This paper is the third of a series of 3, which aims to introduce my PhD. The goal of this research is to explain how capitalism is reproduced and transformed in climate mitigation scenarios for the EU.

In Part 1, I explain why the current organization of the world economy can be said unsustainable, and what I mean by that. Part 2 examines how the economic system is depicted in recent and popular scenarios, and concludes that almost none of them question the current economic instutions. Part 3 lists the potential explanations to this surprising state of things, which constitute some of the hypotheses for my research.

I did not use a systematic method to collect and code the publications I read. This is because the representation of economic institutions in energy and climate scenarios has very rarely been studied before, and thus no normalized vocabulary was available to create queries in research databases.

Main actors and types of scenarios studied

In previous papers, we saw that (1) it is impossible to build a sustainable society without changing the basic rules of our economic system and that (2) there are very few scenarios projecting futures with different economic institutions, and none explored by governments. This leads me to conclude that most scenarios are actually about unsustainable societies, despite the way that they are framed in public communications. I am not the first to say this, although I think I offer a more specific description of the problem than other studies I have encountered.

I am also not the first researcher trying to understand why the organization of the economy is rarely challenged in climate mitigation scenarios. To my knowledge, only 2 studies specifically address this question. Kuhnhenn (2018) describes how Integrated Assessment Models (IAMs) work and are used, and why it's hard to study degrowth with these tools. Cointe and Pottier (2023) provide a more detailed explanation of how economic growth is enshrined in IAMs and modelers' practices, based on interviews, observations and model documentation. Existing knowledge is therefore limited to global scenarios produced by the IAM community, and only adress economic growth. This is not surprising, as economic growth is the main characteristic of capitalism that is challenged in climate mitigation scenarios (Lauer et al. 2024, Lauer et al. 2025), and is the most problematic consequence of capitalist economic institutions from a sustainability perspective.

In this research, I'm more specifically looking at climate mitigation scenarios for the EU. Scenarios and models are very complex objects, and what applies to some might not hold for others. So before listing dozens of hypotheses, I think it's good to be a bit more specific about what I'm studying in this context.

Central actors and scenarios

In my understanding, 5 bodies play a central role in designing the scenarios that inform EU climate policies:

  • DG CLIMA: Directorate-General for Climate of the European Commission
  • DG ENER: Directorate-General for Energy of the European Commission
  • The Secretariat-general of the European Commission.
  • ENTSO-E: European association of electricity transmission system operators
  • ENTSOG: European association of gas transmission system operators

The scenarios used by the European Commission are contained in what they call "impact assessments". These documents are preliminary analyses, that acquired more importance in the policymaking process starting from 2015 (Pircher 2023). They aim to define and assess a range of possible policies for the future, and help choosing a policy proposal that will be discussed by the Parliament. Therefore, scenarios are used for almost every EU policy area, and can be quantified through a wide variety of calculation tools. The DGs for energy and climate therefore have units of modellers, that produce climate mitigation scenarios for different time horizons and sectors, some of which are published in impact assessments.

The Secretariat-general is the body that governs the whole Commission. It therefore does not create impact assessments, but provides the general direction which commissionners and DGs should follow. Most importantly, it publishes each year a strategic foresight report (EU 2023), aimed at anticipating future "sustainability challenges" and providing "key areas for action". This report is based on an analysis performed in collaboration with the Joint Research Center (JRC) of the Commission , which uses foresight methods to build different scenarios to sustainability (Matti et al. 2023).

ENTSO-E and ENTSOG are bodies that allow energy network operators to cooperate on the design and maintenance of the EU energy market. Every 2 years, they publish multiple studies forming their common Ten-Year Network Development Plan (TYNDP). This modelling exercise aims to anticipate which cross-border and offshore infrastructures are needed for the next 30 years, given the EU energy and climate targets. Each candidate project is assessed with a multi-indicator cost-benefit analysis, that will help the Commission decide which should receive the PCI/PMI status. This status allows projects to receive public funding and have accelerated administrative procedures.

Networks of futures and networks of organizations

The scenarios mentioned above have a strong influence on EU climate policies, but they are better understood as part of a network of futures. Here are some of the many examples of interlinkages between scenarios:

  • Every scenario in an impact assessments is compared to a "baseline scenario", which is meant to represent what would happen if no new action is taken by the EU. This scenario is based on the EU reference scenario, a report that is updated regularly (DG CLIMA et al. 2021).
  • This reference scenario is based on a meticulous work of listing all the current and future policies that would affect energy and climate issues in the EU, and assessing their consequences. To do this, analysts rely on multiple national strategies, the most important of which being the National energy and climate plans (NECP) and the Long-term strategies (LTS).
  • These NECPs and LTSs are often an adjusted version (intended for the EU) of national energy and climate strategies, which were developed through dedicated studies.
  • The TYNDP process also uses a baseline scenario, built from NECPs and National Development Plans data (NDPs, dedicated to energy networks).
  • Each quantified scenario results in a number of datasets about the future, which can be used in other studies. This can be as a baseline, to calibrate a model, provide information that the models used do not represent, etc. For example, it's common to rely on scenarios of the International Energy Agency (IEA) to get future fossil fuel prices.
  • Many scientific studies use scenarios to answer a specific research question, such as the influence of hydrogen leaks on the climate benefit of green hydrogen (Hauglustaine et al. 2022), or how much is it possible to lower final energy demand (Grubler et al. 2018). Their conclusions can be used to frame a new scenario, for example by preventing the model to exceed a critical threshold.
  • Many studies made by non-governmental actors aim to contribute to the definition of EU policies, and thus aim to influence the scenarios I mentioned above. Some of them can reach analysts of the Commission through dedicated meetings, scientific conferences, calls for evidence, etc.

This implies that while some organizations are clearly central when it comes to EU policymaking, the authority and efficacy of their scenarios partly relies on cooperations with many other actors. The organisations involved in producing climate mitigation scenarios for the EU can be represented in 3 circles.1 The core are the central actors which I mentioned above. The second circle contains those who have an administrative mandate to support the core (for example, the European Environmental Agency). The third circle are all the organizations that try to make their expertise useful to the core.

Image without description

Moreover, as we have seen in previous papers, scenarios from research also reproduce capitalism. At this point, one can already suspect that the generality of this fact stems the interdependencies between all these actors.2

In the rest of the paper, I will formulate hypotheses based on existing research. I will specify to which actors they apply when it is relevant.

Main research question

The question I am trying to answer in this PhD is therefore: Why do EU-scale climate mitigation scenarios reproduce an unsustainable economic organization?

