Climate change poses a grave threat to the international community, prompting global action on all fronts. Governments are adopting various ‘green energy programmes’, supporting sustainable development and mitigating climate change. Albeit government support is indispensable for promoting environmental-friendly initiatives, it inevitably interferes with the free market. After Canada-Renewable Energy (2013), it is apparent that a clash exists between environmental policy and WTO law. As the Appellate Body shied away from balancing public benefit with trade principles, there is a growing consensus to reform the SCM Agreement to safeguard ‘policy space’ for government support to renewable energy. The proposal of reviving and expanding Art. 8 of the SCM Agreement appears the most effective solution, but faces considerable political and practical hurdles at the present junction.
It is largely uncontested that a healthy environment is essential to building prosperous and resilient economies, however, the linkage between trade and the environment has been intensely debated in the multilateral trading system over the past few decades. Countries, along with the World Trade Organisation, now face an uncomfortable choice: should they be willing to sacrifice certain free trade principles in exchange for increased environmental protection? Or should such principles prevail even if they slow efforts to tackle environmental problems?
The ‘legal acrobatics’ engaged by the Appellate Body in the Canada – Certain Measures Affecting the Renewable Energy Generation Sector/ Measures Relating to the Feed-in Tariff Program (2013) (‘Canada–Renewable Energy’) has been widely criticised for the novel and incoherent analysis in adjudicating the subsidy complaint. In brushing off policy considerations that further industrial transformation and speed up sustainable development, the Appellate Body indicated that the WTO subsidies rules would not accommodate the environmental and societal positive externalities (e.g. climate change mitigation and energy innovation) brought about by renewable energy support measures/ ‘green policy’.
The decision has generated wide debate and resulted in various proposals for reforming the WTO Agreement on Subsidies and Countervailing Measures (SCM Agreement) to shelter support regimes to renewable energy sector from being scrutinised under the SCM Agreement. As is observed:
‘…[T]he essence of the call for reform is that the current SCM disciplines, which were negotiated and designed in the late 1980s and early 1990s, are no longer suitable in the face of the imperative of addressing climate change, among other global public goods.’
This article examines the Canada–Renewable Energy dispute in detail. As the first WTO dispute addressing renewable energy support measures, the case was inevitably complex, involving examination of General Agreement on Tariffs and Trade (GATT), the Agreement on Trade-Related Investment Measures (TRIMs), and the SCM Agreement. Although discussions on GATT and TRIMs offer critical jurisprudence on the WTO law in general, this article confines the analysis to the reasoning on subsidy presented by the Panel and the Appellate Body.
The following of the article begins by examining the policy reasons behind governments’ intervention in the energy industry, emphasising the public benefit provided by renewable energy support regimes and briefly delineating the conflict with trade principles. Part III provides an overview of the WTO subsidies rules, gauging the possibility for governments to introduce and implement renewable energy subsidies that are both environmentally and legally sustainable. Part IV examines the Canada-Renewable Energy dispute, emphasising the lack of cogency in the Appellate Body’s formulation of the benefit test. Part V presents current state practices in trade disputes involving renewable energy support measures, noting the negative impact of unilateral actions on global trade and environmental protection. Part VI concludes the existing WTO subsidies rules are incapable to accommodate legitimate policy consideration.
Climate change is possibly the greatest challenge facing the international community today. While some doubt the need for action, the 1992 United Nations Framework Convention on Climate Change (UNFCCC), with its near-universal membership, represented the global response to this challenge, setting the ultimate objective of stabilizing ‘greenhouse gas emissions concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system’. This commitment has been enshrined into the ground-breaking Paris Agreement, entered into force on 4 November 2016, wherein the overarching goal is set as ‘holding the increase in the global average temperature to well below 2 °C above pre-industrial levels’, calling all signatories to implement measures to curb greenhouse gas (GHG) emissions through ‘nationally determined contributions’.
The energy sector is by far the largest source of GHG emissions — responsible for more than two-thirds of worldwide GHG emissions (of which 80% comes from fossil fuel consumption). Research projects in this area have boosted awareness of and interest in sustainable energy resources development. It is now widely accepted that ‘replacing conventional ‘brown’ or ‘dirty’ energy (generated from fossil fuels such as coal, natural gas and oil) with renewable ‘green’ or ‘clean’ energy (generated from naturally replenished resources such as solar, wind, biomass, geothermal and hydropower) is an important and necessary step to mitigate climate change and achieve the internationally agreed 2 °C target.
Countries are encouraged to join force and replace carbon-intensive fossil fuels with renewable energy sources. With the United Nations launching a global initiative to double the share of renewable energy in the global energy mix by 2030, transition to the ‘low-carbon energy economy’ has become a key theme of many countries’ national energy policies.
Yet, for sustainable solutions to emerge to address climate change, investments are required at various points of the supply chain—from research and development (R&D) to the commercialization of green energy supply. Market failure in the energy industry has caused price distortions and discouraged community’s adoption of renewable energy related products, hindering industrial transition towards sustainable development. Noting inadequate market power, governments across the globe have chosen to step forward, actively and intentionally intervene in the energy sector to encourage the development of renewable energy in the domestic market. This has given rise to a new wave of industrial policy with a green tinge (‘green industrial policy’).
