3.1 Pre-initiation standard of evidence

The Panel ruled in favor of the US on almost all the issues except for the three claims relating to the calculations and explanations in MOFCOM's public notice/disclosure and the decision to apply ‘facts available’ in its calculation of the subsidy rates for the two respondents under certain programs.Footnote ⁶ With regard to the pre-initiation standard of evidence, the Panel found that China acted inconsistently with Article 11.3 of the SCM Agreement in initiating countervailing duty investigations into each of the 11 programs challenged by the US. China did not include this issue in its appeal to the AB.

3.2 The non-confidential summary requirement

The US claimed that MOFCOM had not required the applicants to provide adequate non-confidential summaries of confidential information in the application in violation of Article 12.4.1 of the SCM Agreement and Article 6.5.1 of the ADA.

Disputing the US claim, China made three main arguments. First, it contended that the non-confidential summaries in the application were later supplemented by a non-confidential analysis provided by MOFCOM. Second, China argued that the issue was not substantially challenged by the respondents during the investigation. Third, China reasoned that the exceptional circumstances' exemption applied in this case because there were only two Chinese producers (i.e., WISCO and Baosteel) whose confidential information would have been easily discernible by each other, so disaggregation of the data would disrupt the confidentiality afforded to the applicants.

Concerning the first argument, the Panel considered that the non-confidential summaries needed to be provided before the investigation authority reached its determinations as otherwise an interested party would not have an adequate opportunity to defend its interests. Regarding the second argument, the Panel held that the standard by which confidentiality is granted by an investigating authority cannot be changed irrespective of the sufficiency of the challenge from the respondents. Thus, the fact that the respondents in this case did not seriously contest the matter carried no weight with the Panel. Lastly, it was determined that the record did not establish that the exemption was invoked in this case, because the applicants failed to ‘identify the exceptional circumstances and provide a statement explaining why summarization is not possible’ as required by the Articles, as interpreted by AB and Panel reports.Footnote ⁴¹

The Panel was then left with the question of whether the non-confidential summaries provided permitted ‘a reasonable understanding of the information submitted in confidence’.Footnote ⁴² In evaluating Part II of the application where the non-confidential summaries could be found, the Panel determined that the summaries consisted of ‘minimal descriptions of the nature, rather than the substance’ of the confidential information,Footnote ⁴³ thereby failing to meet the reasonable understanding standard required by Article 12.4.1 of the SCM Agreement and Article 6.5.1 of the ADA. China also argued that non-confidential summaries were provided in Part I of the application since they could have been derived or inferred from the context therein. But the Panel found that at least some of the categories of non-confidential information, such as the output and consumption of GOES in China and the impact on the Chinese domestic industry of the dumping, were deficient even if the Panel agreed with China that the non-confidential summaries could be ‘interspersed throughout the body of an application’ as such practice was not clearly prohibited by the text of the Articles.Footnote ⁴⁴ In addition, China argued that a non-confidential summary exists where an interested party is able ‘to derive, or infer from the context, the possible nature of the confidential information’.Footnote ⁴⁵ The Panel found the argument unconvincing and stated that the interested party providing the information has the explicit obligation to provide the summary under the Articles and that it is not for the others to derive their own summaries. Thus, the Panel concluded that China acted inconsistently with Article 12.4.1 of the SCM Agreement and Article 6.5.1 of the ADA. This issue was not challenged on the appeal by China.

3.3 The use and application of ‘facts available’

3.4 The price effects analysis

In challenging MOFCOM's price effects findings in the injury determination, the US claimed that China failed to meet the two requirements set out for the investigating authority under Articles 3.1 and 3.2 of the ADA and Articles 15.1 and 15.2 of the SCM Agreement in conducting the price effects analysis, which were to base the findings on positive evidence and engage in an objective examination of the evidence provided. More specifically, the US argued that MOFCOM failed to demonstrate that the evidence supported the conclusion that any price depression and suppression during the period of investigation was an effect of imports.

With respect to the price depression findings, the Panel first considered the question of whether there was an obligation for the investigating authority to show that the relevant price depression was an effect of subject imports according to Article 3.2 of the ADA and Article 15.2 of the SCM Agreement. The Panel agreed with the US in finding that the authority must establish that there was a causal link between the imports concerned and the price depression. The Panel then evaluated whether MOFCOM's finding that price depression was an effect of such imports was based on positive evidence and involved an objective examination. In its defense, China primarily relied on the evidence in the application and collected by MOFCOM during the investigation, the latter mainly consisting of price comparisons between the average import prices and domestic sales prices. Regarding the evidence in the application, the Panel found that it could not be considered as positive evidence because it was not incorporated into MOFCOM's Final Determination and even contradictory to MOFCOM's own finding. As for the price comparisons, the Panel noted that MOFCOM's neglect in making any price adjustments to ensure price comparability made the analysis ‘neither objective, nor based on positive evidence’.Footnote ⁵¹ Additionally, China argued that the price depression finding could have been supported by the increase in the volume of imports alone. The Panel rejected this argument and stated that MOFCOM seemed to have shown equal consideration to both volume and price effects in its analysis and therefore the volume effect alone could not be considered sufficient as the primary basis of the finding as contended by China. Based on the evaluation, the Panel concluded that MOFCOM's price depression findings were neither based on positive evidence nor reached through an objective examination. Therefore, they were inconsistent with the relevant Articles.

