Shifting discussions to the supranational level: a narrative discourse analysis on nuclear energy sustainability and the EU Taxonomy

Abstract

Background

While discussions on nuclear energy mostly took place at the national level and the respective media arenas in the past, the questions of regulating sustainability and financing have now moved up to the supranational EU level. In this light, this article analyses the discussions and processes around the inclusion of nuclear energy into the EU Taxonomy, which aims at guiding financial means towards sustainable economic activities.

Results

The research question as to why nuclear energy has become sustainable is examined by applying a narrative discourse analysis. To account for the identified challenges for media-based discourse analysis regarding nuclear energy, an analysis was conducted alongside key events relying on media articles and expert interviews, following the institutional structure of the implementation process. The article contributes to the limited body of literature on sustainability regulation, particularly from a social science perspective. By delving into the specific dimensions of the regulatory process on nuclear energy, it provides a nuanced understanding that enriches existing academic discourse in this emerging field.

Conclusions

In the article, it is argued that four factors contributed mainly to the decision to label nuclear energy sustainable: the delegated act, combined with the decision to connect nuclear energy and gas (1), the publication of the report by the Joint Research Centre (2), the implicit compromise between Germany (gas) and France (nuclear energy) (3) and finally, the Russian war against Ukraine (4). Although narrative discourse analysis still offers valuable insights into ongoing policy processes at the EU level, the different implementation process in the EU by means of the delegated act shapes its explanatory power compared to the national context.

Background

The European Green Deal is the EU Commission’s response to the ongoing climate crisis. In order to further institutionalise the implications of the Paris Agreement within the European Union, the so-called EU Taxonomy promises to be a key cornerstone through which financial resources are redirected towards green and sustainable goals. In the financial sphere, corporate and retail investors are looking for investment opportunities involving some form of sustainability, while companies are adopting sustainable strategies and improving their environmental assessments and their communication regarding corporate social responsibility [1, 2]. It is paramount for the economic transformation to have reliable sources of sustainable finance and with the growing interest in this issue, the call for a clear and binding definition of green finance is attracting more attention [3,4,5]. Some government entities have therefore seen the need to start their own more serious efforts to standardise green financial activity, hoping to give the mobilisation of green financial resources a significant push through the mandatory nature of binding standards. This includes the promise to provide standards for sustainable finance through the EU Taxonomy. Furthermore, the taxonomy intends to reduce greenwashing, help decarbonise the EU economy and thus meet the 2050 net zero goal [6, 7]. Developed through a stakeholder process and overseen by the European Commission, the taxonomy promises to provide clear definitions for assessing whether an economic activity can be considered sustainable. The taxonomy will most likely have widespread economic implications not only for the economic sectors and entities within the EU, but also beyond [8]. It puts other countries under pressure to develop their own approaches if they want to have a voice in the global harmonisation processes for green finance that have already been initiated [9]. The inclusion of nuclear energy within the EU Taxonomy was determined through the use of a delegated act. Delegated acts are non-legislative instruments adopted by the European Commission, designed to amend or supplement the non-essential elements of existing legislation [10]. These acts allow the Commission to fine-tune and update legislative details without the need for a full legislative procedure, thereby ensuring flexibility and responsiveness in regulatory frameworks.

The research question of this article is: why did nuclear energy become sustainable in the EU Taxonomy during the course of the EU Taxonomy implementation process? To this end, an analysis of the narratives alongside the key events that structured the process is conducted. In this light, the following chronology of events unfolded from 2019 until 2022: the blockade of an early version of the EU Taxonomy (1), the publication of relevant research reports (2), the announcement of the inclusion of nuclear energy and gas through a delegated act (3), the outbreak of the Russian war on Ukraine (4), and finally the approval of the inclusion by the European Parliament (5). In light of this line of events, it is argued that four main contributing factors for the decision to include nuclear energy can be identified: the delegated act combined with the decision to connect nuclear energy and gas (1), the publication of the report by the Joint Research Centre (2), the implicit compromise between Germany (gas) and France (nuclear energy) (3) and finally, the Russian war against Ukraine (4). While discussions on nuclear energy mostly took place at the national level and the respective media arenas in the past, the questions of sustainability and financing have now moved up to the supranational EU level. It is further claimed that although discourse analysis still offers valuable insights into ongoing policy processes at the EU level, the different implementation processes in the EU by means of the delegated act limits its explanatory power when focusing on media analysis. In the national case, a distinct public and a clearly delineated media arena exists and functions as a direct resonance body for policymaking processes. This is only the case to a lesser extent at the EU level. Therefore, the change in structural conditions for the implementing actors compared to national policymaking calls for analytical adaptations. Furthermore, it is argued that the final decision for the inclusion has to be understood as a result of various factors which differ depending on the time period. In order to take into account the identified challenges of media-based discourse analysis on nuclear energy (no delineated media arena and distinct public on EU level), an analysis was carried out alongside key events following the implementation process. The article begins by introducing the relevant studies on nuclear energy discourses. This is followed by an introduction to the EU Taxonomy. A theory section then introduces the narrative concept for the analysis, followed by a methods section. Finally, the results are presented, followed by a discussion of the results in light of the literature and theory. The final section concludes.

Literature review on nuclear energy discourse studies

While there was a certain initial enthusiasm for the pledges of the European Green Deal, controversial discussions around issues such as the inclusion of nuclear energy in the taxonomy have emerged. One line of criticism questioned the EC’s decision to assign the task of evaluating the sustainability of nuclear energy to the JRC on the ground of its structural ties to nuclear energy deployment in Europe since the Euratom Treaty in 1957 [11]. Another strand of criticism unfolded around the significantly higher levelised costs of nuclear energy compared to renewables [12, p. 269]. While opponents thus questioned the sustainability and environmental safety of nuclear energy, proponents pointed to the low-carbon electricity generation using nuclear energy and its contribution to climate change mitigation [13]. At the same time as arguments against nuclear energy emerged in the context of Chernobyl and Fukushima, while a “nuclear renaissance” rhetoric appeared at the start of the 2000s, mostly pushed by the associated industry itself [14]. The discussion on accelerated climate change and energy transitions have fed this rhetoric further. Thus, proponents of nuclear energy often point to the contributions of nuclear energy to climate mitigation. This view is also shared by the IEA, the IAEA and the IPCC, who all have influence over global energy decisions. In addition to this, the IAEA and the IPCC engaged in a framework for assessing the sustainability of nuclear energy called INPRO [15, p. 174]. The INPRO effectively assumed prior to the assessment that nuclear energy is necessary. Therefore, the sustainability issue is reduced to best practice scenarios regarding nuclear energy systems, whereas a more comprehensive approach to assessing sustainability is still lacking [16, p. 175–178].

