens the door to computing higher dumping margins. If the upstream subsidy is considered as a particular market situation, it stops being a factor that reduces dumping margins and it actually becomes the trigger to apply higher discretion and compute higher dumping margins.
Other things being equal, any interpretation of the legal text that permits considering B's subsidy to its I industry as a particular market situation valid under Article 2.2 would make AD a relatively more attractive tool for the industry F in country A to tackle B's subsidy to its I industry. On the contrary, any interpretation that would not allow considering country B's subsidy to its I industry as a particular market situation valid under Article 2.2 would make AD a relatively less appealing option.
As a last element in our general framework, we consider the cost related to the requirement to show that the particular market situation 'prevents a proper comparison' as per Article 2.2. A higher cost associated to this requirement decreases the expected payoff of petitioning for AD. In fact, this cost might be high enough to cancel out the benefit accruing when using the particular market condition to get higher dumping margins. The cost parameter becomes relevant only under an interpretation of the 'particular market situation' that does not clearly exclude the possibility of considering upstream subsidies as a particular market situation. As we will see below, this is the case in the Australia-Copy Paper dispute. Figure 1 offers a graphical representation of our framework.

The Panel's Analysis of 'Particular Market Situation'
The panel's legal analysis with respect to the facts of Australia-Copy Paper provides us with an application of our economic framework. In this section, we examine the two key components of its legal analysis of Article 2.2: the interpretation of 'particular market condition' and whether domestic market sales did 'not permit a proper comparison'.

Particular Market Situation Analysis
In Australia-Copy Paper, Indonesia and Australia provided conflicting interpretations of 'particular market situation'. Indonesia argued that this relates to an 'exceptional set of circumstances affecting comparability of domestic market prices in such a way as to affect them unilaterally and, thus, prevent them from being compared to export prices'. Australia, on the other hand, argued that 'particular market situation' means 'any condition, state or combination of circumstances in respect of buying and selling of the like product in the market of the exporting country that is distinguishable and not general' (para. 7.20).
Applying our framework to this situation, Indonesia is the country whose upstream industry (the timber and pulpwood industries) is being subsidized, thus lowering the input price for its (downstream) paper industry. As discussed above, this lowering of input price should reduce both the domestic and export prices of the final goodthe A4 copy paper. To Indonesia, 'particular market situation' should not include this type of situation, but rather should be limited to situations where the domestic market prices are the only ones to be affected, thus requiring a constructed value due to the inability to compare domestic market prices to export prices.
The panel's analysis of Article 2.2 and 'particular market situation' in Australia-Copy Paper is comprehensive and adds significant clarity to an overlooked provision of the AD Agreement. The panel, in interpreting the phrase, focused on its ordinary meaning, centering the operative word 'situation' and noting that 'particular' and 'market' function as adjectives in this context. 6 The situation must have to do with the market, and the market situation must be particular. With the word 'particular' qualifying 'market situation', the panel found, from the dictionary definition, that the market situation must be distinct, individual, single, or specific, and that this requires a fact-specific and case-by-case analysis of the particular market situation. Furthermore, the panel found that the phrase did not 'lend itself to a definition that foresees all the varied situations that an investigating authority may encounter that would fail to permit a "proper comparison"'.
The panel viewed the choice of language by the drafters as deliberate -'particular market situation' was not a haphazard assortment of words. As such, the panel determined that the phrase cannot be broadly interpreted such that it identifies the comprehensive circumstances that make up the situation that the investigating authority may have to consider. This interpretation led the panel to conclude that 'the market situation must be distinct, individual, single, specific but that does not necessarily make it unusual or out of the ordinaryi.e., exceptional' (para. 7.22). With respect to our generalized framework, the panel's findings at this point did nothing to exclude upstream subsidies from the relatively broad set of situations that can qualify as 'particular market situations'.