Subsidy has long been a key instrument in operationalizing such “green policy”. On a general level, government practices aim to lower the price paid by energy consumers (consumption subsidies), lower the cost of energy production, or raise the price received by energy producers (production subsidies). The most common renewable energy subsidies include:
‘(1) Subsidies to lower the cost of production to producers
- Direct financial grant
- Preferential loans/preferential financing
- Provision of inputs on preferential terms
- Tax exemption/tax credit/other tax incentives
- Accelerated depreciation allowances on equipment
- Public supply of R&D
- Land use preferences
- Preferential insurance
(2) Subsidies to increase the productivity of producers
- Direct investment in energy-related infrastructure
- Direct investment in export-related infrastructure specific to given producer(s)
- Incentives to attract human resource expertise specific to the industry (for example, expatriate returnees)
(3) Subsidies to reduce uncertainty of returns for producers
- Feed-in tariffs (FITs)/price guarantees
- Demand guarantees
- Mandated deployment rates
(4) Subsidies to promote consumer adoption of renewable energy technology
- Direct financial grant
- Tax credit
- Rebates for equipment purchases
- Price controls
- Public procurement’
In practice, renewable energy subsidies are more commonly offered to the production side of the energy sector, coming in forms of Feed-in Tariffs (FITs), power purchase agreements, capital grants, soft loans and R&D funding. In recent years, renewable energy subsidies are rising rapidly, reaching USD 140 billion in 2016. The power sector has been the primary target of such government stimulus, with 128 countries implementing various form of fiscal support in 2017 compared to 70 and 24 countries in the transport sector and in the heating and cooling sector, respectively. Government efforts in subsidising the renewable energy sector have promoted significant progress:
‘the electricity sector has performed the fastest in terms of progress towards decarbonization, with renewable energy growing at unprecedented rates in the last decade (e.g. 9.3% growth of RE power capacity was registered only between 2014 and 2015), and it is projected to continue outpacing growth in conventional energy in the years to come, as it has been the case since 2012.’
However, heavy reliance upon subsidies also made the power sector the most varied in terms of public incentives and thereby the most challenging to evaluate from both an international trade law and an environmental policy perspective. This is because under the overly generalised label of ‘renewable energy subsidies’ are a wide range of government support measures whose features and objectives vary considerably. The primary factors driving the fusion of environmental and industrial policy domains have been political economy pressures, technological change, and concerns over energy security. Technological advances of recent years have lowered the cost of renewable energy sources and made their adoption less cost prohibitive. This in turn, makes it difficult for governments to justify public spending to support renewable energy policies on environmental grounds alone. In an era of rising fiscal austerity, the question of “what’s in it for us” takes on greater political salience in every debate over public spending.
To justify funding to support environmental policies, governments increasingly need to ensure some payoff for their constituencies. In order to gain public support, governments are embedding spending on renewable energy projects within an overarching industrial policy designed to create high-paying “green-collar” jobs. In addition, governments seek to emphasise the benefits of renewable energy adoption by associating its importance with national security: it would reduce dependence on energy of foreign sources. Political support for the industrial policy is thus closely related with the purported economic value brought to the domestic market.
Accordingly, environmental considerations are not the only impetus, nor are they necessarily the main driving force, behind government support policies involving subsidies.
It is common for governments to mix political and economic motivations in designing renewable energy subsidies. These may include seeking
‘(1) to develop manufacturing and export competitiveness in an industrial sector where it is believed demand will grow over time;
(2) to increase the number of manufacturing jobs;
(3) to induce the transfer of technologies associated with the manufacturing of green goods;
(4) to capture associated political benefits;
(5) to provide rents to certain interest groups; or
(6) to protect domestic firms from foreign competition.’
Any series of the above rationale can be integrated into the methodology of green industrial policy, and it is not always the case that governments are capable to tailor subsidy regimes properly without inducing negative externalities. At one end of the spectrum, there are renewable energy subsidies that simply incentivize related industries, be it domestic or foreign, to use climate-friendly energy sources. At the other end, there are government support measures which seek to foster competitive domestic industries (i.e. with a protectionism aim) and create local employment opportunities thus garnering political support for the renewable energy implementation and transition. This is mostly the case of renewable energy subsidies incorporating so-called local/domestic content requirements (LCRs/DCRs), which ‘condition eligibility for public support upon the use of a certain percentage of domestically produced inputs into the manufacturing of clean energy technologies, or in the construction of clean power generation facilities.’ It in effect compels industry participants to source a certain percentage of their equipment and other production material inputs from local producers, and suppliers with imported material are thus excluded from receiving the support.
The use of LCRs in renewable energy subsidies should be evaluated against the market condition after the Global Financial Crisis: international economy and the financial system remain distressed and labour markets are showing little progress, with insufficient corporate investments readily available. LCRs can foster economic benefits such as net employment gains and the creation of a domestic industry. In fact, the capability of LCRs to create ‘green jobs’ is often something that helps governments gain political support for green industrial programs: by requiring enterprises to use a certain proportion of inputs from local industries, governments promote green policies as generating new employment opportunities. As for developing and emerging economies where renewable energy remains an infant industry, governments adopt green subsidies with LCRs to protect local enterprises’ development until they can realize their latent comparative advantage (N.B. Art. XVIII(c) GATT is an exception for infant industry, but it is limited for use by developing countries only). It is intended that LCRs will eventually bring more, new mature players to the global renewable energy market, which in turn, increase competition and innovation and thereby lower green technology costs. This accelerates the schedule on which renewable energy reaches grid parity and become capable to compete with fossil fuel and nuclear energy without subsidization.
