Fighting against climate change and for fair trade: finding the EU’s interest in the solar panels dispute with China

The dispute between the EU and China regarding the trade in solar panels has been commonly explained in terms of power politics, whereby a mercantile China exploited European internal divisions to its advantage. But the trade defence case was also criticized for running against European climate policy goals. To which extent does this case illustrate a normative conflict between trade and the environment? The article replaces the dispute in the context of the trade defence procedures, according to which the EU had to decide, first, whether China’s subsidization of its PV industry was illegal, and second, whether Europe’s climate policies warranted against imposing trade defence duties. It finds that, in this case, the familiar competition between divergent European industrial interests was made worse by an important normative cleavage amongst European decision-makers, regarding the appropriate way to achieve global climate change policy goals. Simply applying the law did not settle the dispute. Instead, it plastered a political compromise emerged from a shift in the political narrative of the dispute, from emphasizing competition to emphasizing interdependence, pushing the Commission into a political compromise with China.


Introduction
Climate change has been at the forefront of the EU-China strategic partnership for many years. The dominant rhetoric displayed in bilateral meetings and joint declarations has been one of cooperation, mutual benefits and joint efforts to address this global environmental threat. Therefore, the escalation of the solar panel antidumping case into a major trade dispute, which highlighted both the possible contradiction between the promotion of strategic industrial interests and the pursuit of common global climate goals, and the inadequacy of trade law to resolve it, attracted unprecedented attention.
The dominant narrative of this dispute initially drew from neo-mercantilist interpretations, 1 where the EU and China were represented as defending incompatible economic interests by means of political pressures and threats of "trade war". Mainstream medias on both sides largely relayed this interpretation, which presented the EU and China as single entities competing for economic power globally. In Europe, they portrayed a bullish China playing divide and reign with the Member States, resulting in a European Union (EU) that was divided, and incapable of defending its commercial interests. 2 The Chinese side accused the EU of abusing trade remedies to protect the dominant position of its PV industry. 3 This article argues that the way in which economic interests played out in this dispute cannot be adequately explained with such framework, because it overlooks the internal divergence of economic interests and ignores the impact of global environmental interests. The trade defense case concerned photovoltaic (PV) cells and panels used to produce renewable energy, which is a core element of the European strategy to fight climate change. Environmental NGOs claimed that the action of the EU against cheaper Chinese solar panels in the name of fair trade ran against its global climate policy goals, echoing the concern spelled out by Naomi Klein in her book This Changes Everything: Capitalism versus the Climate that "our economic system and our planetary system are now at war." (Klein 2014) These environmental arguments, which questioned international trade norms and practices, did not side with either the European or the Chinese position. What role did they play alongside industrial interest in the development of the dispute, and to which extent does this case illustrate a wider normative conflict between trade and environmental norms?
The paper replaces the dispute in the context of trade defence procedures carried out in the EU by the European Commission, which follow specific administrative rules partly derives from the global trade regime norms binding on the EU and China under the World Trade Organization (WTO). However, at the time of the dispute these global norms were contested, globally and within the EU. More precisely, the trade defence investigations stumbled on two key normative issues: what does the global trade regime allow governments to do to steer the "low carbon economy?" And does the fight against climate change override the need for trade protection if it would come at the expense of achieving climate policy goals? The solar panel trade defence case turned into a wider political battle when several Member States confronted the European Commission over its proposals as to how to solve these disputes.
The core finding of the paper is that, as the dispute evolved, the legitimacy of the Commission's position on the legal case was challenged and a political compromise emerged outside of the trade defence procedures. It further argues that this political settlement was necessary to reconcile both divergent material interests and divergent norms within the EU, including those that should guide the transition towards a global, interdependent low carbon economy, where China plays a central role.
The rest of the paper develops this finding, based on a parallel analysis of the legal arguments and the broader political narratives involved in different stages of the dispute in the EU. The objective of this paper is to elucidate the evolving relationships between the divergent narratives built around industrial and environmental interests, which were also framed by the trade defence procedures. The empirical material reviewed includes legal and policy documents, news reports, industry papers, interviews as well as personal interactions with stakeholders in Brussels and Beijing in 2013 and 2014.