Having established the definition of 'particular market situation', the panel turned to assessing whether situations that distort input costs, and particularly those that lower input costs, are excluded (para. 7.24-7.32). Indonesia argued that a particular market situation must be capable of preventing a proper comparison of domestic to export prices. The panel disagreed, finding that 'particular market situation' and 'permit a proper comparison' create a condition for disregarding domestic market sales. If domestic sales 'do not permit a proper comparison', then this must be 'because of the particular market situation' (this analysis derives directly from the plain text of Article 2.2). According to the panel, if domestic sales do permit a proper comparison, you cannot disregard them as a basis for normal value, whether or not there is a particular market situation, and regardless of the effects of that situation. From this, the panel found that it is not necessary that a situation be capable of preventing a proper comparison for a situation to constitute a 'particular market situation'. In short, 'particular market situation' is not tied to the inability of 6 Australia-Copy Paper, para. 7.21. The ordinary meaning analysis derives from the general rule of treaty interpretation as codified in the Vienna Convention on the Law of Treaties, 1969, Article 31. domestic sales to permit a proper comparison. Again, the panel's findings on this point continued in the direction of keeping the set of potential particular market situations largely unconstrained.
The panel also provided analysis of whether 'particular market condition' necessarily excludes situations that do not have an exclusively unilateral impact on domestic market sales (para. 7.33-7.40). Indonesia had argued that a situation that primarily affects the domestic market, but that also affects export markets to a lesser degree, does not qualify as a particular market condition. While the Panel found that 'market' in the phrase 'particular market situation' refers to the domestic market, because of the context of the preceding and subsequent clauses in Article 2.2, it also found that just because the situation arising in the domestic market of the exporting country that affects domestic sales also affects export sales does not exclude the applicability of 'particular market situation'. The particular market situation does not, therefore, have to only affect domestic sales. As the panel stated, '[t]he language of Article 2.2 focuses on domestic market sales simply for the reason that the provision is concerned with whether the domestic market sales are an appropriate basis for determining normal value, not because the effects of the underlying phenomena are necessarily exclusively unilateral in nature' (para. 7.39).
The panel's final finding on 'particular market situation' was that a situation arising in whole or in part from government action (as was the case with the Indonesian government's involvement with the timber and pulpwood sector, which constituted the inputs for paper manufacturing), is not necessarily disqualified from constituting the 'particular market situation'. The implications of the panel's final findings are clear: an upstream subsidy can be considered as a 'particular market situation' in the sense of Article 2.2. This begins to suggest incentives for a complaining party to follow the AD route rather than the CD route as per our general framework.

'Proper Comparison' Analysis
The second big piece of the panel's analysis was whether the ADC properly determined that domestic market sales did 'not permit a proper comparison'. This is particularly key to our analysis.
Here, the disagreement between Indonesia and Australia rested on whether distorted domestic sales prices could nevertheless 'permit a proper comparison with export prices' and therefore could not be disregarded as a basis for normal value, with Australia disagreeing that this could be the case (para. 7.62). As a baseline, the panel established that Article 2.2 not only requires a finding that the particular market situation exists, but also requires a further finding that the domestic sales do not permit a proper comparison because of the particular market situation. If there is a particular market situation, the investigating authority has to look to see whether there can be a proper comparison of domestic and export prices (para. 7.73). A proper comparison analysis consequently requires investigating authorities to compare domestic and export prices (para. 7.75). If a proper comparison of domestic and export prices is possible, then the domestic price must be used in determining the normal value.
What then is a 'proper comparison'? In the view of the panel, a purely numerical comparison between the domestic and the export prices is inadequate, and instead, a qualitative comparison of the domestic and export prices is required, since the phrase 'because of the particular market situation' requires focusing on how the particular market situation affects that comparison. In other words, the relative effect of the particular market situation on domestic and export prices must be assessed. To find out if a proper comparison is not permitted because of the particular market situation, 'the investigating authority must examine the domestic sales in order to determine whether a proper comparison between the two prices is permitted in spite of the effect of the particular market situation' (para. 7.76, emphasis added). Any such determination will necessarily be fact-specific and require a case-by-case analysis by the investigating authority, who must give a 'reasoned and adequate explanation of its conclusion' if, because of the particular market situation, a proper comparison of the domestic price and export price is not permitted (para. 7.77).
Of particular interest is the link Indonesia made between the Appellate Body's finding regarding domestic subsidies in the US-Anti-dumping and Countervailing Duties (China) case to the low-priced input that is used identically in producing merchandise for the domestic and export market (para. 7.77). The Appellate Body in that case found that where domestic subsidies are granted in market economies 'both the normal value and the export price will be lowered as a result of the domestic subsidy, so that the dumping margin should not be affected'. 7 The Appellate Body further found that under certain methodologies, a double remedy could, but would not necessarily, arise if both countervailing measures and anti-dumping duties were imposed. In Australia-Copy Paper, the panel noted that there are multiple factors that may affect how domestic prices and export prices are affected, even where there is an equal decrease in input costs (para. 7.80).