The Panel also upheld the US claims regarding the price suppression findings, because the same arguments were put forth by the US and China, with the additional claim from the US that MOFCOM overly relied on the changes in the price–cost ratio in its analysis when these ‘merely reflected changes in the underlying cost structure of the domestic industry’.Footnote ⁵² Scrutinizing the factual circumstances surrounding the Chinese producers (WISCO and Baosteel), the Panel determined that MOFCOM should have considered the underlying changes in the domestic industry in its analysis. However, it was not for the Panel to conduct a de novo review on this complex issue. In any event, the deficiencies found in MOFCOM's price depression findings existed with price suppression as well. Additionally, the Panel agreed with the US that MOFCOM failed to provide an express finding of significant price undercutting by subject imports. Based on these reasons, the Panel held that China acted inconsistently with Articles 3.1 and 3.2 of the ADA and Articles 15.1 and 15.2 of the SCM Agreement.

On appeal, China challenged the Panel's holdings with respect to the price depression and suppression findings, arguing that the investigating authority is not obligated to establish the causal link between the imports concerned and the price depression and suppression. The AB upheld the Panel's findings, noting that the obligation for the investigating authority under Article 3.2 of the ADA and Article 15.2 of the SCM Agreement was to consider the relationship between the imports and domestic prices ‘so as to understand whether subject imports provide explanatory force for the occurrence of significant depression or suppression of domestic prices’.Footnote ⁵³ In focusing on that relationship, the AB further explained that, even though the investigating authority is not required to examine all possible factors causing injury to the domestic industry, it may not disregard evidence which disputes the causal link between the subject imports and the significant depression or suppression of domestic prices.Footnote ⁵⁴

With regard to the interpretation of Articles 3.2 and 15.2, China also alleged that the Panel erred in distinguishing the panel report of EC–Countervailing Measures on DRAM Chips Footnote ⁵⁵ in considering the relevant obligations under the Articles and wrongly interpreted the legal standard through its examination of the MOFCOM's finding of significant price suppression. According to China, the panel report in the EC–Countervailing Measures on DRAM Chips dispute provides that the investigating authority is not required to find the casual link between the subject imports and price effects under Article 15.2 of the SCM Agreement, which the Panel in this dispute failed to recognize. In response, the AB stated that the EC–Countervailing Measures on DRAM Chips dispute was concerned with the effects of price undercutting rather than price depression or suppression caused by subject import and therefore cannot be used to support the proposition put forth by China.

As for the Panel's examination of MOFCOM's finding, China considered that the Panel incorrectly imposed a requirement of determining subject imports as the ‘only reason’ for the price suppression under Articles 3.2 and 15.2 by concluding that ‘MOFCOM's finding was not based on positive evidence and did not involve an objective examination’ because it could not have determined that subject imports were ‘the only reason’ for the price suppression.Footnote ⁵⁶ It was noted by the AB that this argument advanced by China was not directed at the Panel's articulation of the legal standard under the Articles but at the actual practice adopted by the Panel in its examination. The AB then reviewed the Panel's analysis and found that China misconstrued the statement about ‘the only reason prices were not able to rise with costs’ as a standard imposed by the Panel, when it was discussed in reference to the assumption made by MOFCOM in considering the changes in the price–cost ratio of like domestic products.Footnote ⁵⁷ Moreover, the AB reached the conclusion that the Panel had properly focused on the lack of supporting evidence underlying MOFCOM's finding in the analysis.

Having resolved the issues regarding the interpretation of Articles 3.2 and 15.2, the AB considered China's claim with respect to the Panel's assessment of the price effects findings, namely that the Panel had misunderstood MOFCOM's analysis by focusing on the single factor of ‘low price’ in the examination of the Final Determination and failed to properly consider the other factors involved in the analysis, which were a pricing policy aiming at price undercutting, parallel price trends between subject import and domestic prices, and increases in subject import volume. Before turning to the examinations of the factors abovementioned, the AB stated that the Panel appropriately identified the issue as being whether MOFCOM met the standard for price effects analysis conducted on the basis of positive evidence and involving an objective examination.

Additionally, China contended that the Panel had violated its obligations under Article 11 of the Dispute Settlement Understanding (DSU) through ‘misreading of MOFCOM's findings of fact and its misapplication of these findings to the legal standard “so exceeded the proper bounds of review”’,Footnote ⁵⁸ particularly with respect to the ‘low price’ reference in the Final Determination.