The first part traces the emergence of the different environmental arguments in and around the trade defence investigations, leading up to the radical change in the Commission's position in the second half of 2013. The second part analyses the Commission's handling of the environmental arguments in the trade defence case. It shows how the Commission's initial position became contested not only by Member States, but also by segments of the EU industry and civil society. The final part explores how divergent interests were finally reconciled, by showing how the initially dominant narrative of an unfair 'global green competition' between ruleabiding Europe and Mercantile China gave way to an alternative narrative of 'global green production', according to which the EU and China shared interests for better or worse. products. 4 The company that led the campaign against Chinese solar PV in the US, the German manufacturer SolarWorld, formed a coalition of 6 PV producers, 'EU-Prosun', to demand parallel protection in the EU. In response, the Alliance for Affordable Solar Energy (AFASE) was formed to oppose the tariffs. By the end of 2013 AFASE gathered over 800 European companies, some of them with ties to Chinese companies (Curran 2015; Kolk and Curran 2017). 5 These two competing coalitions represented different segments of the PV production chain. EU-Prosun represented manufacturers of solar wafers, cells and modules, while AFASE was a coalition of upstream producers (polysilicon and equipment exported to China) and downstream retailers and installers of PV projects. The European Photovoltaic Industry Association (EPIA), which had members on each side, refrained from taking side until the dispute was settled. However, by the time that the Commission had to decide whether or not to renew its tariff policy in 2015, it did take more clearly position against the tariffs that, according to its judgement, hurt the development of the PV industry in Europe (EPIA 2015). Besides industrial interests, the political actors involved in the dispute were also divided. According to the EU's trade defence rules, the European Commission, as the executive agency of the EU Common Trade Policy, was in charge of conducting the investigations and propose remedies. But then, it had to convince the Council of Ministers representing the EU Member States to adopt the final AD and CVD remedies. 6 As often in TDI cases, the Member States were divided internally and vis a vis the Commission's position. Whether contingent or ideologically motivated (Evenett and Vermulst 2005;Nordström 2011;Bollen 2016), 7 the divisions sparked by the Solar Panel dispute were very salient and critically observed by the Chinese. 8 Some governments opposed the tariffs from the outset. Importantly, Germany's Chancellor Angela Merkel announced in August 2012 that she wanted to solve the dispute politically, even though the main claimant Solarworld was a German company. 9 The Swedish and UK governments also publicly opposed the tariffs. 4 In October 2012, the US Commerce Department set AD ranging from 18.32 to 249.96% on solar cells and modules from China. By comparison, the U.S. case was much smaller than the EU: it covered imports estimated $2.1 billion from China and $513.5 million from Taiwan. 5 Curran and Kolk have analysed the positions and ideological arguments of both coalitions in great details. In Curran's investigations, she also found that Trina Solar, China's most powerful PV producer, was the spokesperson for AFASE campaign and also its main financial supporter. Some could read this as influence, while others would see that it reflects the logical connection between the European companies and their Chinese supplier. 6 This situation changes with the Lisbon Treaty (2009). The Member States no longer vote in the Council. Instead, they give a binding opinion to the Commission through an Examination Committee voting at qualified majority. However, these treaty changes were implemented only after the so-called Omnibus Regulation was adopted on 12 December 2013. Since the Solar Panel case was launched in September 2012, the old procedure still applied and the Council Members had to vote. 7 Bollen demonstrates that there are major empirical and theoretical gaps in the research on how member states' preferences influence TDI cases. 8 The newspaper Fenghuang Jingjinwang represents the mainstream Chinese view that the tariffs would hit the PV industry, PV costumers and threaten EU's energy security and energy strategy (2012). See also the academic discussion hosted by the European Studies Institute of the Chinese Academy of Social Science (CASS 2013). 9 Germany was also the main exporter of polysilicon and manufacturing equipment to China's PV manufacturers.
However, France, Spain, Italy and Portugal initially favoured protection. At the same time, the European Parliament, which only had a consultative role, held a couple of debates in which Members (MEPs) called for a strong defence of the EU's PV industry. 10 For the Chinese, these divisions demonstrated the partiality of trade defence instruments lending support for the Chinese government's stern opposition to the tariffs. 11 From the point of view of the Commission, the involvement of the Member States was perceived as a weakness just awaiting Beijing's exploitation. In response, it portrayed its role as 'to ensure independence and work on behalf of everyone across Europe by seeing the "big picture"' (De Gucht 2013b). However, very different perspectives prevailed in many national capitals, where, as a commentator put it "the geopolitical relationship [with China] has become almost too important to national capitals to entrust to the commission" (Chaffin 2013).

Tracing environmental arguments in the dispute
The strategic importance of China and divergent economic interests were not the only reasons for which the Commission's claim to act on EU's behalf was challenged. From the outset, the Commission's action was contested because the tariffs were aimed at increasing the price of PV panels on the EU market, at a time when a core objective of climate policy was to make them competitive against fossil fuels and affordable to a majority of citizens. Indeed, the probe came against a backdrop of growing concerns about trade rules becoming barriers to meaningful action to combat climate change (Klein 2014;Lewis 2014). The unease in the Commission was evidenced by its unusual public statement within days after launching the probe, where it claimed that 'the EU's investigations into solar panel imports from China does not harm Europe's climate goals' (EC 2012b). This section provides an overview of how the dispute unfolded, and the main steps are summarized in Table 1.