This article also follows in a long tradition of analytical engagement with nuclear energy from a discursive perspective. Häninnen and Yli-Kauhaluoma [17] analysed communication in the form of newsletters within the nuclear energy industry in Finland regarding the construction of “ONKALO, the world’s first repository for the final disposal of spent nuclear fuel”. They found that in order to establish successful communication the local community needs to be included and ideally involved in the process regarding a repository [17]. Bern and Winkel [16] argued that while the positions of the anti-nuclear energy movement in Germany achieved a certain degree of institutionalisation, the French movement remained politically marginalised [16, p. 285]. Thus, legitimacy for the technology only remained stable in France, where anti-nuclear energy positions could not be institutionalised. In a discourse network analysis on the reactions to the Fukushima incident in Japan and Germany, Rinscheid [18] claims that the degree of pre-crisis polarisation within a policy subsystem can serve as a better indicator of change within this subsystem than the crisis itself. This may help explain why Germany decided to phase out nuclear energy and Japan did not [19, p. 62]. Another study analysed a British environmentalist’s shift towards a pro-nuclear energy position after the Fukushima incident from a narrative discourse perspective. The authors also postulated the emergence of a pro-nuclear energy environmentalism in contrast to other fossil fuels and pointed out the connection between narrative analysis and change over time [19]. Balkan-Sahin [20] conducted a critical discourse analysis on the role of the Turkish AKP in promoting a shift towards nuclear energy. Although nuclear energy was framed as the only alternative for economic growth using discursive strategies, the author also s that non-discursive aspects of power laid the foundations for the success of nuclear energy advocation [20, p. 15].

Kratochvil and Mišík [21] conducted a media-based discourse analysis on energy supplies in the Czech Republic and Slovakia against the background of several gas supply crises. They found that nuclear energy was portrayed positively in both countries as contributing to reliable energy supply and the climate goals in being ‘green’ [21, p. 8]. A discursive media analysis in Spain by Mercado-Sáez et al. [22] showed that nuclear energy was mostly portrayed in a positive light in Spanish newspapers and especially in the conservative press, whereas most of the population opposed it. Furthermore, the press mostly covered the opinions of politicians and industry on nuclear energy rather than the positions of scientists and environmentalists. The authors conclude that politicians and industry were more likely to influence the media frame-building process than others, which resulted in nuclear energy being perceived as a clean energy source [22]. Another study based on surveys in the UK, France, Germany and Norway found “that high climate change concern is associated with negative perceptions of nuclear energy” [23, p. 20]. Furthermore, the authors argued that the narrative of ‘nuclear energy as a threat’ still received more public support than the narrative of ‘nuclear energy expansion as a contributor to climate change mitigation’ [23, p. 20]. Vossen [24] highlighted the important role of the media in explaining nuclear energy to the public and to policymakers. In her study on perceptions in the Dutch media, she found that nuclear energy was mostly still framed in terms of safety, risks and failing governance. However, she also found that recent eco-modernist perceptions portray nuclear energy as a solution for climate change mitigation that is linked to a shift in public opinion in favour of nuclear energy [24].

Another recent study analysed public comments in the nuclear energy discussion regarding its inclusion in or exclusion from the EU Taxonomy [25]. The authors linked the comments of pro-inclusion and pro-exclusion arguments to concepts of ecological modernisation, sustainability and technology neutrality. They claimed that “contributors to the discussion directly argued for the inclusion or the exclusion of atomic power in the taxonomy, attempting to influence future EU energy policies” [25, p. 9]. While a considerable number of studies exist on nuclear energy in the national context, literature on nuclear energy discussions at the EU level is still scarce. The EU Taxonomy has shifted the question of nuclear energy and its sustainability from the national level to the supranational EU level, resulting in an additional layer of discussion.

The EU Taxonomy

The “Regulation (EU) 2020/852 on the establishment of a framework to facilitate sustainable investment and amending Regulation (EU) 2019/2088” officially entered into force on 12 July 2020 [6]. This regulation, abbreviated as the EU Taxonomy, was developed in an attempt to provide a definition of whether an economic activity is environmentally sustainable or not. Furthermore, the taxonomy is intertwined with regulations on environmental disclosure by companies such as the Non-Financial Reporting Directive and the Sustainable Finance Disclosure Regulation [26].

But how is the taxonomy actually intended to work? The regulation sets out four essential criteria for economic activities, which (1) should contribute to at least one of six environmental objectives; (2) must not significantly harm any other environmental objective (DNSH); (3) should comply with minimum safeguards, and (4) should comply with technical screening criteria [6]. Under point (1) the taxonomy includes six environmental objectives, namely (a) climate change mitigation; (b) climate change adaptation; (c) the sustainable use and protection of water and marine resourcesl; (d) the transition to a circular economy; (e) pollution prevention and control, and (f) the protection and restoration of biodiversity and ecosystems [6, p. 17]. Technical screening criteria (TSC) were developed in order to assess whether an economic activity is in line with a certain environmental objective according to the taxonomy. Work on the TSCs was most advanced with regard to the first two environmental objectives, namely “climate change mitigation” and “climate change adaptation”, which entered into force on 1st January 2022. The TSCs for the other four objectives entered into force on 1st January 2023 [27, p. 20]. In essence, the TSCs will provide the detailed thresholds to determine whether an economic activity is sustainable in the sense that it “contributes substantially” to an environmental objective [28, p. 5]. Categories form another important part of the taxonomy as different technologies could be categorised as a “substantial contribution”, “transition” or “enabling” [6]. In this regard, substantial contribution means that the economic activity substantially contributes to at least one environmental objective while fulfilling the DNSH principle, the social safeguards and the specific technical screening criteria. Alternatively, an economic activity might be considered transitional if no other more sustainable, technically and economically feasible options are available in the specific sector [6]. Moreover, a transitional activity must not lead to a lock-in of high carbon assets and must not interfere with the development of more sustainable options. The inclusion of nuclear energy into the EU Taxonomy happened on those transitional terms [25, p. 2]. Finally, an activity is regarded as enabling if it is a direct pre-condition for other activities that make a substantial contribution to one of the environmental objectives [6].