The legitimate policy considerations (e.g. to promote renewable energy transition and mitigate climate change) however, are not readily accommodated by the current SCM discipline regime. As discussed below, LCRs are treated as inherently trade distorting thus prohibited under Article 3 of the SCM Agreement unless a WTO Member is exempt on account of Article 27. Also, after Canada-Renewable Energy, there is by now little doubt that the discriminatory effects of LCRs are problematic from a WTO law standpoint.
This part examines the relevant WTO law on subsidies and points out areas of tension with renewable energy initiatives furthering sustainable development.
WTO has long been positioned as the best international forum to take on energy subsidies reform. The interface of trade and climate change mitigation and adaptation is at the heart of contemporary legal developments in the energy sector. Governmental measures that provide financial support for the renewable energy development can distort trade patterns and markets more generally, making it a prime target for the application of WTO rules aimed at liberalizing global trade.
However, the WTO does not contain an energy chapter, nor has it devised any special rules targeting energy subsidies. In fact, the word ‘energy’ is mentioned only once in the SCM agreement (in Footnote 61 regarding inputs consumed in the production process). Renewable energy subsidies are thus subject to the general rules set forth in the SCM Agreement.
In this part, we examine how the SCM rules may apply to renewable energy subsidies in general terms, focusing on the legal elements of the ‘subsidy’ definition prescribed in Article 1 and the conditions in Article 3&5 under which subsidies are either prohibited or actionable. At the end of this section, we introduce the now lapsed Article 8 ‘non-actionable’ subsidies, and briefly examine its utility in sheltering renewable energy subsidies from remedial actions.
It should be noted from the outset that under SCM Agreement, the term ‘subsidy’ has a technical meaning that is limited in certain respects. Government measures that provide financial support to renewable energy sectors may not qualify as subsidies under the SCM Agreement, even if they have some kind of trade distorting impact. Moreover, SCM Agreement applies only to discipline ‘subsidies measures affecting trade in goods.’ Subsidies to service sectors are generally dealt with under Article XV of the General Agreement on Trade in Services (GATS) which imposes minimum standards other than requiring Member States to adhere to Most Favoured Nation (MFN) and National Treatment (NT) Principles. This would ordinarily lead one to wonder whether a particular renewable energy subsidy should be treated as relating to goods or services before applying the relevant WTO rules. Such question arises and is worth considering, mainly because neither the GATT nor the GATS define what constitutes a ‘good’ or a ‘service’, despite the fundamentally different treatment towards subsidies under the respective regimes. Accordingly, answers to the question relies heavily on the factual circumstances surrounding the disputed subsidy, the export/import conduct and requires consideration of industrial norms.
In practice, such debate is limited to non-liquefied natural gas and electricity. There is no question over the fact that energy products such as oil, solid fuels and liquefied natural gas are goods,and despite GATT’s silence on the subject, almost all Member States regarded electricity as a commodity. Moreover, in Canada-Renewable Energy, the Appellate Body upheld the Panel’s finding that the government purchase of electricity under the FIT Program constitutes ‘purchase of goods’ within the meaning of Article 1.1(a)(1)(iii) of the SCM Agreement.
The operation of SCM Agreement essentially involves a two-step analysis:
- Whether there is a ‘subsidy’ within the meaning of Article 1, and
- If so, whether or not it fell within the criteria set out in Article 3 (‘Prohibited’) and Article 5(‘Actionable’).
With regards to the first definitional question, a ‘subsidy’ shall be deemed to exist for the purpose of applying the SCM Agreement only if three cumulative conditions are met: (i) there must be a financial contribution (or income/price support) by a government or public body (or by a private body ‘entrusted’ or ‘directed’ by a government); (ii) it must confer a benefit; (iii) it must be specific to certain enterprises or industries.
As for step-two, only two kinds of subsidies are outright prohibited under the SCM Agreement, namely: export subsidies (i.e. those contingent upon export performance) and import-substitution subsidies (i.e. those contingent upon the use of domestic over imported goods). All other specific subsidies are “actionable” and are permitted under WTO law so long as they do not negatively harm the trade interests of other countries. A government may petition to declare another government’s subsidy illegal if it can demonstrate that the subsidy has ‘adverse effects to the interests of other [WTO] Members.’It may also take unilateral actions in domestic administrative courts against another government’s subsidy if it finds that ‘the effect of the [actionable subsidy] is such as to cause or threaten material injury to an established domestic industry, or is such as to prevent or materially retard the establishment of a domestic industry.’
(a) Financial Contribution or Any Form of Income or Price Support
Article 1 of the SCM Agreement offers an exhaustive list of government interventions that could qualify as a ‘subsidy’ under the SCM Agreement if they confer a benefit.