Regarding China's key argument of the over-emphasis placed by the Panel on the ‘low price’ factor, the AB determined that the Panel properly relied on the price evidence in its assessment because it formed a significant part of the basis for MOFCOM's price effects findings. Though China sought to explain that MOFCOM's reference to ‘low price’ was not the same as the existence of price undercutting, and price comparability was not depended upon in reaching the finding of significant price depression and suppression, China's responses to the Panel demonstrated to the contrary in the view of the AB. In detail, the AB pointed, among others, to China's second response to the Panel's request of 18 November 2011 and the response to Panel Question 69 following the second Panel meeting to show that, even though China did not make a finding of price undercutting, it nevertheless relied on the ‘low price’ of subject imports to find significant price depression and suppression in the Final Determination.Footnote ⁵⁹ In addition, the AB also agreed with the Panel's position that price adjustments were not out of line under Articles 3.2 and 15.2 because, ‘if subject import and domestic prices were not comparable, [it] would defeat the explanatory force that subject import prices might have for the depression or suppression of domestic like products’.Footnote ⁶⁰

However, the AB did find that it was improper for the Panel to dismiss MOFCOM's reliance on the pricing policy because the Panel failed to examine whether, in the absence of price undercutting, the evidence could support a finding of significant price depression and suppression.Footnote ⁶¹ In the AB's view, the relevance of the policy to price depression and suppression should have been examined by the Panel since ‘a policy aimed at price undercutting may very well depress and suppress domestic prices in instances where, as China asserts, “domestic producers were reacting to subject import competition and were lowering domestic prices so as to compete more effectively and minimize any further loss of market share”’.Footnote ⁶² The Panel rejected the pricing policy based on the fact that subject import prices were higher than the prices of domestic like products, but according to the AB the information ‘does not necessarily negate the significance of a policy aimed at price undercutting for findings of price depression and suppression’.Footnote ⁶³ Nonetheless, the AB's disagreement with the Panel's treatment of the pricing policy was not sufficient to reverse the Panel's overall finding of inconsistency under Articles 3.1 and 3.2 of the ADA and Articles 15.1 and 15.2 of the SCM Agreement because there was no ‘foundation to conclude that MOFCOM's analysis of the effects of subject import prices could have been sustained on the basis of the pricing policy alone’.Footnote ⁶⁴

Regarding China's claim of the Panel's failure to consider the parallel trends of subject imports and domestic prices, the AB acknowledged that the existence of parallel trends might signal competition between the subject imports and domestic products, which could help explain the effects of the former on the prices of the latter to some extent. However, there was a lack of sufficient evidence in the Final Determination to draw a conclusion since MOFCOM referred to the price trends as being ‘basically the same’ without providing any further description or explanation.Footnote ⁶⁵ Indeed, the AB noted that it would have expected MOFCOM to provide some reasoning or explanation in the Final Determination as to the analysis of the movements in price trends if it had actually considered the parallel price trends significant for the price effects findings. Therefore, there was no basis to fault the Panel for failing to recognize and discuss the significance of those trends for the analysis.

Lastly, China claimed that the increases in subject import volume were the primary basis for MOFCOM's finding of significant price depression and suppression and the Panel erred in not finding them as such. In evaluating the analysis of the Panel on the issue, the AB held that the focus of the Panel was not whether subject import volume or price was the primary basis for MOFCOM's finding but how the two factors might have interacted and what effects they could have supported with respect to the findings of significant price depression and suppression. In that respect, the AB agreed with the Panel that it was not possible to ‘disentangle the relative contribution of [the volume and price] effects in MOFCOM's Final Determination without substituting its judgment for that of the authority’.Footnote ⁶⁶ According to the AB, the Panel appropriately refrained from conducting the analysis and avoided the risk of engaging in a de novo review in assessing MOFCOM's findings.

In conclusion, the AB held that the Panel did not err in its application of Article 3.2 of the ADA and Article 15.2 of the SCM Agreement.Footnote ⁶⁷ It also determined that the Panel did not act inconsistently with its obligation to make an objective assessment of MOFCOM's Final Determination under Article 11 of the DSU.

3.5 The disclosure of ‘essential facts’

In relation to the claims on the use of ‘facts available’, the US challenged that China failed to disclose the ‘essential facts’ forming the basis for applying the ‘facts available’ to ‘unknown’ US exporters under Articles 6.9, 12.2, and 12.2.2 of the ADA and Articles 12.8, 22.3, and 22.5 of the SCM Agreement. Similarly, with respect to the price effects and causation findings, the US claimed that China acted inconsistently with Articles 6.9 and 12.2.2 of the ADA and Articles 12.8 and 22.5 of the SCM Agreement by reason of the deficiencies in the related ‘essential facts’ disclosure and public notice and explanation.

The Panel upheld US claims regarding the disclosure obligations. Concerning the application of ‘facts available’, the Panel found that the disclosure of the ‘essential facts’ leading to the conclusion to apply ‘facts available’ was warranted because such information was necessary for the US to defend its interests. With regard to the price effects findings, it was considered by the Panel that the ‘essential facts’ supporting the finding of ‘low’ subject import prices were significant to the application of the definitive measures and should have been disclosed by MOFCOM. For the causation analysis, the Panel held that China failed to disclose the data on the non-subject imports as it was one of the ‘essential facts’ and part of the ‘relevant information on the matters of fact and law and reason’ which have led to the imposition of final measures.

On appeal, China's issues were limited to the price effects findings and the disclosure of underlying facts. More specifically, China brought two claims regarding the latter. First, China considered that it adequately disclosed the ‘essential facts’ in the investigation as Article 6.9 of the ADA and Article 12.8 of the SCM Agreement did not require the disclosure of the ‘essential facts’ regarding the ‘low price’ finding in support of its conclusion of significant price suppression and suppression from the subject imports. Second, China argued that the Panel incorrectly focused on the existence of price undercutting in finding that China failed to disclose ‘all relevant information on the matters of fact’ under Article 12.2.2 of the ADA and Article 22.5 of the SCM Agreement.

