Introductory remarks

By way of background, dumping shall be considered to exist if the export price is lower than its normal value. ‘Normal value’ is normally the domestic sales price. However, in certain circumstances, for example when the ‘particular market situation’ in the exporting country does not permit a ‘proper comparison’, investigating authorities may use other values in lieu of sales prices. In those circumstances, investigating authorities most typically calculate a ‘constructed normal value’, that is, a benchmark value calculated on the basis of the cost of production, plus selling, general and administrative expenses, and profits.

In recent years, investigating authorities have increasingly made findings that prices of inputs in the market of the exporting countries are distorted, artificially low due to governmental interventions (for instance, price controls, subsidies, or non-commercial conduct of state-owned companies), or otherwise do not reflect free-market conditions. On that basis, investigating authorities have rejected input prices – even though these prices are duly recorded in the financial records of the investigated companies – and replace them with other ‘correct’ values. These investigation practices vary between in their specific detail and rely on different legal terms and treaty language. However, these practices have in common that they typically increase the calculated normal value and, as a consequence, increase dumping margins or create them in the first place.

These practices are currently a very controversial question in anti-dumping case law and are also politically sensitive, because they are perceived to express (at least indirectly) a judgment about the exporting government's economic policies and because they potentially have a significant commercial impact. Moreover, traditionally, anti-dumping determinations have been perceived to focus on company-specific pricing behaviour, rather than on governmental policies. Panels and the Appellate Body have previously ruled one certain type of this practice to be inconsistent with certain terms of the Anti-Dumping Agreement. However, the caselaw is evolving and several interpretative questions remain unresolved.

This dispute is particularly important, because many investigating authorities increasingly invoke the concept of ‘particular market situation’ and this dispute marks the first time that a WTO panel has had an opportunity to rule on the meaning of this term.

Determination of the ‘particular market situation’

In this dispute, the Australian investigating authority determined that A4 copy paper prices in Indonesia were artificially low due to the Indonesian Government's influence on the timber market (through subsidies and export restrictions) and that therefore a ‘particular market situation’ existed in the Indonesian market. Indonesia contested Australia's determination of the existence of the ‘particular market situation’ under Article 2.2 of the Anti-Dumping Agreement, arguing that there are three situations that can never constitute a ‘particular market situation’ under Article 2.2: (i) circumstances that distort input costs; (ii) situations that do not have an exclusively unilateral impact on domestic market sales (meaning that a situation that impacts sales both on the domestic and export market cannot be a ‘particular market situation); and (iii) situations arising from government action.

The Panel found that none of these three situations is necessarily excluded from constituting the ‘particular market situation’ in the sense of Article 2.2 of the Anti-Dumping Agreement. Given that Indonesia's argument was purely legal, the Panel also found that Indonesia had not established that the Australian investigating authority improperly relied on the decreased cost for an input to determine that a particular market situation existed in the Indonesian market.Footnote ² This finding is important because it permits ‘particular market situation’ findings resulting from alleged government-induced market distortions, reflecting an increasingly widespread practice by national investigating authorities.

Whether the domestic sales affected by the ‘particular market situation’ ‘permit a proper comparison’

Another important question in this dispute concerned the phrase ‘do not permit a proper comparison’ in Article 2.2 of the Anti-Dumping Agreement. Recall that Article 2.2 permits not using domestic sales prices for determining normal value when those prices ‘do not permit a proper comparison’ with the export price. Australia argued that it is sufficient to determine domestic sales are ‘not suitable’ for use as the basis for normal value. In contrast, Indonesia asserted that Australia failed to make, or properly make, a determination that the domestic market sales affected by the particular market situation were not able to be compared with the export price, meaning that the ‘particular market situation’ affected domestic and export sales in different manner.

The Panel ruled in Indonesia's favour. It found that the phrases ‘particular market situation’ and ‘permit a proper comparison’ function together to establish a condition for disregarding domestic market sales as the basis for normal value, and also found that the circumstances in which ‘domestic sales do not permit a proper comparison’ must arise ‘because of the particular market situation’.Footnote ³ Thus, where a particular market situation was found to affect domestic market sales prices solely as a result of a decreased cost for an input and that input is used to produce merchandise for both the domestic and export markets, the investigating authority may not conclude, on that basis alone, that a proper comparison is not possible. This is because the decreased input cost would, in principle, affect the price of both the domestically sold and the exported product in an equal manner. Rather, the investigating authority must assess the effect of the particular market situation on the domestic price in relation to the effect on the export price.Footnote ⁴ Because the Australian investigating authority failed to conduct this analysis, it was found to have acted inconsistently with Article 2.2 of the Anti-Dumping Agreement.