While the Technical Expert Group on sustainable finance (TEG) was responsible for developing the TSCs for the first two environmental objectives, the Platform on sustainable finance (Platform) is responsible for the development of the TSCs for the remaining four objectives [28]. After being advised on the TSCs either by the TEG or the Platform, the Commission then updates the regulation through delegated acts, which will also be used in the future if the taxonomy is updated to include more stringent criteria. This makes it possible to rule out certain economic activities that were initially in line with the taxonomy [28]. In this regard, the work of the TEG and the Platform makes it possible to influence the TSCs in different ways. The current approach to evaluating the sustainability of various energy types is technology agnostic and has an initial emission threshold of 100 g CO2e/kWh [27]. This threshold is to be reviewed and reduced every five years in order to achieve net-zero CO2e by 2050 [28].¹ With regard to the thresholds in general, some argue that the TEG made the mistake of already laying the foundations for political contestation by politicising the thresholds in its reports. The thresholds were therefore nor understood solely as being bound to biophysical realities but rather were treated as objects of future political interest considerations [29].

Finally, the institutional setup of the EU Taxonomy provides the framework within the discourse around nuclear energy takes place. The European Commission is entitled to update the existing law on the EU Taxonomy at a later stage through so-called delegated acts. This has enabled it to implement the detailed TSCs [6]. These delegated acts are sub-legal processes which are intended to supplement the EU Taxonomy law and have specific rules of approval. After the EC adopts a delegated act, the European Parliament and the Council can object to this decision. The delegated act is only rejected if a majority of 20 member states vote against it and/or if the European Parliament vetoes it by a majority of its members [30]. This is especially important in contested cases such as the question of nuclear energy and gas.

Theory: Narrative Discourse Analysis

There is a rich tradition of research on environmental discourses, which exists across many forms and disciplines. This calls for a clear description of the traditions followed in this article [32]. In a narrative discourse analysis, the researcher has the role of interpreting discourses and identify structures of meaning [33, p. 189]. According to Keller [34], discourses have four distinct features that researchers share: firstly, discourse theories and discourse analyses use language and other forms of symbolic interactions that can be found in social practices. Secondly, discourse researchers acknowledge that the actual meaning of phenomena reflected through language is socially constructed. Thirdly, individual interpretations are part of a larger discourse structure. This structure can be subject to change, but is created and temporarily stabilised by the specific institutional context in which the discourse takes place. Fourthly, discourse researchers assume that the use of language follows distinct rules that can be analysed and reconstructed [34, p. 9].

Discourse analysis follows an anti-essentialist ontology with the assumptions that the meanings of certain phenomena are socially constructed and that there is not one specific ‘truth’ that exists [35, p. 176). Hajer and Versteeg [35] further distinguish between discourses and discussions: “’Discourse’ is defined here as an ensemble of ideas, concepts and categories through which meaning is given to social and physical phenomena, and which is produced and reproduced through an identifiable set of practices. The ‘discussion’ in other words, is the object of analysis; discourse analysis sets out to trace a particular linguistic regularity that can be found in discussions or debates” [35, p. 175]. As such, the practices through which a discourse takes place, differ according to the context and the topic of the specific discourse and they can be used to construct identity among the actors involved [36, 37]. According to Viehöver, narratives are part of the discursive ensemble [33, p. 178]. They can be analysed in a specific context and point in time, but also in a process [33, p. 179]. Discursive actors and especially mass media have the capacity to transform the meaning of narratives and select certain aspects or exclude others [33, p. 183]. Thus, a narrative can deviate from the previous intentions of the actors who initiated it. The analysis of mass media publications should therefore be at the centre of an analysis focusing on narratives in policy discourses [33, 38].

In light of the processual element of narratives, key events can have a decisive effect on a discourse as narratives may change, vanish or arise in consequence of an event [39]. The Russian attack on Ukraine is a good example of such an event as it changed the discourse on nuclear energy sustainability and in turn also affected the narratives in use. Therefore, a discourse analysis along events can deepen the understanding of the twists and turns of political processes. Storylines are a condensed form of narrative, for example in the form of a sentence or a catch phrase commonly used to simplify and order complex narratives. This simplification happens when there is limited time or space to set out a narrative in detail [40, p. 69]. The concept of storylines was first introduced by Davies and Harré [41] and re-used and shaped by Hajer [42]. They are ‘flags’ of identification under which different actors can rally up defined as “the ensemble of (1) a set of story-lines; (2) the actors who utter these story-lines; and (3) the practices in which this discursive activity is based” [34, p. 65]. In line with social constructionism, the assumption behind storylines is that they provide agency by simplifying complex discursive systems and the narratives within. Thus, storylines can be understood as giving condensed definitions of problems which functions as a tool to compete in discourses on a specific topic [40], p. 69]. The way people talk about certain phenomena in the form of storylines is also a reflection of the power relationships in this particular sphere [42, p. 57]. If everybody agrees implicitly through their articulations on a certain way of talking about something this narration can be considered as dominant. The term ‘discourse structuration’ applies, whereby actors follow a certain way of talking about an issue in order to retain their credibility in the discourse [42, p. 61, 43, p. 3].

Storylines can be a key element of political change and have a direct impact onto the politics if they reach a certain level of dominance in a discourse [42, p. 56, 65]. However, this does not imply that actors using similar storylines necessarily do so by sharing the same understanding of the issue. On the contrary, Hajer and Versteeg point out that the ‘effect of misunderstanding’ is a major reason why political coalitions form in the first place [35, p. 177]. Storylines can appear in various forms such as metaphors, “analogies, historical references, clichés, appeals to collective fears or senses of guilt” [42, p. 63]. Actors may also decide to challenge a certain storyline through an alternative way of talking about a particular issue. With reference to the non-existence of actors who fully understand every aspect of the acid rain issue, Hajer argues that the “political power of a text is not derived from its consistency (…) but comes from multi-interpretability” [42, p. 61]. Firstly, this means that narrative instruments such as metaphors that subsume a particular issue can provide more discursive power and impact than a detailed but long analysis of that issue. Secondly, this also means that actors can offer different and conflicting explanations of a single event. Discourses in the political sphere usually do not occur between independent individuals, but rather are embedded in an institutional structure [42, p. 57–58].