In examining the first issue raised by China, the AB stated that the terms of ‘essential facts’ under Articles 6.9 and 12.8 ‘refer to those facts that are significant in the process of reaching a decision as to whether or not to apply definitive measures’.Footnote ⁶⁸ In the AB's view, the purpose of the disclosure is to ensure the ability of the interested parties to defend their interests by allowing them to sufficiently understand the basis leading to the decision whether to apply definitive measures under the ADA and SCM Agreement. As discussed above, the AB found that MOFCOM relied on the findings as to the ‘low price’ of subject imports in reaching its conclusions of significant price depression and suppression. Therefore, the AB determined that MOFCOM was required to disclose the ‘essential facts’ relating to the ‘low price’ finding and it failed to do so because no such fact was included in MOFCOM's Preliminary Determination and the Final Injury Determination.Footnote ⁶⁹ Consequently, the AB upheld the Panel's finding that China acted inconsistently with Article 6.9 of the ADA and Article 12.8 of the SCM Agreement.

Regarding the second issue challenged by China, the AB found that the obligations to give public notice of the conclusion (or suspension) of the investigation under Article 12.2.2 of the ADA and Article 22.5 of the SCM Agreement come later in the process than the requirement to disclose the ‘essential facts’ pursuant to Articles 6.9 and 12.8, as the disclosure obligations under Article 12.2.2 and Article 22.5 are only triggered after an affirmative determination has been made for the imposition of definitive duties. The AB also held that what constitutes ‘relevant information on matters of fact’ has ‘to be understood in light of the content of the findings needed to satisfy the substantive requirements with respect to the imposition of final measures under the Anti-Dumping Agreement and the SCM Agreement, as well as the factual circumstances of each case’.Footnote ⁷⁰ On that basis, it was reasoned that the evidence of price undercutting was relevant to the understanding of the ‘low price’ finding and therefore it should have been disclosed by MOFCOM in the Final Determination. Responding to China's argument that it considered those facts immaterial because MOFCOM did not make any findings of significant price undercutting, or that the pricing policy of the producers disclosed by China was sufficient to constitute ‘all relevant information on the matters of fact’ with regard to the ‘low price’ subject imports, the AB stated that it disagreed with China because the arguments were based on the premises that the price undercutting finding was not relied on by MOFCOM for its conclusions of significant price depression and suppression. As a result, the AB upheld the Panel's finding that China failed to disclose the ‘relevant information on the matters of fact’ within the meaning of Article 12.2.2 of the ADA and Article 22.5 of the SCM Agreement.

4.1 Pre-initiation standard of evidence

The Panel agreed with the United States that a CVD application must contain, on a program-by-program basis, evidence of (1) a financial contribution by the government, (2) a benefit to the recipient, and (3) specificity in the case of domestic subsidies. This may seem obvious in light of the legal requirements of the SCM Agreement, but it is often easier said than done. As shown in Table 1, in the GOES case the panel found with regard to the 11 challenged programs that in three instances the application did not contain evidence of a financial contribution, in six instances no evidence of a benefit, and in six instances no evidence of specificity.

Table 1. The 11 subsidy programs evaluated by the Panel in China–GOES

Notes: * For certain programs listed above, the Panel considered it not necessary to evaluate the other evidence of the existence of evidentiary requirements after the evidence for one requirement has been found insufficient, see State of Indiana Steel Industry Advisory Service, Panel Report, at para. 7.98; the Grace Period for Compliance with the Clean Air Act, at para. 7.104; Pennsylvania's Alternative Energy Funding Plan, at para. 7.119; the Provision of Natural Gas, at paras. 7.125, 7.127–7.128; the Provision of Electricity, at paras. 7.135, 7.137–7.138; the Provision of Coal, at paras. 7.141–7.146.

We agree with the Panel's view that while the application need not contain sufficient evidence to allow the investigating authority to draw definitive conclusions regarding the existence of a subsidy, injury, or a causal link between the two, it must have some evidence and not simply contain mere assertions. One possible interpretation is that the Chinese firms were mimicking what they perceive to be the practice of US firms when they file AD/CVD complaints against China. US CVD applications often refer to five-year plans of the Communist Party of China as evidence of illegal government support for the Chinese steel industry. Perhaps the Chinese, frustrated by similar accusations, felt that vague references to long defunct government programs were an acceptable basis for an application.

Ironically, the Panel findings, while going against China in GOES, may help China overall as it is now the main target of CVD investigations by other WTO members. As shown in Table 2, since 2006 China has been the subject of almost half of all CVD cases worldwide and almost three-quarters of all US cases. Many of the underlying applications (in which China is the target) tend to be ‘cut and paste’ jobs of previous applications which means that detailed evidence with regard to one or more of the three elements is often lacking. Thus, it seems only a question of time before China will challenge on similar grounds imposition of CVD measures against it by countries such as the US and the EU.

