Overview of WTO Jurisprudence in 2013

Abstract

This article presents an overview of the reports (judgments) of Panels and the Appellate Body of the World Trade Organization (WTO) circulated in 2013. For each report, we present the key findings on the most salient issues of systemic importance.

The views expressed in this article are entirely the authors’ own and should not be attributed to either the ACWL or to the Government of the Dominican Republic.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

WTO Jurisprudence in 2013 at a Quick Glance

In 2013, only four Panel reports were circulated. Of these four Panel reports, one was appealed, namely, European Communities – Measures Prohibiting the Importation and Marketing of Seal Products (DS400, DS401) (“EC – Seal Products”). The appeal in this dispute was not finalised in 2013 and is therefore not covered in this review. In addition, the Appellate Body in 2013 decided an appeal from a Panel report that had been circulated in 2012, namely, Canada – Certain Measures Affecting the Renewable Energy Generation Sector/Canada – Measures Relating to the Feed-in Tariff Program (DS412, DS426) (“Canada – Feed-in Tariff Program”) disputes. For this dispute, we provide also a summary of the Panel report because it is difficult to fully appreciate the Appellate Body report without a sound understanding of the Panel’s findings.

Of the five circulated Panel or Appellate Body reports, three were trade remedies disputes, namely China – Definitive Anti-Dumping Duties on X-Ray Security Inspection Equipment from the European Union (DS425) (“China – X-Ray Equipment”), China − Anti-Dumping and Countervailing Duty Measures on Broiler Products from the United States (DS427) (China – Broiler Products) as well as United States – Final Anti-Dumping Measures on Stainless Steel From Mexico – Recourse to Article 21.5 of the DSU by Mexico (DS344) (“US – Stainless Steel (Mexico) (21.5)”). This last case, however, was settled and no substantive Panel report was published. Canada – Feed-in Tariff Program related to the Agreement on Subsidies and Countervailing Measures (“SCM Agreement”¹), the General Agreement on Tariff and Trade 1994 (the “GATT 1994”²) and the Trade-Related Investment Measures Agreement (“TRIMS Agreement”³), whereas EC – Seal Products involved the Agreement on Technical Barriers to Trade (“TBT Agreement”⁴) and the GATT 1994, including a defence based on public morals.

Although the number of Panel and Appellate Body reports in 2013 was relatively low, some of these reports have attracted significant attention and have given rise to intense discussions in the international trade community. For instance, the Canada – Feed-in Tariff Program dispute was the first dispute to address a subsidy programme for renewable energy, even if the measure at issue was not that “green subsidy” itself, but rather a discriminatory local content requirement. Nevertheless, concerns about environmental subsidies arguably permeated the Appellate Body’s conspicuous benefit ruling under Article 1 of the SCM Agreement. As another example, the Panel report in EC – Seal Products addressed a number of systemically important issues under both the GATT as well as the TBT Agreement, including a hitherto rare defence on the grounds of public morals.

China – X-Ray Equipment (DS425)

Facts of the Case

The dispute concerned an anti-dumping measure imposed by the People’s Republic of China (“China”) on X-ray security inspection equipment from the European Union. The X-ray security inspection equipment at issue covered a broad range of products (or product types), including both so-called “low-energy” and “high-energy” scanners. “Low-energy scanners” are, for instance, walk-through detectors used in civil airports. In contrast, “high-energy” scanners include industrial scanners able to scan cars, cargo containers, and container trucks at seaports, border crossings and airports.

The petitioner in the Chinese domestic investigation was Nuctech Company Limited (“Nuctech”), a company created under the auspices of Qinghua University in Beijing in 1997. Nuctech claims to be one of the world’s top providers of airport security equipment, and reportedly is the dominant supplier of security equipment in China’s airports.⁵

Only one company participated in the investigation, namely, Smiths Heimann GmbH (“Smiths”). Smiths exported only low-energy scanners to China. The investigation was initiated in August 2009 and was concluded in January 2011. Based on the final determination issued by the Ministry of Foreign Commerce (“MOFCOM”), China imposed a 33.5 % duty on Smiths and a residual rate of 71.8 % on imports from other EU sources.

Salient Legal Findings

The Panel found that China had violated WTO law on a number of counts. Most of the findings are not necessarily systematically significant or surprising. Nevertheless, the report touched on a number of issues that hold significance for further development of the trade remedy case law.

Determination of Price Effects

First, the EU challenged MOFCOM’s price effects determination. Pursuant to Article 3.2 of the Anti-Dumping Agreement, ⁶ an investigating authority is required to “consider” whether dumped imports had resulted in certain price effects—namely, price undercutting, price suppression or price depression. The Panel found that, when analysing such price effects, MOFCOM had failed to distinguish between low-energy and high-energy scanners and had failed to ensure price comparability between the product groups on the domestic and the imported side. Whereas imported products included only low-energy scanners, the group of domestic products included both low-energy and high-energy scanners. In other words, MOFCOM was confronted with two differently-composed product groups.

The key problem was that, rather than comparing imported low-energy with domestic low-energy scanners, MOFCOM calculated a weighted-average unit price for the entire domestic product bundle. MOFCOM then compared that average unit price with the average unit price of the imported products. Not surprisingly, the Panel found this to be an improper price comparison because MOFCOM had failed to consider the fact that each product group had a very different composition. This Panel finding fits into an evolving line of case law that requires investigating authorities to ensure price comparability under Articles 3.2 of the Anti-Dumping Agreement, despite the fact that Article 3.2 does not explicitly require investigating authorities to do so.⁷ At the same time, investigating authorities retain discretion on how precisely to ensure such price comparability—for instance, investigating authorities can choose to make multiple comparisons or, instead, perform price adjustments.

The MOFCOM analysis underscores that the problem of price comparability will arise in particular in situations in which several distinct products are “bundled up” into one single overarching “product under investigation”. Since the early 2000s, several Panel reports have rejected claims challenging an allegedly excessively broad product definition.⁸ These Panels found that investigating authorities have virtually unlimited discretion in how to group products (or product types) into one single “product under investigation”. From a complainant’s perspective, the risk of such unlimited discretion is that the investigating authority may attempt to manipulate the scope of the product—and thereby the scope of the domestic industry producing the like product—to increase the plausibility of an affirmative injury finding. Arguably, in EC – Aircraft, although it did not speak directly to this issue, the Appellate Body indicated obliquely its disagreement with the broad discretion reflected in those Panel reports.⁹ However, even under the permissible Panel case law, investigating authorities pay a price for a broad product definition—that is, they have to pay particular attention to price comparability.

In reaching its finding on price comparability, the Panel also confirmed previous case law that a failure by an investigated party to raise a particular issue during the domestic investigation does not preclude a government from raising the same issue during WTO proceedings.¹⁰ In the domestic investigation, Smiths (the investigated exporter) had failed to raise concerns regarding the price comparability issue described above. However, this failure of the company was held not to preclude the EU from raising this point before a WTO Panel.¹¹

Implicit Investigation of Injury Factors Under Article 3.4 Anti-dumping Agreement

Second, the Panel made a finding that China had failed to examine one of the 15 injury factors under Article 3.4 of the Anti-Dumping Agreement. According to well-established case law, Article 3.4 requires an investigating authority to examine individually each one of the 15 injury factors listed in that provision¹² and requires a process of evaluation and assessment.¹³ In this case, the EU alleged that MOFCOM had failed to examine the factor “magnitude of the margin of dumping”.

China sought to defend itself by invoking the Appellate Body’s 2003 report in EC – Tube or Pipe Fittings, which had agreed that a particular injury factor had been analysed “implicitly”, without an explicit reference, because in that dispute there was sufficient evidence on the record that the factor had nevertheless been evaluated.¹⁴ China argued that MOFCOM’s report contained a reference to the dumping margins, not in the injury section, but rather in the section listing the dumping margins for the various companies. However, the Panel did not accept this defence. Instead, it reasoned that, a mere listing of the dumping margins did not constitute the requisite analysis and assessment. Moreover, if the listing of dumping margins relied on by China were sufficient for purposes of Article 3.4, then the mention of “magnitude of the margin of dumping” in Article 3.4 would be superfluous. This is because every investigating authority report leading to the imposition of anti-dumping measures would, necessarily, contain a mention of margins of dumping and their magnitude.¹⁵

This Panel finding is useful in further limiting the reach of the concept of “implicit” evaluation of injury factors, created by the Appellate Body in EC – Tube or Pipe Fittings. The Appellate Body’s ruling on that point was arguably an unnecessary blunder that complicates the lives of complainants and introduces an entirely unnecessary ambiguity into the process. Findings such as the one by this Panel are a useful limitation of that Appellate Body finding.

Concept of “Start-Up Operations” as a Non-attribution Factor

Third, the Panel provided interesting reflections on the concept of “start-up” operations. The Panel made these remarks as part of its non-attribution analysis. More specifically, as part of its challenge to MOFCOM’s causation and non-attribution analysis, the EU argued that Nuctech was in a “start-up situation” on low-energy scanners, which fact would explain the negative profitability on this product line. As a result, the EU argued, negative profitability should not be attributed to the dumped imports, but rather to the “start-up” scenario.