In the case of the EU Taxonomy, this allows for two basic observations. Firstly, the emergence of the Taxonomy highlights the aspect of institutional reproduction and change through actual discursive practices within the discourse on climate change and the Paris Agreement. Secondly, the process of assessing the sustainability of nuclear energy takes place within a specific institutional mode through the delegated act. Discussions between proponents and opponents of nuclear energy about the neutrality of the Joint Research Centre (JRC) with its long history in nuclear energy safety research emerged. This highlights the contributions of analysing storylines and narratives with regard to the actors who engage in discourses within an institutional structure. If a theoretical concept or an idea translates into a concrete policy, the term discourse institutionalisation applies [42, p. 61]. The analysis of storylines in the environmental sphere has reached a new degree of popularity among researchers in recent years and certain features have become established in the literature over time. One feature is the research focus on different understandings of a single term, such as biodiversity [44] or sustainability [15, 45]. Another feature already introduced by Hajer [42] includes the actual extraction of a number of different storylines.

Discourse analysis and the role of storylines are also getting more attention in the sphere of transition studies [46, p. 316]. The creation of legitimacy for a certain technology through the dominance of transition discourses is becoming important as legitimacy is a vital key to “mobilize financial, human and material resources as well as regulatory support” [47, p. 330]. With reference to nuclear energy, Markard et al. [47, p. 331] point out that the technological legitimacy has faced ups and downs throughout various time periods and regions. Bern and Winkel [16] state that the regional differences seen in discourses on nuclear energy usually depend on the degree of institutionalisation of pro- or anti-nuclear energy positions. This is illustrated in the different degree of institutionalisation of nuclear energy positions in France and Germany [16].

Methods

In terms of methodology, this article relies first and foremost on the analysis of documents, mostly media articles. The main analysis is then supported by seven expert interviews. As mentioned in the theory section, it is important to note that in a narrative discourse analysis, the researcher functions as an interpreter of the discourse and identifies structures of meaning [33, p. 189]. The methodology employed is a simplified version of the procedure developed by Hajer [40] and influenced by the application of Funcke and Ruppert-Winkel [48]. The study by Egres and Sarlós [25] also influenced the analysis in distinguishing between pro- and anti-nuclear energy contributors to the discussion. At the start of the research process, relevant documents regarding the case of nuclear energy under the EU Taxonomy were identified through web-based research in Google’s news section using the search terms “EU Taxonomy nuclear energy”. This was done repeatedly over the course of the research period from 2019 to July 2022. Additionally, government documents and reports focusing on nuclear energy under the EU Taxonomy were identified. In total, 122 documents were identified including mostly articles from various media platforms as well as policy statements, government documents, scientific publications, press releases and position papers from NGOs or companies. A list of all documents which also states the origin of each document is provided in the appendix to the article. These documents were analysed in order to gain an overview of the research field and first indications of storylines that are used. After the document analysis, an initial list of key events and connected storylines was generated and then translated into interview questions. The purpose of the expert interviews was to serve as an additional source of information to get a better understanding of the whole process. However, the media analysis alongside key events still functioned as the main source for the narrative analysis. The expert interviews were all conducted in summer and autumn 2021. The anonymity of the interviewees was assured and consent to record and use of the material was obtained prior to the interviews. The full list of interview questions and the ethical approval is provided in Appendix A. At the same time, the web-based research for documents continued. They were coded for storylines and the specific phase of the process in which the document was published. The focus was on argumentative exchange in the form of questions and debates in the context of the European Parliament and the European Commission. After the identification of the storylines in the documents, these were interpreted by putting the arguments, the sequence of key events and the actors in the discussion on including or excluding nuclear energy in or from the EU Taxonomy, respectively. An overview of the methodological steps is provided in Table 1.

Table 1 Methodological steps.

Based on the literature analysis of nuclear energy discourses and especially the recent study by Egres and Sarlos [25], it was assumed that two main narratives exist, namely a pro-nuclear energy narrative advocating for the inclusion of nuclear energy and an anti-nuclear energy narrative aiming at the exclusion from the EU Taxonomy. The main pro- and anti-nuclear energy storylines were then derived based on the key events. The documents that were analysed stem mostly from European and international sources. Those are displayed in Table 2 with the codes to show the source of origin of the documents. Furthermore, several documents on discussions within European countries were selected, mostly from Germany and France. It goes beyond this article to assess every single discussion in each EU member country.

Table 2 Codes for document and interview analysis

Results

The development of the EU Taxonomy is a very complex and detailed process and as such cannot be laid out within the confines of a scientific article. However, the process of including nuclear energy illustrates the development of a specific aspect of the taxonomy and thus sheds light on the discussions and interests behind the development of the EU Taxonomy. This chapter links the storylines to the key events along the decision-making process. Within this process, several key events structured and changed the discussion on nuclear energy and its storylines and shaped the way in which actors argue. The storylines that appeared in the beginning of the process remained present and did not vanish through the occurrence of a specific event. However, other storylines were moved to the foreground in correspondence to certain events (Table 3). In terms of the time span covered in this analysis, this article begins with the early version of the EU Taxonomy being blocked in December 2019 and ends with the vote on the matter in the European Parliament in July 2022 and the direct reactions to this. Table 3 shows the pro- and anti-nuclear energy storylines alongside key events.

Table 3 Key events and storylines.

