1. Introduction
Whether considered a feature or a bug,Footnote 1 the concept of security defies clear demarcation. This fundamental ambiguity has endowed ‘security’ with a capacity akin to a discourse, underpinning justificatory arguments about a wide range of issues within and beyond the conventional context of national, defence-centred security. Its growing presence in headlines and expanding usage in international affairs have most conspicuously attested to its discursive prominence, with States attaching the label of ‘security’ to matters about public health,Footnote 2 climate change,Footnote 3 energy transition,Footnote 4 and economic policymaking.Footnote 5
These security claims have also entered the hearing rooms of international adjudicative bodies. States brand issues as ‘security’ interests in interpreting established treaty provisions, including those without explicit textual basis or an apparent object and purpose to protect ‘security’. The attempt at introducing unconventional security notions within legal domains where such security parlance has been relatively alien is referred to in this article as ‘security transference’. In the context of interpretation, a State that resorts to security transference argues in essence that, in light of the security interest asserted, a treaty provision should be construed in a specific manner or be subject to a specific legal standard to the effect that leeway is given to a measure contradictory to the State’s existing obligations. Investigating whether and to what extent security transference has been successful in international adjudication is therefore significant in two respects: first, to understand the degree to which existing provisions have been adapted to accommodate contemporary security concerns, and, second, to examine whether appropriate oversight has been maintained to prevent illegitimate claims for policy space, disguised as security interests, from ‘encroaching’ the line between permissible and impermissible deviations from established international rules.
As such, the use of security transference under exception clauses merits particular attention. Existing economic treaties contain roughly two categories of exception clauses: those addressing conventional national security concerns, often expressly incorporating the term ‘essential security interests’ (national security exceptions) and those pertaining to other policy goals that are not originally designed for, and at least textually not concerned with, national security interests (general or public policy exceptions).Footnote 6 Regarding the first category, a growing corpus of literature has examined the use of unconventional security claims by recourse to national security exceptions.Footnote 7 In practice, most transference attempts channelled through national security exceptions have been thus far unsuccessful. Among the relevant cases in the WTO context, the most high-profile ones involve the US invocation Article XXI of the General Agreement on Tariffs and Trade (GATT) – the national security exceptions – to justify measures that are allegedly based on contemporary security concerns, including tariff duties imposed on aluminium and steel imports to counter the adverse impact arising from global excess capacityFootnote 8 and origin marking requirements on goods imported from places suffering deteriorating human rights situations.Footnote 9 Neither of the two situations is found by the panels to fall within the purview of Article XXI of the GATT.Footnote 10 In the context of investment arbitration, it has also been observed that ‘all tribunals have construed essential security interests very narrowly’.Footnote 11 Turning to the second category, while public policy exceptions are textually unrelated to ‘security’, they afford other diverse bases for granting policy space and may therefore provide leeway for security claims as well. However, how security transference attempts would fare under this category of exception clauses – such as Article XX of the GATT and Article XIV of the General Agreement on Trade in Services (GATS), both titled ‘general exceptions’ – has received limited academic attention.Footnote 12
In light of the above, this article investigates the WTO judiciary’s stance on security transference when relevant arguments surface in general exceptions defences. To this end, I seek to ascertain how unconventional security claims have been treated and positioned in the legal parameters of Article XX GATT and Article XIV GATS, and to evaluate whether the current approach effectively distinguishes between legitimate and illegitimate use of policy space.
To be able to treat this inquiry within the length of an article, it is necessary to adopt a micro-perspective within the plethora of contemporary security concepts. I choose energy security as the focus of investigation for three reasons. To begin with, the notion of energy security, as Section 3 will further elaborate, encompasses facets that are substantially removed from the narrow, defence-oriented conception of (national) security, thus representing an ‘unconventional’ security concern. Moreover, energy security is sufficiently versatile and equivocal as a term, indicating that the risks of abuse posed by the transference of energy security claims into public policy exceptions are not merely speculative. The last reason is practical: the WTO judicial bodies’ approach to any type of unconventional security claim can only be examined when relevant disputes have surfaced, which is the case for energy security. Arguably, the emergence of these cases also shows that energy security is gaining traction in the WTO’s dispute settlement system, making it all the more timely to take a careful look at the jurisprudence.
The rest of this article is structured as follows. It first provides a more extensive and contextualized discussion of the conceptual framework, situating security transference in the uptake of security discourse in present-day international economic law (Section 2). It then zooms in on the chosen micro-perspective, commencing with a brief outline of the concept of energy security and the WTO’s general exceptions (Section 3). Sections 4 and 5 dissect relevant WTO disputes and identify the two legal techniques that WTO panels and the Appellate Body adopt for treating energy security claims: the incorporation and consideration of energy security concerns in relation to the framing of regulatory objectives, and the screening of origin-based discriminations that underlie energy security claims. Section 6 presents a normative evaluation of the WTO’s approach to security transference. Section 7 concludes.
2. The Uptake of Security Parlance in International Economic Law
In present-day international economic law, the uptake of security parlance has manifested itself in primarily two ways. First, a State may regard certain interests to be of such importance that they must be made immune from external scrutiny, thus advancing security claims that aim at ‘immunization’ to varying degrees. Through claims of security immunization, the State intends to move relevant interests away from ordinary policy debates,Footnote 13 including those in judicial proceedings, to within its own exclusive powers, often with a view to triggering exceptional measures.Footnote 14 In judicial contexts, security immunization is represented in arguments that seek to invoke extraordinary standards of review; when taken to extremes, it rejects external scrutiny over the very characterization of a measure as security-related and denies any role that may be assumed by tribunals. For instance, it has been the consistent contention of the US that Article XXI of the GATT is entirely ‘self-judging’, asserting that the determination of what constitutes ‘security’ (or ‘essential security interests’ as per the text of Article XXI) rests solely with the invoking Member and is not subject to review by WTO dispute settlement organs.Footnote 15
Security immunization involves chiefly battles over a who question: who – the State or an external authority – has the power to determine the presence of a security situation and decide on the actions to take. A radical answer to this question may leave little room for any meaningful discussion of what constitutes security, when the latter becomes wholly encased in the discretion of a given State. However, demands for complete immunization remain rare occurrences and when raised have hardly been accepted by tribunals. Indeed, the US self-judging argument about Article XXI GATT has been invariably dismissed by WTO panels.Footnote 16 In general, whether expressly or tacitly accepted by States or compelled by external authorities, the appropriateness of security claims, in particular the characterization of an issue as a security interest, is still in large part subject to examination by institutions beyond the unfettered discretion of States. This, in turn, brings the scope of security concerns to the centre of contention and gives rise to a second type of security claim usage, security ‘transference’, as has been defined and briefly discussed in the Introduction.
Security transference has been featured in an extensive body of scholarly writings, particularly those on international law and international relations. Over the years and often contemporaneous with current affairs, scholars have contemplated and debated the conception of security in relation to energy,Footnote 17 the environment,Footnote 18 climate change,Footnote 19 public health,Footnote 20 and so forth. Diverse as they may be, the various strands of discussion share commonality in seeking to ‘decent[re] the role of military affairs’.Footnote 21 The uptake of security parlance has also come to the fore in practice. The trend has certainly reached treaty negotiators. In the recently concluded Agreement on Climate Change, Trade, and Sustainability,Footnote 22 the list of specific exceptions allowing for the use of fossil fuel subsidies encompasses explicit reference to ‘the improvement of energy security’Footnote 23 as well as several other rationales that may be associated with the notion of energy security, such as supporting vulnerable communities and safeguarding disaster resilience.Footnote 24 The US–Mexico–Canada Agreement includes a blanket exception for protecting ‘essential security interests’ without defining the term.Footnote 25 This lack of delimitation, marking a clear contrast with the formulation of the WTO’s national security exceptions, arguably affords space for broader, unconventional security claims.