After providing its detailed analysis of a particular market situation and the 'proper comparison' analysis, the panel evaluated the ADC's AD duties. Ultimately, the panel disagreed with the methodology used by the ADC in determining whether normal value could be determined, since the ADC focused exclusively on domestic sales and domestic prices, without accounting for export prices, and without a determination of whether domestic prices could be compared with export prices, despite the effects of the particular market situation (para. 7.87). Instead, the panel noted that the effect of the particular market situation on the Indonesian market was caused solely by the decreased cost of purchasing or making pulp, a critical input for the paper-making process.
Simply put, the price of A4 paper in Indonesia decreased because of the decrease in the cost of pulp. As the panel found, the ADC should have examined whether domestic sales allowed a proper comparison between domestic prices affected by the decreased cost of pulp with export prices 'for which the pulp cost was presumably equally decreased' (para. 7.89). Whether you make paper for the domestic market or for the export market, if the paper is made using the same pulp, and that input has become less expensive, then presumably the cost of paper for the export market would also decrease. In this respect, the panel found that the ADC's failure to 'assess the effect of the particular market situation on the domestic price in relation to the effect on the export price when determining whether domestic prices permitted a proper comparison with those export prices' was inconsistent with Article 2.2 (para. 7.90-7.91).
In this instance, Australia, or country A within our framework, was not successful in showing that the subsidy prevented a proper comparison as indicated in Article 2.2, because the ADC failed to even examine export prices, and assumed that the mere existence of a subsidy prevented a proper comparison and warranted resorting to constructed values rather than using normal values to calculate the margin of dumping. Had it evaluated export prices and found that there was no possibility of a proper comparison due to differing domestic prices as compared to export prices despite the reduction in the cost of inputs for paper for both the domestic and the export markets, its determination would not have been found to be inconsistent with Article 2.2.
This gives rise to an important question: what would Australia have found had it looked at the relationship between the particular situation in the Indonesian market and export prices? Or, more generally, is it correct to expect an upstream subsidy, under certain circumstances, to have a different effect on downstream domestic and export prices such that a proper comparison between them is prevented precisely because of the subsidy? The answer to the general question is yes, absolutely. Especially in the case of homogenous downstream goods, the world price (the price at which the good is traded in international markets) might be completely unaffected by a shock to the input costs of one country's producers which instead alters the domestic price.
Whenever the interpretation of Article 2.2 allows upstream subsidies to be considered as a 'particular market situation', the possibility of obtaining higher dumping margins rests on the proof of a causal link from the subsidy to the non-capacity of domestic quantities to allow for a proper comparison with export prices.
Assessing the benefit from higher dumping margins is easier as these are more easily quantifiable. Assessing the cost of proving that the particular market situation does not allow a proper comparison is much more challenging. As a consequence, had the panel wished to decrease the relative appeal of using downstream AD to tackle upstream subsidies, it should have opted for an interpretation of 'particular market situation' in Article 2.2 that unambiguously excludes the possibility of considering upstream subsidies as a particular market situation. Keeping the door open to the use of 'particular market situation' referring to a subsidized upstream market in the hope that the expectation of high costs of proving that in fact the upstream subsidy prevents a proper comparison might not work. This because the expected value of such costs for the downstream industry petitioning for AD might be completely different and potentially much lower than that of other agents.

Critique and Implications
In its analysis of Article 2.2 and 'particular market situation', the panel in Australia-Copy Paper arguably opened the door to another avenue of AD investigations by making it relatively more appealing as an option compared to opening an investigation into subsidies. The tension between AD and subsidies in international trade discourse is longstanding. AD measures look only at the price differential, while subsidies investigations require a much more in-depth economic analysis of an actual trade measure.
A general analysis of the welfare impact of choosing AD as a trade policy response to a downstream foreign subsidy lies outside the scope of the present paper. Further research in this direction is necessary, especially when the interpretation of 'particular market situation' under Article 2.2 of the AD Agreement increases the attractiveness for downstream industries to petition for AD investigations. However, considering the specific case under analysis, two points on the welfare implications of using AD can be made.
