At the close of December, 2 WTO panel reports were circulated. First, United States – Certain Measures on Steel and Aluminium Products (DS544/552/556/564) on 9 December 2022. Second, United States-Origin Marking (DS597) on 21 December 2022. This post is in multiple parts. The first, If I were an Appellate Body, discusses the interpretation of GATT Article XXI(b) by focusing on the United States-Origin Marking (DS597) panel’s interpretive approach to the phrase ‘emergency in international relations.’
In its second part, the blog argues that the only way to view the recent reports involving the United States’ invocation of GATT Article XXI is through the lens of the US non-justiciability arguments and, more generally, the United States’ loss of faith in WTO dispute settlement. The Russia – Traffic in Transit (DS512) and Saudi Arabia - Measures concerning the Protection of Intellectual Property Rights (DS567) reports built an analytical framework around the obligation of good faith and stressed balancing considerations when connecting the measures to a party’s essential security interests. What is clear now is that the recent US reports (note also United States – Certain Measures on Steel and Aluminium Products (DS544/552/556/564) seek to severely limit the invocation of the security exceptions, making it crucial that members develop a way to address rising concerns with trade security in a new world that does not see security only in terms of extreme, exceptional situations.
What is an ‘Emergency in International Relations’?
The WTO disputes we have seen so far have respondents invoking GATT Article XXI:b(iii) in times of ‘war or other emergency in international relations.’ Beginning with DS512, the panel focuses on the phrase in ordinary terms, finding the idea of ‘war’ to form one aspect of the broader category of ‘emergency in international relations.’ Finding that ‘war’ refers to ‘armed conflict’ enabled that panel to emphasise situations of ‘danger […] that arises unexpectedly and requires urgent action.’ (DS512, para 7.72). Connecting the phrase in sub-paragraph (iii) to the other subparagraphs (i) and (ii) led the panel to find that all three sub-paragraphs share the same interests. On this basis, the panel connects the situation ‘emergency in international relations’ to ‘all defence and military interests, as well as maintenance of law and public order interests.’ (para 7.74). The panel thus defines ‘emergency in international relations’ as ‘a situation of armed conflict, or of latent armed conflict, or of heightened tension or crisis, or of general instability engulfing or surrounding a state.’ (para 7.76)
Two crucial observations with this strategy flow into all subsequent panels’ assessments. First, the panel avoided temporal issues concerning situations of ‘war or other emergency in international relations’ in the security exceptions. Second, the DS512 panel interconnects a respondent’s articulation of essential security interests to its identification of the situation under sub-paragraph (b)(iii) and – as we later learn – the connection between the measure at issue and the articulated essential security interests.
Imminent War
The DS512 panel bases its interpretation on the US’s initial proposals for the ITO Charter. (para 7.75) In 1946, an initial US proposal was for any ITO member to invoke the exceptions ‘in time of war or imminent threat of war, relating to the protection of the essential security interests of a Member.’ (Pinchis-Paulsen 2020, p.129). The US defence departments rejected the ‘imminent threat’ language because the text would become ‘unduly restrictive’ and could generate ‘serious diplomatic repercussions.’ Ultimately, the final text became the more ambiguous, ‘international emergency’ to prevent the possibility that invoking the trade rules would appear as ‘an act of war.’ After that, the final text of ‘emergency in international relations’ came following internal US concerns that economic emergency could fall under ‘international emergency’ and deviated from the original intention. (Pinchis-Paulsen 2020, p.140-142). The drafters’ focus on how ITO members (could) act before war crystallises, not on broadening the potential situations for invoking the exceptions. That is, to account for the fact that governments must plan for war before they are ‘at war.’ Of course, in hindsight (and with the benefit of the work of excellent historians), we better appreciate the timeliness of war and the tools of economic statecraft.