The Panel examined Article 2.2.1.1 and Annex IV(4) of the SCM Agreement to gain better understanding of the concept of “start-up” operations. Ultimately, the Panel rejected the EU’s claim, among others because there was record evidence that Nuctech had been producing low-energy scanners for three years before the period of investigation and, during the period of investigation, was already selling significant amounts of this type of scanners. The Panel admitted that there may be “room for debate” on the number of years during which a company may be considered to be in a “start-up” phase.¹⁶ Nevertheless, the Panel finding may henceforth serve as an (outer) guiding post on the “start-up” concept, given the lack of other guidance in the treaty text and case law. This may be relevant not only in the context of an investigating authority’s non-attribution analysis, such as in this particular dispute. For instance, under Article 2.2.1.1, an investigating authority is required to consider “start-up operations” when determining appropriate cost allocation methodologies.

Examination of “Impact” of Dumped Imports Under Article 3.4 Anti-dumping Agreement

Fourth, the Panel made a number of findings on MOFCOM’s non-attribution analysis, finding that MOFCOM had not fulfilled the standards of Article 3.5 Anti-Dumping Agreement. In the process, the Panel had to grapple with the consequences of a previous finding by the Appellate Body in China – GOES. Specifically, in China – GOES, the Appellate Body found that Article 3.4 Anti-Dumping Agreement requires not only the traditional factor-by-factor as well as comprehensive overall analysis of injury, but also an “examination” of the “impact” of the subject (dumped) imports on the domestic industry. In the Appellate Body’s view, Article 3.4 Anti-Dumping Agreement (as well as Article 15.4 of the SCM Agreement) do not merely require “an examination of the state of the domestic industry, but contemplate that an investigating authority must derive an understanding of the impact of subject imports on the basis of such an examination.”¹⁷

These Appellate Body findings arguably contradict the long-established analytical approach under which Article 3.4 requires an examination of the individual “injury factors” and an overall finding of whether injury exists, whereas Article 3.5 requires an examination of the causal link between that injury and dumped imports, including a non-attribution analysis of the impact of other causal factors. In other words, Article 3.4 asks whether the domestic industry is in a state that can be described as injurious, regardless of what may have caused that injurious state of affairs. In contrast, Article 3.5 enquires which factor(s) are the cause of this injurious state of affairs.

The problem with the Appellate Body’s findings is that an examination of the “impact” of dumped imports on the domestic industry (under Article 3.4) is difficult to distinguish from an examination of the causal link between dumped imports and injury (under Article 3.5). Thus, despite the Appellate Body’s reassurances to the contrary,¹⁸ its findings necessarily suggest a degree of duplication between the traditional causation analysis under Article 3.5 and the newly required “impact” analysis under Article 3.4.

This duplication or overlap became apparent in the China – X-Ray Equipment Panel report. Before addressing the EU’s claims concerning the non-attribution analysis, the Panel had to resolve the question whether this issue was more appropriately addressed under Article 3.4 or Article 3.5. After pondering the import of the Appellate Body’s findings in China – GOES, the Panel drew a useful distinction between factors that are indicia of the state of the industry and those that may cause injury to the industry. The former, according to the Panel, are to be examined under Article 3.4, whereas the latter are to be examined under Article 3.5.¹⁹ This (logical) approach arguably obviously does not square with the Appellate Body’s findings in China – GOES; rather it serves to highlight the unnecessary confusion that the Appellate Body has injected into the relationship between Articles 3.4 and 3.5. The Panel ultimately decided to examine the EU’s allegations under Article 3.5.

It remains to be seen whether other Panels will follow this approach, which would appear to effectively sideline the Appellate Body’s ill-considered “impact” requirement. This would provide for a degree of predictability in the case law, until the Appellate Body gets another opportunity to explain more in detail how it views the relationship between Articles 3.4 and 3.5.

Treatment of Confidential Information and Preparation of Non-confidential Summaries

Fifth, the Panel found that MOFCOM had violated Article 6.5.1 Anti-Dumping Agreement. Article 6.5.1 is of significant practical relevance, as it requires investigating authorities to ensure that confidentiality of information submitted by interested parties does not deny the due process rights of other parties. Article 6.5 requires investigating authorities to treat as confidential information that is either by its nature confidential, or that has been submitted on a confidential basis. Such information can only be disclosed upon agreement by the party which submitted that information. However, to counterbalance the confidentiality requirement under Article 6.5, Article 6.5.1 requires that the party submitting confidential information provide a non-confidential summary of that information. This summary must be “sufficient detail to permit a reasonable understanding of the substance of the information submitted in confidence”. Nevertheless, Article 6.5.1 also envisages that, “in exceptional circumstances”, confidential information may not be susceptible of summary. In these exceptional circumstances, “a statement of the reasons why summarisation is not possible must be provided.”

The Panel’s findings in this dispute provide a useful contribution to the case law under this provision. Three points can be highlighted. First, the Panel confirmed previous case law that confidential information should, as a general rule, be susceptible of summarisation.²⁰ Second, the Panel also rejected a “curing”-type defence by China. Specifically, the mere fact that interested parties did not complain during the investigation about the inadequate nature of the summaries provided by the domestic industry—and the fact that the parties were able to prepare comments based on the (allegedly deficient) summaries—was deemed not to be legally relevant in a WTO proceeding. The Panel found that the consistency of a non-confidential summary with Article 6.5.1 should be assessed “by reference to the content of that summary, rather than any propensity for respondents to prepare comments on the basis of their best estimate of the substance of the underlying confidential information”.²¹

Third, the Panel found that a domestic governmental agency acting as an interested party could not be exempted whole-sale from the requirement to provide non-confidential summaries. The Panel found that “[t]he mere fact that an authority is required to perform a particular function under its domestic law” [e.g. air transport safety] “does not demonstrate, for the purpose of WTO dispute settlement proceedings, that the authority did actually perform that function.”²² As a result, the confidential nature of information must be demonstrated with regard to that specific information, rather than with regard to the general task of the entity submitting that information.

Findings on “Disclosure of Essential Facts” Under Article 6.9 Anti-dumping Agreement

Finally, sixth, the Panel made relevant findings concerning the sufficiency of “disclosure” under Article 6.9. The Panel’s findings on Article 6.9 add to the evolving standard concerning the disclosure of “essential facts”. This case law has over the years provided incremental insights into what constitutes “essential information”.

In China – X-Ray Equipment, the Panel found insufficiencies in MOFCOM’s disclosure because of the failure to publish certain data. The data at issue underlay MOFCOM’s calculations of the average unit values of the subject products as well as the calculations of the dumping margins. The Panel found that this information was “essential” within the meaning of Article 6.9, because it “constituted the body of facts on which MOFCOM’s determination of price effects was based.” This “body of facts” was “required” to understand the basis for MOFCOM’s price effects analysis as well as MOFCOM’s dumping margin calculations, and therefore had to be disclosed pursuant to Article 6.9.²³ However, the Panel found that the “essential facts” did not include the actual mathematical calculations underpinning the dumping margins, because these calculations were more properly characterised as “consideration” of the facts, rather than the facts themselves.²⁴ As we shall see below, the Panel in China – Broiler Products subsequently addressed this very same issue and arrived at a (arguably) different result.

China – Broiler Products

Facts of the Case

The other trade remedy dispute in 2013 involving China was China – Broiler Products. “Broilers” are chickens raised specifically for meat production.

China had imposed both anti-dumping and countervailing duties on broiler products from the United States. The two underlying investigations were both initiated on 27 September 2009, upon petition by the China Animal Agriculture Association (“CAAA”). Preliminary determinations were issued in February 2010 (Anti-Dumping) and April 2010 (Countervailing Duties, CVDs), final determinations in August 2010 (antidumping duties) and September 2010 (CVDs). MOFCOM imposed individual anti-dumping duties ranging from 50.3 to 53.4 %, and individual CVDs ranging from 4 to 12.5 %. Non-investigated firms that had come forward and registered themselves at initiation received an 51.8 and 7.4 % rate, respectively. However, “all other” companies received very high rates, namely, an AD rate of 105.4 and a CVD rate of 30.3 %.

Salient Legal Findings

The Panel report covers a broad range of issues under the Anti-dumping Agreement and the SCM Agreement. The Panel found a number of violations by MOFCOM, the Chinese investigating authority. The following five sections highlight the most salient legal findings from a systemic perspective.

Application of Facts Available to Unknown Exporters: Article 6.8 and Annex II of the Anti-dumping Agreement

The first key legal issue in this dispute was the application of “facts available” to unknown exporters. This is an issue that many investigating authorities regard as a key policy tool to encourage cooperation by foreign companies. The Panel ruling explicitly contradicts the Panel ruling in China – GOES on this very issue, creating a tension within the case law that can ultimately be resolved only by the Appellate Body.

The finding in this dispute was as follows: MOFCOM applied a very high anti-dumping duty (105.4 %) and CVD (30.3 %) to companies that did not make themselves known to MOFCOM at the outset of the investigation. These companies are generally referred to as “unknown” exporters.