Apart from international law-making and, perhaps more frequently, security transference emerges in the use of exceptions. The rendition of security transference varies depending on the type of exception clause invoked. Under national security exceptions, as the term ‘security’ is expressly incorporated into the legal text, its interpretation naturally takes centre stage in legal contentions. What disputing parties centre arguments around, and what tribunals are called upon to determine, is whether the security-related term prescribed in a given national exception may be construed to justify certain circumstances that the respondent claims to be within the exception’s coverage. For instance, the two WTO disputes challenging US measures mentioned in the Introduction turn on whether situations of global excess capacity or deteriorating human rights can be interpreted as ‘essential security interests’ that trigger measures ‘taken in time of war or other emergency in international relations’, which are elements specified under Article XXI(b) of the GATT. In the context of investment arbitration, Henckels has comprehensively surveyed the variety of ‘securitizing moves’ adopted by States to convince tribunals that their measures fall within the coverage of ‘essential security interests’ exceptions, some of which come ‘close to eliding the distinction between security and other aspects of public policy’.Footnote 26
With respect to public policy exceptions, the absence of any textual ‘hook’ for security interests results in a different form of security transference. Here, a State typically claims that certain security considerations should have a bearing on the interpretation of the legal elements specified by a public policy exception, which are textually unrelated to security. Importantly, the security claims invoked are often beyond the conventional boundaries of defensive security. For example, in one of the WTO cases discussed in the subsequent sections, India argues that concerns about energy security should be taken into account when determining whether the product at issue is ‘in general or local short supply’.Footnote 27 The question therefore arises as to how public policy exceptions, whose parameters and standards have largely derived from case law contexts devoid of security connotations, should attend to arguments based on emerging security concepts.
I do not consider the use of security transference to be necessarily problematic. By way of repurposing established legal norms, transference can play an instrumental role in addressing contemporary concerns that had not arisen, or were overlooked, when existing treaties were initially negotiated, particularly when their amendment or modification lacked political momentum. In that light, references to unconventional security interests in international economic disputes may well facilitate post-date accommodation of emerging legitimate values. However, security transference can be susceptible to abuse. As per the theory of ‘securitization’. coined by the Copenhagen School, the social construction of an issue as a security concern necessarily triggers a rationality of emergency and imposition of exceptional measures that override otherwise binding rules.Footnote 28 The Copenhagen School therefore maintains a rather vigilant attitude towards securitization. This proposition has been critiqued as overly definite by subsequent scholarly works, which contend that securitization processes may follow logics different from exceptionalism and retain the role of ordinary law.Footnote 29 In my view, despite their contrasting positions, the opinion on securitization held by the latter strand of literature rests on the same basis of assessment as the opposite verdict given by the Copenhagen School: whether, and to what extent, exceptionalist logics – or, under the present conceptual framework, extreme claims for immunization – would prevail upon the securitization of an issue. However, as noted above, demands for complete immunization that seek radical power shifts from international institutions to States remain scarce in practice, meaning that security transference would arise largely in the routine operation of international law. The appearance of business-as-usual, however, does not necessarily guarantee that security transference serves benign ends. As a device for accommodating rising concerns that were previously foreign to established legal parameters, security transference can potentially disrupt the boundaries of legitimacy that exception clauses have been set to preserve. Without appropriate guardrails, dubious claims disguised as security interests could transgress the demarcation line between legitimate and illegitimate use of legal exceptions.
Such a risk of abuse is present with respect to both national security exceptions and public policy exceptions. However, given the relatively scarce scholarship addressing the latter scenario, security transference channelled through WTO public policy exceptions merits a closer look. The remainder of this article attends to this investigation and, given reasons set out in the Introduction, turns on WTO disputes related to energy security as the central focus.
3. The Concept of Energy Security and the WTO General Exceptions
To facilitate a close look at the treatment of energy security claims in WTO disputes, this section attends to two preliminary matters: the concept of energy security and the legal text of the WTO’s general exceptions. With a view to setting the scene for the following sections, the analysis here is necessarily brief and limited to what is essential for the purpose of this article.
3.1 The Concept of Energy Security
The concept of energy security has been characterized as ‘polysemic’Footnote 30 for its capability of encompassing divergent meanings and perspectives. From the viewpoint of energy exporters, ‘security’ hinges on the reliability of demand, as stable returns from energy projects are necessary for sustaining infrastructure investment in the production and transportation of energy.Footnote 31 However, the term is more often used in reference to the security of energy supplies, representing the perspective of energy importers. In this aspect, energy security has been associated with the availability and affordability of energy sources,Footnote 32 particularly in the face of external shocks. For instance, the International Energy Agency (IEA) defines energy security as ‘the uninterrupted availability of energy sources at an affordable price’,Footnote 33 which entails ‘ensuring reliable sources of supply and designing and maintaining infrastructure and systems that are resilient against physical and cyber threats’.Footnote 34 Similarly, the World Energy Council conceptualizes energy security as ‘a nation’s capacity to meet current and future energy demand reliably’ and ‘withstand and bounce back swiftly from system shocks with minimal disruption to supplies’.Footnote 35
The notion of energy security has also been predicated on the origin of suppliers. The argument that a country’s energy supply is secure and reliable only when it is provided through domestic manufacturing underlies the growing trend of industrial policies in energy and energy-related sectors worldwide. In a similar manner, States may adopt ‘friend-shoring’ regulationsFootnote 36 that afford comparatively favourable treatment to both domestic industries and producers from like-minded trade partners in the name of energy security. A prominent example is the US Inflation Reduction Act 2022, which limits the scope of electric vehicles eligible for tax credits to those manufactured using critical minerals that are either extracted or processed in the US or a country with which the US has an effective Free Trade Agreement, or have been recycled in North America.Footnote 37 An origin-based notion of energy security may also be relatively more removed from the context of industrial policy and exhibit conceptual proximity to conventional, defence-oriented national security.Footnote 38 The European Union’s energy policy schemes to reduce reliance on natural gas supplies from Russia amidst the latter’s expansive military operations in Eastern Europe – which became the focus of contention in one of the WTO disputes discussed in the subsequent Footnote sections 39 – may be appreciated as an example.Footnote 40
Energy security, when understood as the availability of supply, is also frequently approached in a triangular framework alongside two other elements: equity and sustainability. The former represents universal access to affordable and abundant energy sources, while the latter concerns energy transition to mitigate and avoid potential environmental harm and climate change impacts.Footnote 41 The three elements are understood by some to be in tension with each otherFootnote 42 and constitute an ‘energy trilemma’.Footnote 43 Challenging this view, some conceptions regard environmental sustainability as integral to a modern construction of energy security, interlinked with the traditional elements of availability and affordability.Footnote 44 Conceptual frameworks that endorse this proposition, however, demonstrate nuanced differences in their approaches to sustainability. On the one hand, sustainability may be perceived as a means to an end, serving to ensure availability by diversifying and expanding the energy matrix through the introduction of new energy sources.Footnote 45 On the other hand, some have argued that sustainability should be the foremost objective, rather than a secondary consideration, of energy policymaking.Footnote 46 Here, the central position granted to sustainability derives from the understanding that the threats posed by climate change constitute a major source of insecurity.Footnote 47 The exploitation and use of environmentally unsustainable sources would therefore be deemed undesirable despite their enhancement of (short-term) availability.
3.2 The WTO General Exceptions
In the legal architecture of the WTO, ‘general exceptions’ are specified in both the GATT and the GATS. The two provisions – Article XX of the GATT and Article XIV of the GATS – adopt identical structures, with several subparagraphs grouped under an introductory paragraph commonly referred to as the ‘chapeau’.