In reaction, the Chinese government immediately announced a probe into European (mainly German) imports of polysilicon, which it used to manufacture the contested cells and panels. 12 Soon, it also brought an action against Italian and Greek renewable feed-in-tariff policies at the WTO. These actions significantly broadened the scope of the trade dispute for EU member states, while also pressing the EU on the issue at the core of the dispute: the subsidization of low carbon technologies development by the state (see below). 13 The issue of the sustainability of the energy transition also had a crucial employment dimension. Both the EU and China have repeatedly emphasized that the Energy transition is about growth and jobs. Yet, while EU-Prosun claimed that     The position of the Commission became increasingly contested within Europe, notably by a group of environmental NGOs, who denounced this 'rude and fully undeserved action' as 'the last thing that Europe needed to address climate change' (WWF et al. 2013). But the tug of war nonetheless kept escalating. Trade Commissioner Karel De Gucht announced his intention to launch and ex officio probe (without official industry complaint) into Chinese telecom services (De Gucht 2013a) and attempts to negotiate by a delegation of Chinese industries in May, ahead of the release of the Commission's AD investigations findings, were rejected. 14 When the Commission announced in June that it had found ample evidence of dumping from the Chinese exporters and proposed to impose provisional duties between 37.3 and 67.9%, only 4 Member States voted in favour, while 18 voted against and five abstained (Choudhury 2013; Shi 2013).
However, Trade Commissioner De Gucht, to show his determination, decided to use his legal prerogative and impose the provisional duties. 15 Some governments reacted strongly against this unilateral action, even though De Gucht claimed that he was acting to protect the Member States who feared retaliation from China (Chaffin  (Martin 2013). By mid-July, even more Member States seemed inclined to vote against the final measures scheduled in December. 19 Meanwhile, the Commission granted a temporary discount at 11.8%, until the official publication of AD duties on 6 August, to help facilitate the negotiations of a deal with the Chinese PV industries. This decision was immediately challenged (though unsuccessfully) by EU-Prosun, who denounced a "manifest error of assessment of the facts" by the Commission and a "serious violation of its duty of care and good administration". 20 The de-escalation process started with the negotiation of an amiable "price undertaking" 21 following two weeks of negotiations in Beijing in July 2013. The negotiated settlement received almost unanimous support by the Member States, even though the minimum import price agreed was substantially lower than what the Commission initially proposed (Solar Server 2013). 22 The Chinese accepted the deal, instead of gambling on a better outcome when the Member States' final vote came in December. 23 Commissioner De Gucht justified the decision in the light of the global nature of the PV industry (De Gucht 2013c). But EU-Prosun was enraged by what it denounced as a "dumped price" (EU-Prosun 2013). It brought a series of (thus far unsuccessful) judicial actions, denouncing a lack of due process and 16 The letter was reported by the specialized PV magazine and the Chinese media, such as the Nanfang Ribao. 17 Mofcom Notice No. 2013/36 and 2013/37. Rumors also reported lobbying actions towards smaller Central and Eastern Europe Member States (Interview, 2 December 2014). 18 According to Mofcom's Bai Ming, Premier Li Keqiang was determined to do everything possible to avoid the tariffs and get a win-win deal with the EU. Xinlangwang (2013). 19 Portugal, which had been a supporter of the duties, was reverting its position, Interview, Brussels. 24 September 2013. 20 EU-Prosun action was dismissed by the General Court (case T-393/13) also on appeal (case C-312/15 P). 21 Price undertakings are minimum import conditions that Chinese exporters agreed to abide by when selling their products to the EU. 22 The minimum import price was set at €0, 56/WP for modules and 0.29€/WP for cells. According EU-Prosun, this required "almost no concession" from the Chinese manufacturers, even with import quotas of 7 GW and 2.3 GW per year, respectively (roughly 60-70% of the expected EU market in 2013). 23 Interview Chinese Embassy. 27.11.2014. transparency in the negotiation of the price undertaking and the fact that it was, according to them, "manifestly inadequate" to remove the injury to EU producers. 24 The political 'détente' nevertheless continued throughout the autumn. The Commission decided to follow the usual practice under the 'lesser duty rule' 25 of subsuming CVD tariffs duties under the price agreement, even though Commissioner De Gucht would have preferred to forego this practice in cases involving what he called 'capitalist states' like China (De Gucht 2013d). The EU-China annual summit on 19-21 November, themed "Green Growth", celebrated the diplomatic victory, resumed bilateral high-level trade dialogue, which had been suspended for 3 years, and even launched the negotiations of a bilateral investment treaty (Xinhua 2013a, b; Van Rompuy 2013). Finally, the official adoption of AD and CVD measures in December eliminated wafers from the scope of the probe and reduced the length of the measures to 2 years instead of the normal five, because of the 'volatility in the sector' (the very rapid decrease in global PV prices). 26 The Commission abandoned the threatened probe into Chinese telecoms (EC 2014), while the Chinese probes into German polysilicon and French wine rapidly found settlements. 27 Beijing did not press further on its WTO actions against European feed-in-tariffs policies and quietly accepted the EU imposition of double AD and AS tariffs on its exports of solar PV glass.