Tightening the meaning of ‘emergency in international relations’ feeds into a broader effort to address pretexts for war. Legal scholars studying the pretext model observe how states use formulaic language ‘in terms of generally accepted international principles.’ State leaders persuade domestic or international audiences about their actions by speaking the language of international law, even while actively breaking it (a point Gerry Simpson makes about the 2022 Russian invasion of Ukraine). However, this critical question of pretext does not account for changes to security patterns or power relations over time – planning for war by actions that come by creep or by tsunamis. Wartime is a political argument. (Dudziak, 136) The GATT security exceptions fail to explain how the security exceptions work in a ‘forever war’ or what time looks like before ‘wartime’ or following the end of ‘wartime’. This has massive implications for the future shaping of trade policies and how governments may attend to their WTO negotiation, notification, and surveillance commitments, but I digress.
None of the four panels has directly engaged with the question of wartime. The DS512 panel found it an ‘objective fact’ that Russia’s actions were taken ‘during’ the found situation. (para. 7.70). Likewise, in DS567, in an ongoing conflict, it was clear the measures were ‘of a continuing nature’ and taken ‘during’ an ‘emergency in international relations. (para. 7.268, 7.269). The DS544 panel finds the phrase ‘in time of’ creates a ‘temporal link’ between the action taken and the situation at issue. (7.140).
The DS597 panel did not consider how time flows through a situation. Instead, its interpretation of ‘emergency in international relations’ requires evidence of a ‘breakdown’ of relations. This interpretation equates GATT Article XXI with a specific point in time. The panel does not address the durability or longevity of this breakdown. Therefore, it remains unclear whether the panel renders GATT Article XXI into a ‘non-application’ clause compared to Articles XXXIII and XIII of the WTO Agreement, which otherwise only applies at the time of accession.
Avoiding discretion and finding a breakdown
As I will elaborate more in the next post, the DS597 panel report includes a detailed assessment of the history and plain meaning of the GATT Article XXI construction to find that, contrary to the US submissions, the security exceptions are neither non-justiciable nor self-judging in interpretation. Appreciating the predominant attention to the US arguments helps set the context for the DS597 panel’s approach that bypasses the earlier DS512 panel’s attention to how certain situations ‘give rise to particular types of interests.’ (DS512 para 7.76).
To evaluate the US invocation of the security exceptions, the DS597 panel (and Hong Kong, China) work hard to get the US to identify a particular ‘situation’ – that is, to do more than invoke GATT Article XXI and instead identify one specific sub-paragraph. Based upon some reluctant responses from the US in the proceedings, the panel proceeds to interpret sub-paragraph (iii) of Article XXI(b) and the phrase ‘emergency in international relations.’ (paras. 7.257, 7.258)
Citing dictionary meanings, the DS597 panel finds that the terms ‘emergency’ refers to ‘danger or conflict’ beyond ‘the ordinary’ and suggests a ‘degree or magnitude of seriousness, as reflected by the need for urgent or immediate action.’ The panel does not unpack the degree idea, for it would require considering a member’s assessment of its security concerns. Identifying a spectrum implies that a government must determine what is serious and when. The DS597 panel produces a ‘definition’ as a situation of ‘utmost gravity’ and ‘representing a breakdown or near-breakdown in those relations.’ (para 7.289, 7.290, 7.310, 7.315). Although the panel observes a ‘thermometer’ for gauging international relations, it rejects the role of the invoking Member’s appreciation of the situation. (7.309).
The panel treats the exceptions like a light switch. When toggled on, members will no longer comply with international commitments, including WTO commitments. (7.298, 7.311). According to the panel, evidence of international cooperation is evidence that a member cannot invoke GATT Article XXI! The exact panel text (without sourcing) may be helpful here:
As we know from daily media reports, we live in a world driven by a range of political, economic, social, and environmental tensions and divergences. At the same time, in the midst of these tensions and divergences Members will in most cases continue to manage their relationships within a range of international legal frameworks aimed at ensuring predictability and stability within the international system. Article XXI(b)(iii) stands for the principle that situations of war or other emergency in international relations represent an exception to this. If the existence of tensions or policy divergences of any magnitude, however, were to constitute an emergency in international relations, the character of Article XXI as an exception to apply in the gravest of circumstances would fundamentally change.