Whether a particular government intervention constitutes a financial contribution (or any form of income or price support) largely depends on the specific form it takes. Article 1.1(a)(1) of the SCM Agreement sets out in four categories of measures that would constitute a financial contribution. In the US—Final Countervailing Duty Determination with Respect to Softwood Lumber from Canada (US-Softwood Lumber IV), the Appellate Body observed:
An evaluation of the existence of a financial contribution involves consideration of the nature of the transaction through which something of economic value is transferred by a government. A wide range of transactions falls within the meaning of “financial contribution” in Article 1.1(a)(1). According to paragraphs (i) and (ii) of Article 1.1(a)(1), a financial contribution may be made through a direct transfer of funds by a government, or the foregoing of government revenue that is otherwise due. Paragraph (iii) of Article 1.1(a)(1) recognizes that, in addition to such monetary contributions, a contribution having financial value can also be made in kind through governments providing goods or services, or through government purchases. Paragraph (iv) of Article 1.1(a)(1) recognizes that paragraphs (i)–(iii) could be circumvented by a government making payments to a funding mechanism or through entrusting or directing a private body to make a financial contribution. It accordingly specifies that these kinds of actions are financial contributions as well. This range of government measures capable of providing subsidies is broadened still further by the concept of “income or price support” in paragraph (2) of Article 1.1(a).
Generally speaking, the case law offered a rather expansive interpretation of the three types of financial contribution (Article 1.1(a)(1)), whereas they endorsed a narrow reading of ‘income or price support’ (Article 1.1(b)(2)). Governments may ‘subsidize’ by positive action when transferring monetary or non-monetary resources directly or indirectly through private actors. Alternatively, they could ‘subsidize’ by negative action when refraining from collecting revenue otherwise due. Only two types of government intervention are explicitly excluded from the subsidy definition: the provision of general infrastructure (which in essence relates to the specificity element as discussed below), and border tax adjustments on indirect taxes and import duties.
It is evident that the exhaustive list in Article 1.1 is sufficiently broad to capture most support measures widely adopted by countries’ green policy initiatives (e.g. power purchase agreements, capital grants, soft loans and R&D funding) as these interventions would normally provide economic value to the recipient thus qualify as ‘a financial contribution’.
(b) The Financial Contribution Must Confer a Benefit
Article 1.1(b) requires the ‘financial contribution’ to ‘confer a benefit’. The Appellate Body has explained that determination of whether a ‘benefit’ exists implies some kind of comparison, and that the ‘marketplace’ provides an appropriate basis for comparison. As stated in the Canada—Measures Affecting the Export of Civilian Aircraft (Canada-Aircraft):
…The word “benefit”, as used in Article 1.1(b), implies some kind of comparison…[F]or there can be no “benefit” to the recipient unless the “financial contribution” makes the recipient “better off” than it would otherwise have been, absent that contribution. …[T]he marketplace provides an appropriate basis for comparison in determining whether a “benefit” has been “conferred”, because the trade-distorting potential of a “financial contribution” can be identified by determining whether the recipient has received a “financial contribution” on terms more favourable than those available to [other] recipient in the market.
The above principles apply differently according to the types of financial contribution identified in each dispute. Benefits conferred are thus difficult to identify in some circumstances where no direct cash transfer takes place. For example, in Canada–Renewable Energy, the ‘financial contribution’ was found to be taking the form of ‘government purchasing goods’. Analysing ‘benefits conferred’ thus requires evaluation of the purchase price or terms of provision as compared to what is available in the market. A key issue however was how to define the relevant benchmark when government intervention has long distorted the market. The energy industry has been largely dominated by state run monopolies and governed by strict territorial allocation. International trade in energy resources and products are also heavily concentrated, cartelised and controlled by a few multinational companies. The ‘market benchmark’ test presupposes the existence of sufficient market power and faces particular challenges, associated with the existence of natural monopolies, as well as with the role of state-owned enterprises that dominate some national energy markets, in adjudicating support measure provided to the energy sector (renewable energy subsidies in specific).
(c) The Subsidy Must be Specific
Article 1.2 provides that subsidies are ‘subject to’ SCM discipline only if they are ‘specific’ as defined in Article 2. Article 2.1 stipulates that only subsidies that are either de jure (Article 2.1(a)&(b)) or de facto (Article 2.1(c)) ‘specific’ to certain enterprises or industries are subject to the disciplines of the SCM Agreement. The Appellate Body in the United States — Countervailing Duty Measures on Certain Products from China, explained the relationship of the three sub-provisions as:
…[T]he specificity analysis should normally begin by examining the evidence that is relevant for determining de jure specificity pursuant to subparagraphs (a) and (b) of Article 2.1…
… In the context of determining whether a subsidy is de jure specific, the language in subparagraphs (a) and (b) directs an investigating authority to scrutinize any explicit limitation of access to a subsidy or look for the existence of objective conditions or criteria governing eligibility for a subsidy. In cases where an examination of the nature and content of the challenged measure indicates that access to a subsidy is explicitly limited to certain enterprises or, alternatively, if there are “criteria or conditions” governing the eligibility for a subsidy that are spelled out in law, regulation, or other official document, an investigating authority will normally begin by examining this evidence in the light of subparagraphs (a) and (b) in order to determine whether the subsidy is de jure specific….