In short, although direct actions by the Chinese government did enrage some European governments, overall it fails to fully explain why Europeans were so publicly divided and why the Commission changed its position so radically. Disagreements on the impact of AD and CVD on Europe's green growth goals, including whether fighting China was the appropriate strategy to achieve them, contributed to 'outflank' the trade Commissioner (Chaffin 2013). This sets the solar panels dispute apart from the many other trade disputes with China. 28 3 The demise of trade rules as the arbiter of European interests in the solar PV case The EU's trade defence cases against China have been amongst the most challenging and controversial, due to the application of the "non-market economy status" rules (Snyder 2001;Vermust and Gatta 2012). This section examines how, in 24 EU-Prosun action against the Price Taking Agreement itself was considered inadmissible (Case T-507-13, C-142/15P). EU-Prosun then requested the annulation of the articles in the final AD and CVD regulations referring to it, but were dismissed on ground of non-severability (cases T-141/14 and T-142/ 14) Their appeal is still pending (cases C-204/16 P and C-205/16 P). 25 The Lesser Duty rule refers to a practice of the EC to lower the total remedy to a level sufficient to remove the injury, when it appears to be lower than the dumping or subsidy margin. 26 Solar wafers, which are slices of crystalline silicon, were ultimately withdrawn on the ground that it was a different product from cells and modules, which can produce electricity. 27  the solar panel case, trade defence procedures became additionally politicized by environmental arguments, and how this contributed to undermining the Commission's legitimacy as the institution responsible for applying trade rules fairly and to Europe's best interest. The arguments are summarized in Table 2.

Determining the fair level of government intervention in the green economy
The central claim made by EU-Prosun was that Chinese PV manufacturers could dump their products on the EU market because of the aggressive industrial policies of the Chinese government (EU-Prosun 2012). This issue was central to the whole dispute. Yet, the fact that government support for renewables could be considered illegitimate under international trade rules was itself controversial. Moreover, the action seemed inconsistent with the praise that China received in climate circles for its unprecedented and unmatched efforts to build a renewable energy capacity and for lowering the costs of these technologies. Yet, as mentioned above, the dispute broke against the background of an important legal controversy about the incompatibility of support programs for renewable energy with global trade law, which seemed at odds with the claims by the self-same governments that they were committed to combat climate change (Wilke 2011;Lewis 2014). Different types of subsidies are those given to producers and those given to consumers. One major issue brought to the WTO concerned the help provided to consumers through "feed-in-tariffs". But the DSB Appellate Body's first ruling in the pioneer Canada Renewable Energy 29 case (2013) only decidedly concluded on the illegality of "local content requirements" in Canada's Feed-in-Tariff, but declined to conclude on whether Feed-in-Tariffs schemes actually qualify as subsidy.
However, the subsidies provided to producers were even more likely to be considered countervailable under WTO law and pose a challenge from the point of view of the promotion of environmental goods (Rubini 2011). A nuanced view on this problem was that It would be clearly inacceptable if one country was proven to benefit economically at the expense of others from industrial policies that violate WTO rules. But it would similarly be unacceptable if current WTO rules prevented the kind of government intervention needed to efficiently fight against climate change, bearing in mind the need for urgent action (Guérin and Schiavo 2011).
And yet, in the solar panel case, the Commission examined the situation of China's PV industry without consideration for their environmental value. It found that 'the PV industry in the PRC receives preferential treatment 30 ' amounting to an average aggregate subsidy margins of 6.4%, by applying a similar methodology as  in the 2011 Coated Fine Paper Case, 31 its first concurrent AD and AS investigations against China. A crucial elements of AS cases in general is the necessity to establish the specificity of the targeted measures for them to give rise to lawful countermeasures. The Agreement on Subsidies and Countervailing Measures (SCM Agreement) specifies that only measures that specifically benefit to a certain enterprise, group of enterprises, industry or specific territory can give rise to legal protection if they cause arm to foreign markets.