Table 2. Worldwide CVD activity, 2006–2011Footnote ⁷¹

4.2 Facts available

Article 12.7 of the SCM Agreement provides that:

This article is frequently invoked by administering authorities in CVD cases, among others, because the massive amount of information requested by them often does not make it possible for respondents, whether governmental or non-governmental, to fully respond to the questionnaire. To some extent, broad data requests are unavoidable in CVD cases, especially as regards non-recurring subsidies. This is because non-recurring subsidies are normally allocated over time, depending on the useful life of the assets to which they relate. Suppose, for example, that in 1980a government provides a plot of land for free. Suppose further the useful life of the land is 50 years, the company that received the land for free will enjoy the benefit of that land for 50 years, in other words until 2030. Therefore, in the course of a CVD investigation, administering authorities may decide to request information relating to events that occurred many years ago in order to capture such benefits.

The other side of the coin, of course, is that overly broad information requests may make it de facto impossible for interested parties to provide all the information requested which will then lead the administering authorities to use facts available. A good example of this is the practice of some administering authorities to request the exporting country government to provide transaction-by-transaction information relating to all transactions during the investigation period between input suppliers and producers of the product under investigation. From a legal perspective, the issue would seem to be whether such information is ‘necessary’ within the meaning of Article 12.7 SCM Agreement. This issue was raised by the US, but rejected by the Panel. The US did not appeal this Panel finding likely because the US did not wish to tie its hands in its own CVD practice. Nevertheless, a good argument can be made that Panels should more rigorously examine whether information requests by administering authorities in CVD cases are ‘necessary’ within the meaning of Article 12.7 SCM, particularly as the failure to provide requested information is so often used in CVD practice as ‘the stick to beat the dog’.Footnote ⁷²

There are typically two issues at stake then. First, whether resort to facts available is justified or, in other words, whether the conditions of Article 12.7 SCM Agreement are satisfied. Second, if the first question is answered in the affirmative, authorities must determine which facts are available to use. As regards the two respondents AK Steel and ATI, the facts available issue came up because the two respondents had not provided transaction data on domestic sales which supposedly would have enabled MOFCOM to determine whether they had benefitted – directly or indirectly – from government procurement at more than adequate remuneration.

Regarding the first question, the Panel considered that MOFCOM properly found that the respondents failed to cooperate with MOFCOM's requests (notably in the deficiency letter) to provide transaction data for domestic sales of GOES and non-GOES products during the period of investigation (POI). As a result, MOFCOM had by way of facts available assumed that AK Steel and ATI had sold all of their output to the government under programs requiring the payment of a 25% price premium (referred to in the panel report as a 100% utilization rate). The Panel referred to the AB report in Mexico–Rice Footnote ⁷³ for the propositions that facts available means ‘best’ facts and that any resulting findings must be based on facts. The Panel considered that MOFCOM's finding was not based on facts and rejected MOFCOM's argument that it could draw adverse inferences from the respondents' failure to cooperate.

The Panel could have left it at that, but went on in great detail to explain why MOFCOM's finding in fact was illogical and contradicted by information provided in the case.Footnote ⁷⁵

The US also challenged China's use of facts available to calculate a dumping margin of 64.8% for unknown exporters as being inconsistent with Article 6.8 and paragraph 1 of Annex II of the ADA.

Paragraph 1 of Annex II of the ADA provides that:

When initiating the GOES anti-dumping and countervailing duty investigations, MOFCOM had issued public notices of initiation in which it had indicated that any interested party should register for investigation and that failure to register would result in the application of best information available. MOFCOM provided the initiation notices to AK Steel and ATI, the two known US GOES exporters/producers, and placed the notices on its website and in its public reading room. MOFCOM had also notified the United States Embassy in Beijing of the initiations and requested that the Embassy notify the relevant exporters and producers. However, apart from AK Steel and ATI, no US GOES producers/exporters registered for the investigation. Therefore, MOFCOM eventually applied facts available for the ‘all others’ rate.

The Panel found that MOFCOM had violated this paragraph because the notice of initiation, relied upon by China as providing the requisite notification, did not specify in detail the information required of the interested parties for the purposes of the anti-dumping investigation.Footnote ⁷⁶ The Panel found support for its finding in the rulings of the AB in Mexico–Rice and Argentina–Ceramic Tiles.Footnote ⁷⁷

The finding of the Panel seems legally correct, but it raises difficult operational issues for administering authorities. To the best of our knowledge the steps taken by China are the ones that most authorities would undertake to deal with the problem of unknown exporters. The initial information in a trade case comes from the application in which the domestic industry will normally provide a list of known exporters. Once the case is initiated, the authorities will then send the exporters' questionnaire to such known exporters.

It is basically this questionnaire that specifies ‘in detail the information required from any interested party, and the manner in which that information should be structured by the interested party in its response’ within the meaning of paragraph 1 of annex II. Of course the questionnaire cannot be sent to unknown exporters as their identities, let alone their mailing coordinates, are not known. Therefore, the published notice of initiation will typically specify a time period during which unknown exporters can come forward, make themselves known, and request a copy of the questionnaire.

At the end of the investigation, sampled cooperating producers will normally receive individual duties, based on company-specific findings (as was the case for AK Steel and ATI), non-sampled cooperating producers will receive a weighted-average duty based on the duties imposed with respect to the cooperating sampled producers (excluding duties based on facts available and/or de minimis margins) and the authorities will then normally impose a residual or ‘all others’ duty applicable to non-cooperating producers. The ‘all others’ rate will typically be at least a high as the duty imposed with respect to any cooperating sampled producer to avoid that non-cooperating producers get a bonus for non-cooperation. The ‘all others’ rate will apply to all parties that did not cooperate in the investigation, no matter the reason for their non-cooperation. This rate will therefore be applicable to unknown exporters and even to parties that did not export (or produce!) during the original POI. The latter (but not the unknown exporters that did export during the period off investigation and therefore could have cooperated) can then have recourse to the special ‘newcomer review’ possibility laid down in Article 9.5 of the ADA.