Blocking of the early version of the EU Taxonomy

Probably one of the most important events in the early stages of the discussion was the blocking of a preliminary version of the EU Taxonomy in mid-December 2019 by nine member states (France, United Kingdom, Czech Republic, Poland, Hungary, Slovakia, Bulgaria, Romania and Slovenia) due to a planned exclusion of nuclear energy and gas (1a_MD). After intense negotiations, a compromise was reached stating that the question on nuclear energy and gas should be decided at a later stage (32_MD). This decision had a significant impact onto the publication of the final TEG report in March 2020. The compromise was then to neither include nor exclude nuclear energy from the taxonomy at this stage but rather to evaluate the contested issue based on its contribution to one environmental objective and the DNSH principle (1a_MD). The EC originally assigned the task of evaluating taxonomy alignment of the first two environmental objectives to the Technical Expert Group on sustainable finance (TEG) established in July 2018. In the technical annex to the final report, the TEG stated that nuclear energy could contribute to climate mitigation. However, the TEG also stated that nuclear energy could not fulfil the DNSH requirements, especially due to the lack of a “viable, safe and long-term underground repository” for nuclear energy waste anywhere in the world [31, p. 210]. The TEG thus concluded that they could not recommend the inclusion of nuclear energy at this stage and that more technical work on the issue was required [31, p. 211]. Furthermore, the TEG did not evaluate the whole life cycle of nuclear energy and limited itself to the aspect of carbon intensity during the energy generation. This was because the TEG realised early on that the question of gas and nuclear energy in the EU Taxonomy would become very politicised. They therefore refrained from evaluating this question in detail and instead focused on other, less controversial aspects (122_IN_GV). Despite the concern about its looming political nature, the TEG report fuelled the debate regarding nuclear energy as the technical annex argued that more research with regard to the DNSH and especially on the fifth environmental objective (circular economy) was required [31, p. 208]. Moreover, the technical annex mentioned that in the feedback round for the TEG report, some stakeholders expressed their desire for greater acknowledgement and inclusion of nuclear energy [31, p. 208]. As the discussion on nuclear energy intensified, the task to evaluate the role of nuclear energy and gas was taken away from the TEG. In September 2020, the EC then decided to assign the JRC to evaluating whether nuclear energy was sustainable enough to be taxonomy-aligned [27, p. 13].

Regarding the general sustainability discussion, one major storyline in favour of nuclear energy is its positive contribution to climate mitigation as a low-carbon energy source and the inability to reach the climate targets without nuclear energy (26_FO). In terms of sustainable finance regulation, this argument further states that finance for nuclear energy also needs to be provided in a reliable manner also by private investors (26_FO). Even in these early stages of the discourse this perspective was challenged by opponents of nuclear energy arguing that, from a life cycle perspective, nuclear energy was not carbon-free (11_AN, p. 18). Furthermore, it was pointed out that the nuclear energy waste issue still has not been solved and there is no long-term depository (56_AN). This perspective was challenged by proponents who claimed that the nuclear energy waste sector is highly regulated and well-managed (3_PN; 124_IN_PN).

Furthermore, proponents of nuclear energy argued that new technologies such as transmutation and the widespread deployment of small modular reactors (SMRs) could provide a clean and safe energy source (121_IN_SC). These technologies promise solutions to several problems in the nuclear energy sector: for example, the issue of nuclear energy waste (transmutation) and the high costs due to the current inability to construct blueprinted nuclear energy power plants on an industrial scale (121_IN_SC). One of the main arguments against inclusion is safety problems regarding radioactivity and the possibility for a serious incident (11_AN). The second and closely related argument evolves around the problem of processing nuclear energy waste, as no long-term solution is in place anywhere in the world at this stage (22_AN). Furthermore, it is pointed out that nuclear energy is far more expensive than coal, gas, solar and wind energy (8_SC). Nuclear energy, with its long construction phases and nascent technologies such as SMRs and transmutation, has come too late to play a major part in combatting climate change as these technologies are not expected to reach maturity and industrial scalability for several decades (121_IN_SC). Another argument was made in terms of technological neutrality. Nuclear energy proponents claim that an exclusion of nuclear energy would lead to a disadvantage against other energy types and hence to market distortion (31_PN).

“In other words, all economic activities should be screened against the same set of transparent technology-neutral technical criteria and subject to the same assessment process. This is the only way to restore the credibility of the taxonomy and to make it a strong and efficient tool to achieving carbon neutrality and sustainable development in the EU and worldwide.” (14_PN).

With regard to the second process-related discussion, the focus was now on the actual credibility of the JRC to make an objective recommendation. Opponents of nuclear energy questioned that credibility:

“However, the JRC’s structural links with the Euratom Treaty, its relations with the nuclear industry and the views expressed publicly by JRC members on nuclear energy call into question the JRC’s ability to conduct an objective assessment of the sustainability of nuclear energy.” (22_AN).

On the other hand, nuclear energy proponents applauded to the decision by the EC to appoint the JRC with the evaluation. They argued that, in regard to credibility, it was imperative for the EU Taxonomy to be based on science (14_PN). It was further claimed that the only effective way to slow down climate change is to rely on both, nuclear energy and renewables while phasing out fossil fuels (123_IN_SC).

The publication of the JRC, TEG and Scheer reports

While there was little discussion of the comparatively low emissions caused by nuclear energy during the energy generation process, the main task of the JRC was to evaluate whether it is possible to handle nuclear energy waste safely. The important aspect here are the aforementioned DNSH requirements, which should ensure that the economic activity of producing energy through atomic fission does not significantly harm any environmental objective (17_MD). The JRC then published their “Technical assessment of nuclear energy with respect to the ‘do no significant harm’ criteria of Regulation (EU) 2020/852 (‘Taxonomy Regulation’)” in late March 2021 [49]. Essentially, the report concluded that nuclear energy does not do more environmental or health-related harm than other energy sources which are already taxonomy-aligned [49, p. 7].

The reactions to this report illustrate the multi-interpretability aspect of storylines as mentioned earlier in the theory section [42, p. 61]. While proponents emphasised the neutral and science-based position of the research institute (4_PN), this view was challenged by opponents who argued that the JRC could not be neutral due to its long history of nuclear energy safety research (5_MD; 23_MD). Thus, different storylines attempted to condense the narratives around nuclear energy, in order to either legitimise or undermine the JRC report. The JRC report shaped the discourse in the sense that it became more prominent and accessible through mass media. It became the main focal point for discussion and opponents of nuclear energy claimed that the decision was made due to heavy lobbying by nuclear energy proponents, emphasising that civil society organisations should have been included in the evaluation study (126_IN_AN). After the decision to assign the JRC to report on nuclear energy and the taxonomy, the storyline of science-based reasoning was also used, for example in headlines such as: “EU Returns to Science-Based Decision-Making in Landmark Nuclear Report” (25_PN). This argument was, first, intended to underline the integrity and objectivity of the JRC report and, second, suggested that debates prior to this were largely driven by emotional subjectivity rather than scientific reasoning.