The first point concerns the motives behind Australian Paper seeking protection. As noted above, Australian Paper is the only domestic manufacturer of A4 copy papers (ADC, 2017), accounting for slightly less than 50% of the A4 paper supply in Australia during the investigation period. Imports from the four countries -Brazil, China, and Thailand, in addition to Indonesiawhich were the object of ADC investigations covered 52% of the market, leaving imports from other countries a very small share of around 2%. Given the structure of the Australian domestic A4 copy paper supply, one potential rationale behind AD would be the protection of Australian Paper's dominant position. A similar situation of AD implemented in an imperfectly competitive market is described by Messerlin (1990).
Looking at the European Community (EC) chemical industry in the 1980s, Messerlin demonstrates that the two EC cartels of polyvinyl chloride (PVC) and low-density polyethylene (LdPE) manufacturers 'have used antidumping cases as de facto substitutes to exceptions from competition rules in markets difficult to cartelize because of foreign competition' and that antidumping actions turned out to be decisive for the survival of the cartels with clear negative welfare effects. If the same rationale motivatesmutatis mutandis -Australian Paper's petition for an AD case against foreign competitors, the implementation of AD duties would be likely associated with welfare losses. Under these circumstances, any factor potentially limiting the attractiveness of AD, would have positive welfare implications. An interpretation of the particular market situation that explicitly excludes upstream subsidies would act in that direction.
The second point relates to the effectiveness of using AD to rectify the negative effects of a subsidy issue. This is not the first trade dispute to involve partially integrated industries with subsidized components. 8 In Canada-Softwood Lumber, the US authorities took both AD actions and CM actions against the alleged Canadian subsidies in the same case. Similarly, in US-Anti-dumping and Countervailing Duties (China) the possibility that a dual remedy might arise if CM were imposed simultaneously along with AD duties was raised.
As we have discussed, Indonesia's paper industry has been the subject of another, related dispute: US-Coated Paper (Indonesia), where Indonesia challenged the imposition by the US of AD and CM on Indonesian paper imports. The US Department of Commerce determined in that case that Indonesia's government confers a benefit in violation of the SCM Agreement by providing standing timber for less than adequate remuneration and through its log export ban. The panel upheld the Department of Commerce's findings, indicating that the subsidies are ultimately the real issue at the heart of this case as well.
It is important to remember that while we have posited an economic framework that can be broadly applied to future situations where particular market situations may be invoked, the facts of the case have important implications, not just from an anti-dumping perspective, but also as a reflection of growing tensions with respect to the environmental pressures of the paper industry. The Indonesian government has taken intentional steps to develop its paper industry, including licensing out over 10 million hectares of forest lands for pulpwood plantations over the last thirty years (Environmental Paper Network, 2018, 39). These efforts have displaced communities, destroyed local ecosystems, and wiped out millions of hectares of rainforest. Since the 1990s, Indonesia has been draining peat swamps and converting them into pulp plantations, which has significant climate change consequences, due to the high levels of greenhouse gas emissions arising from drained peat (Environmental Paper Network, 2018, 39). The extent of illegal logging that is taking place in Indonesia is also of significant concern (Environmental Paper Network, 2018, 39). While subsidies can often be beneficial, and the SCM Agreement arguably takes an overly restrictive approach to subsidies, in the case where government subsidies drive an industry that is wreaking environmental havoc, such subsidies are problematic.
In situations like this, it is particularly important that the challenge be to the subsidies rather than using AD investigations as a means of rectifying situations where the price of exports from another country appears artificially low. An AD investigation will never be able to address the problematic aspects of government subsidies and following the approach taken by the panel in Australia-Copy Paper, there is an incentive for parties to choose an AD approach rather than a CD approach, since a particular market situation can exist even where a subsidy has an effect on both domestic sales and export sales.
For those two reasons, we would tend to welcome an interpretation of the 'particular market situation' that clearly decreases the incentives to use downstream AD as a trade policy response to an upstream subsidy issue, since this panel's finding compounds existing issues with AD as a trade remedy by further incentivizing states to pursue AD investigations rather than more laborious, but ultimately, arguably, more economically sounds subsidies investigations. AD has always been a surrogate remedythis decision only emphasizes that.