In the solar panel case, the Commission applied the same broad interpretation of specificity it had adopted in the in the 2011 Coated Fine Paper Case. More precisely, only one of the subsidies investigated by the Commission was specifically targeted at the solar PV industry: the 'Golden Sun Demonstration Programme' (金 太阳示范工程, jintaiyang shifan gongcheng), which it considered a de facto (though negligible) subsidy, because in some cases it found evidence that solar demonstration projects' grants had been paid directly to the PV manufacturers instead of the solar power producers. 32 The rest of the subsidies reviewed by the Commission could have applied similarly to a range of other industrial sectors targeted for development in Chinese policy documents. They included, for instance, preferential loans afforded by Chinese state-owned commercial and policy banks to the category of 'encouraged' industries' 33 ; preferential export credit insurances afforded to 'certain enterprises', as well as various tax breaks granted to 'foreign investment firms' and 'New-Technology Enterprises'; dodgy land acquisition deals were also included, even though the evidence and methodology to quantify them was contested. 34 This interpretation by the Commission was in line with that adopted by the United States in several anti-dumping cases against Chinese imports, which was validated by the WTO panel and Appellate Body ruling in the 2011 case US-Anti-Dumping and Countervailing Duties (China), 35 but this interpretation has been criticized by trade lawyers who wondered what then would not be construed as an actionable subsidy (Vermust and Gatta 2012 34 To calculate the subsidy, the Commission used Taiwan's land prices. For loans, it used a "constructed price". Both techniques were permitted by the TDI regulations and the WTO because of China's nonmarket economy status, which is due to expire in December 2016. The use of these methodologies was nonetheless contested by Chinese producers (cases T-157/14, T-160/14, T-161/14, T-162/14, and T-163/ 14), who believed that it inflates the dumping margin. Interview Chinese Embassy 27.11.2014. 35 The Appellate Body upheld the Panel's findings that the fact that large segments of industry were broadly targeted by Chinese industrial policy documents to receive preferential loans, or the fact that certain economic zones received preferential conditions, or that land rights provisions, were specific in the meaning of article 2.1 of the SCM agreement. WT/DS379/AB/R para 401, 414, 424. issues, such as the dumping of steel or the refusal to grant China Market Economy Status, the Member States were unanimous in resisting to Chinese pressures. But in the solar panel case, the environmental dimension of the goods at stake undermined this legal approach. Hence, the environmental lobby worried that if the EU, the US and potentially others applied such broad definition of subsidization to call into question each other's renewable policies, it would undermine the efforts deemed necessary to support the deployment of renewable energy, and foreclose the kind of policy measures that some pioneer European governments had used in the past and which included not only Feed-in-tariffs programs, but also widespread tax and regulatory support measures (Rubini 2011;Lewis 2014;Kolk and Curran 2017). At that time, anti-subsidy cases were very rare and nobody had thought to contest the legality of these policies, even as the companies that benefited from them expanded globally. Grau et al. (2011) and Dunford et al. (2013) for instance demonstrated how German industrial policies, using public grants and preferential treatment, had supported the expansion of the pioneer German PV manufacturing in the 1990s and 2000s. Ironically, the beneficiaries included Solar AG, which later became Solarworld, the main plaintiff in the EU-China solar dispute. But whereas 'a cheeky Chinese policymaker could make much of European state intervention' (Evenett 2013), the adequacy of the policies that EU member states could adopt to support the growth of the renewable energy sector was at the core of EU's internal debates on how to implement EU renewable energy targets, which were ongoing at the time of the dispute (Buchan and Keay 2015; Goron and Freeman 2017). The solar panel case notably resonated with the worries expressed by European environmental NGOs' in that context, that a strict application of the EU's liberal State Aid rules (which are stricter than WTO rules to avoid distortions on the

Trade defence, climate change policy and 'the union interest'
Whereas the Commission's findings on Chinese subsidies were in line with the WTO ruling, but debatable with regards global Climate Change policy goals, they had no direct implications for the EU's domestic PV market. However, AFASE also claimed that tariffs would significantly and adversely impact the development of Europe's PV market, and thereby impede on its ability to achieve its climate policy targets. Their core argument was that the tariffs would increase the price of PV cells and panels in the EU market and decrease the demand from costumers already weakened by the withdrawal of support schemes in several EU countries (notably in Spain, Buchan and Keay 2015). According to them increasing the cost of renewables on the EU market contradicted EU's climate change policy objectives. The inability of the anti-dumping proceedings to address these concerns adequately probably more directly contributed to the politicization of the case and the necessity to settle it through negotiation.
AFASE's claim was addressed under a unique provision in the EU trade defence regulations, according to which the Commission must verify that the imposition of AD or AS measures would not run against the EU 'general interest'. 37 As the Commission enjoys wide administrative discretion in interpreting this provision, which does not exist in WTO trade law, the different parties strived to convince the Commission to share their definition of what constituted this 'Union Interest'.