As far as we know, this system functions well in practice, not in the least because news about a trade case normally travels fast in the industry concerned, thanks to trade associations and other parties with an interest in the case. Therefore, while there may often be producers/exporters ‘unknown’ to the authorities, there will rarely be producers/exporters that do not ‘know’ about the case. Any decision not to come forward and make themselves known to the authorities in accordance with the instructions in the notice of initiation will then often be based on a conscious decision not to cooperate, typically on the basis of a cost–benefit analysis.

It is therefore mysterious why the US decided to challenge this issue, particularly as it seemed relatively clear that there were in fact no other exporters during the POI.Footnote ⁷⁸

4.3 Non-confidential summaries

A typical application will contain a mix of both confidential business information (such as the domestic firms' pricing, production, and profits, evidence of lost sales, and often proprietary industry information) and non-confidential discussions of the industry and trends. Petitioners must be careful when preparing the non-confidential summaries so that they do not inadvertently reveal confidential information about any individual producer/exporter. The issue is most clear-cut when there are only two firms because then even an aggregate summary will convey important information.

Suppose, for example, an application were to report that domestic output fell from 100 to 90 tons. Suppose further that there are two producers (A and B) and that one of the firms (Firm A) produced 50 tons both years. Without a non-confidential public summary, Firm A could immediately deduce proprietary information about its domestic competitor (Firm B); namely that B's production fell from 50 to 40 tons.Footnote ⁷⁹ Firm B could also deduce A's production levels.

The agreements allow an ‘exceptional circumstances’ exemption under Article 12.4.1 of the SCM Agreement and Article 6.5.1 of the ADA to be claimed in such situations. While these provisions do not provide further detail on what circumstances may qualify as ‘exceptional’, supposedly a situation in which only two petitioners bring a case might qualify as an exceptional circumstance allowing for extra leeway. However, China did not explicitly make the claim; nor did the application report even general summaries of the data (e.g., unlabeled trend lines, year-over-year percentage changes without any reported absolute levels). In China's view, the fact that there were only two domestic producers meant that the ‘exceptional circumstances’ exemption was understood to be invoked.

While there is some validity to China's view, we feel the Panel was correct in rejecting China's position. First of all, claiming the exemption and providing a statement explaining why summarization is not possible is not an onerous burden on either the petitioner or the investigating authorities. Second, China's position would create a vast opening for investigating authorities to limit what they include in their reports. Effectively China seems to be saying ‘you should have known what I meant’. Had the Panel accepted China's position, respondents' ability to mount an adequate defense would require advanced degrees not just in law and economics but also in mind-reading. The requirement that the non-confidential summaries ‘permit a reasonable understanding of the substance of the information submitted in confidence‘Footnote ⁸⁰ unless the exemption is explicitly made is reasonable.

4.4 Price effects analysis of evidence

The Panel also found serious shortcomings with China's price effects' analysis. To a large extent, the core issue was that China did very little to document its basis for determining that subject imports had caused price depression and/or price suppression. Said differently, one wonders how much analysis was actually involved in its price effects ‘analysis’. The Panel's report makes it clear that reporting conclusions without any analysis and justification is insufficient.

The one piece of price effects analysis that is commonly done by investigative authorities (price–cost margin) was criticized by the Panel as being insufficient because it ‘merely reflected changes in the underlying cost structure of the domestic industry’.Footnote ⁸¹ This conclusion potentially poses a complication for the US International Trade Commission (USITC) as it often cites a declining price–cost margin as an indicator of injury. The notion is that import competition limits the ability of the domestic industry to raise its sales price as input costs rise. Consequently, its profits will fall. The Panel is correct that the price–cost margin metric is only informative if the investigating authority simultaneously controls for the impact of changes in the domestic market (e.g., new domestic competitors can also reduce the price–cost margin). It seems inevitable that the US will eventually find its injury analysis challenged on this exact point because the USITC generally does not distinguish domestic and foreign influences.

A deeper question that the Panel did not address – and certainly one question the US did not want specifically addressed – is what constitutes a proper price effects analysis. What does China (or, for that matter, the US) need to do in future cases in order to produce a satisfactory price analysis? On the one hand, China can look to EU and US AD/CVD proceedings and mimic the approaches proffered by their agencies. EU and US authorities present a mix of data reporting (e.g., charts of prices), trends analysis, and simple summary statistics (e.g., counting the number of quarters during the POI where the subject product undersold the comparable domestic product). Their facility with public reporting largely obscures the fact that neither the US nor the EU rigorously assess the impact that alternative factors might have played in the evolving pricing dynamics. The term ‘rigorous’ is defined as ‘extremely thorough, exhaustive, or accurate’ and must be interpreted in the context of modern econometric methods.