By way of background, in a typical anti-dumping investigation, an investigating authority may calculate and apply three kinds of duty: (1) individual anti-dumping duties for each of the investigated (sampled) exporters; (2) an average rate—calculated based on the individual rates under (1)—that will be applied to all companies that the investigated authority knew of but decided not to investigate individually²⁵; and finally (3) a residual rate applied to all other (unknown) companies. In an investigation involving only a few foreign exports, each exporter can be investigated and can receive an individual dumping margin, and there will be no duty type (2). In contrast, in investigations involving a large number of companies, the investigating authority may decide to create a sample of exporters, attribute an individual margin to each exporter in the sample (rate (1)), but then apply an average rate to the known companies left outside the sample (rate (2)) and a residual rate to all unknown remaining companies (rate (3)). Under the practice of many investigating authorities, (3) is higher than (2). For instance, rate (3) will be equal to the highest individual rate calculated under (1). This differential treatment of known non-sampled companies and unknown companies is intended to create an incentive for exporters to come forward at the beginning of the investigation, rather than to “hide” from the investigation.²⁶

The Anti-dumping Agreement explicitly regulates rates (1)²⁷ and (2).²⁸ However, concerning rate (3)—the dumping duty to be applied to unknown exporters—the legal situation is unclear. According to a widely held view, rate (3) is not regulated by the Anti-dumping Agreement and thus investigating authorities have discretion.²⁹ This also appears to have been the view espoused by the Panel in EC – Salmon.³⁰ According to a contrary view—not reflected in existing case-law—the same provision that governs rate (2)—that is, Article 9.4—applies also to rate (3).³¹

There are two more pieces to the puzzle. The first piece is the Appellate Body’s ruling in Mexico – Rice, which establishes that the attribution of a dumping margin to unknown companies constitutes application of facts available.³² Second, the Appellate Body has arguably suggested that the application of facts available is not intended to be per se punitive.³³

In keeping with the Appellate Body’s decision in Mexico – Anti-Dumping Measures on Rice, the Panel in China – Broiler Products deemed the application of the 105.4 % anti-dumping duty to the unknown companies an instance of “facts available”. The Panel then proceeded to examine whether MOFCOM had satisfied the substantive and procedural requirements for applying facts available. Article 6.8 provides that facts available may be applied when an interested party “(i) refuses access to necessary information within a reasonable period; (ii) otherwise fails to provide necessary information within a reasonable period; or (iii) significantly impedes an investigation”. These conditions are further fleshed out in Annex 2 to the Anti-dumping Agreement. The thrust of the provisions of Annex 2 is to impose substantive and procedural disciplines on the investigating authority’s recourse to facts available. For instance, pursuant to Annex 2(1), an investigating authority can apply facts available only if it “specif[ies] in detail the information required from any interested party.” Moreover, the authority must also “ensure that the party is aware that, if information is not supplied within a reasonable time, the authorities will be free to make determinations on the basis of the facts available”.³⁴ In other words, facts available can be applied only if an interested party was aware of the information required of it, and if it subsequently does not supply that information.

A typical situation in which facts available may be applied is when the investigating authority sends a questionnaire to an investigated company and the company subsequently fails to respond adequately to the questionnaire. In these circumstances, provided other requirements are satisfied, the investigating authority may ultimately apply facts available.

The problem with unknown exporters, of course, is that these exporters logically cannot be directly apprised of the information required of them—by definition, they are unknown. The question therefore is how the requirement in Article 2(1) can be satisfied with respect to them. The Panel in China – GOES, which was decided in 2012, found that a certain set of notifications by MOFCOM were insufficient in this regard. MOFCOM’s action in that case consisted of (1) placing the notice of initiation on the internet (2) placing the notice in a public reading room and (3) requesting the US Embassy in Beijing to notify US exporters. In any event, the Panel in that dispute relied on the fact that the notice of initiation did not specify the information to be provided by exporters. The Panel also found the case at hand to be very similar to Mexico – Anti-Dumping Measures on Rice and rejected China’s attempts to differentiate these two disputes. Therefore, the application of facts available to those unknown exporters was in violation of WTO law.³⁵ This finding was not appealed.

The Panel in China – Broiler Products reached a radically different conclusion from essentially identical facts. Again, China had placed the public notice of initiation on the internet as well as in a MOFCOM reading room and had requested assistance of the US Embassy in Beijing. The Panel reasoned that it was “generally recognised and accepted” that the manner to inform and communicate with interested parties in administrative proceedings was by way of public notices.³⁶ Although MOFCOM’s notice did not request information underlying the calculation of a dumping margin (that is, information about normal value and export price), the Panel nevertheless reasoned that the notice, had it been properly answered by the unknown parties, would have subsequently led the authority to request that information.³⁷ The Panel also distinguished this case from the Appellate Body’s decision in Mexico – Anti-Dumping Measures on Rice, based on differences between the content of the public notice. The Panel also reasoned that not accepting China’s approach as sufficient would render it impossible to apply a distinct anti-dumping rate to unknown exporters, thereby creating an incentive for non-cooperation.

Ultimately, the Panel still found a violation in how MOFCOM had treated unknown exporters. However, this violation was not related to MOFCOM’s notifications. The Panel found that MOFCOM had not provided an explanation of how and based on what information it had calculated the very high 105.4 % “all others”. This was a violation of Annex II(7), because it failed to reflect the requisite “special circumspection” demanded of investigating authorities when applying facts available.³⁸ Thus, in the end, China lost on this particular claim. However, the Panel’s legal interpretation and its finding on principle opens the door for investigating authorities to apply “residual” rates to unknown exporters.

Because of this ruling, WTO case law on this important issue points in two radically different directions. A decision by the Appellate Body appears to be the only solution. Until the Appellate Body has the opportunity to do so, investigating authorities may seek to rely on the Panel ruling in China – Broiler products, and emulate precisely what MOFCOM did in the Broiler investigation. However, even these authorities will have to worry that a WTO Panel may instead decide to follow the China – GOES approach.

Investigating authorities may also “play it safe” and apply rate (2)—calculated pursuant to Article 9.4—to unknown exporters. This approach is highly unlikely to be challenged or found WTO-inconsistent, given that nobody would seriously contest the authority’s right to apply some form of duty to unknown exporters. The Article 9.4 rate is an objective and non-controversial tool to impose duties on non-investigated companies. At the same time, investigating authorities would lose their much-vaunted policy tool to encourage cooperation by exporters. However, while the need to encourage cooperation is accepted as legitimate by a number of investigating authorities and practitioners, it is not explicitly enshrined in the Anti-dumping Agreement and may not be a valid objective that should inform the interpretation of the agreement.

Definition of the Domestic Industry

The second noteworthy finding from the China – Broilers Products case relates to the definition of the domestic industry. Objective parameters for defining the domestic industry are important, because both the Anti-dumping Agreement and SCM Agreement permit the investigating authority to define the “domestic industry” as less than the totality of the domestic producers. Instead, the “domestic industry” can be defined as a “major proportion” of the domestic producers producing the like product.³⁹

Adequate interpretation requires ensuring the appropriate margin of discretion for investigating authorities, all the while ensuring that the authorities exercise this discretion in an even-handed manner. Once the investigating authority starts selecting producers to be included and excluded from the scope of the domestic industry, necessarily a risk of bias arises and the choices in the authority’s definition must be subject to review by a WTO Panel. Disputes in which the domestic industry had been defined in an inconsistent manner include EC – Salmon (Norway), where the Commission failed to include in the domestic industry the producers of a sub-section of the product under consideration.⁴⁰ A more recent instance is EC – Fasteners, where the domestic industry included only domestic companies that had expressed a willingness to be included in a sample. This willingness—while undoubtedly important from the practical perspective of collecting data—was held to be a biased criterion for defining the industry. This is because this form of self-selection made an affirmative injury finding more likely, thereby creating a “material risk of distortion”.⁴¹

In the dispute at hand, the Panel found that the domestic industry had been properly defined and rejected an analogy to EC – Fasteners. As an important principle, the Panel confirmed previous findings that there is no hierarchy between the two options granted by both the Anti-dumping Agreement and the SCM Agreement, that is, the option to (1) define the domestic industry as all producers or the option to (2) using “only” a “major proportion”. As a result, the investigating authority is not required, at the outset, to seek out all domestic producers and use the “major proportion” option only as a fallback of sorts, in case information on all producers cannot be obtained. Rather, the authority can from the outset choose the “major proportion” option. However, the authority must have sufficient information concerning the total volume of production of the domestic industry.⁴²

Moreover, on the facts of this particular case, the Panel rejected the United States’ allegations that MOFCOM ignored certain information about certain producers and thereby made it more likely that companies supporting the petition would be included in the scope of the domestic industry. The Panel accepted that, although MOFCOM indeed failed to contact a company after receiving information about its existence, this did not lead to a WTO inconsistency. The Panel reasoned that, at that point in time, information from producers accounting for approximately 50 % of the domestic production had already been received.⁴³

This finding arguably demonstrates that, absent very cogent evidence of obvious bias in the industry definition process, Panels will be reluctant to find that the selection of producers for the “major proportion” is inconsistent with WTO law.