The argument of AFASE was that the 'Union Interest' should include climate policies. The Swedish National Board of Trade shared this interpretation, which it argued stemmed from the Lisbon treaty provisions, according to which the EU's common commercial policy must be conducted 'in the context of the principles and objectives of the Union's external action, which include environmental objectives (SNBT 2013). This approach, however, entailed a significant change in the Commission's classical application of this provision, which specifically excluded considering wider policy aspects (EC 2013a) and merely consisted in "balancing the economic interests of the different operators on a given market", with the presumption that only compelling reasons could justify not imposing tariffs once dumping was found (Vermust 2005).
The Commission upheld this classical approach and dismissed AFASE's interpretation. Therefore, it construed climate change policy narrowly, mainly as the interest of downstream industries and European consumers. It concluded that 'the overall positive effects [of the duties] for the union industry (solar PV manufacturers) outweigh the likely negative impact on other operators on the PV 36 The state aid rules were revised to bring more flexibility to the Member State's use of state aid "granted for environmental protection or energy objectives" (European Commission Guidelines on State aid for environmental protection and energy 2014-2020, para. 13, COM 2014/C200/01). 37 This provision is found in Article 21(1) Council Regulation No 1225/2009 (AD) and Article 31(1) Council Regulation No 597/2009 (CVD). It is an addition in EU from WTO obligations. market including end-users.' 38 More precisely, it denied that a plausible price increase for PV panels on the European market would affect demand for renewables. On the contrary, it concluded that it would be partly absorbed by operators and that the deployment of renewable energy would not be affected because (ironically considering the arguments on subsidies examined above) it relied also on policies that shielded it from fluctuations in market prices. It further argued that demand for solar energy would not be curtailed because the tariffs would improve the situation of the EU industry, who could ramp up production, lower prices and meet rising demand. 39 It must be underlined that such conclusions endorsed a contested appreciation of the development of the European solar energy market. Premised on the belief that unfair subsidies and dumping had caused EU industry's dramatic decline 40 and not, for instance, the withdrawal of support policies or problems within EU's PV manufacturing industry itself as AFASE argued, it assumed that the status quo ante would be instantly restored with the imposition of tariffs. Other segments of the PV industry and final consumers could not avail themselves of the 'Union Interest' provision to overturn the investigators' conclusion that trade defence measures were necessary to resist Chinese dumping.

Green competition versus global green production: different narratives of a climate friendly trade regime
These conclusions reached by the Commission's AD investigations could have led it to impose high-level tariffs, following the precedent set by the US and its own preference for sending a 'strong' message to China, as the imposition of interim measures despite the Council's majority opposition showed. Yet, the solution finally adopted was a price undertaking negotiated with Chinese exporters, even though this legal option was reputed for being difficult to enforce in China (Steinbach 2014). 41 This section aims to demonstrate that this compromise emerged from the fact that, in the politics leading to resolution of the dispute, the narrative of 'global green competition', with which the Commission's findings became identified, progressively lost ground in favour of a 'global green production' narrative. These alternative narratives are summarized and displayed in Table 3. 38 This argumentation can be found in the provisional AD regulation published in June 2013 (OJ L152) § 259. 39 Idem § 257-259. 40 Chinese companies challenged the Commission's weak consideration of domestic factors. 41 Steinbach shows that the EU has been reluctant to use price undertakings in cases against China because of difficulties to monitor their implementation and cooperate effectively with Chinese authorities to enforce them. The price undertaking was subsequently withdrawn for several Chinese companies who violated the agreement (see the chronology for details). A circumvention case led to the extension of the TDI measures to PVs and cells consigned in Taiwan and Malaysia in 2015.
Fighting against climate change and for fair trade: finding… 117

Green competition: zero sum game in the competitive global green economy
The scenario that dominated European discussions in September 2012 portrayed the EU and China as competitors embarked in a zero-sum game race for green markets.
In a context of souring trade relations, Members of the European Parliament for instance urged the Commission to 'send China a strong message' that it would defend its industry and would not 'let it threaten its businesses with retaliation'. 42 EU-Prosun claimed that Chinese exporters had lowered markets prices to such a level that European manufacturers could not sustain the competition. The Commission's investigations indeed showed that Chinese PV production, which was virtually inexistent in 2005, had ballooned to account for about 60% of global production in 2011. Meanwhile, it estimated that less than 10% of this production was sold in China and 70% was exported to Europe, where market demand was supported by feed-in-tariffs in several Member States. For EU-Prosun, this meant that 'European taxpayers money was used to ramp up solar panel production in China'. 43 According to the Commission's findings, between 2009 and 2011 the share of EU's market taken by Chinese companies had increased from 63 to 80% for modules and 8-25% for cells.