Such an assessment significantly departs from the complexity and ambiguity of wartime. The panel did not seek to opine on war or security, but its interpretation of the provision creates significant implications for the future deliberation of trade security. Recent lessons from the 2022 Russian invasion of Ukraine and the battle over semiconductor controls exemplify the complexity of economic interdependence in the modern globalised economy. The panel’s assessment makes sense if trying to thicken the exceptions’ boundaries to catch pretext or opaque self-judging arguments. But it also eradicates the safety valve function of GATT Article XXI. It washes the good away with the bad.
What is the emergency at issue in DS597?
Adding to the complexity of this dispute is that there are competing emergencies at play. At issue was the United States’ relations with China and Hong Kong, China and the relations of Hong Kong, China and China. According to the US, the situation in Hong Kong, China, resulting from China’s actions (including the adoption of the National Security Law) and actions of Hong Kong, China's authorities, creates a threat to the US’s essential security interests. The panel then adds the additional situation of ‘an erosion of freedoms and rights of the people in Hong Kong, China, as well as the institutional degradation of democracy in Hong Kong, China.’ (para 7.323).
The two identified situations became muddled as the panel evaluates the evidence presented by the United States. Such evidence fell into one of four categories: US domestic instruments; reports and statements by United States officials on the situation in Hong Kong, China; other Members’ views; and press reports. Executive Order 13936 of 14 July 2020 confirmed the US President’s conclusion that the situation in Hong Kong ‘constitutes an unusual and extraordinary threat […] to the national security, foreign policy, and economy of the United States’ and on this basis, the President declared ‘a national emergency with respect to that threat.’ (DS 597 para 7.331). Subsequently, the US enacted the 2020 Hong Kong Autonomy Act ‘to impose sanctions with respect to foreign persons involved in the erosion of certain obligations of China with respect to Hong Kong.’ The panel notes that the sanctions on the importation of goods from Hong Kong, China are ‘expressly excluded from the authority of the Act.’ (7.334).
Additional evidence from the State Department’s report of the situation in Hong Kong, China aids the DS597 panel’s analysis of whether an ‘emergency in international relations’ existed. First, there is an examination of the extent Hong Kong, China’s ‘autonomy’ for implementing commercial agreements. This autonomy examination extends to other structural policies that address state control over its national economy from China, including (but not limited to) the continuation of Hong Kong China’s legal system (despite questions of judicial independence), the practice of protecting property rights, the maintenance of Hong Kong, China’s currency pegged to the US dollar, and Hong Kong, China’s participation ‘separately from China’ in 24 international organisations, including the WTO (para 7.336). Second, the panel observes that the US and Hong Kong, China maintain some bilateral agreements.