Sufficiency of Disclosure

The third systemically interesting finding relates to the sufficiency of the investigating authority’s disclosure. Similarly to the China – X-Ray Equipment case, the China – Broiler Products case included a claim on the sufficiency of the disclosure of the dumping margin calculations.

The Panel found that MOFCOM had disclosed insufficient information to satisfy that requirement. The Panel’s reasoning started with the premise that the “essential facts” to be disclosed under Article 6.9 Anti-Dumping Agreement include the data that are the basis for the dumping determination. Therefore, a declaration of the weighted-average dumping margin for a particular model does not satisfy Article 6.9, unless it is “accompanied by the data relied on to reach that conclusion”. The Panel referred to the finding of the Panel in China – X-Ray Equipment, described above, which had held that the actual mathematical determination of the dumping margins was not part of the “facts … under consideration”, but rather only part of the “consideration” of those facts.⁴⁴ The Broiler Products Panel agreed with the China – X-Ray Equipment Panel that presenting the actual calculations in the form of spreadsheets, while useful and efficient to disclose the essential facts, was not required under Article 6.9. However, the Panel also reasoned that, if the X-Ray Equipment Panel ruling were to be read as suggesting that the “formula” used to make those calculations need not be disclosed, then it would “respectfully disagree”.⁴⁵ As a way of disclosing the “formula”, without disclosing e.g. the spreadsheet calculations, the Panel mentioned a narrative explanation and references to questionnaire responses, “as long as the respondent would be able to defend its interests based on the information disclosed”.⁴⁶

The Panel then found that MOFCOM had not satisfied this legal standard, because it had failed to provide information about which sales prices had been used to calculate the aggregate normal values and export prices. The Panel reasoned that, without the information concerning the sales prices used to make all these calculations, the respondents would be unable to ascertain the accuracy of MOFCOM’s calculations and properly defend their interests.

It is not clear whether the Panels in China – Broiler Products and China – X-Ray Equipment set out two distinct standards for the Article 6.9 disclosure. In any event, the thrust of both rulings is to create a fairly comprehensive and far-reaching disclosure requirement. From a policy perspective, there would seem to be little basis to argue against providing as much information as possible to investigated companies concerning the calculations of the dumping margins.

Cost Allocation Methodologies

Perhaps the most colourful finding in this dispute—involving a rare cross-cultural aspect of consumption patterns—pertains to chicken feet. A delicacy in China, chicken feet are considered offal in the United States and presumably in most other countries around the world. The key relevant issue in the anti-dumping investigation as well as in the WTO proceedings was the question of allocation of costs. Which production costs should be allocated to the chicken feet in the companies’ financial records? Broadly speaking, the investigated US companies in their financial statements followed a value-based approach. This approach was based on the value of the various parts of the chicken in the US market; because that value was very low, the production cost of chicken feet were correspondingly allocated a very small percentage of the total production cost. In contrast, in its investigation, MOFCOM favoured a weight-based approach. Under that approach, if chicken feet represented X per cent of the total weight of the broiler product, a corresponding X percentage of the production costs should be allocated to chicken feet.

Under the companies’ approach, no dumping margin existed, because the relatively high prices that chicken feet commanded in the Chinese market (export price) exceeded the relatively low production costs assigned to these feet in the companies’ books (normal value). To the contrary, MOFCOM’s weight-based approach would generate a dumping margin, because it resulted in higher production costs that, in this particular instances, exceeded the value (price) of the chicken feet on the Chinese market.

The US companies’ value-based approach was one that these companies traditionally used and that was consistent with United States general accepted accounting principles (GAAP). Pursuant to Article 2.2.1.1 of the Anti-dumping Agreement, this should mean, in principle, that MOFCOM had to accept the allocation methodology, unless it determined that the allocated costs do not “reasonably reflect” the costs associated with the production of the product at issue. In principle, the investigating authority would then have to provide a thorough, reasoned and adequate explanation for why it rejected the costs as allocated in the companies’ books.

The resolution of this case hinged on the applicable standard of review in WTO trade remedy disputes. Pursuant to well-established case law, compliance with the majority of the substantive obligations in the Anti-dumping Agreement depends on the existence of a reasoned and adequate explanation in the investigating authority’s published report.⁴⁷ In the instant case, MOFCOM was found not to have properly explained why it rejected the companies’ value-based allocation methods, and why it instead relied on its own weight-based allocation method. The Panel rightly rejected China’s explanation during the panel proceedings as being too late and outside the Panel’s terms of reference—the Panel was entitled to review only the explanations contained in MOFCOM’s published determination. However, the Panel also signalled that it considered (or would have considered) pertinent the fact that the US investigated companies had based their value-based allocation methods exclusively on values derived from the US market. The Panel signalled sympathy for China’s argument that, although the companies sold their products also in markets other than the US, including in markets in which the product values were different (e.g. China), these other markets were not taken into consideration for determining the proper value-based allocation.⁴⁸

While this dispute is particularly idiosyncratic, in the light of the very different attitudes to the product “chicken feet” in different cultural contexts, the underlying issue of how to value products across different markets is interesting and may present itself in future cases.

Other Findings of Interest

In addition, the Broiler products Panel report contains findings of interest, including on the adequacy of disclosure under Articles 6.9 (AD) and the adequacy of the public notice under Article 12 (AD) and 22 (SCM), the sufficiency of non-confidential summaries (Article 6.5.1 AD) and the grant of opportunity for parties to meet with parties with adverse interests under Article 6.2 (AD).

EC – Seal Products (DS400, 401)

The EC – Seal Products dispute was arguably the politically most sensitive dispute of 2013, attracting the attention of not only the traditional trade community, but also of animal welfare groups and even prominent Hollywood actors sympathetic to animal welfare concerns.⁴⁹ As of the timing of writing, the case was under appeal.

Facts of the Case

The measure at issue was an EU ban on the placing on the market of products derived from seals. The measure was ostensibly driven by ethical and animal welfare concerns surrounding the commercial hunt of seals and the “inhumane killing” methods occurring during at least parts of such commercial hunts. However, the EU regime also contains exceptions that seem to run, at least in part, counter to those concerns.

The regime consists of a main regulation and an implementing regulation. Jointly, these two measures impose a ban on seal products unless one of the following three situations occurs: (1) the seal products at issue “result from hunts traditionally conducted by Inuit and other indigenous communities and contribute to their subsistence” (the “IC condition”); (2) the seal products are obtained from seals hunted for marine resource management purposes, i.e. seal population control (the “MRM exception”) and (3) the seal products are imported by travellers in limited circumstances (for personal use) (the “Travellers condition”).

The implementing regulation specifies the precise conditions and provides for procedural requirements that must be met to place the seal products on the market. For instance, the seal product must be accompanied by an attesting document issued by a recognised body.

Overview of the Key Findings

The complainants presented claims under the TBT Agreement and the GATT 1994. The Panel first examined the claims under the TBT Agreement. It found that the EU’s seals regime was a technical regulation within the meaning of Annex 1.1 of the TBT Agreement. It then found that the IC and MRM exceptions were inconsistent with Article 2.1, because the IC exemption was available de facto exclusively to Greenland and therefore was not designed and applied in an even-handed manner. The Panel found the MRM exception not justifiable given that, concerning the incidence of “inhumane killing” methods, there was no practical difference between standard commercial hunts and hunts for MRM purposes.

However, the Panel rejected the claim that the EU seals regime was more trade restrictive than necessary within the meaning of Article 2.2 of the TBT Agreement. The Panel found that the complainants’ proposed alternative measure(s) were not “reasonably” available, in that they did not adequately ensure that products derived from inhumanely killed seals would not reach the EU market.

The Panel also concluded that certain provisions of the implementing regulation constitute a “conformity assessment procedure” (“CAP”) within the meaning of Annex 1.3 of the TBT Agreement. It then went on to find that the CAP had created unnecessary obstacles to international trade and therefore violated Article 5.1.2 of the TBT Agreement. This was because the EU had failed to create a conformity assessment body by the time the seals regime entered into effect; therefore, there was no body to verify compliance with the statutory exceptions, the result being that trade could not practically occur.

Concerning the claims under the GATT 1994, the Panel found that the IC and MRM exceptions violated Article I:1 and III:4. Although both exceptions were provisionally justified under Article XX(a), they were found to be inconsistent with the chapeau of Article XX and therefore could ultimately not be justified. The reasons underpinning that finding were the same as those underlying the finding on the IC and MRM exceptions under Article 2.1.

The Panel report was appealed by Canada and Norway and a decision is expected in late May 2014.

Salient Findings

The following is a discussion of certain salient aspects of the Panel’s findings.