Another key argument of EU-Prosun was about 'jobs', a major concern for EU leaders in the aftermath of the economic crisis. It argued that 25,000 industrial highend jobs in research and skilled production were threatened as European manufacturers closed doors. Moreover, as noted by Curran (2015), even more compelling was the argument that, ultimately, what was at stake was the preservation of Europe's competitiveness in the global green economy. Without manufacturing capacity, EU-Prosun spokesperson Milan Nitzschke argued, Europe would lose its innovative capacity to China, who would enjoy a de facto global monopoly (Nitzschke 2013). The chairman of the Parliament's international trade committee relayed this view when he argued that: "the devastating Chinese competition is also affecting some of the European Union's most innovative industries, including the solar panel industry" (Moreira 2012). The 'Green competition' narrative build on the broader economic challenges that some Europeans identified with the rise of China (INTA 2012;De Gucht 2013b).
As for global climate change, this narrative emphasized that China's unfair trade practices were harmful to the global PV industry development. By 2013 the Chinese production capacity alone amounted to 150% of global consumption (EC 2013b). The sharp fall in European PV prices, from 2100 EUR/kW in 2009 to merely 764 EUR/kW in 2011, 44 was presented by EU-Prosun as a "race to death" provoked by the irresponsible behaviour of Chinese manufacturers. 45 Consequently, the EU's action was portrayed as the best strategy to rescue the global PV industry. Commissioner De Gucht affirmed that Even those arguing that cheap solar panels are good for sustainable energy and the environment must realize that you need to keep the solar panel industry 'sustainable' in the first place if you are going to see any benefits over time.

Global green production: embracing the green "world factory"
These were compelling arguments, especially since the Chinese PV industry recognized that it suffered from serious overcapacity problems domestically, which were already leading to large numbers of bankruptcies (Gao 2012;Economic Observer 2013). Nonetheless, AFASE and others thought that protectionist interests had profoundly misunderstood the structure of the emerging PV market, and offered a simplistic solution to the wrong problem in their attempt to foster Europe's

Solution to the conflict between environment and trade
The environment is best protected when the European industry is protected from "unfair" competition and an unsustainable "race to the bottom" in production costs is avoided This sentence typically illustrates what is called here a 'green global production' narrative, which argued that tariffs would only damage competitive segments of the European PV industry and curtail the potential offered by economic globalization to scale up solar energy markets in time to address climate change (Green 2011). The Swedish Board of Trade for instance argued that The EU's TDI measures on renewable energy sources are clearly acting in opposition to the goals of the EU's climate policy, which recognizes the potential in creating an open, competitive and demand-driven renewable energy sector (SNBT 2013). This argument was supported by several academic analyses besides those provided by AFASE. They first emphasized the highly-fragmented nature of the PV production chain at the global level (Grau et al. 2011;Gallagher and Zhang 2013). In this chain, moreover, they pointed out that while Chinese companies dominated the manufacturing segment, European, Japanese and American firms were still largely dominant in most other segments, including R&D and expensive manufacturing equipment (Dunford et al. 2013;Gallagher and Zhang 2013). Unsurprisingly, then, equipment manufacturers in England lobbied their MEPs and the government to voice their concerns about the duties in Brussels (Watson 2012). 46 Commissioner De Gucht acknowledged this challenge as a principal reason for the EU to accept a price undertaking agreement in August 2013 (De Gucht 2013d).
Another dimension of this situation was that, according to Gallagher, most Chinese PV technology had been purchased, in the form of licenses, from western technology providers who continued to draw down substantial dividends while maintaining their technological edge (Gallagher 2014). Furthermore, these studies stressed the extreme sensitivity of the global PV sector to both financial markets and policy signals compared with the established fossil energy industries (Curran 2015). 47 Thus, they argued that the EU tariffs would hit harder the most 'freemarket' oriented companies in China and perversely consolidate the Chinese PV industry around companies-like the bankrupt Suntech-who did benefit from obvious local political support. 48 46 MEP Graham Watson's Newsletter of 8 February 2013 elaborates on these concerns (Interview, 26.11.2014 AFASE also had strong arguments on jobs. According to the study they commissioned from Prognos, duties threatened up to 200,000 jobs in competitive segments of the industry across Europe (Prognos 2013).
At a time of declining demand for solar PVs in Europe, a simultaneous crisis of Chinese over-production (Xie 2013;Laurent 2013) and rapidly falling production costs regardless of dumping, the international renewable energy industry claimed that a climate friendly solution was not trade barriers, but rather, it was to expand markets for PV panels, especially in China and other developing countries. This position was echoed by European climate NGOs, who argued that At a time of economic crisis, concrete actions to grow EU-China low carbon markets would lower costs, create employment and generate momentum for greater mutual ambition in tackling climate change. (WWF et al. 2013).