Following the JRC report, two other groups were tasked with an evaluating the taxonomy alignment of nuclear energy: The Scientific Committee on Health, Environmental and Emerging Risks (SCHEER) and the Group of Experts referred to in Article 31 of the Euratom Treaty (TEG31) [50, 51]. The latter report broadly confirmed the evaluation in the JRC report and thus the taxonomy alignment of nuclear energy against one opposing opinion [51]. While the SCHEER report confirmed the results as well, it expressed doubts regarding the JRC report’s comparative approach with other energy types in terms of the DNSH principle. Instead, a more nuanced evaluation of the impacts of nuclear energy on the six environmental objectives would be needed, especially in terms of a circular economy and in light of the early stage of nuclear energy waste recycling which is not yet operative on an industrial scale at the moment [50, p. 12].

Meanwhile, at the state level, a pro-nuclear energy faction primarily involving France, Bulgaria, Cyprus, Czech Republic, Malta, Poland, Romania and Slovakia articulated their interests. They either expressed their concerns regarding the delayed decision on nuclear energy (and gas) through a public letter (63_AN) or they pointed towards the importance of nuclear energy for CO₂ neutrality in Europe (37_FO). In addition to this, a group of MEPs drafted a letter to the EC in support of its inclusion (53_PN). One group of states expressed their opposition to inclusion in a letter sent by five environment ministers from Germany, Denmark, Luxemburg, Austria and Spain (46_MD). The letter summarised various criticisms of the decision at this stage of the process. It opposed the JRC decision by arguing that “nuclear power is a high-risk technology and wind energy is not” (47_MD). This was a reaction to the JRC finding which stated that nuclear energy did not do more significant harm than other energy types included in the EU Taxonomy [49]. Furthermore, the letter argued that the decision to include nuclear energy would harm the credibility of the EU Taxonomy (46_MD). This storyline corresponds to the actual intention of the EU Taxonomy to provide guidance for private investments in sustainable activities. The argument here is that if nuclear energy (and gas) were included, investors would continue to be uncertain regarding sustainable investments, which would defeat the whole purpose of the EU Taxonomy (125_IN_SC).

The involvement of German environmental minister Svenja Schulze in this letter also shed light on the conflict between the German Federal Ministry for Economic Affairs and the Federal Ministry for the Environment regarding the role of gas in the taxonomy. This conflict was not resolved until the new election in September 2021, which resulted in the German government taking an inactive position regarding the delegated act (90_MD). Furthermore, Austria and Luxemburg announced that they would file a lawsuit with the European Court of Justice against a possible decision to include nuclear energy into the taxonomy (97_FO). Meanwhile, the French President Emmanuel Macron argued that: “nuclear is among the energy sources that “emit” the least CO2. Gas produces more CO2; nuclear is much better” (84a_MD). This reflects the French position of favouring nuclear energy over gas in the EU Taxonomy while the German position was reversed on this matter.

After the German parliamentary elections in September 2021, while the new coalition was still being formed, the acting government under Chancellor Angela Merkel refrained from taking an active stand against the inclusion of nuclear energy into the taxonomy during the European Council meeting. Instead, it advocated for the rapid approval of the new taxonomy, which included nuclear energy and gas (97_FO). At a press conference, Merkel herself claimed that it would be very complicated to take action against the delegated act as it required a majority of 20 out of the 27 member states to stop such an act at the European Council level (66_GV). In the aftermath of the Council meeting, the President of the European Commission, Ursula von der Leyen announced that gas and nuclear energy were a requirement for the EU if it wanted to pursue a low-carbon economy. Moreover, the French president Emmanuel Macron advocated strongly for the use of nuclear energy and its role in climate change mitigation (66_PN). Meanwhile, the coalition talks in Germany were affected not so much by the question of what role nuclear energy should play but by the role of gas in the taxonomy. While the Green Party was against the inclusion, the Social Democrats were in favour. This resulted in an overall neutral or inactive official position on the part of the forming government regarding the delegated act, as both energy types were supposed to be included in combination (80_MD). Despite this state of abeyance, the new government’s green ministers expressed their opposition to the impending decision to include nuclear energy (88_MD). Moreover, von der Leyen took matters into her own hand regarding the delegated act, which further indicated the highly politicised nature of the decision within the EU (86_MD).

Announcement of the inclusion of nuclear energy by the European Commission

On the 1st of January 2022, the European Commission announced the Complementary Delegated Act to the EU Taxonomy, which included nuclear energy as compliant with Article 10(2) of the taxonomy regulation, thereby categorising it as a transitional activity (102_GV, p. 10). In order to justify this decision, the European Commission argued that a technical assessment of nuclear energy was carried out by the JRC which came to a positive conclusion. Therefore, a more detailed impact assessment would not be needed (100_FO, p. 5). After the complementary delegated act was announced, the German government was divided regarding the opposition to the act (97_FO). While the government spokesperson emphasised the legal capability of the EC to regulate in this regard, the Green Party ministers announced that they would join the lawsuit by Luxemburg and Austria if a further challenge to the legal aspects has a chance of success (97_FO). Moreover, through the institutional setup of combining nuclear energy and gas in one delegated act, the disagreement around the role of gas in the German government had major impacts also on the inclusion of nuclear energy (80_MD). This again demonstrated the power play behind the inclusion of nuclear energy, with Germany not opposing the act to its fullest capacity because the inclusion of gas was tied to the inclusion of nuclear energy in the EU Taxonomy (90_MD).

In terms of criticism of the decision, it was stated that climate neutrality in the EU would be under threat if nuclear energy and gas were included (94_MD). In addition to this, voices from the financial sphere were concerned because a taxonomy including gas and nuclear energy might undermine investors’ trust into the taxonomy and lead to lower financial volumes for sustainable economic activities (100_FO). A member of the Green Party in the German Parliament also argued that an exclusion from the EU Taxonomy does not stop any country from investing in nuclear energy or gas: “If France wants to build new nuclear plants, they can. If Germany wants to invest in fossil gas, it can. And I would say, even in Eastern Europe, the capital will be found (…). Interest rates are at zero for OECD countries, big companies can refinance themselves equally cheaply on the market” (80_MD).

In this light, EU Parliamentarians announced that they would veto the delegated act in the European Parliament (94_MD). Finally, the politicisation of the process was criticised and it was claimed that nation state power play would triumph over the idea of a taxonomy that reflects sustainability after biophysical realities (99_AN). Nuclear energy proponents mostly welcomed the decision, although some voices emphasised that nuclear energy should be more than a transition technology, as it is currently categorised in the taxonomy. Stricter regulations regarding nuclear energy waste and the use of ‘accident tolerant fuel’ by 2025 were also criticised (96_PN).