Asking this question is assuming that there would be reasons why this economic organization would not be reproduced in scenarios. In fact, a scientific reasoning is not sufficient to change society, coalitions of actors need to be assembled around it before it becomes a social force. Here are some reasons that led me to think that could happen:

  • The unsustainability of economic growth is extremely well documented, and this research was carried out mainly in economics. A large part of the analysts working on scenarios at the European Commission, other EU agencies, and research institutes, are from this discipline. There is therefore a good chance that they have heard of this research.
  • EU administrations frequently assert that their policymaking is and should be based on science,3 and expertise plays a very important role in decision-making (Saurugger 2002). Of course, that argument only holds if you think narratives on green growth are to be taken literally,4 and are not a publicly acceptable way to defend specific interests (Peters and Nagel 2020).
  • The Commission funds a dozen of large research projects on degrowth. In addition, the European Research Council granted 10 million euros to the project A Post-Growth Deal, which aims to create degrowth and decolonial mitigation scenarios.
  • The inhabitants of EU countries are generally open to ambitious sufficiency and regulatory measures. Lage et al. (2023) show a strong divergence between national plans and the recommendations of citizen assemblies on sufficiency and regulation. Implementing these measures would not necessarily reduce the political capital of politicians.
  • Researchers, economists and engineers who work on degrowth and sufficiency are today able to produce their own models and scenarios. Some are quite active in advocacy with EU authorities, as shown by the Beyond Growth conference (European Parliament 2023), articles and open letters (Hickel et al. 2021, Parrique et al. 2023a, Parrique et al. 2023b, EC 2023a).

The last point illustrate why it is important for my research to follow these actors, in order to understand how their point of view is marginalized.

Preliminary hypotheses

In this section, I list possible explanations for why EU-scale climate mitigation scenarios reproduce an unsustainable economic organization. I currently know a lot more about this question thanks to my fieldwork, but this is only accounting for previous research. As I mentioned, research on this exact topic is very scarce, so a lot of these hypotheses are generalizations or extensions of previous research results, and some are only based on grey literature.

Some hypotheses are very well documented, and I will consider them proven for the rest of my PhD, although my research could lead to refine and nuance their formulation. Some have rarely been assessed, and I will try to evaluate them in my research.

European environmental objectives are insufficient, and used in most EU climate mitigation scenarios

There are at least 5 steps to go from a scientific assessment of sustainability to public policies :

  1. Research (and sometimes consensus) on critical sustainability thresholds.
  2. International agreements on some thresholds, which are not necessarily those recommended by researchers.
  3. National and EU laws on targets, that are often less ambitious and constraining than international agreements.
  4. Scenarios to explore possible ways to achieve the targets.
  5. Public policies chosen and refined among the scenarios.

This means that climate mitigation scenarios used for the EU will often, if not always, aim for the "wrong" target (again, if sustainability is the goal) or at least one that is ridiculously underestimated. For example, the Paris Agreement 1.5°C threshold was once deemed too high by some researchers (Morena 2023), even if today it appears like the pinnacle of international climate governance. Depending on how you split the global remaining (and very thin) carbon budget to stay under that threshold, you can find immensely different values per country. But you have to adopt very far-fetched and unfair sharing principles if you want the carbon budget of rich countries to be positive, an exercise in which EU administrations and their allies have become masters.

One of the strategies used by rich countries to make their budget higher is to only account for territorial emissions and not footprints (as exemplified in Regulation (EU) 2018/842, Regulation (EU) 2023/857, Directive (EU) 2016/2284, Regulation (EU) 2021/1119). That is also the case for all targets set during the COPs (Ala-Mantila 2023 et al.).

But even if one accepts this distribution criterion — which can be deemed unfair — these objectives are largely insufficient to remain under 1.5°C of global warming (Cuny and Parrique 2024, Spragg Nilsson 2024). For example, an assessment using that principle finds that the carbon budget of Germany has been exhausted since the 1930s, while that of the UK was exhausted even before 1900 (Li et al. 2025). One of the only ways to argue that the EU has not exhausted its carbon budget is to ignore all emissions prior to 2015 (Pelz et al. 2023, ESABCC and EEA 2023).

And that only holds if the global carbon budgets from the UNFCCC can be trusted. A recent estimate shows that only 70% of actual GHG emissions are declared in national emissions inventories (Pearce 2024), which means the world could already be over 1.5°C of global warming (Jarvis and Foster 2024).5 Moreover, there appears to be very few quantified European objective for reducing other environmental pressures, such as the material footprint (Parrique et al. 2023b).

Why does this matter for our question? Simple: many if not all climate mitigation scenarios seek to reach targets and/or stay under thresholds of environmental pressures. If this threshold is insufficient to reach sustainability, then the futures depicted are necessarily unsustainable.6 And for some reason, it seems that EU environmental objectives are used in the vast majority of EU climate mitigation scenarios. This can explain why analysts and researchers are very rarely pushed to challenge the main economic institutions.

Anyone can read the official documents that contain the environmental targets of the EU, and see how they are used in scenarios, so I hold this hypothesis to be true. However, the extent to which these targets are used outside of the Commission, and why, is another question which would need to be investigated.

Existing laws and plans in the EU prevent economic institutions to be challenged

Continuing with the obvious stuff, our legal systems have been built for capitalism for hundreds of years. As I explained before, private property, free trade and entrepreneurship freedom are enshrined in laws and constitutions.

Impact assessments and the TYNDP must comply with existing laws and plans, just like they have to meet certain environmental targets. In principle, this makes it near impossible to challenge the current economic institutions in scenarios. For example, enacting a degrowth strategy could be seen as contravening the Treaty of Lisbon (2007), which makes economic growth a fundamental objective of the EU. As a consequence, Mauger (2023) describes the European Union's laws regarding energy and the environment as an "ocean of green growth", and struggles to find provisions favorable to degrowth policies.

Of course, when they are designed for governements, scenarios are often meant to change the law, so they can theoretically deviate from existing laws. But as radically changing many laws can be more difficult, such pathways might not be considered.

Moreover, member States are sovereign, and the legitimacy of the EU relies for a large part on the willingness to cooperate from member states. This means impact assessments cannot go too far in imagining a different economy without interfering with national policymaking. Thus EU scenarios tend to be more detailed on aspects where the EU has the most responsibilities.

As with the previous hypothesis, all this information is easily accessible, so I consider it proved. Impact assessments and TYNDP scenarios often describe which laws and plans they take into account, and the criteria that were used to select them. However the reproduction of fundamental legal principles (such as free trade) is often implicit, especially in research papers.

Organizational constraints

Political and administrative elites that have an interest in preserving the current organization of the economy can constrain the process and outcome of scenarios

Analysts of the European Commission are under the authority of commissionners, who are elected in a process involving the European Council, the Parliament and the Presidency of the Commission. And as there are very little political parties who advocate for a planned reduction of economic output and strong social programs such as those I have described in previous papers, it is unlikely that Commissionners would see a point to explore them in scenarios. Research in the sociology and history of energy planning has shown that elected officials and civil servants who commission studies can intervene at different levels in the design of scenario studies so that they conform to their objectives:

  • Problem and questions framing (Süsser et al. 2021, Royston et al. 2023, Baumgartner and Midttun 1987): scenarios are designed to answer specific questions, which are often determined by political elites. This is especially true in EU administrations, where backcasting7 is the most common approach, which implies a clear predefinition of the goals to be reached by scenarios (Royston et al. 2023).
  • Solutions and scenarios (Royston et al. 2023, Baumgartner and Midttun 1987): political elites can pre-define the kind of solutions used to reach the objectives. Policies can even be decided on before they are tested in models, and marginally changed aftewards (Süsser et al. 2021).
  • Model design, theories used (Süsser et al. 2021, Royston et al. 2023): analysts can be pushed to use certain models which present favorably the output of a given policy.
  • Evaluation, validation and calibration (Süsser et al. 2021, Royston et al. 2023): participation of political elites in the assessment of models can lead to adjusting them to the pre-existing beliefs of these elites (Royston et al. 2023).
  • Input data (Süsser et al. 2021, Royston et al. 2023, Lunde and Middtun 1987): technology costs, energy efficiency rates, productivity rates and the like are subject to controversies and are often chosen for political reasons: justifying a specific policy, facilitating acceptance by some member states, etc. (Royston et al. 2023).