The subparagraphs provide an exhaustive list of situations in which GATT- or GATS-inconsistent measures may nonetheless be justified. Although they cover a diversity of policy domains, the subparagraphs exhibit some commonality in formulation with each comprising a regulatory objective that concerns the preservation of a certain non-trade value as well as a nexus requirement that specifies the degree of connection, which a disputed measure must possess in relation to the protected value in order to be justified.Footnote 48 Some of the subparagraphs are common to both Article XX GATT and Article XIV GATS. Among the regulatory objectives stipulated, four appear particularly relevant to energy security concerns. To the extent that security is taken to denote the availability of energy supplies, security claims may involve Article XX(j) GATT, which provides for measures that are ‘essential to the acquisition or distribution of products in general or local short supply’. In addition, both lists of general exceptions contain subparagraphs that address the protection of public morals or the maintenance of public order (Article XX(a) GATT and Article XIV(a) GATS), which may be associated with a risk of social disruptions when the availability or affordability of energy supplies is substantially undermined. Finally, the exceptions concerning measures conserving exhaustible natural resources (Article XX(g) GATT) or protecting public health (Article XX(b) GATT and Article XIV(b) GATS) may be relevant when a security claim centres on sustainability, since clean air has been recognized in WTO jurisprudence as an ‘exhaustible natural resource’,Footnote 49 and the reduction of CO2 emissions an objective within the scope of the public health exception.Footnote 50 Two more recent panel reports have also acknowledged measures addressing climate change to fall within the scope of these two subparagraphs.Footnote 51
The nexus requirements pertain to the relation between ends and means, specifying that a measure at issue must be ‘necessary’, ‘essential’, or ‘relating’ to achieving the concerned non-trade objective. Necessity and essentiality have been interpreted by the Appellate Body and WTO panels to entail a relatively stringent legal standard. The applicable legal test, referred to in WTO jurisprudence as the ‘weighing and balancing’ approach,Footnote 52 involves a holistic consideration of four factors. Notwithstanding minor variances in iteration, they generally consist of the importance of the non-trade values protected, the contribution of the disputed measure to the identified objective, the trade restrictiveness of the measure, and the existence of reasonably available and less trade-restrictive alternatives at the chosen level of protection.Footnote 53
Using the jargon of WTO disputes, a measure is ‘provisionally justified’Footnote 54 if it satisfies the relevant requirements of a specific subparagraph. A two-tier approach is well-established in WTO jurisprudence with respect to general exceptions analysis, according to which a disputed measure is only examined against the chapeau requirements when provisional justification has been established.Footnote 55 The chapeau, specifying additional conditions horizontal to all subparagraphs, requires a given measure not to be applied in a manner constituting ‘arbitrary or unjustifiable discrimination between countries where the same conditions prevail’ or ‘a disguised restriction on international trade’. WTO case law has elicited much more discussion on the former element, which dispute settlement organs draw on to examine discriminatory aspects in the application of challenged measures.
The next two sections turn to dissecting the use and treatment of energy security claims in WTO disputes invoking Article XX GATT or Article XIV GATS. Section 4 addresses energy security claims made in relation to the regulatory objectives of disputed measures, the determination of which comprises part of the step of provisional justification. In this respect, the WTO judiciary has given regard to energy security claims, either by characterizing energy security as the very objective of a contested measure or by considering energy security a factor in establishing a different objective stipulated under a subparagraph. Section 5 discusses other legal tests that apply specifically to energy security claims with origin-based discriminatory aspects, namely those asserting that products or services of certain origins are energy ‘secure’ whereas others are not. Although the justifiability of discrimination has been addressed by WTO adjudicative bodies as a routine requirement under the chapeau, origin-based discrimination has been scrutinized across the steps of provisional justification and chapeau analysis in energy security-related disputes.
None of these approaches is exactly new or unique to energy security – as will be set out in the subsequent analysis, they are reflected in a long line of WTO cases across various contexts before energy security claims emerge in general exceptions defences. However, the manner in which these existing legal devices have been applied gives important clues for understanding WTO judicial bodies’ perception of, and stance on, energy security concerns.
4. Framing of Regulatory Objectives
Presented with defences citing general exceptions that are in se unrelated to security, the WTO judiciary must determine whether energy security claims should be given a foothold at all and, if so, to what extent. This determination is partly channelled through the characterization of a given measure’s regulatory objective – an analytical exercise central to the step of provisional justification that WTO adjudicative bodies consistently carry out. For ease of reference, I term this process ‘objective framing’ in the discussion below.
Depending on the exception invoked, a regulatory objective may be framed with different degrees of specificity. This section first exemplifies, drawing on general exceptions jurisprudence, the exercise of objective framing with respect to different subparagraphs, before focusing on how panels and the Appellate Body have treated energy security claims in the framing process thus far.
4.1 Objective Framing: Immediate Goals or Broader Purposes?
To determine whether a measure violating the GATT or the GATS may be provisionally justified by a general exception, an initial step is to identify and characterize the measure’s objective. Under certain subparagraphs, this characterization exercise is dictated by legal text. Examples include Article XX(g) GATT, which speaks to ‘the conservation of exhaustible natural resources’, and Article XX(j) GATT, which concerns ‘the acquisition or distribution of products in general or local short supply’. The protected interests in these subparagraphs are formulated with predetermined and rather specific parameters, which WTO panels and the Appellate Body must apply to gauge the fit of any disputed measure. In another sense, a predetermined framing embedded in the legal text provides a direct analytical pathway into dissecting a challenged measure, rendering it somewhat redundant to construct a regulatory objective afresh. When applying Articles XX(g) and (j) of the GATT in practice, WTO judicial organs do frequently follow the former approach, structuring their analysis by discerning relevant subparagraph elements from a challenged measure in turn, and without necessarily engaging with the disputing parties’ representation of the measure’s objective or any broader regulatory context in which the measure is situated.