Definition of a Technical Regulation

Summary of the Finding

The Panel found that the EU seals regime was a “technical regulation” under Annex 1.1 of the TBT Agreement. This, of course, opened the door to Canada’s claim under Article 2.1 and Norway’s claim under Article 2.2. The Panel relied on the Appellate Body’s ruling in EC – Asbestos, to find that the applicability of the TBT Agreement had to be assessed against the totality of the measure—that is, both the general ban and the exceptions taken together or, put differently, “taking into account both the prohibitive and permissive aspects”.⁵⁰ The Panel also found that a ban on all products containing seal elements was a measure that prescribed physical characteristics of an identifiable group of products, namely, all products. Moreover, in any event, numerous product categories to which the seal regime was applicable were explicitly listed in a “Technical Guidance Note” issued by the EU Commission. The Panel also found that the criteria of the exceptions identified the seal products that may be legally placed on the EU market, because they define the categories of seal that can be used as an input for such products.

With regard to the implementing regulation, the Panel found that certain provisions in those regulations constituted “conformity assessment procedures” under Annex 1.3 of the TBT Agreement. The basis for this finding was that particular provisions of the implementing regulation “established the procedure for determining whether the specific requirements under the EU Seal Regime are fulfilled.”⁵¹

Observations

The applicability of the TBT Agreement to measures that might qualify essentially as product or input bans remains a controversial issue in academic discussions, and it will be interesting to learn about the Appellate Body’s current approach to this important question. The leading precedent is of course EC – Asbestos and the Panel faithfully attempted to apply that case law. However, despite that precedent, veterans of GATT and WTO law are sometimes heard to allege that the TBT Agreement was not designed to deal with bans, but instead were designed to deal with product specifications that do not altogether “ban” particular aspects. According to this view, the requirement that a technical regulation apply to an “identifiable product”, in Annex 1.1, bears testimony to that intention of the drafters. A counter-argument is that, in EC – Asbestos, the Appellate Body distinguished between “identifiable” and “given” products⁵² and found that “all” products was an “identifiable” product category. This distinction appears conceptually entirely respectable, all the more as it would seem a mere formality whether a legislator would or would not include in a legal document a list of the typical products affected by the prohibition of a particular input or component. At the same time, admittedly, there is an obvious factual difference between, for instance, technical specifications of electronic products, on the one hand, and a ban on a particular input or raw material, on the other hand. The TBT Agreement could be interpreted to draw legal distinctions between such measures. The search for an adequate response is further complicated when, such as in EC – Asbestos and in EC – Seal Products, a ban is complemented by exceptions. An adjudicator must decide whether the exceptions should be assessed on their own or together with the general ban.

The Appellate Body’s decision in EC – Asbestos gave the definition of a technical regulation a broad reading. At the same time, this ruling dates back over ten years and it is entirely possible that in this appeal the Appellate Body, with an entirely different set of Members, may review its approach. What is at stake with the applicability of the TBT Agreement are the more far-reaching disciplines that go beyond the GATT strictures.⁵³

Moreover, an interesting aspect of this case is that the IC and MRM exceptions to the ban refer to issues unrelated to the physical characteristics of the product, namely, the identity of the hunter or the purpose of the hunt. There is ample academic debate on whether the TBT Agreement applies to what is sometimes referred to as product-unrelated process and production methods. The Panel does not seem to have addressed this aspect at all. If the “ban” on seal as raw material is assessed separately and held to fall under the TBT Agreement, perhaps the product-unrelated nature of the exceptions is not material. However, if the Appellate Body were to consider either that a ban per se does not fall under the TBT Agreement, or if the Appellate Body were to consider the ban and the exceptions as one single integrated measure—as the Panel claims to have done—then the applicability of the TBT Agreement to the exceptions (and perhaps to the entire measure, considered as a whole) may become a genuine question.

“Overriding” the Objective of a Measure with Another Objective Under Article 2.1

Summary of the Panel’s Findings

An interesting aspect of the case is the Panel’s approach to the multiple objectives of the EU’s regime or plurality of interests protected by that legislation. To recall, the EU argued that its measure was concerned with seal welfare and protected public morals. This objective was achieved by prohibiting the marketing of seal products harvested through inhumane killings of seals, in the context of commercial-style hunts. However, the EU seals regime contains important exceptions to this general rule. Specifically, the EU tolerated on its market seals hunted by traditional Inuit communities and seals hunted in the context of maritime resource management (i.e. seal population control) to reach the EU market. The Panel found that both these exceptions could lead to the marketing of seal products obtained by inhumane killings. This was because neither of these two conditions guaranteed that only humanely killed seals would reach the EU market. In other words, the exceptions made possible the very same scenario that the measure was allegedly aiming to prohibit.

The EU’s response to this problem was that, in particular circumstances, the initially stated objective—protection of public morals—could be partially overridden by another unconnected concern. In this particular dispute, these other concerns were the respect for traditional forms of hunting and subsistence activities of indigenous peoples as well as the need to preserve an ecologically-balanced environment.

The Panel accepted this approach in principle. It found that the concerns underlying the IC and the MRM exception were not the objective of the EU seals regime in the same way as the public moral and seal welfare objectives. Rather, these were concerns or “interests” that, while reflected in the measure, did not rise to the level of being the measure’s official “objectives”. Moreover, these “interests” were not connected to the objective. Indeed, the Panel went as far as explicitly finding that the IC and MRM distinctions did not bear a “rational relationship to the objective of addressing the moral concerns of the EU public on seal welfare”.⁵⁴ However, the Panel determined that the distinction drawn by the EU between commercial and non-commercial (subsistence Inuit) hunts was “justifiable”, largely because the interests of indigenous communities—including the Inuit—were recognised at the international level. The Panel then went on to examine whether the distinction was designed and applied in an even-handed manner. Based on the available evidence, the Panel found that the IC exemption was available de facto exclusively to Greenland and therefore had not been designed and applied in an even-handed manner.

With regard to the MRM exception, the Panel found that the distinction between standard and MRM hunts was not justified, because no material difference existed between these two types of hunt. It also found that the MRM exception was not applied in an even-handed manner, because only one EU Member State had registered certain entities as recognised bodies for conformity assessment and because the MRM exception had been designed with the situation of EU Member States in mind.

Observations

It is perhaps too easy to criticise the Panel’s approach to the multiple objectives of the EU regime, especially the distinction between the “objectives” of a measure and other “interests”, given that limited guidance exists in the treaty language or from the Appellate Body. In the real world, legislative acts are driven by a range of concerns, rather than one single concern, and synthesising all of these concerns into a coherent analytical edifice under a legal provision may not be easy. Moreover, the Panel may be correct that concerns about Inuit communities and marine resource management were different from the public morals and animal welfare objective, because they may not have been the initial drivers of the EU’s ban. Rather, these concerns were triggered only once the decision had been taken to prohibit seal products on the EU market.

At the same time, it is not clear why this should mean that addressing these concerns was not considered to be an “objective” of the measure. These interests were the drivers of an important part of the final legislation, namely the IC and MRM exceptions. Moreover, certain other parts of the Panel’s reasoning seem highly questionable—for instance, the fact that certain interests were not “grounded in the concerns of […] citizens”⁵⁵ is hardly a reason for denying the existence of an objective. Arguably, most technical regulations respond to technical concerns of which citizens will in most instances be blissfully unaware, let alone be concerned about. Similarly, denying the “interest” the status of an “objective” on the grounds that it “appear[s] to have been included in the course of the legislative process”⁵⁶ is rather puzzling. Legislators can be expected to form their views, articulate objectives and enshrining them in legislation during the legislative process—it is not clear when else this should occur.

Be that as it may, a key consequence flowing from the Panel’s approach cannot be denied. The Panel developed an arguably new concept, whereby a “justifiable” distinction warrants a partial departure from the official objective of a measure. However, and this is the key concern, this distinction does not, at the same time, constitute a distinct, separate and competing “objective” of the measure. This approach of course carries the risk of limiting the reach of WTO legal disciplines. This is because the TBT Agreement—as well as Article XX of the GATT 1994, in a somewhat different manner—require that the objective of a measure be “legitimate”. This requirement is important, because not every legislative objective or accommodation of an “interest” is legitimate. Moreover, not every legislative objective or accommodation of an “interest” is legitimate as a regulatory concern underlying a trade measure. Nevertheless, the Panel’s interpretative approach implies, for instance, that the EU’s second-tier “interest” of promoting Inuit traditional ways of hunting is not subject to the “legitimacy” requirement. Rather, this “interest” appears to have been examined only for the coherence of its operationalisation, rationality and even-handedness in justifying a departure from that overall, stated “objective”. This is a potential loophole in the disciplines.

Admittedly, the Panel found that the protection of interests of indigenous communities is a concern present in international law and policy. However, this type of endorsement is not the same as arguing that the protection of such interests via trade measures is a legitimate objective under Article 2.1 or 2.2 of the TBT Agreement. It is not clear, for instance, whether WTO members should have a free-hand in determining that particular indigenous communities within certain countries (e.g. developing countries) are worthy of promotion via special trade measures and whether these members should then be permitted to introduce differentiated trade concessions for the benefit of such unilaterally defined special interest groups.

A similar problem arguably arises under Article XX of the GATT 1994. Under Article XX(a), the Panel examined the EU seals regime as a whole, rather than the IC and the MRM exception separately. Admittedly, there is some degree of controversy whether the provisional justification under the sub-paragraphs of Article XX must exist concerning the offending aspect of the measure (e.g. discrimination) or rather the measure as a whole. The Panel relied on the Appellate Body’s ruling in US – Gasoline to examine the measure as a whole. The Panel then examined the justifiability and even-handedness of the IC and MRM exceptions under the chapeau and reached the same conclusions as under the TBT Agreement.