…Turning to an examination of de facto specificity… the word “if”, at the beginning of the first sentence of Article 2.1(c), is ordinarily used to introduce a condition that may either be fulfilled or not fulfilled. By contrast, the preposition “notwithstanding”, which introduces the subordinate clause in the first sentence of Article 2.1(c), is relevantly defined as “[i]n spite of, without regard to or prevention by”. In the context of the first sentence of Article 2.1(c), this clause suggests that, in spite of any appearance that a subsidy is not specific following an application of the principles set out in subparagraphs (a) and (b), an investigating authority may still examine “other factors” and find that the subsidy at issue is de facto specific. For instance, in a situation where the evidence suggests that a subsidy is not de jure specific because the conditions set out in subparagraph (b) are satisfied, … a subsidy may nevertheless be found to be “in fact” specific.
An inquiry into whether a particular subsidy measure is de facto specific under Article 2.1(c) is fact-incentive and depends on the condition/eligibility criteria affixed to the measure and actions by the government in the implementation.
In the context of renewable energy subsidies, establishing specificity is rather straightforward where the incentives are offered to the production side (as discussed in Part II). Moreover, given that the renewable energy industry remains a specific portion of the overall energy market, it is arguable any industrialised subsidies would still meet the de facto test under Article 2.1(c), regardless of whether their target being certain technologies, or renewable sources in general. Subsidies to promote consumer adoption of renewable energy are non-specific and have positive social externalities (‘energy welfare’). However, their adoption is largely restricted by fiscal prospect and has produced minimum progress.
(d) Subsidies are Either Prohibited or Actionable
The current SCM Agreement prohibits two forms of subsidies outright: (i) subsidies contingent on the use of domestic over imported goods (that is, subsidies with local content requirements [LCRs]), and (ii) subsidies contingent on export performance. All others are permissible so long as they do not have an “adverse effect” on trading partners.
Article 3 singles out export subsidies and domestic content subsidies for harsher treatment under Article 4 because of their egregious trade-distorting effects: export subsidies can harm foreign competitors in markets around the world; and domestic content subsidies limit foreign companies’ access to a Member’s domestic market. Rules governing prohibited subsidies do not require any demonstration of actual adverse effects: so long as the elements in Article 3.1 are established, the disputed subsidies qualify as ‘prohibited’and must be ‘withdrawn without delay’. As noted in Part II, although LCRs/DCRs have been quite popular in the renewable energy sector, in the absence of any express exemption, ‘subsidies’ with LCRs component face elimination regardless of their underlying policy objective.
Provided that they meet the requirements of specificity and adverse effects, subsidies that fall outside the definition of a prohibited subsidy could still be challenged under the SCM Agreement as ‘actionable subsidies’. Adverse effects may occur in the form of: ‘(i) injury to the domestic industry, (ii) nullification or impairment of benefits, and (iii) series prejudice to the interests of other members.’ Any member, adversely affected by a specific subsidy, may take action either unilaterally through countervailing measures or multilaterally through the dispute settlement system.
(e) Non-actionable Subsidies
For the first five years of the WTO (1995–99), the SCM Agreement provided a third category: the ‘non-actionable’ subsidies as defined in Article 8. It was essentially a safe harbor available to all Members to devise and report to the SCM Committee particular forms of subsidies permissible under WTO rules: (1) certain types of research subsidies, (2) aiding subsidies to disadvantaged regions, and (3) subsidies promoting the adaptation of existing facilities to environmental requirements. The non-actionable subsidies would be immune from remedial actions initiated pursuant to Article 7 of the SCM Agreement, even if such subsidies produced adverse trade effects. Similarly, where such subsidies caused injury to a domestic industry, the aggrieved member state would be prohibited from unilaterally imposing countervailing duties as a response. This provision would have provided some legal certainty to shelter renewable energy subsidies that attain environmental objectives with controllable/acceptable trade-restrictive impacts (i.e. compliant with Article 3). Many trade academics purport to revive and extend Article 8 to provide a specific exemption for ‘good’ renewable energy subsidies while adding into SCM Agreement some space for policy consideration. While this proposition sounds reasonable as a matter of principle, the practical challenge is to solicit support from WTO member states to unite in designing an exemption clause that effectively shields ‘good’ subsidies, and is ‘free from being abused’. The never-ending Doha Round negotiation arguably has demonstrated the difficulty facing this proposal.
Given the current jurisprudence on SCM discipline, where interpretations of financial contribution, benefit, and specificity are confined to rigid textual and linguistic analysis, it is arguable that almost all renewable energy subsidies are vulnerable to challenge and may face elimination in certain circumstances, regardless of the legitimate policy consideration and environmental benefits behind their implementation.
IV Canada–Renewable Energy
This section examines the Canada-Renewable Energy dispute. As the first challenge involving renewable energy support measures’ compatibility with SCM Agreement, it presented the Dispute Settlement Body (DSB) a rare opportunity to soften the technicality of subsidies rules with policy considerations which would provide wide public benefit.
Two parallel WTO complaints were initiate by Japan and the European Union (EU) with respect to certain domestic content requirements in the feed-in tariff programme (FIT Programme) established by the Canadian Province of Ontario. The provincial government enacted a law in 2009 to incentivize the production of electricity from wind or solar generators through the widely used instrument of a feed-in tariff (FIT).
A FIT is a long-term contract by a government agency to secure wholesale electricity at a set price that reflects a rate of return attractive to investors and developers. The incentive occurs because the price in the contract is generally above the wholesale price of power in the region (otherwise, there would be no incentive).FITs, as typically designed, are effective in addressing technology ‘spillovers’ (the excess social benefit above that which can be appropriated by private innovators through patents and the like), as they can tailor the degree of support according to learning potential. Technologies at earlier stages of development can be afforded larger premiums.