Some other subparagraphs, in contrast, are couched in rather ambiguous language. Designating protected interests with generic terms such as ‘public morals’, ‘public order’, and ‘human, animal or plant life or health’, subparagraphs such as Articles XX(a) and (b) of the GATT contain no or rather loose embedded framings within their legal text. As a result, WTO panels and the Appellate Body enjoy considerable discretion in characterizing the objective of a challenged measure. Particularly when the measure is but one component of a broader, multiplicitous regulatory scheme, the objective may be framed either narrowly to reflect the immediate goal of the specific measure or broadly to capture the wider purpose of the overarching scheme. As the panel in EU–Palm Oil (Malaysia) observes, ‘[t]here is nothing inherent in the concept of identifying a measure’s objective that dictates the level of immediacy or specificity at which it must be defined’.Footnote 56 This observation is initially made in connection with Article 2.2 of the Technical Barriers to Trade (TBT) Agreement, where the panel is called upon to identify whether the disputed measures pursue a ‘legitimate objective’. It remains pertinent for present purposes given the overlapping of legal elements between Article 2.2 of the TBT Agreement and Article XX of the GATT and as the analysis regarding the former is extensively referenced in the examination of the EU’s general exceptions defence.Footnote 57
The varying degrees of immediacy in objective framing has been reflected in a number of contrasting WTO cases. In Brazil–Taxation, one of the Brazilian measures challenged by the EU is a tax exemption, contingent upon the use of domestic goods, for producers of digital television transmission equipment.Footnote 58 In analysing Brazil’s public morals defence under Article XX(a) GATT, the panel recognizes that two levels of objectives may be associated with the measure: an immediate, specific goal to ‘ensure domestic supply of digital television’ and the broad, overarching purpose of ‘bridging the digital divide and promoting social inclusion’.Footnote 59 However, the former is regarded by the panel as ‘merely intermediate’ and ‘not in and of itself a public moral objective’.Footnote 60 The necessity analysis under Article XX(a) GATT therefore centres entirely on the overarching purpose.Footnote 61 A noteworthy point is the EU’s counterargument before the panel, which challenges the panel’s objective framing on the basis that such a general characterization would effectively allow any government to circumvent GATT obligations by simply asserting that the action was taken in the name of public interest.Footnote 62 The panel expressly dismisses this contention, considering the identified purpose to be of sufficient particularity.Footnote 63
A similar approach to objective framing can be found in Australia–Tobacco Plain Packaging. At issue is an Australian measure that requires tobacco products and their retail packaging to conform to a uniform appearance (TPP measure),Footnote 64 the objective of which is broadly formulated by the panel as ‘improv[ing] public health by reducing the use of, and exposure to, tobacco products’.Footnote 65 At issue is Article 2.2 of the TBT Agreement, which entails a necessity analysis similar to that under some GATT and GATS general exceptions. In determining contribution, the panel considers it apt to situate the TPP measure among ‘a broader suite of complementary tobacco control measures’, maintained by Australia, rather than evaluating its impact as a discrete component.Footnote 66 On this basis the panel finds that the TPP measure makes a meaningful contribution ‘in combination with other tobacco-control measures’ to the broadly framed public health object.Footnote 67 In assessing the importance of the protected values – which corresponds to ‘the risks non-fulfilment [of a legitimate objective] would create’ under Article 2.2 TBT Agreement – the panel recognizes the ‘particularly grave’Footnote 68 public health consequences arising from the use of, and exposure to, tobacco products. The two findings are either upheld by the Appellate BodyFootnote 69 or not challenged on appeal.Footnote 70
The relevance of the two cases to the present subject lies in the exemplifying value of their analytical exercise: framing objectives along broad-stroke purposes, in itself, presents no fundamental hindrance to the determination of necessity. This point is of particular importance when considered, again, in light of the approach favoured by the EU–Palm Oil (Malaysia) panel. The EU measures challenged in the latter dispute set a number of floor levels for the overall share of renewable energy sources that the EU and individual Member States are required to reach within specified time frames.Footnote 71 One challenged measure qualifies the scope of and the extent to which biofuels may count towards the achievement of these targets; specifically, the eligible contribution of ‘high indirect land use change (ILUC)-risk biofuels’ – defined in the disputed measures as conventional biofuels made from food or feed crops, the production area of which risks a significant expansion into high-carbon stock land – is capped at specified levels for EU Member States.Footnote 72 As per the formula set by the EU, palm oil-based biofuel is the only biofuel that qualifies as a ‘high ILUC-risk’ and thereby subject to these restrictions.Footnote 73
The panel finds the measure to be GATT-inconsistent and turns to examining the EU’s invocation of Article XX GATT.Footnote 74 According to the EU, the wider EU regime regulating the consumption of biofuels and the use of renewable energy sources forms a ‘composite whole’, from which the challenged measures cannot be isolated.Footnote 75 Hence, the objective of the challenged measures should be framed in line with the policy aims pursued by the wider regime, including ‘mitigating climate change, preserving biodiversity, and addressing the associated moral concerns of the EU public’.Footnote 76 While the panel considers it possible, in principle, to characterize the objective of the contested measures ‘in terms of one or more higher level objectives’,Footnote 77 it chooses to define the objective in a ‘relatively narrow and direct’Footnote 78 manner as ‘limit[ing] the risk of ILUC-related GHG emissions associated with crop-based biofuels’.Footnote 79 The narrow formulation shapes the panel’s analysis significantly. Notably, the risk of non-fulfilment – or the importance of the protected values, per the general exceptions lexicon – is identified as ‘GHG emissions savings resulting from the promotion of conventional biofuels … [being] partially undermined, or even completely negated, by their ILUC-related GHG emissions’,Footnote 80 appearing almost paraphrastic to the framed objective itself and extending no regard to any broader considerations that may pertain to the gravity of non-fulfilment consequences.
In doing so, the panel reasons that a wider framing would render it difficult to evaluate the contribution and the risk of non-fulfilment.Footnote 81 This argumentation seems untenable when put in perspective with the analyses in Australia–Tobacco Plain Packaging and Brazil–Taxation. It is clear that existing WTO cases do not spell out a definite position on the level of immediacy and specificity in the framing of regulatory objectives, indicating that the WTO judiciary possesses and exercises certain discretion over objective framing when permitted by the legal text.
4.2 Energy Security as the Frame
The discretion over objective framing affords an initial entry point to energy security claims in the invocation of public policy exceptions. In other words, energy security may be characterized as the very objective of a given measure and hence the frame within which a public policy defence is examined. The EU–Energy Package dispute represents precisely this scenario.
The dispute emerged in the context of the EU’s liberalization and diversification of gas and electricity sectors, seeking to reduce the degree of control that a single undertaking can wield over the entire energy supply chain.Footnote 82 Forming part of this dispute is the EU’s Third Energy Package, a main component of which sets ‘the criteria and procedure applicable to the granting of authorization for transmission, distribution, supply and storage of natural gas and the operation of systems’.Footnote 83 The dispute, initiated by Russia in 2014, challenged a number of measures adopted in this regard. Of particular relevance to the present discussion is the third-country certification scheme, under which energy transmission system operators (TSOs) or transmission system owners are assessed in terms of the potential risks they would pose to the EU’s ‘security of energy supply’.Footnote 84 If they fail to demonstrate that ‘granting certification will not put at risk the security of energy supply of the Member State and the [EU]’, the competent authority ‘will’ refuse the certification.Footnote 85 In practical terms, the measure imposes energy security assessment as a prerequisite for TSOs controlled by third-country entities to operate within the EU. TSOs controlled by domestic EU entities, however, are not subject to similar ex ante screening processes.Footnote 86
The third-country certification measure is found to be inconsistent with the EU’s national treatment obligation under the GATS.Footnote 87 The EU invokes Article XIV(a) of the GATS, arguing that the measure serves to ensure energy security and is therefore necessary for maintaining public order.Footnote 88 Before the panel, the definition of ‘security of energy supply’ given by the EU is multifaceted and involves various dimensions discussed in Section 3. The EU cites the IEA’s definition of energy security,Footnote 89 which appeals to the availability and affordability of energy from an importer’s perspective,Footnote 90 while also recognizing energy security to encompass ‘the ability to respond to sudden changes’ caused by ‘infrastructure breakdown, natural disasters, social unrest, political action or terrorism’.Footnote 91 The latter aspect relates not so much to energy availability in a state of normalcy as resilience to external shocks, including those arising from geopolitical tension.
The panel report reproduces these conceptual facets set out by the EU, without pinning down whether the notion of energy security in the present dispute pertains more to one than the other. However, it seems evident that the panel’s understanding of energy security in this case involves geopolitical elements. In establishing energy security as ‘a fundamental interest of society’ under ‘a genuine and sufficiently serious threat’,Footnote 92 the panel examines the possible concurrence of several circumstances to assess the likelihood that such a threat may arise from foreign control of TSOs as follows:Footnote 93 the incentive of a foreign government to undermine the EU’s security of energy supply due to conflicts of important economic or political interests; the ability of that foreign government to require or induce foreign-controlled TSOs to undermine the EU’s security of energy supply; and the ability of foreign-controlled TSOs to achieve that by, inter alia, ‘acting in a manner that is not in their own commercial interest’.Footnote 94 The panel’s focus on the threat posed by foreign government control to the EU’s energy security interests, as well as its recognition of all the three circumstances as ‘reasonable inferences’ and of the ‘real and true’ possibility that they may be simultaneous occurrences,Footnote 95 clearly demonstrate a perception of energy security with geopolitical and defensive connotations.