I consider this hypothesis to have sufficient empirical support to be true, especially since Royston et al. (2023) specifically studied EU energy policymaking. It remains however to be explored how economic elites' imperatives can conflict with those of administrations, in the case of the TYNDP. I should also investigate the influence of elites in other scenario building processes (NGOs, research, administrations supporting the Commission, etc.).

EU administrations filter knowledge and policy ideas that question the current organization of the economy

Peters and Nagel (2020) explain that external actors produce too much knowledge and policy ideas for an administration to process. In order to cope with that, they inevitably create filtering processes for ideas, which could exclude radical policies such as those necessary to create a sustainable society. For example, position papers submitted in consultations for impact assessments are rarely all studied: 120 out of 237 in the scenarios for the 2040 climate targets (EC 2024a).

The nature and degree of intentionality of these filtering processes remain an open question in the case of energy and climate policies. It could probably be assessed with observations during "stakeholder workshops", and interviews with the workers that analyse received positions.

Analysts are pressured to be productive, which excludes spending time to question the basic assumptions of their work

Peters and Nagels (2020) highlight that, like most workers, policymakers and analysts have to find solutions quickly. This could leave no time to learn new models, have long discussions over the purpose of their work or redefine their methods. This could be reinforced by the mainly theoretical image that degrowth policies still have. Which is understandable, given that historical examples of such policies are scarce, and that research on this topic tend to lack empirical evidence (Engler et al. 2024 ; Savin and van den Bergh 2024).8

This hypothesis could be assessed for impact assessments and the TYNDP, by asking analysts about their working conditions, and observing how and when methods and tools do change.

Scientific controversies

The scientific controversy over green growth is not closed, despite the overwhelming evidence showing its impossibility

There are mutliple reasons to believe that this is the case. First, one would expect the number of articles on decoupling to decrease as literature reviews are written and settle what seems to be the consensus on this theory. But that is not happening: the amount of papers on the environmental Kuznets curve and decoupling are still rapidly increasing (Menezes et al. 2023, Vàden et al. 2020). Second, Latour and Woolgar (1979) show that when a consensus has been reached on a scientific matter, it is less and less justified when asserted in research papers. But even scholars that know very well the literature on decoupling seem to consider its impossibility as controversial. For example, Mauger (2023) uses 4 introductory pages to re-affirm the impossibility of green growth, which shows that this assertion is still at a low degree of facticity (Latour and Woolgar 1979). The first paper of this series is another proof of that. This seemingly artificial continuation of the controversy could allow States, european administrations and other organizations to promote green growth strategies while still maintaining an appearance of "science-based policy".

In that sense, green growth can be qualified as a zombie idea (Peters and Nagel 2020): a theory that is continually disproven by research and practical experience, but is still guiding policymaking and tested in research. For now, this is just an empirical observation and I cannot explain why research papers on decoupling are still being published. Latour (1987) affirms that scientific controversies close when no actor is any longer ready to bear the cost of doubt (be it cognitive, human, material, financial or symbolic). Thus, it could be explained by the fact that a lot of actors are still willing to fund research on decoupling that does not disqualify green growth, as has already been documented for many other public problems (Bonds 2011 ; Pinto 2017).

The fact that the scientific controversy over green growth is not closed seems established, but I could try to understand what actors are bearing the cost of doubt, and why. This could be assessed with scientometric methods and documentary research.

Studies showing the unsustainability of current economic institutions do not rely on the same proof standards than scenarios using models, and thus are hard to compare

Models can produce outputs (numbers) that are useful to assess if a set of changes (policies) will achieve an expected effect (a certain level of GHG emissions for example). These outputs largely depend on the theories that the models were built with. However, these theories can be flawed, incomplete, insufficient, or valid for very specific situations. Moreover, the sociotechnical context for which models were built is rapidly evolving, which sparks further doubt on their reliability (Floyd et al. 2020). Some researchers argue that model results should always be confronted to other types of knowledge (Anderson and Jewell 2019).

I don't think any of these statements are controversial. But still, scenarios produced with models seem to have a large rhetorical power, and can hardly be countered with other kinds of proofs (Saltelli et al. 2024). The articulation between different types of knowledge has been an intractable epistemological problem for centuries, especially between quantitative and qualitative (Desrosières 2010). Different disciplines have different hierarchies of proof (Sovacool et al. 2018), and therefore analysts could give little credit to arguments based on qualitative empirical social science. In fact, Royston and Foulds (2021) show that civil servants writing the calls for projects in the Horizon 2020 program expect social scientists to adopt quantitative methods from economics.

Therefore, the controversy on green growth could seem closed or open depending on the epistemic community one belongs to. Controversies on other economic institutions could not even be considered to exist by engineers and economists, because a lot of critiques come from qualitative social sciences. To further assess this hypothesis, I could observe the practical consequences of different types of arguments during the design of scenarios, and after they are published. This could be done through observation during meetings, conferences and consultations.

Professional cultures of analysts

Analysts read very specialized papers and do not know much about alternatives to growth and capitalism

For them, there could be no consensus that green growth will never exist. Pottier (2016) explains that environmental economists that work in international organizations and States mainly come from the "center" of the discipline, which is committed to neoclassical theory. In economics, most critiques of growth and capitalism come from peripheral sub-disciplines, such as ecological economics, biophysical economics, post-keynesianism and institutional economics. They publish in specific journals which might not be frequently read by the analysts from the Commission and the ENTSOs.

To assess this hypothesis, I could analyze the careers of analysts, and create citation networks based on reports and research papers they published.

Analysts consider a radical reorganization of the economy to be unrealistic or undesirable

Even if they were aware of the limits of current policies to reduce environmental pressures from the EU economy, they might not be able to imagine how a post-capitalist society would work. Research in sociology of economists shows that they have been mostly trained to design markets (Lebaron 2009), and to view society as a sum of market exchanges (Lebaron 2009). It is therefore particularly difficult for them to imagine a society that is not dominated by markets, even though markets were marginal institutions for almost all of human history (Polanyi 2009).

The same could be argued for the desirability of such a society. In a global survey on instruments of climate policy, Drews et al. (2024) show that economists (excluding ecological economists) tend to be less favorable to direct regulation than any other discipline. In a survey involving 112 experts on the energy transition in Estonia, most of them considered measures impeding market and consumption freedom to be immoral, even when they were aware that market-based solutions might be insufficient (Pahker et al. 2024).

Although this hypothesis is most likely true, it seems to have insufficient support by the current literature, as to my knowledge it has not been assessed in the context of scenario building. Observations and interviews seem appropriate to do that.

Analysts do not take environmental problems seriously

Pottier (2016) argues that the idea of a stable nature and society is a precondition for the optimality of markets, which is a fundamental axiom for a lot of economists. Thus, the existence of ecological problems is fundamentally incompatible with their worldview. This is well documented for the field of environmental economics, in which the use of cost-benefit analysis and irrealistic assumptions led to the illusion that climate change is not damaging because of its theoretically small impact on GDP (Pottier 2016, Azar 2007, Hourcade 2007, Vieille Blanchard 2007).