FIT has long constituted the lion share of renewable energy subsidies: according to the latest Global Status Report of the Renewable Energy Network (REN 21), 113 FIT schemes were in place in 2017 at either state, province, or country levels., while at the time of the dispute, 99 FIT schemes were in place across the globe.
The Panel closely studied Ontario’s FIT programme and observed:
‘The FIT Programme
The FIT Programme has very clearly two fundamental objectives: First, to encourage the participation of new generation facilities using renewable sources of energy into Ontario’s electricity system in order to diversify Ontario’s supply-mix and help replace the generation capacity that has been (and will be) lost as a result of the closure of Ontario’s coal-fired facilities by 2014, and thereby also reduce greenhouse gas emissions; and secondly, to stimulate local investment in the production of renewable energy generation equipment needed to design and construct qualifying generation facilities using solar PV and wind power technologies. These objectives are pursued through the execution of the FIT and microFIT Contracts, which involve an exchange of performance obligations on the part of the OPA and qualifying Suppliers. There is no inherent grant element to the FIT and microFIT transactions.
The FIT and microFIT Contracts
In essence, the FIT and microFIT Contracts envisage an exchange of the following core performance obligations between Suppliers and the OPA:
A Supplier must:
i. design, construct, own (or lease) and operate a qualifying facility in accordance with all relevant IESO Market Rules, laws and regulations;
ii. comply with the “Minimum Required Domestic Content Level” when designing and constructing a solar PV or a microFIT wind power facility;
iii. deliver the electricity that is produced into the Ontario electricity system in accordance with all relevant IESO Market Rules, laws and regulations;
iv. participate in a defined electricity payment processes to settle Contract Payments that is not unlike that used generally in Ontario’s electricity system; and
v. assign all Environmental Attributes associated with the Contract Facility to the OPA, pay the OPA 50% of all payments received by the Supplier under the “ecoENERGY for Renewable Power Program”, and effectively transfer to the OPA 80% of total net revenues from the sale of Future Contract Related Products.
In return, the OPA agrees to make the Contract Payments, which are defined in such a way that ensures each Supplier will be remunerated via defined settlement processes at the guaranteed FIT Contract Price for each kWh of Delivered Electricity for 20 years.’
The Appellate Body understood the measures’ practical effect to be:
‘An entity that enters into a FIT or microFIT Contract is required to, inter alia, build, operate, and maintain the approved generation facility in accordance with all relevant laws and regulations, and deliver the electricity produced into the Ontario electricity system. In return for performing these and other contractual obligations, such entity will be remunerated, over the term of the particular contract, in accordance with a formula that is based on a standard “Contract Price” established by the OPA.
In addition to these obligations, the FIT Programme imposes “Minimum Required Domestic Content Levels” that must be satisfied in the development and construction of solar PV electricity generation facilities participating in both streams of the FIT Programme and of windpower electricity generation facilities taking part in the FIT stream. The Minimum Required Domestic Content Levels do not apply to qualifying projects using any of the other renewable energy sources covered by the FIT Programme…’
It was thus rather straightforward for the Appellate Body to find the guaranteed payment (Feed-in Tarif) to be conditional upon the use of local content machinery producing renewable energy.
In a nutshell, Japan and EU were claiming that:
(a). The “Minimum Required Domestic Content Level” prescribed by the Ontario’s FIT regime was inconsistent with Article III.4 GATT 1994 (National Treatment Obligation), and Art. 2.1 TRIMs, if the regime was found to be a ‘trade-related measure’;
(b). Alternatively, if the FIT regime constituted a subsidy, it contravened Article 3.1(b) and 3.2 of the SCM Agreement and should be prohibited.
There was no difficult for the Panel to conclude that Ontario’s FIT program and the related contract were trade-related investment measures which accorded less favourable treatment to imported equipment (which would be unable to comply with the local content requirement). The Panel thus found — as upheld by the Appellate Body — that the measures were inconsistent with Art. III.4 GATT 1994, and Art. 2.1 TRIMs.
Canada had sought to argue that the measure was “government procurement”, to invoke Article III.8 GATT 1994 as grounds to justify violations of the national treatment obligations. This argument was rejected by the Panel and the Appellate Body, on slightly different grounds. Jurisprudence on this matter, although influential, is beyond the scope of this paper.
For Ontario’s FIT programme to constitute a subsidy, it must, as per Article 1 & 2 of the SCM Agreement, cumulatively satisfy three conditions: (a) a financial contribution by the government must (b) confer a benefit (c) to a specific recipient.
(a) Financial contribution?
As noted in Part III, both the Panel and the Appellate Body concluded that the purchase of electricity under the FIT programme was a ‘purchase of goods’ rather than a ‘grant’:
‘…[T]he Panel found that the appropriate legal characterization of the FIT Programme and Contracts is as “financial contribution[s]” in the form of government “purchases [of] goods” within the meaning of Article 1.1(a)(1)(iii) of the SCM Agreement. The Panel’s reasoning in reaching this conclusion was based on three key elements. First, it noted that the OPA transfers funds to FIT suppliers for “delivered electricity” into Ontario’s electricity grid. It is by paying a FIT Contract Price for delivered electricity that the Government of Ontario seeks to achieve the objective of securing investment in new generation facilities for the purposes of diversifying Ontario’s supply-mix. Thus, in the Panel’s view, there is no grant element inherent in the design and operation of the FIT Programme. The Panel highlighted that, while FIT and microFIT Contracts facilitate suppliers’ search for project financing, it would be wrong to characterize the Contract Payments themselves as finance payments for the construction of a generation facility.