Having identified security of energy supply as the regulatory objective, the panel proceeds to determine whether the third-country certification measure is ‘necessary to’ achieve this end. What is notable about the panel’s necessity analysis, as already noted by other scholars, is the unusually cursory manner in which the weighing and balancing test is conducted, and, consequently, the significant amount of deference granted to the EU.Footnote 96 I argue that what shapes the necessity analysis to that effect is the panel’s recognition of energy security itself as the objective frame. Regarding the factor of societal importance, having identified security of energy supply as an objective on its own terms concerning ‘one of the most basic necessities of modern societies,’Footnote 97 it would indeed be peculiar to assign this protected interest with anything less than ‘fundamental importance’.Footnote 98 Similarly, since the identified objective is literally the statutory criterion for granting or denying market access to TSOs under the screening mechanism, it is only logical to find the measure apt to make a material contribution to the EU’s security of energy supply.Footnote 99 With a broadly defined objective not much more specific to ‘public order’ itself and an appraisal mechanism plainly comprising that objective as an assessment criterion, the task of demonstrating necessity could not be exacting. Indeed, the third-country certification measure is found to pass the necessity test.Footnote 100
The panel’s approach to energy security claims is further manifested in its response to one of the arguments raised by Russia. Russia submits that the lack of a ‘clear and consistent’ definition of ‘security of energy supply’ in the challenged measures is a deliberate tactic to maximize the EU’s discretion, so that the screening mechanism can be manoeuvred to reduce reliance on Russian natural gas imports.Footnote 101 This contention, reminiscent of the EU’s in Brazil–Taxation (discussed in Section 4.1), speaks precisely to the risk of abuse harboured within security transference.Footnote 102 The panel, however, finds the EU’s definition to be sufficiently clear for ‘assess[ing] its defence in a meaningful manner’.Footnote 103 It therefore seems that, for the panel, the equivocalness of energy security is neither analytically prohibitive nor normally problematic for assessing the justifiability of the concerned measure.
4.3 Energy Security Claims within the Frame
I now turn to situations where energy security per se is not defined as the objective frame for assessing general exceptions defences. As discussed in Section 4.1, where a general exception is formulated with rather specific parameters and requirements, the applicable objective frame is predetermined by legal text rather than constructed afresh case-by-case by WTO dispute settlement organs. Article XX(j) of the GATT, which exempts measures ‘essential to the acquisition or distribution of products in general or local short supply’, is one of these specifically worded public policy exceptions. While this subparagraph has not been frequently invoked in WTO dispute settlement procedures, the two disputes involving Article XX(j) GATT so far are India–Solar Cells and EU–Energy Package and have afforded abundant material for ascertaining the WTO judiciary’s approach to energy security claims within predetermined objective frames.
4.3.1 Energy Security Considerations and the Specified Non-Trade Value
The measure challenged in India–Solar Cells concerns the mandatory domestic content requirement under a government initiative to promote the generation of solar power, which conditions the participation of solar power developers upon the use of certain cells and modules manufactured in India.Footnote 104 The domestic content requirement is found inconsistent with India’s national treatment obligation under Article III:4 of the GATT.Footnote 105 Invoking Article XX(j) GATT to justify the measure, India argues that the determination of a situation of ‘general or local short supply’ must take into account the measure’s underlying objectives, including, in the context of this dispute, energy security.Footnote 106 Here, the security claim presented by India draws on the need to ‘achieve domestic resilience’ to the ‘risks associated with supply side vulnerabilities and fluctuations’, which allegedly arise from ‘sole dependence on imported solar cells and modules’.Footnote 107
The panel seems somewhat uncomfortable with introducing energy security concerns into the application of Article XX GATT. While acknowledging that the immediate objective of the challenged measure – to ensure ‘access to a continuous and affordable supply of the solar cells and modules needed to generate solar power’Footnote 108 – does not ‘exist[] in isolation’ from the broader aim of energy security, the panel does not consider energy security ‘legally relevant’ for assessing an Article XX(j) defence.Footnote 109 India’s energy security claim is therefore categorically dismissed.
The Appellate Body adopts a strikingly different stance, expressly recognizing the legal relevance of broader policy considerations.Footnote 110 Evaluation of an Article XX(j) defence can take into account concerns of energy security, but not as the very framing of analysis; rather, they are considered within the specific parameters imposed by legal text. As Article XX(j) GATT pertains to ‘products in general or local short supply’, energy security concerns ‘may inform the nature and extent of supply and demand’ and thereby bear on the existence and degree of a situation of supply shortage.Footnote 111
The Appellate Body’s interpretation of ‘products in general or local short supply’ provides further insights into how energy security considerations may be incorporated within a given objective frame. The phrasing is construed by the Appellate Body to entail a case-by-case ‘balanc[ing] of different considerations’ to determine the amount of available supply relative to demand.Footnote 112 It can be credibly argued that various factors listed open up channels for energy security considerations to feed into the assessment of Article XX(j) defences. For instance, when disruptions to supply stem from undermined energy security situations, such situations ought to be taken into account in evaluating ‘the total quantity of imports that may be “available” to meet demand in a particular geographical area or market’ and ‘the reliability of local or transnational supply chains’.Footnote 113 An invoking Member could, for example, demonstrate that the risks of supply disruptions arising from imported solar cells are of such a degree that any reliance on imports cannot be reasonably expected, which then exclude imported solar cells from the pool of ‘available’ supply.Footnote 114 These inferences are not far-fetched; indeed, the panel in the EU–Energy Package expressly recognizes that supply chain reliability may be compromised of supply disruptions due to energy security issues.Footnote 115
Another aspect relating to energy security claims in India–Solar Cells concerns the temporal dimension of a supply shortage. In brief, the relevant point of contention is whether a supply shortage must be actual, rather than only potential, to be eligible for invoking Article XX(j) GATT. As energy security concerns often emerge before material damage is actually registered, not to mention the considerable time cost of dispute settlement processes, limiting the scope of Article XX(j) to situations of ongoing supply shortage would effectively preclude its availability to measures taken for energy security reasons. The Appellate Body’s pronouncements in this dispute have indeed been taken by some to mean such a constraint, including by the panel in EU–Energy Package.Footnote 116 Understood in this manner, the Appellate Body report has attracted criticism for overlooking energy security considerations.Footnote 117
I contend, however, that no such restraint has been adopted by the Appellate Body in India–Solar Cells. Contrasting the panel’s and the Appellate Body’s findings would once again shine a helpful light. The term ‘products in general or local short supply’ is interpreted by the panel to cover, in principle, only situations of an actual shortage of supply; and ‘even assuming for the sake of argument that the concept … could be interpreted to include products at risk of being in short supply’, the scope could not extend beyond products in ‘imminent risks’ of short supply.Footnote 118 The concessive assumption and the threshold of imminence clearly reflect the panel’s inclination to endorse a restraint of actuality.
The Appellate Body adopts a different approach. On the one hand, it does highlight the lack of evidence for ‘any actual disruptions in imports of solar cells and modules’ experienced by solar power developers in India.Footnote 119 On the other hand, and in the same paragraph, it recognizes that ‘risks inherent to the continued dependence on [imports]’ are relevant for assessing ‘whether a situation of “short supply” exists’.Footnote 120 In my view, a coherent reading of these pronouncements must be based on a distinction between actual disruptions to supply, risks of short supply, and an actual situation of short supply. By citing the panel’s finding that India has not identified any actual disruptions to supply, the Appellate Body by no means conditions the invocation of Article XX(j) GATT on the existence of an actual situation of short supply. Rather, it simply requires the invoking Member to prove, by evidencing the occurrence of such less systemic and less material situations as episodes of disruptions, that the risk that an actual state of supply shortage should emerge is genuine and not merely hypothetical. Simply put, Article XX(j) can be invoked for a potential state of short supply that has not yet materialized, for which evidence of actual disruptions to supply need be shown. To equate this to a categorical constraint of actuality would be an undue stretch of the legal threshold articulated by the Appellate Body.
4.3.2 Energy Security Considerations and the Assessment of Necessity or Essentiality
No Appellate Body report has addressed energy security claims in relation to another important requirement of provisional justification: the nexus threshold. With respect to the relevant requirement under Article XX(j) GATT that the measure must be ‘essential to the acquisition or distribution’ of the product concerned, the Appellate Body took the opportunity in India–Solar Cells to articulate the legal standard of essentiality, finding it to set a bar at least as high as necessity and that the weighing and balancing test remains applicable.Footnote 121 However, having determined that India did not demonstrate solar cells to be ‘products in general or local short supply’, the Appellate Body did not proceed to conduct an essentiality analysis on the specific facts of this case.Footnote 122 From a normative perspective, there is no cogent reason for treating energy security claims differently between one legal element and another under the same exception. Rather, it may be argued that the same approach should be maintained for the sake of legal coherence and predictability. Since it has been recognized that energy security claims can inform the examination of supply availability,Footnote 123 they should likewise be taken into account in the determination of essentiality.