I have not read research on the environmental concerns of engineers and other specialists involved in the design of scenarios. This hypothesis could be assessed with a more robust literature review on the sociology of engineers and policy analysts, and possibly interviews.

Analysts favor models that tend to preserve the current organization of the economy, because they know how to use them

Cherrier (2023) argues that many choices in the history of economic modelling can be explained by ease of use, especially when one tries to understand why unrealistic assumptions were chosen. She calls this tendency tractability, and argues that tractability standards "differ across time and fields".

This phenomenon has certainly taken place in the formation of IAM research field. In the 1980s, at least two competing model types could have been favored by the emerging IAM community: those created after system dynamics, and those from environmental economics. IMAGE was created after the World 3 model used in the Limits to Growth report, and its underlying theory was system dynamics. DICE and MARKAL were created by very renowned neoclassical economists who turned to environmental issues in the 1970s, mainly to defend economic growth (Vieille Blanchard 2007). Matarasso (2007) shows that despite the initial success of the Limits to Growth report, DICE and MARKAL have seen much greater use than IMAGE. He argues that this is because the models from environmental economics are based on theories and mathematical tools that had been developed in economics for 50 years. On the opposite, system dynamics had no theoretical tradition or epistemic community, which means very few people could easily use and extend the IMAGE model.

The same might be true for the field of energy-economy-climate modelling as a whole, which IAM is only a part of. But I have not found a detailed study of the academic and professional background of individuals involved in scenario building for the EU. This hypothesis could be evaluated with prosopographic methods, for example by scraping data on careers and academic background of analysts on Linkedin.

Energy and climate policies have been constituted as a specific field of expertise, which is not intended to reorganize the entire economy

Categories such as "energy policy" and "climate mitigation scenarios" could be associated with a restricted set of measures perceived as belonging to that field. This could mean that social security policies, although they contribute to sustainability as a whole, could be viewed as irrelevant to this field. In fact, the category of "energy policy" only emerged in the 1970s (Mitchell 2018), and its prerogatives have substantially changed over time.9

This is also apparent in climate mitigation scenario reports, where indicators such as inequalities or biodiversity are rarely quantified. This allows to ignore the problem shifts linked to green growth policies. In response, some modellers have been trying to define "sustainability guardrails for energy scenarios" (Child et al. 2018).

This hypothesis could be assessed by observing the interaction between researchers and policy analysts, for example in conferences and workshops. In these contexts, the delineation of energy and climate policies might be brought up to justify the exclusion and inclusion of certain changes and indicators.

External actors involved in the production of scenarios

External experts anticipate what will be deemed realistic and acceptable, and thus often exclude radical change

Producing tools and studies that can be used by other powerful actors is a valuable asset in the career of a researcher or analyst. Conversely, if one produces studies to assess policies that the EU show no will to implement (such as degrowth policies), no policymaker would call on their expertise and few researchers might be ready to extend their work.

There are numerous examples of that in social sciences. Greenberger and Hogan (1987) show that the energy studies that had the most resonance in the United States in the 1970s were those that coincided with the interests of political elites. Consequently, their authors tended to conform to the futures that they believed to be consensual. Süsser et al. (2021) argue that, as researchers are increasingly urged to have an "impact" on policymaking, they only produce energy futures that can be used by policymakers, and therefore do not radically diverge from current policies.

The IPCC reports frame the space of possibilities for climate mitigation pathways (Beck and Oomen 2021), but their authors also seem to anticipate the expectations of political elites. Dahan (2007) explains that each chapter of IPCC reports is reviewed by governments, and that the summary for decision-makers must be validated "line by line" by each State. IPCC researchers can then only anticipate what will be deemed acceptable if they want their reports to be published.

There is sufficient proof in different contexts for this hypothesis to be held true, but the exact way it happens in the EU context is worth studying. Interviews and participant observation seem to be the most promising methods here, as in the normal course of things, "unrealistic" policies are never brought up.

Researchers who work in research institutes and think tanks depend on external and conditional funding

The dominance of private-funded institutes on knowledge production has been documented for a wide range of environmental issues (Bonds 2011, Beder 2001, Kinchy and Schaffer 2018). Plehwe (2022) shows that in the 2010s, the rise of renewables in the energy mix of Germany has been stalled by the increasing domination of neoliberal discourse produced in think-tanks, funded by companies that had interests in oil and gas. This led to prioritize "competitiveness" over environmental concerns.

For some topics, government agencies and public companies are dominant. One of the most striking example is that of nuclear energy in France (Schweitzer and Mix 2022, Puiseux 1987), where the Atomic Energy Committee (CEA) employs on its own two thirds as much people as the National Scientific Research Center (CNRS) (respectively 21·000 and 32·000). Of course, this incredible amount of knowledge production power cannot be countered by civil society organizations with small funds. NGOs are themselves indirectly funded by some of the richest capitalists on Earth, who have been vocal advocates of green capitalism (Morena 2023).

This hypothesis seems established, but could be studied for university researchers and NGOs that work on EU climate mitigation. I could look at the funding sources of the most active research institutes and NGOs, and have interviews with grant managers in foundations and EU agencies.

Most actors participating in consultations and workshops do not question the current organization of the economy

Companies, governments and lobbies have vast financial resources which can be used to influence the process of scenario building. Lobbyists in particular can obtain private meetings with analysts, write detailed position papers in calls for evidence, receive draft legislation in advance, be part of expert groups, etc. (Laurens 2015) They can spend thousands on studies, which can be handed out and presented to civil servants, especially when an administration is missing the data or knowledge to decide on its own (Saurugger 2002). Or, as is the case with ENTSOs, companies' interests can be directly represented through the TYNDP building process.

This hypothesis is certainly true, but I have not found research on what this imbalance means for climate mitigation scenarios. It could be more precisely assessed with analyses of contributions to consultations, observations in workshops, and interviews with lobbyists and NGO workers.

Critical actors do not have the same resources as the others to defend their point of view

This hypothesis is an extension of the former one: critical actors might try to take part in the same kind of advocacy as others, but might not have the same resources to do so. Resources can be money and time, but also tools, computing power or social networks.

Again, it certainly is true but would need a detailed description. This could be done through interview and an analysis of their contributions.

Degrowth and sufficiency communities have not built coalitions with powerful actors

Aykut (2019) argues that, in highly specialized policy fields such as energy, actors need to create predictive policy assemblages to be able to influence public policies. Those are networks of actors, discourses and practices, that strive for the same future. In other words, degrowth scholars alone might not be able to change much to EU scenarios on their own, but have much bigger chances to do so if they have detailed models and strong political allies.

Assessing that hypothesis could entail asking critical actors about their relationship with policymakers, civil society organizations, companies, etc.