How big of a role have non-subject imports played in the domestic industry's deteriorating performance? How have changes in domestic competition (via either entry or exit) or changes in input supply conditions affected the market? Both EU and US authorities seem to consider that it is usually sufficient to simply refer to these other possible causes of injury in their reports; there is no requirement that they explicitly quantify the impact of the alternative factors. If a determination were ever appealed, such references provide evidence that the investigating authorities did consider the alternative factors and ultimately concluded that subject imports were nevertheless a cause of material injury.

Presentation of data, if not explicit quantitative analysis, is what investigative authorities need to do under the EU/US approach. For instance, neither the US nor the EU have ever offered any standard for how many quarters of underselling are necessary to conclude price suppression or depression. Their reasoning, of course, is that depending on the case (the product, the subject country) the authorities want flexibility to come to different conclusions. There are instances where underselling is never present and yet the authorities find price effects and – much more rare of course – other cases where underselling is present every quarter and yet the authorities find no price effects. It is fair to question the value of the underselling analysis when the authorities have such discretion to interpret the results in any way they like.

On the other hand, China could look to the large body of statistical and econometric literature for guidance. Is determining the extent to which subject imports affect domestic prices that different from a myriad of policy questions where econometrics has helped provide an answer? To what extent is a worker's lower pay due to the fact she is a female versus other measurable qualities? To what extent is a person's diabetes caused by cigarette smoking versus other factors such as weight or diet? These are questions that econometric techniques are well suited to answering and are widely accepted by courts and administrative bodies. Yet, to our knowledge, no investigating authority does price effects analysis using any methods resembling those found in econometric books. For instance, neither the US nor the EU determine the extent to which prices are lower due to subject imports rather than, say, lower iron ore costs, weak demand, or any number of other alternative causes.

It is clear that this latter approach goes beyond what Panels have determined in prior disputes regarding whether Article 3.2 requires an investigating authority to consider an explicit quantitative determination must be made in this regard.Footnote ⁸² Thus, to a large extent the deficiency with China's approach in this case was that it failed to ‘paper’ its price effects analysis as well as the US does. The Panel and AB generally defer to the investigating authority when it is clear that the agency has considered the alternative effects and provided some basic quantitative discussion. However, because China failed to document its consideration of these other factors, the Panel concluded that it did not fulfill its obligations.

4.5 Disclosure of essential facts

The Panel found that China failed to fulfill its disclosure obligations under Articles 6.9, 12.2., and 12.2.2 of the ADA and Articles 12.8, 22.3, and 22.5 of the SCM Agreement regarding the application of ‘facts available’, and also under Articles 6.9 and 12.2.2 of the ADA and Articles 12.8 and 22.5 of the SCM Agreement with respect to the price effects and causation findings. China chose to appeal only the issues with respect to the price effects findings to the AB.

China's arguments on appeal mainly focused on the question of what evidence falls within the meaning of ‘essential facts’ and ‘relevant information on the matters of fact’ under Articles 6.9 and 12.2.2 of the ADA and Articles 12.8 and 22.5 of the SCM Agreement. In China's view, it was not obligated to disclose the evidence in support of the ‘low price’ subject imports finding because it was only required to consider the existence of adverse price effects, which did not include any facts about the ‘comparison’ between the subject import prices and domestic prices, or the causal relationship between the two variables. In other words, China understood the obligations to be that it was solely required to disclose the information which it considered ‘essential’ to the price effects analysis. The AB disagreed with China by holding that the disclosure obligations have to be appreciated in the context of the application of definitive measures under the ADA and SCM Agreement, rather than from the mindset of the investigating authority. Given that the evidence of ‘low price’ subject imports in relation to the significant price depression and suppression was relied on by MOFCOM in reaching the decision to apply definitive measures, the AB found it necessary for the disclosure and subsequently upheld the Panel's findings.

While we think it is correct for the AB to safeguard the interested parties' rights to defend themselves by requiring the investigating authority to disclose the ‘essential facts’ and ‘all relevant information on the matters of fact and law and reasons' regarding the imposition of definitive measures, and to interpret that understanding with respect to the ADA and the SCM Agreement and factual circumstances of each case, we consider that the AB in this dispute did not offer any further clarification on the disclosure obligations under Articles 6.9 and 12.2.2 of the ADA and Articles 12.8 and 22.5 of the SCM Agreement. What the AB essentially did was to adopt a case-by-case analysis approach when it comes to finding the institutional balance between guaranteeing due process through adequate disclosure and permitted discretion for the investigating authority. It was no surprise in this investigation that the AB agreed with the Panel in concluding that the ‘low price’ subject imports was ‘essential’ and must be disclosed due to MOFCOM's reliance on that factor to support the finding of significant price depression and suppression, but that was a relatively easy call. The explanation offered in the AB report, however, does not help much with the more muddled situations where it is less clear whether the facts have been relied on by the investigating authority or necessary for the interested parties to understand the imposition of definitive measures.

5.1 Why GOES?

With respect to the first issue, given how China handled its investigation, we cannot help but speculate why it pursued GOES in the way it did. One hypothesis is that China is learning how to conduct an AD/CVD case. After all, China is a relatively new AD/CVD user, having WTO-consistent regulations for only about ten years. While there is likely some truth to this interpretation, we find it difficult to believe the issues in this case were solely, or even primarily, due to China's inexperience. According to the Global Anti-dumping DatabaseFootnote ⁸³ the GOES investigation was the 57th investigation conducted by MOFCOM.Footnote ⁸⁴ It seems unlikely that MOFCOM had not considered most of the issues in this dispute in previous investigations; issues such as non-confidential summaries, facts available, and price effects are present in every case.