Second, the Panel found that the Government of Ontario takes possession over electricity and thus “purchases electricity”. The Panel found that government “purchases [of] goods” will arise under the terms of Article 1.1(a)(1)(iii) of the SCM Agreement when a “government” or “public body” obtains possession (including in the form of an entitlement) over a good by making a payment of some kind (monetary or otherwise). In particular, given the specific characteristics of electricity, the Panel preferred to characterize a purchase of electricity as involving the transfer of an entitlement to electricity, rather than the taking of physical possession of the electricity. Moreover, the Panel rejected the European Union’s argument that the notion of government “purchases [of] goods” implies that the government is the entity being supplied with something for its use…’
By categorising the government payments as “purchase of goods” rather than ‘grant’, the Panel and the Appellate Body sought themselves some discretionary grounds to evaluate the “benefit”. This is because in the case of ‘grants’, benefit is normally assumed to exist as grants are not readily available in the market.
As foreshadowed above, the issue of benefit in the context of government purchasing electricity was particularly controversial. The Panel held the evidence provided was insufficient for it to conclude that a benefit had indeed been conferred. It observed that:
‘…[T]o determine whether the challenged measures confer a benefit within the meaning of Article 1.1(b) of the SCM Agreement would involve testing them against the types of arm’s length purchase transactions that would exist in a wholesale electricity market whose broad parameters are defined by the Government of Ontario. In the present set of circumstances, this could be done by comparing the terms and conditions of the challenged FIT and microFIT Contracts with the terms and conditions that would be offered by commercial distributors of electricity acting under a government-imposed obligation to acquire electricity from generators operating solar PV and windpower plants of a comparable scale to those functioning under the FIT Programme.’
One member of the Panel, however, issued a dissenting opinion and held that a benefit had indeed been conferred: ‘Facilitating the entry of certain technologies into the market …by way of a financial contribution can itself be considered to confer a benefit.’ 
The Appellate Body sought to reconcile the Panel members’ conflicting findings by further breaking up the ‘market benchmark’ test. It started by defining the relevant product market as ‘it is a necessary pre-condition’ in order to decide whether a benefit had been conferred to a specific recipient. It was found there existed two separate markets: one for conventional energy and the other for renewable energy. To reach this conclusion, the Appellate Body dismissed the Panel’s formulation of the ‘market’, arguing that the substitutability analysis requires consideration of consumers’ perspective (demand-side substitutability) and producers’ perspective (supply-side substitutability). The Appellate Body sought its previous analysis in EC and certain member States – Large Civil Aircraft as support:
‘Demand-side substitutability – that is, when two products are considered substitutable by consumers – is an indispensable, but not the only relevant, criterion to consider when assessing whether two products are in a single market. Rather, a consideration of substitutability on the supply-side may also be required. For example, evidence on whether a supplier can switch its production at limited or prohibitive cost from one product to another in a short period of time may also inform the question of whether two products are in a single market’
Thus, by emphasising supply-side factors that argued in favour of separating renewable energy market from conventional energy market (i.e. the Ontario government’s regulation and administrative guidelines), the Appellate Body concluded ‘the benefit comparison under Article 1.1(b) should not be conducted within the competitive wholesale electricity market as a whole, but within competitive markets for wind- and solar PV-generated electricity, which are created by the government definition of the energy supply-mix.’
The Appellate Body then turned to identify the ‘appropriate benchmark for the benefit comparison under Article 1.1(b) of the SCM Agreement.’ It was observed:
‘…[W]hile introducing legitimate policy considerations into the determination of benefit cannot be reconciled with Article 1.1(b) of the SCM Agreement, we do not think that a market-based approach to benefit benchmarks excludes taking into account situations where governments intervene to create markets that would otherwise not exist… By regulating the quantity and the type of electricity that is supplied through the network (base-load, intermediate-load, or peak-load) and the timing of such supply, governments ensure that there is a continuous supply-demand balance between generators and consumers, thus avoiding imbalances that would destabilize the network and cause interruptions of power supply. Although this type of intervention has an effect on market prices, as opposed to a situation where prices are determined by unconstrained forces of supply and demand, it does not exclude per se treating the resulting prices as market prices for the purposes of a benefit analysis under Article 1.1(b) of the SCM Agreement. In fact, in the absence of such government intervention, there could not be a market with a constant and reliable supply of electricity.
Nevertheless, a distinction should be drawn between … government interventions that create markets that would otherwise not exist and … other types of government interventions in support of certain players in markets that already exist, or to correct market distortions therein. Where a government creates a market, it cannot be said that the government intervention distorts the market, as there would not be a market if the government had not created it. While the creation of markets by a government does not in and of itself give rise to subsidies within the meaning of the SCM Agreement, government interventions in existing markets may amount to subsidies when they take the form of a financial contribution, or income or price support, and confer a benefit to specific enterprises or industries.’