In this light, the essentiality analysis in EU–Energy Package under Article XX(j) GATT is worth reflection. The EU’s invocation of the short supply exception in this dispute concerns a regulation adopted to complement the Third Energy Package. This so-called ‘Trans-European Networks for Energy’ measure,Footnote 124 or the TEN-E measure, seeks to facilitate investments in the cross-border energy infrastructure to ameliorate the insufficient interconnectedness of national energy networks.Footnote 125 A less fragmented European energy infrastructure network would then ‘enable switching promptly to alternative sources or routes of gas supply in case of disruption’.Footnote 126 To this end, the TEN-E measure provides various incentives to infrastructure projects that are designated as ‘projects of common interests’. The criteria for designation, as characterized by the panel, centre on the objective to diversify natural gas supply, which in practice favours projects that develop infrastructure for the transportation of natural gas of non-Russian origin.Footnote 127
The panel characterizes the TEN-E measure’s societal value as preserving the EU’s security of natural gas supply and, cross-referencing its own analysis regarding the third-country certification measure under Article XIV(a) GATS, reaffirms this value to be ‘of a high importance’.Footnote 128 But its contribution analysis is less forgiving with the EU’s energy security claim. The relevant finding is grounded on a distinction between the ‘acquisition and distribution’ of natural gas – a requirement specified in Article XX(j) GATT – and the security of natural gas supply. The means of contribution in respect of an objective as broad as energy security may likewise encompass a wide range of measures, but, as the reasoning goes, ‘not all such measures will be “essential to the acquisition or distribution” of natural gas’.Footnote 129 As such, the panel observes that while diversifying the origin of natural gas suppliers is capable of promoting energy security,Footnote 130 the EU has not established how exactly the TEN-E measure contributes to the acquisition and distribution of natural gas.Footnote 131
The panel’s narrow construction of ‘acquisition and distribution’ may have stemmed from the fact that energy security is not, in its own right, a protected non-trade value under Article XX(j) GATT. In other words, the panel may have intended to emphasize that the burden to satisfy the specified legal requirements cannot be relieved by simply resorting to an argument of energy security, so as to avoid Article XX(j) GATT being turned into a mere ‘vessel’ for energy security claims. In that case, a cautious approach to ‘acquisition and distribution’ is not in itself problematic. However, whether this requirement has been appropriately assessed against the specific facts in EU–Energy Package is debatable. In my view, in the sense that diversifying the origin of suppliers contributes to energy security by ensuring a more stable energy supply in times of highly anticipated disruption, such diversification naturally facilitates the acquisition and distribution of energy sources. It should be mentioned that the panel ‘note[s]’, without expressing agreement or disagreement, Russia’s contention that the acquisition and distribution of natural gas necessarily entails building more infrastructure projects.Footnote 132 To the extent that this citation is meant to signal endorsement, I am critical of the panel’s position. If reality is any indication, a more diversified and securely interconnected network of natural gas supply, even with fewer infrastructure projects, arguably contributes to a greater extent to the acquisition and distribution of natural gas than a larger number of projects highly concentrated in terms of origin.
5. Scrutiny of Origin-Based Discrimination
I now proceed to address another legal technique for examining energy security claims that has emerged from the relevant jurisprudence, which targets security claims raised to justify regulatory distinctions between products of different national origins. It should be clarified at the outset that, while the WTO judiciary’s approach to such security claims exhibits discernible similarities with certain legal tests for assessing ‘arbitrary or unjustifiable discrimination’ under the chapeau – which is why the following analysis turns immediately to the latter – scrutiny of origin-based discrimination in the present context is carried out not only in relation to the chapeau elements, but also in the step of provisional justification.
As well-established in WTO case law, the chapeau entails an examination of the ‘cause’, or ‘rationale’, that underlies the discriminatory elements of a disputed measure.Footnote 133 The regulatory objective identified in the step of provisional justification, therefore, continues to be relevant in the chapeau analysis, serving as a point of reference for assessing the justifiability of the rationale of discrimination. Two legal tests in this respect are particularly notable for the purpose of the present discussion. First, a justifiable discrimination must arise from a rationale that ‘bear[s] rational connection’Footnote 134 to the regulatory objective of the disputed measure. To establish ‘rational connection’, the cause underlying the discrimination must ‘fall within the purview’ of the prescribed non-trade value in the concerned subparagraph and be ‘reconcil[able]’ with and not ‘go against’ the identified regulatory objective.Footnote 135 Second, a justifiable discrimination should be even-handed, in the sense that products that are equivalent with regard to the regulatory objective should be subject to equivalent treatment. A requirement of ‘calibration’ is the other side of the same coin: a discrimination is only even-handed when it is calibrated between products that vary in conditions relevant to the regulatory objective concerned.Footnote 136 An even-handedness test therefore involves gauging the alignment between the objective of the disputed measure and the differentiating criteria of its discriminatory elements. For instance, the disputed labelling standard in US–Tuna II (Mexico) claims to protect dolphins from the risk of harm arising from tuna fishing. The Appellate Body finds the measure not ‘calibrated to the risks to dolphins arising from different fishing methods in different areas of the ocean’,Footnote 137 as it excludes only the tuna products caught by a specific fishing method in a specific area, but not those by other fishing methods outside the given area that may cause similar adverse effects to dolphins, from bearing the dolphin-safe label.Footnote 138
Turning to the present context, in both EU–Energy Package and India–Solar Cells, the respondents seek to justify certain origin-based discrimination through the use of security claims. In EU–Energy Package, the third-country certification scheme subjects only foreign-controlled TSOs to energy security assessment as a precondition for market access, while no similar requirements are imposed on domestically controlled TSOs. In defending this discrimination, the EU argues that foreign-controlled TSOs, due to their commercial, personal, and family links abroad, are ‘much more vulnerable’ to inducements and requirements by the relevant foreign governments than EU-controlled TSOs.Footnote 139 In India–Solar Cells, distinction between foreign and domestic products is cited in the step of provisional justification to assert a situation of short supply. The crux of India’s security claim is that imported solar cells and modules cannot be considered ‘supply’ due to potential fluctuations and disruptions in their availability and that, consequently, low domestic production alone demonstrates a shortage of supply.Footnote 140
Neither of the security claims results in a successful general exceptions defence. In EU–Energy Package, the panel reckons that commercial, personal, or familial connections in foreign countries, which render TSOs susceptible to inducements or requirements from relevant governments, may be possessed by EU-controlled TSOs as well.Footnote 141 As such, both domestic and foreign entities can expose the EU’s energy supply to security risks. The panel therefore finds the ‘lack of any assessment’Footnote 142 in the certification measure of security risks posed by domestic entities to constitute a lack of calibration, though ‘not mean[ing] to imply that the same regulatory scheme should necessarily be applied with respect to foreign and domestically controlled TSOs’.Footnote 143 The EU’s certification measure is then determined to be unjustified under the chapeau.
A similar line of reasoning emerged from the India–Solar Cells Appellate Body report, but this time in relation to provisional justification rather than chapeau analysis. The Appellate Body contends that availability is not inherently related to national origin, reasoning that the list of factors relevant for assessing supply and demand, as discussed in Section 4.3, may affect the availability of imported products as much as that of domestically manufactured ones.Footnote 144 Just as products manufactured at home may be primarily exported and therefore scarcely available on the domestic market,Footnote 145 it would be unreasonable to presume the unavailability of imports in regards to such intensively traded goods as solar cells and modules.Footnote 146 India is therefore found to not have established provisional justification under Article XX(j) GATT.