Models and knowledge infrastructures

In this part, I talk about how the structure of models can cause the reproduction of the current economic institutions, either by hiding their consequences or by preventing alternatives to be explored. There are a lot of classifications of models, but these 3 should be enough for what follows:

  • Energy system models (ESMs): models mostly designed by engineers, which represent the machines and infrastructures used for energy production, transport and consumption. These models are useful to know what happens when you change those machines and infrastructures, or when their use changes. They can also integrate market dynamics, as is the case with the PRIMES model used by the European Commission (DG CLIMA et al. 2021). They are then called hybrid models (Labussière et Nadaï 2015, Hourcade 2017) or equilibrium models (Koppelaar et al. 2016).

    All energy system models I will talk about in what follows belong to the subclass of capacity expansion models. These can design a new energy system to meet a given energy demand, based on cost assumptions and other constraints (Boyd 2016).

  • Macroeconomic models: models mostly designed by economists, which represent all sectors of an economy. They can anticipate how prices, production and consumption will evolve given certain public policies. This is useful to derive an energy consumption from a forecasted level of production in different sectors, given energy/money coefficients for each sector.

    Thus, usually in a scenario building process, production levels are first forecasted by a macroeconomic model, translated to energy demand, which is provided to a capacity expansion model to design an energy system that meets the demand.

  • Integrated assessment models (IAMs): models that represent interactions between the global economy, the energy system and the climate.10 The former 2 classes do not represent the climate, but they can take into account climate targets through GHG emission levels. As climate change only makes sense on a global scale, IAMs are mostly used to produce global scenarios, which form the basis of the IPCC studies on climate change mitigation.

For that reasons, IAMs are rarely used to design scenarios on the EU level, although some include world regions. But they can influence EU scenarios in multiple ways:

  • Some IAMs have geographical regions, which allows to produce EU scenarios. This is for example the case of WILIAM (LOCOMOTION H2020 project 2019).
  • The IPCC has an enormous scientific and political legitimacy, which implies that the content of IPCC reports can be used as authoritative arguments (Beck and Oomen 2021).
  • Global scenarios produced with IAMs are sometimes directly used to frame the space of possibilities in EU scenarios. This was the case for the European Commission's (2024a) scenarios to define a climate target for 2040, which was partially framed using a study from the ESABCC that relied on an assessment of hundreds of scenarios produced by the IAM community (ESABCC & EEA 2023).

Given what precedes, it makes sense to include IAMs in my review of factors that could prevent alternative organizations of the economy in EU scenarios.

In this section, I will only talk about problems regarding the normative aspects of models. I will discuss their realism in the section on "Bad science and undone science".

Most energy system models are designed to minimize the cost of an energy system, which makes it difficult to use other valuation principles

Research on energy system models (Lund et al. 2017, Koppelaar et al. 2016) distinguishes three paradigms:

  • Optimization models provide a least-cost solution for an energy system, given some constraints (for example GHG emission levels). They provide a single solution, presented as the best one.
  • Equilibrium models rely on a representation of markets, where every actor tries to minimize their cost. The result is again a single solution, presented as the most likely and best one in a market economy. The equilibrium paradigm is only present in hybrid models.
  • Simulation models just compute how a pre-designed energy system or pathway performs regarding different indicators, such as system costs, GHG emissions, energy prices, etc. They are better suited to debate of the advantages of different energy systems.

As Moglen et al. (2023) state, the dominance of optimization models has led to disregarding potential alternatives where costs would not be the only decision criterion. They stress that minimizing system costs can "disproportionately disadvantage certain demographics", which could also be argued for other social and environmental realities. Unfortunately, not all those realities can be easily quantified, which means they cannot be integrated as constraints for optimization and equilibrium models.

Simulation models thus seem more adequate to debate the sustainability of an energy system. They allow to put more development efforts on the diversity of indicators, and to politicize the discussion on scenarios. This could in turn favor discussions over the sustainability of the economy as a whole. Other modelling techniques can also help generating multiple alternatives and use multiple decision criteria (Moglen et al. 2023).

The problems associated with cost minimization in the optimization and equilibrium paradigms are easy to understand. I could investigate to what extent the ESMs used by the European Commission and the ENTSOs rely on these paradigms, and how much their results are used authoritatively. This could be done by reading model documentation, and interviewing their designers and users.

Most macroeconomic models and IAMs presuppose a market economy and are designed to maximise consumption.

The equilibrium mechanisms I mentioned before come from economic theory, they are thus very often used in macroeconomic models. All 3 macroeconomic models in use in the European Commission (QUEST, GEM-E3 and GM) rely on an equilibrium framework, even though they come from different schools of economic thought (Souffron and Jacques 2024). I will outline the problems posed by the lack of realism of these models in another section, but this also poses normative issues.

First, in an equilibrium framework, all actors try to maximise their consumption. This implies that whatever the policies projected, the model will tend to produce as much goods and services as possible, which is as we know deeply unsustainble. This problem is identified by multiple papers on IAMs (Cointe and Pottier 2023, Keyßer and Lenzen 2021, Kuhnhenn 2018, Pottier 2016, Fremstad and Paul 2022), which for the most part also use general equilibrium theory.

Second, it could be argued that equilibrium theory is not prescriptive, and is just a description of how a market economy works. But even if that was true, the fact that macroeconomic models and IAMs incorporate it so often prevents from modelling alternative organizations of the economy (Anderson and Jewell 2019). For example, this mechanism alone cannot represent how public universal basic services, or a commons-based economy, would affect consumption and production. Ultimately, because it is very unlikely that a sustainable future can be built in a market-based economy, most of the current macroeconomic models and IAMs can be considered useless in that regard.

This hypothesis has seems established for macroeconomic models and IAMs. However, I would need to better understand the implications of trying to model non-market economies. This can be done by reading model documentation and extending my literature review.

Most models used to design scenarios cannot explore alternative economic systems

A generalization of the observation that markets are embedded in macroeconomic models is that ESMs, IAMs and macroeconomic models are not designed to explore alternative organizations of the economy.

Koppelaar et al. (2016) compare 11 electric system models for Germany, and finds that they are only useful to compare different policies, for a predefined policy instrument. They cannot be used to discover new problems, even less to explore other "political paradigms". The authors mean by that the use of policy instruments that are not commonly used today, such as planning.

Dekker et al. (2023) compare the outputs of 7 IAMs and 1 ESM for 10 scenarios in Europe, to better understand how they behave. They underline that their comparison is relative, and thus does not allow to detect "structural bias in the community". Indeed, they do not note that among the 4 levers they identify as "mitigation strategies", even sufficiency is absent. This means that 8 models which are widely used for building scenarios (the PRIMES model used by the European Commission is included) do not integrate the reduction of demand as an option. Kuhnhenn (2018) comes to the same conclusion for IAMs in general.

Beck and Oomen (2021) underline that IAMs tend to favor changes that can be easily quantified. Thus, carbon dioxyde removal (CDR) techniques have been included in IPCC reports, even if they are unproved at scale and dangerous, whereas much more technically feasible alternatives that implied a change in economic institutions have not.

It seems established that neither ESMs, IAMs or macroeconomic models can explore alternative organizations of the economy. This is easily understandable, as alternative institutions have rarely been seriously considered in climate policies, and thus the demand for their modelling by policymakers remain weak. Yet this is far from unfeasible: as an example, economic research on commons have long used agent-based models to study the merits of different institutions for resource sustainability (Rouchier 2020).