Chinese NGOs, many of whom were supported by the EU through environmental protection cooperation projects, also argued that the dispute compromised their and also EU's efforts to promote green development in China (Chen and Bai 2013), notably by threatening 40,000 jobs there, which would strengthen the position of the powerful fossil fuel lobby and support those in China who opposed climate action as western plot to contain China's rise (Gou 2010). 49 Moreover, it was clear, by the time the dispute erupted, that the Chinese central government was already taking measures to ramp up domestic PV installation capacity. The targets were raised consecutively from 21 GW in 2011 to 35 GW in 2013 to keep up with the speed of PV projects developments (Gao 2014). 50 This rose prospects for a more balanced trade relationship and opportunities for further cooperation with EU companies. Furthermore, two new global initiatives emerged in parallel to build alternative mechanisms to manage trade frictions in the global renewable energy sector. First, the Semiconductor Equipment and Materials International (SEMI) launched an initiative in Brussels in February 2013 to establish a 'World Solar Council' (SEMI 2013). 51 Secondly, a political initiative for a 'Global Agreement on Free Trade in Environmental Goods', launched at the 2011 APEC summit, was discussed at a series of hearings in the European Parliament in the spring of 2013. By 2014 the EU, China, and many other countries were ready to start negotiating a multilateral agreement that would "eliminate tariffs for goods we all need to protect our environment and address climate change", including PV panels. Several rounds of negotiations have followed and preliminary conclusions were reached at the G20 Summit in Shanghai in July 2016, even though negotiators failed to conclude the deal in December as foreseen (Sell 2017). 52 49 These arguments prevailed amongst Chinese grassroots NGOs such as Greennovation Hub, who cooperated under EU-funded projects with the delegation in Beijing (Li. interview 21.02.2013). Antiwestern "low carbon imperialism" populist had gained a lot of credit in China in the aftermath of Copenhagen, with opuses like Gou's "Low Carbon Plot" (2010) selling volumes. 50 By 2014 the target was scaled up to 50 GW and Chinese experts discussed a 100 GW target for 2020. 51 Initiative presented in Brussels on 21 February 2013 in the presence of EU officials (Commission and European Parliament), industry representatives (SEMI, AFASE, EU-Prosun) and NGOs (WWF). 52 The "Joint Statement Regarding Trade in Environmental Goods", was signed at the World Economic Forum on 24 January 2014 in Davos. By 2016, 18 WTO members were negotiating on behalf of 46. The dispute on trade in solar PV brought the EU and China to the brink of a trade war. In Brussels, initially the dominant narrative was that the Chinese solar PV industry, with the support of 'state capitalism', was intent on eliminating competition through dumping to create a global monopoly for itself. From this perspective, trade defence was the only way to re-establish fair trade conditions for the European industry and promote a healthy development of the global PV industry. However, by the end of 2013, the political momentum for tariffs had weakened and a return to cooperative solutions was building up. The alternative narrative supporting this shift argued that Europe should instead harness China's comparative advantage to scale up PV panels production and consolidate industrial support for climate policy globally, while pushing for increased market access for European technology providers and market building policies for solar energy globally. In this view, trade protection was an outdated instrument, ill-suited to address the issues of the born global PV industry. Worse, it sent the wrong political signal for EU-China cooperation on climate change. Ultimately, the Commission was convinced to make full use of the flexibility provided by its trade defence regulations to strike a political compromise, which enabled it to resume cooperation with China (De Gucht 2013b). As shown in the chronology provided early in Table 1, and in the Annexed list of cases, the legal actions that followed the price settlement, notably the withdrawal of the price undertaking for several Chinese companies, found in violation of the agreement, remained under the political radar. Then, in a complete reversal from earlier statements, a cross-political coalition of MEPs even called on the Commission to terminate the measures which 'armed the European PV market' in October 2015 (EP 2015). In 2016, the Commission decided to prolong, but also progressively phase out the duties to reflect the sharp decrease in PV prices globally.
The solar panel dispute thus showed a specific manifestation of a normative conflict between trade law and the environment, where measures aimed at protecting industries from the adverse impact of trade liberalization were questioned on the ground that they challenged government interventions that were deemed beneficial for the development of green technologies and the promotion of a capitalist low-carbon future. In the solar panel dispute, fending off China's 'state capitalism' may have been a bigger concern for European leaders than the Climate. But they disagreed on the best strategy to achieve this goal, considering the globalized and interdependent structure of the PV industry. Such tensions might be the essence of economic globalization. But the ability of existing trade rules to manage them was called into questions and the jury is still out on whether the solution lies with a free trade agreements for environmental goods, or nongovernmental initiatives such as the 'Global Solar Council' finally launched at the COP 21 in December 2015 (Sell 2017). 53