What this means, once again, is that there is no analysis whether the two rationales underlying the IC and MRM exceptions qualify as policy justifications under the sub-paragraphs of Article XX. The test under the chapeau of Article XX—at least for most measures—is whether there is abuse or arbitrariness in the application of a measure, not in its basic design, object and purpose.⁵⁷ Thus, a partial objective of a measure can sail through at least some of the legal tests without being sufficiently-well scrutinised.

Application of Article 2.2 TBT Agreement

Summary of the Panel’s Findings

Under Article 2.2, the complainants argued that the EU measure was unnecessarily trade restrictive. They submitted that observation of strict animal welfare standards in commercial hunts and corresponding labelling of the products would be an equally effective, but less trade restrictive alternative to the EU’s outright ban.

The Panel found that the objective of the EU measure was to address the moral concerns of the EU public with regard to the welfare of seals. This objective had two aspects—the incidence of inhumane killing of seals and the EU citizens’ individual and collective participation as consumers in, and exposure to, the economic activity that sustains the market for seal products derived from inhumane hunts. The Panel also found that these objectives were “legitimate”, within the meaning of Article 2.2. However, as noted above, the Panel rejected the argument that the protection “interests” accommodated in the three exceptions (Inuit communities, travelers and marine resource management) were also “objectives” of the measure. Rather, these “interests” occupied some undefined and unspecified lower-level significance and role.

The Panel then examined the contribution that the EU’s seals regime made to its objectives. The Panel emphasised that it was required to examine the “actual” contribution. It found, respectively, that the ban was “capable” of making a contribution and actually made “some contribution”.⁵⁸ However, the IC and MRM exceptions “diminish[ed] the degree of the actual contribution made by the ban”,⁵⁹ because they permitted consumers to be exposed to seal products derived from seals killed inhumanely. At the same time, because some Inuit communities have been adversely impacted and had not been able to benefit from the exceptions, the negative impact of the IC exception on the EU’s ethical and animal welfare objective was less than it could have been.

The Panel then examined the contribution of the complainants’ proposed alternatives. In essence, these alternatives were that market access for seal products would be conditioned on compliance with animal welfare standards combined with certification and labelling requirements. The Panel found that there were uncertainties, subjectivity and divergence of opinions among experts with regard to the prescription of animal welfare criteria, for instance, delays in the seal killing process. It found that the alternative measures “could possibly span a range of different levels of stringency and leniency”.⁶⁰ It also appeared to doubt that the enforcement of the standards could be effectively assessed and that distinctions between humane and inhumane killings could be readily undertaken. It concluded that an alternative measure within the range of measures proposed “may give rise to an increase in the number of seals hunted with the accompanying risks to seal welfare through restored market opportunities within the EU”,⁶¹ which would undermine the EU’s objective. It also found that “[t]he complainants do not deny […] and the evidence […] confirms […] that killing and poor animal welfare do occur in seal hunts”.⁶² The Panel also found that the complainants had not precisely defined the exact welfare standard to be imposed and thus had not clearly defined an alternative measure “in respect of its separate components and their [sic] cumulative capability to address the moral concerns of the EU public”.⁶³ On that basis, the Panel concluded that the alternative measure proposed by the complainants was not reasonably available, within the meaning of Article 2.2.

Observations

This is yet another example of an unsuccessful claim under Article 2.2, following the initial “trinity” of disputes under the TBT Agreement, namely, US – COOL, ⁶⁴ US – Tuna II (Mexico), ⁶⁵ and US – Clove Cigarettes. ⁶⁶ This series of findings has led some observers to wonder if an Article 2.2 claim effectively has a chance of ever succeeding. This is an important question, because if the bar to findings of Article 2.2 violations is set unattainably high, a very important added-value of the TBT Agreement over the GATT 1994 will be compromised.

The Seal Products Panel finding is arguably problematic, for two reasons. The first reason is, as already previously explained, the characterisation of the IC and MRM concerns not as “objectives”, but rather as “interests” that do not require justification as “legitimate objectives”. One can understand the Panel’s view these concerns may not have been the key driving force of the legislation, in the same manner as the concerns about seal welfare. However, placing these concerns, for this reason, at a level below “objectives” seems a disproportionate analytical conclusion to draw from this fact. Consequently, the Panel effectively exempted these “interests” from scrutiny other than non-discrimination or even-handedness. This is a problem, because—as previously noted—while certain concerns are very worthy causes in general, it is far from clear that such concerns should be legitimate drivers of trade measures.⁶⁷

The second problem is the Panel’s approach to comparing the contribution of the challenged measure and the proposed alternative. Although the Panel emphasised that it had to examine the “actual” contribution of the EU’s measure to its stated objective,⁶⁸ the Panel’s analysis remained entirely conceptual and abstract. In the Panel’s view, the ban made some (unspecified) contribution to its stated objective; however, that contribution was then to some (unspecified) extent diminished because of the IC and MRM exceptions, which permitted inhumanely killed seals and their derivate products to reach the EU market. Finally, however, that diminution was reversed to some (unspecified) extent because Inuit communities were unable to fully benefit from the IC exception. The analysis of the complainants’ alternatives remained similarly vague.

Admittedly, it is difficult to do full justice to the Panel’s analysis without detailed knowledge of the complainants’ arguments. However, at the same time, the Panel’s criticism that the complainants did not sufficiently specify the particular animal welfare standard under the alternatives; and that the complainants had only indicated a range of measures of varying leniency and stringency also may not be fair. This is because that precise standard would be linked to the EU’s detailed choice, driven by its own understanding of public moral concerns and their effective operationalisation within EU domestic law.

One of the Panel’s ultimate conclusions, presumably intended to point to an important perceived shortcoming of the alternative measure, that the complainants “do not deny […] that inhumane killing and poor animal welfare outcomes do occur in seal hunts” even when animal welfare standards are applied, rings rather hollow.⁶⁹ That conclusion ultimately means only that the alternative measure does not ensure to 100 % that inhumanely killed seals will never reach the EU market. However, this was precisely the Panel’s conclusion also with respect to the current EU regime. In fact, one could even draw the contrary conclusion that the proposed alternative provided for a higher level of protection. In fact, the Panel’s skepticism as to whether welfare standards, certification and labelling—under the proposed alternative measures—could be sufficiently well administered seems misplaced, because the EU’s measure had no such standards, certification and labelling at all.

The problem is, to a large extent, that the Panel nowhere specified and adequately compared the precise degrees of contribution of the applicable and the alternative measures. However, this of course is precisely the point of the analysis under Article 2.2, so the Panel arguably failed to perform a crucial aspect of its work. Taking the Panel finding at face value, the existence of an alternative that—just like the measure at issue—makes only a partial contribution to the stated objective, but is less trade restrictive (which the proposed alternative measure undoubtedly was) should lead to a finding of violation of Article 2.2. The Appellate Body’s decision may thus be an important clarification of how precisely a Panel should specify, quantify and compare the contribution of a measure and the proposed alternative to the stated objective.

US – Stainless Steel (Mexico) (21.5) (DS344)

Facts and Salient Aspects

In this dispute, Mexico claimed that the United States had failed to implement the findings in the original dispute, US – Stainless Steel (Mexico). The original case—decided by the Appellate Body in 2008—was one of several disputes addressing the United States’ practice of calculating dumping margins in anti-dumping determinations, using the so-called “zeroing” methodology.⁷⁰ The original Panel decided to set aside a series of previous Appellate Body rulings and found that “zeroing” was consistent with the Anti-dumping Agreement. The Appellate Body overruled the Panel, confirmed its rejection of the zeroing methodology and expressed deep systemic concerns about the Panel’s refusal to follow clear Appellate Body precedent. That Appellate Body report is often quoted far beyond the anti-dumping context, because it established a new standard for precedent in WTO dispute settlement. In brief, Panels may depart from previous Appellate Body rulings only where “cogent reasons” exist for doing so.⁷¹

In 2009, Mexico initiated this Article 21.5 of the Dispute Settlement Understanding (DSU⁷²) dispute, arguing that the United States had not implemented the rulings in that dispute and continued to be in violation of its WTO obligations. The 21.5 Panel was established in September 2010 and issued its final report to the parties in March of 2012. In April 2012, at the request of Mexico, the Panel suspended its work. The suspension was subsequently extended, until the parties in April 2013 notified the Panel that they had reached a mutually agreeable solution.⁷³ The mutual agreed solution identifies a number of United States Department of Commerce determinations that appear to be recalculations of dumping margins without the use of the zeroing methodology.