Construction of the normal value of the exporters

Indonesia also challenged the Australian investigating authority's determination under Article 2.2.1.1 of the Anti-Dumping Agreement, which guides an investigating authority when it calculates cost benchmarks, in particular, as here, in constructing normal value. This ‘constructed normal value’ is based on the cost of production in the country of origin plus reasonable amount for administrative, selling and general costs, and for profits. The first sentence of Article 2.2.2.1 of the Anti-Dumping Agreement stipulates that these costs ‘shall normally be calculated on the basis of records kept by the exporter or producer under investigation’. Indonesia argued that an investigating authority is required to use the actual costs of the exporters (that is, those recorded in their financial and accounting records) unless these costs fail to meet one of the two conditions set out in the first sentence of Article 2.2.2.1. These two conditions are that (i) the financial records are in accordance with the generally accepted accounting principles (GAAP) of the exporting country, and that (ii) these financial records reasonably reflect the costs associated with the production and sale of the product under consideration.

In contrast, Australia argued that, because Article 2.2.1.1 contains the term ‘normally’, an investigating authority may disregard an exporter or producer's records even if the two conditions above are satisfied. The Panel agreed and found that, even where two conditions in the first sentence of Article 2.2.2.1 are satisfied, the recorded costs may nevertheless be rejected in some circumstances, on the basis of the term ‘normally’. However, in order to rely on the term ‘normally’ in a WTO-consistent manner, the investigating authority is required to ‘examine whether the records satisfy the two explicit conditions’ and, if they do, must also ‘provide a satisfactory explanation as to why, nonetheless, it finds compelling reason to disregard’ those records.Footnote ⁵ In this case, the Panel found that the Australian investigating authority had not examined whether both the first and second conditions set out in the first sentence were satisfied.Footnote ⁶ The Panel therefore found a violation of Article 2.2.1.1.

Construction of the cost of production

Indonesia challenged the rejection of the actual cost of hardwood pulp recorded in the two exporters' records not only under Article 2.2.1.1, but also under Article 2.2. This claim under Article 2.2 was distinct and separate from Indonesia's previously discussed claim under Article 2.2. that the investigating authority had improperly determined the existence of a ‘particular market situation’.

For this claim, Indonesia argued that – assuming that the investigating authority was entitled to disregard the exporter's recorded costs – the use of third-country export prices as a proxy for the actual costs of pulp of the Indonesian producers was inconsistent with Article 2.2, because it did not reflect the ‘cost of production in the country of origin’. Previous case law has found that this phrase does not necessarily preclude reliance on out-of-country benchmarks (for instance, international commodity price indices), provided that these benchmarks are then adjusted to simulate the price in the country of production (origin).Footnote ⁷ The Panel confirmed this case law, citing a previous finding of the Appellate Body Report in EU–Biodiesel on this point.Footnote ⁸ But Indonesia argued that, even though the Australian investigating authority had made adjustments to the Brazilian/South American export price, those adjustments did not yield a value that reflected the cost of production in Indonesia, contrary to Article 2.2. In contrast, Australia argued that its adjustments were adequate to reflect the cost of production of paper in Indonesia.

The Panel's finding under Article 2.2. was in two parts. First, the Panel made an apparently purely consequential finding that Article 2.2 had been violated as a direct consequence of the investigating authority's failure to comply with Article 2.2.1.1. Due to improperly rejecting recorded costs, the investigating authority was not entitled to resort to an out-of-country benchmark.Footnote ⁹ Second, the Panel then stated that, therefore, it was no longer necessary to make a finding on Indonesia's claim on whether the adjusted Brazilian other South American exporters' prices accurately reflected prices in Indonesia. Nevertheless, the Panel offered observations on this issue, in order to ‘assist the parties in resolving their dispute’.

With respect to one of the exporters, the Panel opined that the Australian investigating authority failed to explain why it had not deducted a profit margin from the Brazilian export prices, to approximate the price/cost of pulp produced by that producer ‘in-house’. For that same exporter also, the Panel also found that the investigating authority should have analyzed whether it was possible to adjust the cost of woodchips (the ‘upstream’ input for pulp), rather than the cost of pulp. This was because the alleged distortion of the domestic market on which the ‘particular market situation’ determination was based related to the timber market (upstream input market) and not the pulp market. In response to certain arguments by Australia, the Panel found that, even if data for this analysis were not easily obtained, the investigating authority should have addressed and explained these facts in its published report.Footnote ¹⁰ These panel findings are useful in that they may provide some guidance for investigating authorities when adjusting out-of-country benchmarks to reflect cost of production in the country of exportation.