The outbreak of war and the vote in the European Parliament

The Russian invasion of Ukraine also had an impact on the nuclear energy discussion. The war broke out just as the EU was initiating a major energy transition towards renewables and low-carbon energy generation. The need for greater independence from Russian fossil fuel imports and energy sovereignty became immediately important (108_MD). The argument was made that nuclear energy should henceforth play an even greater role within the EU in order to ensure energy security (106_MD). However, this argument focuses mainly on the long-term perspective. Because of the long planning and construction phases for new nuclear energy power plants, the energy type cannot contribute immediately to alleviating to the current energy crunch in the EU (108_MD). Furthermore, the exposure of Ukrainian nuclear energy power plants to the actions of war shed light again on the possibility of serious incidents, a concern which was also shared by the IEA (101_FO). After the European Commission’s decision to include nuclear energy in the EU Taxonomy, the debate turned to the remaining hurdles, namely pending approval by the European Parliament and the member states. Approval by the member states is not considered to be a major hurdle as 20 of the 27 member states would have to oppose the decision. However, approval by the European Parliament is less certain as the opinions on nuclear energy vary within different factions, groups and member countries (94_MD). Furthermore, some trade unions across Europe urged the members of the European Parliament (MEPs) not to vote against the delegated act, calling it the “best compromise that could be reached” (109_PN).

On the 14th of June 2022, two European Parliament committees (Environment and Economic) voted against the delegated act thus in favour of excluding nuclear energy and gas from the taxonomy (110_GV). However, on the 6th of July 2022, in a historic vote on the delegated act the European Parliament decided not to object to the delegated act, thereby categorising nuclear energy and gas as sustainable (114_GV). The vote was preceded by an intense debate on the role of, arguing that categorising gas as sustainable would benefit Russia. The importing of uranium from Russia also played a role in the discussion. Voices on the Ukrainian side were first to express their opposition to the delegated act. However, on the 5th of July the Ukrainian Minister of Energy expressed support for the delegated act and suggested that it be implemented (113_AN). Despite many voices against the inclusion from Ukraine and elsewhere in Europe, the letter from the Ukrainian minister was arguably the decisive moment at which previously undecided MEPs took the decision not to veto the delegated act (117_MD). European Commissioner Mairead McGuinness cited the letter from the Ukrainian minister: “I strongly believe that the inclusion of gas and nuclear in the taxonomy is an important element of the energy security in Europe, especially with a view to replacing Russian gas” (117_MD). The French state-owned Électricité de France (EDF) that manages 56 nuclear energy power plants in France expressed its relief after the decision: “We welcome this decision. It is in line with the international scientific consensus” (114a_MD). They further argued that: “including nuclear in the taxonomy would help finance the company’s debt with `preferential rates` from lenders and the possibility of subscribing to green bonds” (114a_MD). This shows that the science-based and the financial argument did not vanish through the process and remained important. As a response to the inclusion of nuclear energy and gas, environmental groups along with Austria and Luxembourg challenged the decision legally (117_FO). The argument against the inclusion was mainly based on the DNSH principle emphasising that other environmental goals are significantly harmed by the unclear perspective on nuclear energy waste and the use of vast amounts of fresh water at nuclear energy power plants (115_MD). Although several court decisions are still pending, the inclusion of nuclear energy has already had an impact internationally. In the South Korean draft taxonomy, nuclear energy was initially excluded and then included again with specific reference to the decision on the EU Taxonomy (119_FO). This is just one example of the far-reaching consequences of this highly contested political decision.

Discussion

In this article, the question as to why nuclear energy became sustainable in the EU Taxonomy was examined. Thus, it is asked why the narrative that nuclear energy is sustainable became hegemonic and institutionalised. I argue that this comes down to four major factors explained in detail below. To this end, a narrative discourse analysis was conducted which aimed at the identification of storylines, key events (Fig. 1), and the institutional setup governing policymakers. In terms of methodology, the analysis in this article was mainly based on the examination of media articles. Expert interviews served as an additional source of information to get a better understanding of the process. Based on the literature analysis of nuclear energy discourses and especially the study by Egres and Sarlós [25], this research worked under the assumption that two main narratives exist: firstly, a pro-nuclear energy narrative which aimed at the inclusion of nuclear energy and an anti-nuclear energy narrative advocating for an exclusion of nuclear energy from the EU Taxonomy. While the article by Egres and Sarlós [25] focused on the verbalisation of arguments in the inclusion process of nuclear energy and analysed them according to three conceptual frameworks, this article linked those with the key events and focused on the transformation of the discussion.

Fig. 1

Own compilation

Timeline of key events.

It became clear that the question as to whether nuclear energy should be labelled sustainable was negotiated and discussed in a highly politicised context. This is shown also in a statement by the Climate Action Network Europa: “(The EU Commission) sacrifices the scientific integrity of the taxonomy on the altar of fossil gas and nuclear lobbies (and failed to) reorient financial flows towards genuinely climate-positive investments” (94_MD). The statement further shows that doubts exist whether the EU Taxonomy is able to provide financial means for the transformation. Meanwhile proponents underlined that nuclear energy should be part of the solution: "Given the scale of the challenge to transform the global generation mix, it is likely that nuclear will have to play a more important role in certain power markets, especially where there is clear and adequate government support" (43_PN).

As such, the discussion on nuclear energy and its role in the EU Taxonomy was also a highly polarised one between two main narratives, which left little room for mediating positions. In terms of discourse structuration, proponents and opponents of nuclear energy repeatedly used arguments to support or discredit nuclear energy and its future role. Key events structured and changed the discourse, which was initially dominated by arguments regarding sustainability and the use of nuclear energy for decarbonisation. The delegated act and the decision to bind gas and nuclear energy together governed the entire process. This institutional setup was identified as the first major contributing factor to the decision to label nuclear energy as sustainable. As the process intensified and the JRC produced its report, the general discussion regarding the sustainability of nuclear energy stagnated and the process-related and political discussion took precedence. This does not mean that the arguments around technical issues of sustainability vanished. Instead, it seemed to be more important at this stage to undermine the credibility of the JRC and its decision among nuclear energy opponents.