Redesigning models to represent alternative economic systems requires significant time and funding

As I said in the previous sections, representing alternative economic institutions in models require specific developments. As Cointe and Pottier (2023) show, economic growth is deeply embedded in IAMs, because agents will always try to maximise their consumption. Degrowth researchers argue that current models often require significant developments to properly simulate degrowth scenarios.

D'Alessandro et al. (2020) designed a degrowth scenario for France, which relies on a job guarantee, working time reductions, a reduction in consumption and export, alongside other traditional measures. To show the benefits of these policies, they developed the EUROGREEN macroeconomic model, which for example allows for differentiated employment rates depending on worker skills. Li et al. (2023) and Kikstra et al. (2024) analysed degrowth scenarios for Australia, and they underline that their MESSAGEix model version would need to be further modified to account for the consequences. For example, modelling products instead of economic sectors would allow to differentiate how useful they are, and thus change their production accordingly. They also argue that changes in how goods are allocated influence their consumption, but this has not been modelled (Kikstra et al. 2024).

These papers as well as many others help envisioning societies where institutions such as consumption freedom, the labor market, entrepreneurial freedom or international trade are less dominant (Lauer et al. 2025). Nevertheless, they remain dominated by markets and private property, and those institutions could be more difficult to challenge. Durand et al. (2024) highlight that until now, a lot of macroeconomic modelling efforts for degrowth have been devoted to represent alternative market societes. Research to model an economy where democratic planning and commons are the norm is only in its beginning.

Therefore, the difficulty of redesigning macroeconomic models and IAMs is certainly hampering the possibility of presenting EU administrations with credible alternatives. This is partly due to the fact that it's easier to study societies which currently exist, and have large administrations producing economic data (Lauer et al. 2025).

In the models used to design scenarios, economic growth has "positive" consequences

Cointe and Pottier (2023) show that in GCAM and WITCH, GDP growth facilitates investment, which allows to transform the energy system more quickly. Reducing GDP would therefore hamper the model's ability to reduce GHG emissions. Moreover, consumption levels are often used by analysts as a proxies for social wellbeing, which explains why it's so commonly measured in reports.11

Although these relationships can be deemed valid under certain circumstances, these models and theories ignore other factors. For example, industrialized economies have become more energy and carbon efficient in periods of recession (Bersalli et al. 2023), a phenomenon which I do not believe is commonly modelled.

This hypothesis could be true of all the model types mentioned above, since they all use some kind of economic activity indicator to find out how much energy will be consumed.

It could be investigated in more detail for the models and reasonings used by the actors doing energy and climate scenarios for the EU. I could do this by reading model documentation, extending my literature review, and observing model-building sessions.

Models used to design scenarios produce insufficient environmental and social indicators to assess sustainability

Many papers underline that the indicators produced by models used in scenarios are insufficient to assess sustainability. Süsser et al. (2022) highlight that IAMs and ESMs often only include territorial GHG emissions and land use as sustainability indicators. Phenomena such as material use, biodiversity, water use, carbon footprint and others are rarely quantified. Salin et al. (2024) assess how 18 IAMs represent biodiversity, and find that the impacts on nature and feedbacks to the economy are insufficiently represented to assess the evolution of biodiversity.

To my knowledge, most macroeconomic models use employment and economic output as proxies for social wellbeing, but ignore more reliable indicators such as the Gini curve (used in the EUROGREEN model, see d'Alessandro et al. 2020) that represents inequalities. Dioha et al. (2023) underline that important social indicators should be added to ESMs, such as measures of environmental justice and energy poverty.

This seriously undermines the very possibility of using multi-criteria decision analysis (Moglen et al. 2023) to assess the merits of different scenarios, as recommended by Vettese and Pendergrass in their book on ecosocialist planning (Bickhardt et al. 2023). Models simply do not output enough indicators to assess sustainability. Of course, it would be incredibly difficult to represent all the realities mentioned above, as even modellers specifically working on the relations between biodiversity and the economy struggle to integrate most of the relevant interactions (Salin et al. 2024).

This hypothesis seems confirmed. I could however ask if other types of knowledge are used to complement and correct model outputs, potentially leading to more complete assessments. This can be done with observations and interviews.

Models are complex and need a lot of time to learn, which reduces the propensity of analysts to adopt new ones

For people who do not use or create models, it is hard to picture how complex these algorithms are. The design of the World Energy Model from IIASA mobilised 140 researchers during 8 years, and each of its 2 modules necessitate thousands of input variables. Keepin and Wynne (1987) argue that the complexity of this model is so enourmous that peer review becomes almost impossible. Moreover, this was in the 1980s, but models have been significanlty complexified since then. Models live for a long time, and are frequently amended based on needs identified by researchers and policymakers (Kolkman 2022). When they are open source, they can aggregate entire communities working with them and developing new versions. Because of their complexity, learning to use a model is an investment in the professional life of an expert.

There appears to be no straightforward way to simplify models. Moglen et al. (2023), in a review of the common critiques made to ESMs and IAMs, explains that choosing a complexity level is often a hard task for modelers, as it implies to find a middle ground between precision and running time. Many modellers also underline that a simple model can lead to "systematic bias" (features of the modelled system are ignored), whereas a complex one may lead to more "parametric uncertainty" (greater risk of having errors in the input data) (Saltelli et al. 2024).

As a consequence, adding a new set of indicators to better account for sustainability (as suggested in the last section) might actually reduce the reliability of the model as a whole. These problems are common to all disciplines that use models, and their severity has grown with the increasing availability of computing power. Guillemot (2007) explains that climate models have become much more complex in the last 70 years, but that the field also sees movements of simplification as climatologists try to isolate the most relevant factors.

Given this complexity, Kolkman (2022) shows that only a small number of workers in an administration fully understand the models used for policymaking. Models use their own abbreviations and jargon, which makes switching to another more difficult. Even for expert modelers, learning to use a model correctly can take up to 1 year. Several of the analysts interviewed by Kolkman argue that anyone whose job is not to work with models might never understand the full functioning of a model.

Complexity and learning time certainly contribute to the reluctance of the Commission and the ENTSOs to adopt models that better account for the limitations of green capitalism, and could represent other economic institutions. However, the relative importance of this factor could be investigated, for example by studying past and present changes in the tools used by analysts. This could be done with archive work and interviews.

Using knowledge infrastructures to compare models and scenarios prevents or slows down innovations

Many disciplines and professions use common categories, softwares and data inputs, in order to compare data across many different contexts. These knowledge infrastructures allow to create much more general knowledge, but they are often very costly and difficult to modify. Because the categories in knowledge infrastructures determine what can be studied by the discipline, this can in turn prevent or slow down innovation in science (Bowker and Star 2023).

Innovation in the IAM community is very constrained by the IPCC process. During the 2000s, the IPCC group 3 stopped producing their own scenarios, and instead assessed those of the emerging IAM community (Beck and Oomen 2021). A very refined knowledge infrastructure was developed to facilitate the comparison of scenarios: scenario databases, model intercomparison exercises, standardised data templates for models, etc. (Cointe and Pottier 2023) 5 Shared Socio-economic Pathways (SSPs) were defined, which include assumptions that scenarios have to use in order to be included in the IPCC meta-analyses. All these pathways have an economic growth between 1 and 3% per year, and perpetuate existing international inequalities.