The Appellate Body seemed to suggest in the event certain government initiative creates a new market, there would hardly be any benefit conferred (thus no subsidy) given the absence of a prior ‘market benchmark’ to compare with. This circular reasoning lacked cogency: the Appellate Body’s two-step market analysis could be boiled down into one line of reasoning — when the government defines the energy supply mix, such definition and regulatory regime creates a brand new ‘competitive market’ for products fit into the regime (i.e. wind- and solar-produced electricity), and the benefit conferred to these products shall be evaluated against the market value available to other products fitting the regime which are receiving the same consideration from the government.
In the present case, noting that the Panel had failed to accurately define the ‘relevant market’, the Appellate Body refrained from concluding its benefit analysis: ‘in the light of the complexity of the issues and in the absence of full exploration … before the panel … we cannot determine whether the challenged measures confer a benefit within the meaning of Article 1.1(b) of the SCM Agreement.’
As Appellate Body could not find that a benefit had been conferred, it did not examine whether the FIT programme would satisfy the specificity-requirement (Article 2 of SCM Agreement). It is worth noting however, had the Appellate Body followed the dissenting opinion and found benefit conferred, the FIT programme would be deemed specific (Article 2.3) as it fell squarely under Article 3.1(b) as ‘prohibited subsidies’ (for the incorporated LCRs).
The ‘new market vs existing market’ analysis has perplexed many. This novel test was disparaged by Member States immediately: in the DSB debate surrounding the adoption of the reports, the US government expressed concerns about the Appellate Body’s approach, stating that it was not clear how the new definition of the relevant market ‘would continue to maintain the SCM Agreement disciplines on a wide range of potential subsidies’.
Although some commentator welcomed the relaxed principle as ‘adapting the current rules to serious challenges arising from global climate change’, by ‘carving out some policy space’ for Members’ effort to create an environment-friendly market, most academics expressed concern over the unpredictability stemming from the decision: Rajib Pal warns that the decision ‘permits WTO Members to parse an existing product market into separate markets defined according to production technology, and provide support to chosen higher cost producers … without fear of repercussions under the SCM Agreement,’ and that it ‘potentially removes a wide range of trade-distorting government support programs from the scope of the SCM Agreement.’ Rubini argues that the vagueness of the Appellate Body’s language opens the door for ‘dangerous analogic reasoning in cases to come, and not necessarily in the clean energy sector only’.
Some criticised the Appellate Body for venturing too far from its previous jurisprudence in vitiating the LCR without declaring the FIT itself a WTO violation merely to avoid confronting conflicts between trade and environmental policy. As Cosbey observed:
‘The convoluted reasoning of the Appellate Body and the Panel are, in our view, motivated by a desire to find the LCR in breach of WTO obligations, but to avoid finding the FIT itself to be a subsidy. The former result was guaranteed in any case by the finding that the LCR was a prohibited performance requirement under TRIMs. As to the latter, the acrobatics involved in both the Panel and AB reports are testimony to the challenge this entailed under existing law.’
Trade tensions between major countries are undoubtedly increasing on account of green industrial policies. Current disputes involving renewable energy subsidies are being litigated not only at the WTO but also in trade remedies proceedings before domestic administrative agencies: under the SCM Agreement, Members can elect between multilateral (i.e. through the WTO dispute settlement system) and unilateral (i.e. through domestic anti-dumping and countervailing duty investigations) actions to challenge the industrial support regime of another trading partner. This election is only available in the case of subsidized imports causing ‘material injury’ to the domestic producers of ‘like’ (or competing) products.
According to a global survey conducted by the United Nations Conference on Trade and Development (UNCTAD) ‘the number of anti-dumping (AD) and countervailing duty (CVD) cases in the renewable energy sector ‘far outnumbers the number of [renewable energy] disputes that have arrived at the WTO’. The survey recorded that between 2008 and 2014, ‘a total of 41 trade remedy investigations initiated in the renewable energy sector: 26 anti-dumping and 15 parallel countervailing duty proceedings. Almost half of these cases (18) targeted solar technology products (e.g. solar cells and modules and solar glass), whereas 7 cases involved wind technology products (i.e. wind towers) and the other 16 cases instead concerned biofuels (i.e. biodiesel and ethanol).’ UNCTAD estimates that anti- dumping and countervailing duties could translate into a global trade loss at around USD 14 billion annually, as well as obstructing renewable energy development.
While unilateral proceedings are subject to oversight by WTO rules, they have the advantage of allowing governments to take quick remedial action without waiting for costly and time-consuming multilateral review. But this development of a second channel for trade dispute litigation presents its own risks. Without a neutral multilateral body serving as an impartial adjudicator, the outcome of these administrative proceedings may be seen as politically motivated. Aggrieved parties (usually the import-competing domestic industry) will put pressure on their own government to respond in kind. This gives rise to an increased risk of an unilateral action sparking a tit-for-tat trade dispute. The Sino–American and Sino–European trade rows over solar panels provide a disturbing lens into how such disputes can evolve into a brewing trade war.
The article attempted to find space within current WTO subsidies discipline regime for policy considerations to be evaluated as a balancing factor in adjudicating disputes involving renewable energy support measures. It is observed the text of the SCM Agreement and the formalist approach adopted by the Appellate Body prohibits such discretion. The article acknowledges the utility of reviving and extending Article 8 of the SCM Agreement, yet remains sceptical about the proposal’s prospect.