We think there are at least two other possible explanations for China's actions in this dispute. First, China may well have wanted to send the US a message about its unhappiness about the way the US has treated China in its AD/CVD proceedings. Since the US changed its policy allowing simultaneous AD and CVD cases against NME countries in 2007,Footnote ⁸⁵ China is easily the leading target for US CVD allegations, having been subject to over 30 CVD investigations (Table 2).Footnote ⁸⁶ Many of these CVD cases include subsidy allegations based on long-passed five-year plans of the Communist Party of China. Further, US applications have referred to historical Central Planning documents pertaining to the Chinese steel industry. Much to the frustration of the Chinese firms, the US Department of Commerce has accepted these claims as sufficient evidence to begin its own investigation. From this perspective, China may have felt the 11 challenged programs were not dissimilar from the type of programs the US has invoked in its CVD proceedings. It strikes us as quite probable that the Chinese wanted to do a case so it could obtain Panel and AB rulings on what constitutes sufficient evidence in the application.

A second motivation may well be related to China's willingness to use its administered protection in response to other countries' use of administered protection. Said differently, China decision to initiate this case may have partially been driven by retaliatory motivates.Footnote ⁸⁷ The US steel industry has long been the heaviest user of AD and CVD. Over the past decade two-thirds of US AD investigations have involved China and almost half of these involve steel or steel products. In fact, just two weeks before China initiated the GOES case the US imposed AD duties and CVD duties on Chinese line pipe producers of 74% and 31–40%, respectively.Footnote ⁸⁹ Despite all of the US steel industry's activity aimed toward China, the GOES case was the first time China filed an AD or CVD case against steel imports from the US.

The fact that GOES is a niche product may have made it a more attractive product to target. By flat-rolled standards GOES is a small volume product and its production requires sophisticated chemistry and production techniques. GOES is an unusual product.Footnote ⁹⁰ Consequently, it is a product produced by just a handful of steel mills. Despite the modest volume, GOES is generally considered a high profit margin product and is therefore quite important to the firms who do produce it. Further, given the combination and production complexity and small volume, GOES happens to be a product that China must import to meet domestic demand.

As shown in Table 3, despite the fact that GOES is a niche steel product, it accounts for over 10% of all of China's steel imports and over 60% of China's steel imports from the US. Given that effective economic retaliation requires the target to feel a serious economic impact, GOES was a good steel product for the Chinese to target in order to effectively retaliate against the US steel industry's aggressive use of AD/CVD against China. Moreover, despite the fact that GOES is a large export product for US steel producers, the US is not a major supplier of GOES to China. In fact, as shown in Table 4 despite the large reduction in shipments from the two subject countries following the orders (from over 140,000 MT to less than 20,000 MT), China was able to turn to alternative suppliers, most notably Japan, to offset the decrease from the US and Russia.

Table 3. Flat-rolled imports by China, 2008 (metric tons)Footnote ⁸⁸

Table 4. Import supply, GOES, 2008 vs. 2011 (metric tons)Footnote ⁹¹

Taking all of these factors together suggests credible motivation for China to target US GOES producers, both to draw attention to some of the questionable allegations it has faced and also to raise the US industry's awareness of the potential trade losses due to AD/CVD actions.

5.2 Why China might ultimately benefit from losing

China is easily the world's largest target of AD and CVD actions, not just in the US but around the world. As shown in Table 2, since 2006 China has been the subject of almost half of all CVD cases worldwide. The same story is true for AD – China is easily the leading target of AD cases worldwide.

In Table 5, we report AD cases targeting China. In the left-half of the table, we report the number of cases targeting China by new and traditional AD users.Footnote ⁹⁴ As shown, China accounts for about one-quarter of all AD cases by both new and traditional users.

Table 5. Anti-dumping cases initiated against China, 2000–2011Footnote ⁹²

In the right-half of the table, we follow Chad Bown's approach and report the number of investigations involving China. Footnote ⁹⁵ The difference between the two is that many applications allege dumping by multiple countries. The Global Anti-dumping Database records each country as a case but considers the set of countries a single investigation. Using the investigation metric, we see that China is even a larger target. For both new and traditional users, China is a subject country in over half the AD investigations since 2000.

In Table 6, we contrast these numbers with the number of AD cases initiated by China since 2000. Whereas Chinese producers have been subject to 666 AD cases, China has initiated only 184 cases and relatively evenly spread geographically.

Table 6. Anti-dumping cases initiated by China, 2000–2011Footnote ⁹³

Consequently, China has much to gain by having the WTO DSU clarify the rules for applications and investigations. This is all the more so in the CVD area where many countries are following the US lead in aggressively pursuing perceived Chinese subsidization, often heavily based on facts available.

Time will tell whether the US will succeed in exporting its AD/CVD methodologies to other WTO members or whether the GOES case will come back to haunt the US in the future. After all, in the trade remedies' area, no country has clean hands and, if the number of WTO disputes filed against it is an indicator, US hands are dirtier than most.