This process unified the hypotheses and data sources used in models, and made it impossible to depict post-growth economies in IPCC reports. As the IAM community is built around the IPCC process, this had strong implications for the careers of researchers. The IAM community is theoretically open to participation (if you have the right degrees and affiliations), but changing the SSPs will anyway come at a strong cost.

Such infrastructure does not yet exist on this scale for the ESM and macroeconomic communities, but could be developed in the future. Since 1976, the Stanford Energy Modelling Forum runs model intercomparison exercises for ESMs in the USA (Greenberger and Hogan 1987), and the European Climate and Energy Modelling Forum has started doing that for Europe (Dekker et al. 2023). The project's EU funding is now over, and it is still unsure how the community and infrastructure will evolve in the future.

In conclusion, the IAM knowledge infrastructure contributes to prevent the representation of alternative economic institutions in scenarios. A closer look at knowledge infrastructures in the ESM and macroeconomic communities is needed. Observations in scientific conferences and interviews with researchers seem appropriate to do that.

Methods and modelling practices

Models are not the only tools used to draft a scenario report. Before doing any model runs, analysts imagine different possible futures with tables of public policies, lists of events, or narratives. These futures are then translated into quantitative assumptions that can be used by models. The process of building those futures is a whole domain of research, and incorporates very strong value judgments. This section also touches on modelling practices: the way analysts build and use models.

Methods useful to explore alternative economic institutions are rarely used in EU climate mitigation scenarios

Andersson (2018) shows that different schools of future research have had different ways of thinking about the future: it can be closed, open, multiple, uncertain, etc. These views imply methodological differences. For example :

  • Futurology is a field that developed between the 50s and 60s at the RAND corporation, and aimed to discover general laws of collective behavior based on operations research and game theory. These mathematicians thus gave little methodological reflexion to the design of narratives, as the future was framed as unique and knowable (Andersson 2018).
  • The Delphi method is a way to stabilize multiple plausible futures with rounds of surveys given to experts, and quantitative clustering methods. It is thus very descriptive, but open to a plurality of futures (Andersson 2018).
  • Structural analysis can be used in the famous "scenario method". It uses tables to describe the relationships between variables judged relevant for the system studied, which can then be used to create differentiated scenarios (Roubelat 1993).
  • As I explained before, "backcasting" is an explicitly prescriptive approach, but does not create very differentiated scenarios.

The scenario method seems to be used by some researchers, leading to a few highly differentiated sets of scenarios. In fact, 14% of the 342 scenarios studied by Lauer et al. (2024) diverged from the current status quo, on their economic system, governance or relationship with nature.12

However, this does not seem to be the case for the ENTSOs and the Commission. The method to define futures in the TYNDP and impact assessments is rarely described in detail, and the changes envisioned are very homogeneous. "Narratives" are most often exposed in the form of succinct tables, which also holds true for the scenarios of international agencies, NGOs and energy companies (The Shift Project 2019).

For example, the TYNDP 2024 uses 2 narratives that were differentiated based on 3 factors: "driving force of the transition", "energy intensity", "technologies" (ENTSO-E and ENTSOG 2023). Although these narratives are framed as an exploration of "uncertainties" and "a wide range of possible future evolutions", their economic institutions are not diffentiated. Similarly, the main differentiating criteria for Commission scenarios are GHG or energy targets (see for example EC 2024a).

In general, it seems that methods to build highly differentiated narratives are very rarely used for EU climate mitigation scenarios, especially in the Commission and ENTSOs. This could be linked to the fact that futures research has largely fallen out of fashion before the 21st century, partly because of the severe criticism it has received from other scientific disciplines (Andersson 2018). It would be interesting to better understand how the scenarios to be analysed are defined, if not through these methods. Observations and interviews about scenario building processes would be appropriate to do that.

IAMs and macroeconomic models are often calibrated with data that leads to posit economic growth

Cointe et Pottier (2023) highlight that common modelling practices of analysts lead to reproduce economic growth. SSPs and OECD forecasts on economic growth are systematically used to calibrate models, which means the model will always tend to follow this pathway. Therefore, shocks cannot significantly alter economic activity. This was also documented by Salin et al. (2024) on the impact of biodiversity loss on GDP in IAMs.

This is likely to be the case as well for macroeconomic modellers, and energy system modellers. Moreover, since energy system models do not represent all sectors, GDP pathways are a mere input that the model cannot alter. To assess how generalized these practices are, I could have interviews with analysts and researchers.

The models used by the Commission and the ENTSOs are not open source, which prevents other actors from contesting them

The ESM and IAM communities have been publishing more and more of their source code (Morrison 2018 ; Moglen et al. 2023), which facilitates critical evaluation from the outside (Kolkman 2022 ; Hodencq 2022). Still, this does not seem to be common practice in governments and EU administrations:

  • GEM-E3, QUEST and PRIMES, the 3 main macroeconomic models and ESM used by the European Commission, are not open source. Even within the European Commission, not all Directorate-General understand how PRIMES works (Royston et al. 2023).
  • The ENTSOs use the Energy Transition Model for their demand projections, which is open source. But they also use the proprietary software PLEXOS as a capacity expansion model, and the DFT to produce demand profiles, a software that is not open source as of 2019 (Heymann 2019).

Without access to the code, actors that might want to criticize the realism or value judgments included in models are at a big disadvantage. This hypothesis thus seems valid, although the specific effects of this closure of models could be evaluated during my fieldwork.

Bad science and undone science

This section groups the hypotheses related to the quality of the knowledge used to design scenarios and models, and the possibilities for alternative knowledge to be produced and used.

In IAMs and macroeconomic models, economic output is often poorly modelled, which leads to systematically posit economic growth

Multiple papers suggest that economic growth is not realistically represented in a lot of IAMs and macroeconomic models, and thus that the GDP curves that those models output do not represent what would happen for a certain set of input policies.

Cointe and Pottier (2023) show that in the models GCAM and WITCH, GDP depends on a few variables entered by the analysts: population and work productivity for GCAM, total factor productivity and energy efficiency for WITCH. Moreover, those productivity rates often cannot be affected by public policies, and are calibrated to grow continuously. In other words, economic output cannot be influenced by actors in the model, and its growth is framed as an immutable natural phenomenon. This is of course not the case in the real world, as specific policies can influence productivity and GDP growth. Souffron and Jacques (2024) show that the macroeconomic models used by the European Commission suffer from similar problems: they do not sufficiently take into account the influence of economic fluctuations or transition policies on GDP. Sakai et al. (2019) argue that no macroeconomic model accounts for energy efficiency on a physical level, which leads to understimate the role of energy in GDP growth.

In all three cases, GDP growth is insufficiently influenced by different policies or contexts. This potentially allows analysts to postulate a continuously increasing production, with little regard to how this would happen. This could contribute to the illusion that economic growth can be maintained in a sustainable society, whereas more scientifically sound models would call it into question.

Therefore, the representation of economic output can be lacking both in IAMs and macroeconomic models, although in different ways depending on the model. Still, the exact consequences of those shortfalls regarding the reproduction of economic growth in scenarios is uncertain: some lead to overestimation while others lead to underestimation. For example, Souffron and Jacques (2024) argue that public investments might actually foster private investment and GDP growth, instead of reducing them as is often posited in macroeconomic models. To further assess this hypothesis, I could to read the documentation of macroeconomic models used by EU administrations, and extend my literature review.