In terms of the research question, the EC’s decision to task the JRC with the sustainability evaluation and the publication of the JRC report was identified as the second major contributing factor to the final decision in the early stages of the process. The discussion became more driven by the political arguments of actors in higher positions with more visibility as soon as a decision was looming. The question as to whether nuclear energy was ultimately included could not be answered without focusing on the political interests of the EU member states, especially France and Germany. France and other likeminded European countries actively advocated for the inclusion of nuclear energy, especially after the JRC report. Germany, on the other hand chose to not to oppose this decision to its fullest capacity as German actors wanted gas to be included in the taxonomy. The decision to remain passive—apart from expressing some rhetorical opposition in statements—marked an active political stand in favour of the taxonomy in its current form.

In addition, anti-nuclear energy policymakers within the EU and especially in the European Parliament were still fond of the European Green Deal and the EU Taxonomy and had no interest in the whole policy package failing. This shows that the setup of the decision process had a great impact on the actors involved. The implicit compromise between France and Germany on nuclear energy and gas can only be understood in the light of the delegated act which tied the decision on both energy types together in one policy process. I argue that this is the third major contributing factor to the question why nuclear energy became sustainable. Finally, the Russian invasion of Ukraine marked another key shift in the discourse, with new arguments and motives emerging in favour of or against the inclusion of nuclear energy. The pro-nuclear energy narrative was enriched with storylines around the increased importance of energy security and sovereignty through nuclear energy: “The question of how nuclear power may come back onto the scene was already being discussed because of climate goals (…). Now we have the whole Russian gas question. And again, it’s an answer.” (106_MD).

In contrast, nuclear energy opponents claimed that serious incidents remained a threat especially in times of war: “We have plenty of evidence of how dangerous nuclear power can be” (106_MD). They further argued against the dependence on Russian imports, especially gas but also uranium. Finally, a letter by the Ukrainian Minister of Energy was arguably tipping the scales in favour of the inclusion. Thus, the outbreak of the war is identified as the fourth major contributing factor to the decision. After the decision in the European Parliament, fierce opposition to the inclusion of nuclear energy came from environmental organisations, as well as Luxembourg and Austria, who decided to challenge the decision in court. The original purpose of the EU Taxonomy was to unlock financial means for decarbonisation and an economic transformation towards sustainability. The discussion around the inclusion of nuclear energy showed how politically contested questions of sustainability can become when they threaten the financial basis of an entire industry: “Soon to be 100% state-controlled Électricité de France (EDF) breathed a sigh of relief on Wednesday (6 July) after EU lawmakers voted in favour of including nuclear and gas in the EU’s sustainable finance taxonomy by 328 votes to 278.” (114a_MD).

The narrative that nuclear energy is sustainable became dominant when looking at the result of the process. Thus, pro-nuclear energy storylines such as nuclear energy is climate-neutral and the waste sector is well regulated prevailed. Anti-nuclear energy storylines on the other hand, such as the lack of a long-term repository for nuclear energy waste and the threatened credibility of the EU Taxonomy through this decision could not influence the process to the same extent. However, the actual influence of those storylines on the decision is limited since the decision was mostly driven by the four major factors described earlier. In addition, most previous discourse analyses regarding nuclear energy were conducted on the national level, which leaves a gap for more research on the European level. The fact that the nuclear energy discussion and policymaking has now moved up to the EU level called for methodological adjustments. This is due to the lack of a specifically demarcated EU public with a delineated media arena. As a result, it was argued that the need to create a dominant narrative for political change is less present in the supranational sphere, especially in the specific context of the EU Taxonomy and the delegated act. However, narratives remain important in the national context and shape a country’s position at the EU level. This was shown here especially for the case of Germany.

Conclusions

This article explored why nuclear energy was included into the EU Taxonomy as a transitional activity and examined how the narrative of nuclear sustainability became dominant and institutionalised. This classification is explained by four main factors, analysed through a narrative discourse approach focusing on storylines, key events, and the institutional framework. The study relied primarily on media articles, supplemented by expert interviews, to understand the process better. The four main factors that influenced the decision to classify nuclear energy as sustainable were: the delegated act linking gas and nuclear energy, framing the entire process (1). The European Commission's decision to task the JRC with assessing sustainability, with its report being a pivotal early influence (2). The implicit compromise between France and Germany regarding nuclear energy and gas, shaped by political negotiations (3). The Russian invasion of Ukraine, which shifted the discourse to emphasise energy security and sovereignty, strengthening the pro-nuclear narrative (4).

The nuclear energy discussion around the EU Taxonomy highlights how politicised sustainability issues can become when financial interests are at stake. Pro-nuclear storylines like "nuclear energy is climate-neutral" prevailed over anti-nuclear arguments, such as waste storage concerns, but the decision was primarily shaped by institutional and geopolitical factors. Future regulatory processes must consider institutional dynamics and the potential impact of unforeseen events, as stakeholders with high stakes can influence the outcomes significantly. In terms of policy recommendations, stakeholders of future regulations processes are best advised to analyse the institutional setup closely if they want to assert their interests. The line of events showed that regardless of how regulation is designed, some unforeseen events can still change the outcome (e.g. close decision in EU Parliament). Moreover, the article has shown that actors who have a lot too loose through a regulation (e.g. EDF or French state and access to “cheap” nuclear energy finance) should be taken into account. Thus, seemingly technical process can turn into political contestations if major interests are neglected or challenged.

Future research should also explore the changing relationships between policymakers in the EU, their respective national publics and the relation to EU media discussions. This article has further shown that the analysis of discussions around sustainable finance regulation processes can offer valuable insights with regard to financial interests. It is therefore suggested to intensify this research. It would be also fruitful to further examine the role of polarisation in discussions and how it potentially limits argumentative exchange [52]. In terms of research approaches in sociotechnical transitions, the concept of sociotechnical imaginaries (STI) [53] could also be applied to the case of nuclear energy and the EU Taxonomy as was already done for the energy transitions in China and Taiwan [54, 55]. In light of sustainable and sociotechnical transitions, the role of a shift in financial means should also be further researched. As such, the case of the EU Taxonomy and nuclear energy has demonstrated that discussions can become highly political and contested when and entire industry feels threatened in its financial basis. Further research is also needed when it comes to the analysis of internal decision-making processes among key institutions in the process such as the JRC. More research is also required on discourses regarding other regulation processes and the international impact of regulations such as the EU Taxonomy on other countries.