Several countries raise significant concerns about the events in Hong Kong, China. The panel notes a ‘concern in public opinion’ regarding undermining human rights and democracy in Hong Kong, China. However, the panel does not find that this evidence meets its defined ‘breakdown’ of relations. While other US government reports evidence US consideration of ‘the human rights situation in Hong Kong, China’, the panel finds there is no ‘meaningful information on the specific issue of how the human rights situation in Hong Kong, China has affected international relations between the United States, Hong Kong, China and China.’ (emphasis added, para 7.340). Does this statement help the panel identify the situation? Its focus is on how a situation impacts the respondent in the dispute but not the situation itself. To this end, the panel sets a new threshold of a ‘requisite level of gravity’ defined as the ‘point of breakdown.’ (7.353). The panel points to but does not mention the elephant in the room. The United States spoke to measures ‘vis-à-vis Hong Kong, China’ but does not provide any evidence on measures taken ‘vis-à-vis China.’ It is difficult to square this panel conclusion with an earlier observation that an ‘emergency does not necessarily have to originate in the invoking Member's own territory and bilateral relations.’ (para. 7.297)
Additionally, the DS597 panel points to the factual circumstances of DS512 and DS567 to highlight the differences between them and the US concerns with the ‘human rights situation in Hong Kong, China.’ Recall, however, that the assessments of those situations were thin. The DS512 panel relies on evidence that the dispute is ongoing, ‘publicly known’, involves the disputing parties, and that other Members imposed sanctions on Russia. (DS512, para.7.119-7.120) The DS597 panel dwells on different aspects, noting that the General Assembly resolution considered the situation of ‘armed conflict’ and sanctions imposed on Russia following the annexation of Crimea. For the DS597 panel, it is significant that goods were marked ‘China’ but not subject to different tariff treatment. (7.356). In contrast, the DS567 panel observes that it was a matter of fact that Saudi Arabia ‘severed diplomatic and consular relations with [Qatar].’ (DS567 para 7.258) The DS597 panel adds that there was a complete collapse of ‘economic relations’ between Saudi Arabia and Qatar. In sum, a future invocation of GATT Article XXI(b)(iii) would require demonstrating a ‘breakdown’ by showing international recognition of the situation, trade restrictions and evidence of the severance of diplomatic, consular, or economic relations between the parties.
The DS597 panel report strategically chooses to incorporate the factual circumstances of DS512 and DS567 to require evidence of a ‘breakdown’ in relations to identify a ‘war or other emergency in international relations.’ The effort here appears to render the assessment value-neutral while ensuring a member cannot self-judge the circumstances by which it deviates from WTO rights and obligations. Evidence of a breakdown seems easy to quantify without requiring relativism. In reading the report, it appears that the panel sacrifices the important balancing the DS512 panel injects into the interpretation of GATT Article XXI. The genuine concern is that the US now understands what it takes to invoke the security exceptions – a clear, demonstrated breakdown of international relations. But is it the function of a WTO panel to consider the long-term implications of this finding? Suppose security guides US economic policy (for how long is unclear). What does it mean for a report to say the invocation was improper because the respondent did not do enough to fracture international relations?
I hear Emily Lydgate’s assessment of EC Seal Products in the back of my mind, as she cautions against a rationality test that tried too hard to impose ‘a consistency requirement onto a pluralistic, democratic society as it does not match the reality of how regulation emerges.’ Jo Langille and Rob Howse have a forthcoming AJIL article that deeply tackles this.
As I will explain in the next post, the panel should have attended to the DS512 panel’s original inquiry and asked whether there was an urgent and serious situation that gave rise to a member’s essential security interests. The DS597 panel, having crafted its definition, does not examine whether the situation in Hong Kong, China, could be understood to be ‘a situation of armed conflict, or of latent armed conflict, or of heightened tension or crisis, or of general instability engulfing or surrounding a state.’ They highlight the limits of past panels’ interpretation.
The issue here comes down to how this affects the US. This is precisely the purpose of the third and fourth steps of the DS512’s legal interpretive approach (and accepted by the DS567 panel) for analysing GATT Article XXI. Namely, whether a respondent has sufficiently articulated its essential security interests as dependent on the situation in question and then whether, while meeting an obligation of good faith, the respondent demonstrates that the measures meet a minimum requirement of plausibility in relation to the proffered essential security interests. In this case, how the different markings identifying “Hong Kong, China” versus “China” plausibly relate to the US essential security interests – but not as identifying the existence of a situation. (see 7.220, 7.221). Yet, how the situation impacts the US becomes key to identifying the situation. The panel asks for evidence that a situation in China breaks down the relations between the US, Hong Kong, China, and China. In focusing on GATT Article XXI, I put to the side more complex questions about the assessment of GATT Article IX.
Recent Comments