Despite the similarity in outcome, there are important nuances between the adjudicators’ treatment of origin-based security claims in the two disputes. Regarding the EU’s security claim, it is critical to note that the panel does not regard all discrimination in the energy security screening of domestic and foreign TSOs as unjustifiable. The finding of unjustifiability hinges on the complete lack of calibration, which can be fulfilled by establishing a commensurate screening scheme that takes account of the energy security risks arising from domestically controlled TSOs caving in to or colluding with foreign governments. The need for calibration does not, however, require that identical screening schemes be imposed on domestic and foreign entities. In other words, the EU’s origin-based security claim is accepted to some extent: different national origins of service providers give rise to distinct energy security risks, though not in a dichotomous manner dividing home and abroad. Tying back to the legal tests outlined earlier, it can be inferred from the reasoning that origin per se is recognized as rationally connected to energy security – the regulatory objective of the certification measure identified in EU–Energy Package – and as a relevant condition for classifying service providers to which equivalent treatment should be accorded. Arguably, had the measure incorporated some less strict variants of the certification scheme appropriate for domestically controlled TSOs, the EU’s public order defence would have been successful.
As India’s origin-based security claim in India–Solar Cells is raised in relation to provisional justification, the legal standards for chapeau analysis are technically irrelevant. However, as the line of logic followed by the Appellate Body in dismissing India’s security claim echoes the tests of rational connection and even-handedness, these tests still provide convenient analytical anchors for elucidating the judicial organ’s position. Judging from the finding that India’s arguments are unacceptable to the extent that they assume ‘that all imports, in and of themselves, entail supply-related risks’ and ‘that such risks are intolerable, as long as a sufficient level of domestic manufacturing capacity of solar cells and modules has not been met’,Footnote 147 the Appellate Body does not consider the origin of products determinative of their associated energy security risks, nor dispositive of their availability. Given that energy security concerns can bear on the assessment of supply and demand in the step of objective framing,Footnote 148 what the Appellate Body faults in the security claim is the presumed unavailability of imports: there can be a rational connection between origin and availability if the presence of energy security risks justifies excluding imported products from the pool of available supply, but such a trail of relation is not intrinsic and must be demonstrated by the respondent. If it is established, for instance, that energy security issues have fundamentally undermined the reliability of transnational supply chains, a case may be credibly made that imported products should not be deemed available for the purpose of Article XX(j) GATT. India’s security claim fails as it affords no tangible evidence to prove that the energy security concerns at issue have generated such an impact, but instead posits the connection between product origin, energy security, and availability as a self-evident premise of analysis.
6. A Balance Struck?
Taking stock of the discussion above, a first observation is that transference of unconventional security concepts is certainly not precluded from the legal terrain of the WTO general exceptions. While the attempt at transference in India–Solar Cells was once categorically dismissed at the panel stage, the panel’s approach has been expressly overturned by the Appellate Body. Overall, claims of energy security have been addressed and accepted to different degrees in the use of public policy exceptions, even though relevant provisions are textually unconcerned with conventional or unconventional security notions.
Beyond this general proposition, there are a few important points to unpack further. Ambiguous as the concept may be, it is only possible to accurately represent the adjudicators’ position on security transference by identifying the accent of ‘energy security’ placed in each dispute. From this aspect, taking into account the different circumstances underlying the energy security considerations in EU–Energy Package and India–Solar Cells, a comparison of the approaches adopted by the WTO judiciary seems to reveal a more sympathetic stance towards energy security claims arising from a narrower, defence-oriented security notion. Some critical particularities are present in EU–Energy Package with regards to the third-country certification measure, which is designed plainly for evaluation of energy security risks arising from foreign control. One cannot miss the unique geopolitical context of EU–Energy Package: the case was initiated after Russia’s annexation of Crimea in 2014, and the panel report was handed down after Russia banned Ukrainian goods from transiting through its territory, which instigated another WTO dispute.Footnote 149 Russia’s expanding military operations in Eastern Europe at the time, and how energy imports were increasingly felt to be a leverage for restraining the EU’s strategic responses, cast an inescapable shade of conventional security over the energy security claims raised in this dispute. As noted in Section 4.2, the panel is apparently aware of, and does place emphasis on, this geopolitical dimension. In contrast, the energy security claims in India–Solar Cells are cited to justify government support measures for renewable energy generation equipment, adopted in a context with limited defensive connotations.
The treatment of energy security claims in the two cases has been discussed side by side in the preceding sections; a brief summary is in order here to recapitulate and highlight their differences. In relation to the third-country certification scheme in EU–Energy Package, energy security is framed as its very objective. As the measure is designed plainly for evaluation of energy security risks arising from foreign control, this framing arguably eases the evidentiary burden arising from the necessity requirement under Article XIV (a) GATS. Energy security is also in and by itself recognized as a basis for discriminating between service providers of different origins in this case, though the rationale cannot go as far as to justify a complete disregard of any energy security risks associated with domestic providers. Both the framing of regulatory objective and the scrutiny of origin-based discrimination fare differently in India–Solar Cells. Energy security is treated not as the objective frame per se, but a factor that may bear on the regulatory objective stipulated in Article XX(j) to acquire or distribute ‘products in general or local short supply’. In other words, the effect of energy security considerations on the availability of certain supplies cannot be presumed, but rather must be demonstrated by the invoking Member. Similarly, energy security alone, without more to prove its impact on the availability of imports, falls short of justifying measures that discriminate between domestically manufactured and imported goods.
Since the most permissive position on energy security claims has emerged in the context of perceptible conventional security concerns, it appears that the existing WTO jurisprudence – despite the expanding landscape of unconventional security concepts – leans towards re-centring transference claims around a traditional, defence-oriented notion. At the same time, two important qualifications must be attached to this statement. First, this statement does not amount to precluding unconventional security concepts from the general exception defences. Even though the invocation of Article XX(j) GATT was unsuccessful in India–Solar Cells, it must not be overlooked that the Appellate Body has recognized energy security considerations as legally relevant in respect of the prescribed requirements, including the determination of ‘general or local short supply’. The more lenient interpretation endorsed by the Appellate Body for the temporal dimension of this component, extending to not only actual but also potential situations of supply shortage, also ensures that it would not constitute a virtually absolute bar to energy security claims. It is hence clear that security concepts that are more distant from conventional notions can bear on the interpretation of general exceptions as well. Second, the recognition of security claims that possess more defence-oriented attributes is also subject to limits. Such limits can be discerned from EU–Energy Package, where the panel emphasizes that energy security concerns do not relieve the EU of the burden to demonstrate ‘acquisition and distribution’ in the essentiality analysis – although I contend, as argued in Section 4.3, that this interpretation appears unduly restrictive. Another example in the same case is the panel’s assessment of the third-country certification measure under the chapeau: even when a frame considerably forgiving with energy security concerns is adopted in the process of provisional justification, any origin-based discrimination remains under scrutiny at the stage of chapeau analysis.
Overall, as the more contemporary energy security notion in India–Solar Cells is indeed constrained by legal confines that are more restrictive, it may be inferred that energy security claims that are more removed from the conventional confines of national security have been addressed by the WTO judiciary with greater caution than those exhibiting traits of a defence-oriented security concept. There is some validity to this distinction: arguably, it is with the former type of transference attempts that illegitimate protectionist measures would more likely come about behind the façade of security concerns. However, it will be advisable for the WTO dispute settlement organs not to implement this distinction as a strict dichotomy. Recalling one unconventional conception of energy security centring around sustainability in the context of green transition in Section 3, a rigid application of the distinction could overlook the relevance of security claims underpinned by genuine sustainability concerns. Insofar as public money and government support – often requiring measures inconsistent with GATT or GATS disciplines – are more indispensable for developing economies in renewable energy deployment, a mechanical differentiation between conventional and unconventional security concerns could result in systemic and disproportionate restraints on the policy space of developing Members. WTO dispute settlement organs should exercise caution in this regard. However, it will be a travesty to represent India–Solar Cells as an example of such bias, given the untenable assumptions about supply availability in India’s security claims.