Additional Observations

It is not very common that WTO disputes are settled at such a relatively late stage, namely, after a Panel report has been issued to the parties. To the authors’ knowledge, a settlement after the conclusion of significant procedural steps has occurred only in two previous disputes.⁷⁴ At the same time, the initial issuance of the final reports to the parties—before circulation to the Membership at large and the global public—does serve, inter alia, precisely the purpose of permitting the parties to reach a settlement. The fact that such settlements do not occur very often is presumably because, at that stage, there is certainty about the distribution of gains and losses, and the winning party may require a very high price for a settlement. The losing party, despite its wish to suppress the publication of an unfavourable judgment, may simply be unable or unwilling to pay that price. Moreover, with political stakes on both sides in a typical WTO dispute, there is strong domestic pressure on the complainant to take the WTO dispute settlement process to its conclusion. In this dispute, the technical nature of the issue and the previous extensive case law may have contributed to the parties’ willingness for a mutually agreed solution.

Canada – Feed-in Tariff Program (DS 412, DS 426)

These disputes received significant attention for being the first Panel and Appellate Body reports dealing with renewable energy subsidies. A frequently-heard concern in policy and academic circles is that WTO subsidy rules may be excessively restrictive in tying the hands of governments wishing to promote renewable energy. However, in this dispute, the challenged measure was a local content subsidy that—although formally embedded in a governmental renewable electricity generation scheme—was itself unrelated to strictly ecological concerns.

Facts of the Case

In 2009, the Government of Ontario implemented a scheme to increase the supply of electricity generated from renewable sources called the Feed-in Tariff Programme (the “FIT Programme”). Renewable electricity suppliers who participated in the FIT Programme would feed their electricity into the electricity system under 20-year or 40-year contracts with the Ontario Power Authority (OPA) and were paid a guaranteed price per kWh. These participants were located in Ontario and produced electricity from several energy sources, i.e. wind, solar photovoltaic, renewable biomass, biogas, landfill gas, and waterpower. Each renewable energy supplier that entered into the FIT Programme was remunerated, in accordance with a formula based on a standard “Contract Price” established by the OPA.

Additionally, and crucially for this dispute, the FIT Programme imposed a “Minimum Required Domestic Content Levels”. Under these rules, a renewable energy supplier was eligible for participating in the FIT Programme only if its solar or wind power generating facilities contained a minimum percentage of locally produced goods, for instance, wind turbines.

The Appellate Body report in this dispute was issued in May 2013. The preceding Panel report was circulated in late December 2012, was covered in last year’s EYIEL issue and is therefore technically outside the scope of this dispute. However, because an understanding of the Panel report is required to fully appreciate the Appellate Body report we include the Panel report again in this year’s overview.

General Overview of the Panel and Appellate Body Findings

The Panel found that the requirement to use a certain minimum percentage of local content fell within the scope of Paragraph 1(a) of the Illustrative List of the TRIMs Agreement. As such, the local content requirement was inconsistent with Article III:4 of the GATT 1994 and with Article 2.1 of the TRIMs Agreement. In reaching that finding, the Panel rejected Canada’s argument that the measures were government procurement activities covered by Article III:8(a) of the GATT 1994 and as such exempted from the scope of Article III:4.

However, the Panel rejected the complainants’ claims under Articles 3.1(b) and 3.2 of the SCM Agreement. The Panel accepted that a subsidy existed, because the government provided a “financial contribution” under Article 1.1(a)(1)(iii) in the form of “government purchases of goods” (electricity). However, the Panel found that the complainants failed to establish the existence of a “benefit” within the meaning of Article 1.1(b), because they had not identified a proper comparison benchmark against which to compare the electricity prices paid by the OPA to the electricity suppliers.

The Appellate Body upheld some of the Panel’s findings and reversed others. It confirmed the Panel’s conclusion related to Article III:4 of the GATT 1994 and Article 2.1 of the TRIMs Agreement. In reaching those findings, the Appellate Body also confirmed that the measures at issue were not covered by the government procurement exemption under Article III:8(a) of the GATT 1994. However, the Appellate Body relied on different reasons than the Panel. The Appellate Body also upheld the Panel’s finding regarding financial contribution. However, the Appellate Body reversed the Panel’s finding on benefit. It adopted a different analytical approach to benefit than the Panel, but was unable to complete the analysis. Thus, the Article 3.1(b) and 3.2 claims ultimately remained unresolved.

Salient Legal Findings

Article III:8(a) of the GATT 1994

Summary of the Panel and Appellate Body Findings

Article III:8(a) of the GATT 1994 exempts from the national treatment obligation those laws, regulations or requirements that govern government procurement. In other words, procurement that favours domestic products and disadvantage imported goods is carved out from the scope of Article III:4. However, this exception applies only when such procurement is undertaken “for governmental purposes and not with a view to commercial resale or with a view to use in the production of goods for commercial sale”.

The Panel agreed with Canada that the FIT Programme and Contracts “governed” procurement, because it saw a “close relationship” between the products affected by the domestic content requirements (the products used to build the energy-generating infrastructure, e.g. wind turbines) and the product procured (electricity).⁷⁵ However, the Government of Ontario and municipal governments earned profits from the resale of electricity under the FIT Programme and Contracts. Therefore, the purchase of electricity by the Government of Ontario was undertaken “with a view to commercial resale” and was therefore not government procurement within the meaning of Article III:8(a). As a result, the disciplines of Article III:4 of the GATT 1994 (and of the TRIMS Agreement) remained applicable.

The Appellate Body reached the same conclusion as the Panel, but for different reasons. The Appellate Body disagreed with the Panel’s finding that the Canadian local content requirement was a legal requirement “governing” the procurement of electricity. To recall, the Panel had to address the fact that the procurement activity concerned a product (electricity) that was different from the product subject to the local content requirement (e.g. wind turbines). The Panel bridged this gap by claiming that there was a “close relationship” between the two products and considered this relationship a sufficient trigger for Article III:8(a). In contrast, the Appellate Body required a significantly tighter link, in that the product procured of foreign origin must be in a competitive relationship with the product purchased. Because renewable energy generation equipment and electricity cannot be said to be in a competitive relationship, Article III:8 was not triggered and Canada could not benefit from the exemption on the national treatment requirement. The Panel finding was thus reversed, but the ultimate result under Article III:8 remained the same.

Observations

These disputes were the first time that the Appellate Body interpreted Article III:8(a) of the GATT 1994. The Appellate Body’s clarification of the term “governing” tightens the requirements for a measure to benefit from the exemption under Article III:8(a). The Panel’s “close link” standard appears as both unnecessarily generous to defendants, as well as too vague to provide sufficient ex ante guidance. The Appellate Body’s focus on likeness and competitive relationship arguably provides a clearer and more predictable framework.

However, the Panel’s approach appears to have been rooted—at least in part—in a concern regarding inputs. For instance, a government may be procuring cars and accepts both domestic and foreign car. However, the government requires that the tyres of the cars procured be of domestic production only. It may make sense, as a matter of policy, to permit the government to employ this form of discrimination in its procurement activities because it may be deemed a “lesser evil” than requiring that the entire car be of domestic production. However, cars and tyres are not in direct competition, such that the Appellate Body’s “competitive relationship” standard would not be satisfied and Article III:8(a) therefore not triggered. Nevertheless, perhaps conscious of this matter, the Appellate Body explicitly left unresolved the question whether the derogation in Article III:8(a) can also refer to “inputs” or “processes of production” used in respect of products purchased by way of procurement.⁷⁶

Article 1.1(a) SCM Agreement: “Financial Contribution”

The complainants challenged the local content requirement as a prohibited subsidy under Article 3.1(b) of the SCM Agreement. To reach Article 3.1(b), the Panel first had to find whether the requirement conferred a subsidy. Article 1 requires two elements for a subsidy—a “financial contribution” (or income or price support) and “benefit”.

Summary of the Panel and Appellate Body Findings

The complainants argued that a “financial contribution” existed, because the government both purchased goods as well as provided a “direct transfer of funds”. The Panel accepted that the government was purchasing goods, but it rejected the argument that a “direct transfer of funds” existed. The Panel’s reasoning was, in essence, that the various forms of financial contributions under the SCM Agreement were mutually exclusive. For that reason, a transaction properly characterised as a “purchase of goods” could not, at the same time, constitute a “direct transfer of funds”, otherwise the principle of effective treaty interpretation would be violated.⁷⁷ The Panel also found that the measure in this case was properly characterised as a governmental purchase of goods.

The Appellate Body confirmed the Panel’s finding that the transactions between the government and the energy suppliers were appropriately characterised as governmental purchase of goods. It also agreed that there were no additional, independent elements that would also permit characterising these transactions as a “direct transfer of funds”. However, the Appellate Body disagreed with the Panel that the forms of financials contribution under Article 1 were mutually exclusive. It also rejected the Panel’s argument that characterising the same transactions as both a governmental purchase of goods and a direct transfer of funds would necessarily violate the principle of effective treaty interpretation.⁷⁸

Observations

The Appellate Body’s finding is a useful clarification of the relationship between the various forms of financial contribution under Article 1 SCM Agreement. This is important, because different types of financial contributions have different implications for the “benefit” test, e.g. they may imply a lesser evidentiary threshold for the complainant. The Panel’s approach whereby all forms of financial contribution would be mutually exclusive is more constraining for the complainant than the Appellate Body’s approach. At the same time, it is not entirely clear how exactly the “direct transfer of funds” differs from the governmental purchase of goods and which “different” characteristics the Appellate Body had in mind.⁷⁹

Article 1.1(a) SCM Agreement: “Benefit”

Summary of the Panel and Appellate Body Findings

The complainants argued that the challenged measures conferred a benefit, based on two distinct and independent benchmarks. They argued (1) that the guaranteed price paid to the renewable energy generators exceeded the price on the wholesale and retail markets in Ontario (or, alternatively, in four alternative comparison markets); and (2) that the inherent nature of the FIT Programme was to facilitate private investment in new renewable energy generation that the market, on its own, would be incapable of attracting. Under this second benchmark, the mere existence and operation of energy suppliers that would have otherwise not entered the market suggested the existence of benefit.

The Panel rejected both benchmarks. As regards (1)—that is, the price on the Ontario market—a key finding of the Panel was that electricity and competitive wholesale electricity markets exhibit certain idiosyncratic features, in particular the inability to attract sufficient investment in the generation capacity infrastructure (referred to as the “missing money” problem). As a result, strong governmental intervention would often be necessary to secure an electricity supply that is safe, reliable and sustainable. Moreover, the equilibrium price in the electricity whole sale market was not determined through forces of supply and demand, but rather was based on pervasive government intervention, and reflected the government’s pricing policy as well as the desired mix of energy sources. Hence, no “natural” market with sufficiently free forces of supply and demand existed in Ontario. For these reasons, the Panel did not consider these parameters as an appropriate market benchmark.

The Panel also rejected arguments regarding wholesale electricity markets outside of Ontario, because these markets either also failed to attract sufficient generation capacity investment or because they were not comparable to Ontario.

The Panel also rejected benchmark (2), namely, the allegation that the mere existence of renewable energy generation demonstrated a benefit. The Panel pointed once more to the inherent challenges faced by electricity markets as well as the particular situation in Ontario. Because the amount of investment in generating capacity needed to secure a reliable electricity supply could only be secured via government intervention, and because competitive market outcomes would not be acceptable as an alternative, the Panel found the proposed benchmark not to be appropriate.

In sum, the Panel found that the “competitive wholesale electricity market that is at the centre of the complainants’ main submissions cannot be the appropriate focus of the benefit analysis in these disputes”.⁸⁰

Nevertheless, at the request of the complainants, the Panel also identified a benefit benchmark that it considered would have been appropriate and acceptable. That benchmark, according to the Panel, was the rate of return of projects with a comparable risk profile in the same period. This rate of return could be compared with the rate of return achieved by the FIT programme beneficiaries. However, the Panel saw insufficient facts on the record to determine that appropriate benchmark rate of return.

One Panelist dissented from the majority’s view. Overall, the dissenter revealed less skepticism about the inherent failures of a freely competitive market for electricity and found that a potential competitive market could in principle be an appropriate benchmark. He found that “the fact that a market is imperfect in its operation or does not meet the objectives that a government might have for the goods or services [at issue] does not shield financial contributions […] in th[at] market from the benefit analysis […] under the SCM Agreement.”⁸¹ Like the majority, the dissenter ultimately rejected the price benchmarks proposed by the complainants. However, he accepted the proposed benchmark (2)—namely, that by bringing high cost renewable electricity producers into the wholesale electricity market through government-sponsored programmes, when these producers otherwise would not be present, the Government of Ontario’s purchases of electricity from these generators conferred a benefit upon those very same generators.⁸²

The Appellate Body reversed the Panel. It found that the appropriate market was not the overall wholesale electricity market (as the Panel had found) but rather only the market for renewable energy—that is, wind and solar energy. One reason given by the Appellate Body was that both on the demand side and supply side, renewable and non-renewable electricity were distinct. On the demand side, the Appellate Body mentioned criteria such as type of contract, customer size and standard vs. peak electricity needs. On the supply side, the Appellate Body highlighted inter alia different cost structures and operating costs of the different categories of producers. Another factor was the government’s determination of what the Appellate Body called the “supply-mix” for electricity. The government’s policy of fostering and including renewable energy among the desired sources of electricity meant—at least in this case—that the government had created a market for wind- and solar PV-generated electricity.

Having decided that the relevant market was the market only for wind and solar energy, the Appellate Body then set out to define the proper benchmark for that market. It emphasised that the fact that the government had created the market at issue, it did not preclude a search for a market-based benchmark within that market. In justifying this approach, the Appellate Body drew a distinction between markets that, although created by the government, could still function in a competitive fashion and where a benchmark could be found; and, in contrast, pre-existing markets that were distorted by governmental intervention, to the point where no benchmark could be identified in that market.

The Appellate Body then rejected both benchmarks put forward by the complainants. The first, price-based benchmark was rejected because it related to the overall market—energy from all sources—and not solely to the solar and wind power market. The second benchmark—the mere presence of otherwise non-competitive high-cost renewable energy suppliers in the overall market—was also rejected. The relevant question according to the Appellate Body was whether, absent the FIT programme, the renewable energy suppliers would have entered the wind and solar market, not whether they would have entered the overall wholesale market.

The Appellate Body then attempted to complete the legal analysis and to identify the proper market benchmark itself. Not surprisingly, the government-administered FIT price was not considered a market-based benchmark. The Appellate Body then attempted to identify a benchmark in competitive bidding prices offered by renewable energy suppliers under a predecessor programme of the FIT. However, it found that there were insufficient facts on the record. The Appellate Body saw some indications that the FIT prices for wind power were above the wind power prices under the predecessor programme, but these facts did not in its view suffice to reach a proper finding.

Observations

Perhaps the biggest question surrounding the Appellate Body’s benefit ruling is whether its logic is confined to the idiosyncratic situation of electricity markets or whether it can be applied to other industries as well. Previous Appellate Body caselaw establishes that a government that intervenes strongly in a market may distort that market to the point where a market-based benefit benchmark can no longer be found; in that scenario. The benchmark can only be identified in outside, proxy markets (such as in US – Softwood Lumber IV ⁸³). However, the present case demonstrates that, if the government intervenes in the market to such an extent that an entirely new category of high-cost suppliers is enabled to participate, then the goods offered by these suppliers are suddenly considered to be a free-standing, separate market and a benchmark must be found within that narrower market. This approach is by no means obvious and raises questions about how the benefit requirement will be applied in the future, in cases involving different types and different degrees of governmental intervention. The Appellate Body’s approach could possibly mean that, where a government through a far-reaching industrial policy intervention creates a new industry that otherwise would not exist, no benefit is conferred merely by virtue of the fact that the producers would not otherwise exit. The consequence would be, for instance, that the monies disbursed to create and enable these enterprises to operate would not constitute a subsidy and no countervailable duties could be imposed.

In market definition, the Appellate Body’s list of demand and supply-side factors that allegedly differentiate the renewable from the non-renewable electricity markets is also remarkable for its attention to minute detail. Arguably, with such a microscopic eye for differences, one wonders whether the Appellate Body would have found, for instance, in US – Upland Cotton that there was a single world market for cotton.⁸⁴

Perhaps all of these aspects suggest that the Appellate Body’s ruling is indeed particular to electricity markets. The Appellate Body’s reasoning may have been driven by the wish to provide governments with policy space in the area of renewable energy. The complainants’ and dissenter’s approach would affirm the existence of a benefit whenever the government enables otherwise non-competitive suppliers to compete in a market. In contrast, the Appellate Body considers these otherwise non-competitive suppliers to be operating in an entirely separate narrower market, for which rules have to be found within that narrower market. This makes it more challenging for complainants to demonstrate the existence of benefit. Perhaps raising the benefit bar in this manner shields renewable energy and perhaps other socially desirable goods from what some may consider excessive scrutiny under the SCM Agreement.

However, one may wonder whether this concern—assuming this was indeed the Appellate Body’s concern, echoing the concerns voiced in scholarly writing—about the vulnerability of green energy subsidies under WTO law is unnecessary. Even if benefit is affirmed under Article 1 of the SCM Agreement, a genuine green energy subsidy may very likely not be specific. Specifically, a government that promotes the use of green energy by any and all users (or objectively defined users, such as large or small industries) can rest assured that no WTO challenge will succeed. If the challenge were directed directly at the energy producers themselves, then the subsidy would have to be found to cause serious prejudice to trade in energy of other members under Articles 5 and 6 of the SCM Agreement to make this subsidy vulnerable under WTO law. This is arguably quite unlikely. Energy is traded under very particular conditions and it is doubtful whether a serious prejudice finding would ever be more than just a theoretical possibility. In any event, as shown in the present case, the complainant’s grievance was with the local content requirement, not with the renewable energy policy itself. Another similar case has been brought to the WTO, and again the target are local content requirements.⁸⁵ There is thus no indication so far that (1) genuine renewable energy subsidies are in the cross-hairs of potential complainants and (2) that genuine renewable energy subsidies would run afoul of WTO law. One can legitimately ask whether creating new concepts under the benefit analysis, or going to great lengths to modifying the existing concepts, to further shield (genuine) green energy subsidies from WTO challenges, is worth the risk that these new and untested concepts entail for other industries and other contexts.