It should be noted that there has yet to be a test case for sustainability oriented energy security claims. Although it remains to be seen whether and to what extent such claims would be recognized by WTO dispute settlement organs, the factual aspects of existing WTO disputes may be drawn on as a hypothetical case to aid projections. Using India–Solar Cells as an example, an energy security claim that situates sustainability at the centre may also give rise to a defence under Article XX(j) GATT. Instead of seeking to establish a situation of short supply based on the unreliability of imports – which was the argumentation used by India, unsuccessfully, in the dispute – it could be contended that clean energy supply is exposed to credible risks of global, ‘general’ shortage, and that India, a major player in the energy transition process worldwide, must adopt measures to enhance its capacity in manufacturing clean energy generation equipment. In addition, under Article XX(g) GATT, a sustainability oriented security claim could appeal to the role of green energy transition and hence government support in preserving clean air and a liveable climate as ‘exhaustible natural resources’. The characterization of clean air as an exhaustible natural resource in WTO case law,Footnote 150 in combination with the recognized need for government intervention to counter market failure in the production and consumption of renewable energy,Footnote 151 could give substance to these contentions. However, given the stringent scrutiny that WTO panels and the Appellate Body have exercised over origin-based discrimination and the mixed empirical evidence demonstrating their benefits for the deployment of clean energy,Footnote 152 any domestic content requirements would remain difficult to defend.
To summarize, in the context of energy security, the WTO’s approach to attempts of security transference under the general exceptions is not all-or-nothing. WTO dispute settlement organs have recognized the pertinence of energy security and do not exclude relevant arguments from legal analysis. At the same time, energy security claims are more or less moulded and circumscribed by the established parameters of the general exceptions, whether prescribed by legal texts or derived from case law. Judging from the discussion above, these ‘old’ parameters have generally served as effective but not overly restrictive guardrails in the treatment of security transference, allowing for reasonable consideration of unconventional security claims while safeguarding against attempts to expand unduly the outer limits of public policy exceptions. In my view, this is a balance properly struck.
The current Appellate Body impasse merits a few final remarks. In the present context, the absence of a functioning Appellate Body is probably more detrimental to complainants and to the WTO dispute settlement system as a whole more than it is to any respondents attempting at security transference. As explained in the scholarship and by way of logic, respondents gain by appealing unfavourable rulings (findings of inconsistency) into the void. Complainants, however, cannot benefit in a similar way, as either accepting or appealing unfavourable rulings (findings of consistency) into the void would end up keeping the disputed measure in place.Footnote 153 Respondents have therefore a stronger incentive to appeal into the void, which has been shown by empirical research to be consistent with the pattern of appeals since 2020 across WTO disputes.Footnote 154 In regards to the WTO judiciary’s treatment of security transference, this asymmetry means that overly stringent approaches would have less chance to impair the interests of respondents than overly expansive approaches do to complainants. Furthermore, the present context affords (but) one example of how the ongoing institutional paralysis could cost the WTO dispute settlement system its contemporary relevance regarding concepts of international law in flux. Recalling the important mitigating role played by the Appellate Body in India–Solar Cells, what is lost from its absence are crucial interpretive and discursive moments for the WTO judiciary to remain a part in shaping the meaning of ‘security’ as the concept evolves and expands.
7. Conclusion
In present-day international affairs, a wide range of issues removed fairly from the conventional conception of national security have been labelled as ‘security’ interests. The uptake of security parlance has also reached international economic disputes, including through the invocation of exception clauses in existing treaties. Apart from those containing express security-related phrasings, transference of security language to contexts beyond military and defence affairs has also been attempted under the so-called public policy exceptions, whose parameters and legal standards have largely developed from non-security-related contexts.
The article focuses on the latter scenario through a case-specific lens, investigating the treatment by WTO panels and the Appellate Body of energy security claims raised under the general exceptions – Article XX of the GATT and Article XIV of the GATS. As a quintessential example of contemporary security constructs, ‘energy security’ can denote a diversity of issues ranging from supply availability, origin reliability, defensive concerns, to environmental sustainability. Various types of energy security concerns have been cited by WTO Members in constructing general exceptions defences, giving rise to a not yet extensive but illuminative body of cases for establishing the dispute settlement organs’ approach to security transference.
Two legal techniques can be discerned from existing case law in this respect. Under generally-worded subparagraphs, such as Article XX(a) GATT and Article XIV(a) GATS, it is logically possible to characterize energy security as the very objective frame for the analysis of provisional justification. The adoption of such a broad-stroke framing could give rise to a rather lenient legal standard for nexus analysis, including requirements that are often considered exacting to establish, such as ‘necessary to’ and ‘essential to’. Specifically worded subparagraphs, however, give rise to predetermined objective frames. As clarified by the Appellate Body in India–Solar Cells, their existence does not merit the categorical dismissal of energy security claims either. Instead, they factor in the interpretation of – though not automatically relieving the burden to establish – the legal requirements prescribed. Contrasting these enabling aspects, the other legal technique is restraining in character. Throughout the steps of provisional justification and chapeau analysis, WTO judicial bodies attentively identify and review energy security claims that seek to justify discriminations between domestic and foreign products or services based solely on their origins. Unless rationally connected to the regulatory objective and applied in an even-handed manner, such distinctions are always dismissed as unjustified.
Neither of these techniques is exactly unique to energy security, as they stem from a long trajectory of WTO jurisprudence outside the context of energy security. However, how these techniques have been deployed by WTO panels and the Appellate Body can be indicative of their stand on energy security claims. Bringing the relevant case-law contexts into the picture helps elucidate the varying degrees of recognition that the WTO judiciary has accorded to contemporary security transference claims: more permissive when the security claims are more closely related to the traditional, defence-oriented security notion, and more cautious when that connection is more remote. However, importantly, the practice of caution does not give rise to categorical dismissal, and the exercise of leniency is still accompanied by guardrails against abuse. Basically, what WTO judicial bodies uphold is that the use of security transference must not automatically relieve responding Members from the burden of establishing the specified legal requirements of relevant general exceptions. Lastly, a sustainability-based energy security claim has not yet been tested in WTO disputes. If and when such a claim emerges, its interplay with Article XX(g) and (j) of the GATT would be especially noteworthy.
Overall, WTO dispute settlement organs seem to have thus far effectively contained the risk of abuse that security claims may bring about to public policy exceptions. However, they have also accorded much regard to unconventional security claims, a point worth highlighting as calls for granting more policy space to Members have been gaining prominence in the research and practice of WTO law. The WTO judiciary’s treatment of unconventional security concerns is not always without flaws; as critiqued in Section 4.3, the EU–Energy Package panel’s assessment of the TEN-E measure’s contribution in the necessity analysis under Article XX(j) GATT seems problematic, and its interpretation of the temporal dimension of the same subparagraph appears inconsistent with the Appellate Body’s reading in India–Solar Cells. But these do not undermine the appropriateness of the general stance that WTO judicial bodies have taken on contemporary security concerns, and it is hoped that the misinterpretation and inaccuracies can be addressed in future case law.
Acknowledgements
I am most grateful to the two anonymous reviewers for their constructive and insightful comments, and to the editors for their professionalism and efficiency in handling the review process. I also wish to extend my sincere thanks to my PhD supervisors, Professor Piet Eeckhout and Professor Gracia Marín Durán, for their detailed feedback on an earlier version of this article, and to Dr Olabisi Akinkugbe, Dr Kim Anh Dao, Dr Caroline Henckels, Dr Chen Yu, Julie Szu-Yi Lee, and other attendees at the Ninth Society of International Economic Law Biennial Global Conference for their engaging discussion. All errors are mine.
Open access