Energy efficiency assumptions and modelling in scenarios is unrealistic, which leads to show green growth is possible

Energy efficiency is the amount of energy needed to produce an effect, such as heating, moving, making a computer work, etc. It is often approximated by dividing the value added of a sector by the energy it uses. Theoretically, increasing energy efficiency makes the economy use less energy for the same amount of production. If it went fast enough, it could lead to an absolute decoupling between GDP and energy consumption. It is thus a very common lever used to project green growth scenarios. But many scholars in ecological economics, biophysical economics and history of energy agree that there are some serious problems in the way most scenarios and models represent energy efficiency.

First, the vast majority of global and EU scenarios use energy efficiency rates that have never been observed in the past and seem very unlikely to happen. Of course, energy efficiency has been steadily increasing for centuries for many activities (Fressoz 2024), so the debate is about growth speed. In order to reach -55% of GHG emissions by 2030 in the EU,13 the EU economy would have to reduce its energy intensity 2 times faster than for the last 20 years, while reducing the carbon content of its energy 4.2 times faster (Cuny and Parrique 2024). Keyßer and Lenzen (2021) analyse 22 global scenarios, and show that all green growth scenarios need to postulate energy efficiency growth rates much greater than historical rates. Degrowth scenarios are the only ones to have efficiency rates similar to the historical ones. The Shift Project (2019) find that most global scenarios use energy efficiency rates that seem insufficiently substantiated. Moreover, Palmer (2018) observes that the progress in energy efficiency has actually been slowing global since the 2000s, which poses even higher challenges. For Hickel et al. (2021), there is no reason to keep using these unrealistic efficiency rates in scenarios, other than making economic growth seem sustainable.

Second, an increase in energy efficiency systematically causes multiple rebound effects. This means that as less energy is used to produce a good of service, its production cost lowers, and the saved money can be invested in producing and consuming more. This sometimes leads to use even more energy than before the efficiency gain. Parrique et al. (2019) distinguish between :

  • Direct rebound effects: the use of the same commodity increases.
  • Indirect rebound effects: the money saved is used to acquire other commodities.
  • Macroeconomic rebound effects: reducing the resource intensity of commodities transforms markets and increases resource consumption in general.

Floyd et al. (2020) highlight that most energy scenarios do not account for rebound effects. Brockway et al. (2021) review 33 model-based and empirical studies, and find that on average, macroeconomic rebound effects could cancel between 55% and 71% of the energy savings. 3 empirical studies find a rebound effect superior to 100%. Analysing the structure of IAMs and ESMs, they determine that only a small fraction of the key mechanisms contributing to macroeconomic rebound effects are modelled. Interestingly, they argue that those effects can only be accounted for using computable general equilibrium (CGE) macroeconomic models with detailed sectors. Sakai et al. (2019) take a more general view on this issue and try to model the links between energy efficiency and economic growth. Their model suggests that energy efficiency is a "key engine of economic growth", and that the massive increase in energy consumption in the last centuries has been possible because of — and not despite — energy efficiency gains. This is also what leads Palmer (2018) to predict a lower GDP growth in the future, because of the decline in efficiency gains.

Finally, Fressoz (2024) shows that the adverse effects of efficiency are also true for materials in general. As material efficiency progresses, artifacts become more complex. More and more materials are used in their production, and an increase in the use of any material leads to increase the use of other materials. In fact, Magee and Devezas (2017) show that the use of 69 materials only increased between 1960 to 2010, except for wool and tellurium.

It can thus be concluded that most global and EU scenarios rely on unrealistic energy efficiency rates, and that even if those rates were possible, they would still lead to unsustainable levels of energy consumption. The same would probably hold for material efficiency. If they follow common practices, the Commission and the ENTSOs can therefore create the illusion that continuing economic growth is sustainable. I could confirm this hypothesis by looking for efficiency rates and rebound effects in model documentation.14

Most models do not take into account the availability of raw materials

Süsser et al. (2022) highlight that ESMs rarely integrate the availability of materials while designing future energy systems, and that still holds for materials identified as critical. That leads for example to project unrealistic production rates for electric cars. Souffron and Jacques (2024) argue that the availability of minerals is not accounted for in the Commission's macroeconomic models, although research is very clear that it limits the speed on an energy transition (International Energy Agency 2022, Hache et al. 2018, Pitron 2018). Parpan et al. (2025) find that even under optimistic assumptions, global climate mitigation scenarios based on SSPs imply a copper shortage after 2050, and before 2040 for 3 of the 5 SSPs. This suggests that these scenarios are very unlikely to be feasible.

This hypothesis is validated, although that could change in the coming years. The EU is very concerned about the availability of raw materials for the development of "green" technologies, and has been taking measures to accelerate materials extraction and refining in member States (EC 2023b). Models could then be modified to account for materials needs.

Many scenarios integrate unrealistic and dangerous levels of carbon dioxyde removals

Since the IPCC AR5 report in 2014, researchers have integrated gigantic quantities of carbon dioxyde removal (CDR) in their scenarios, in order to meet climate targets (Fressoz 2024, Pottier 2016). For instance, the latest report from IPCC group 3 states that bioenergy with carbon capture and storage (BECCS) could technically absorb 11.3Gt CO2/year, and afforestation/reforestation (A/R) 10 Gt. This would use 3 times the area of the USA just for CDR (Deprez et al. 2024).

IAMs use high discount rates (around 5%), which means that money in the future has less value. This leads to a significantly underestimate the cost of technologies implemented in the future. Consequently, most models often come to the conclusion that it is much more cost-effective to implement carbon dioxyde removal later than to reduce GHG emissions now (Beck and Oomen 2021). CDR is also a last resort to keep economic growth in global scenarios, as even with unrealistic energy efficiency rates, climate targets cannot be reached without it (Kuhnhenn 2018).

There are multiple ways to capture and store carbon, some aiming to restore and improve natural sinks (A/R), others to create sinks with machines (often called "CCS"). But CDR as a whole is still experimental: its technical and economic feasibility at an industrial scale is doubtful, and it could cause very significant environmental and social damage. Juaied and Whitmore (2023) find that direct air capture (DACCS), one of the most discussed technologies, is very unlikely to remove more than 1% of global emissions in 2060. Beyond 2.8Gt/year for BECCS, and 5.1 Gt/year for A/R, CDR induces high risks for biodiversity, water availability, biogeochemical cycles and food security (Deprez et al. 2024). This is less than half the levels estimated by the IPCC. The IPCC recognized that the large-scale feasibility of CDR technologies is not proven, and that relying on them could prevent the achievement of global climate targets (Beck and Oomen 2021). Even if it were possible and did not cause problem shifting, the late implementation of large carbon removals is a very bad choice, since the temporary overshoot will cause death, suffering, and might trigger tipping points in the Earth system (Dyke et al. 2024).

The hype for CDR has very concrete consequences for scenarios and policies in the EU. For example, the European Commission (2024b) published an Industrial carbon management strategy in early 2024, and all of its recent climate scenarios include industrial carbon capture and storage. The inclusion of CDR therefore allows analysts to avoid challenging the current economic institutions. However, it could be asked what incites modellers to continue ignoring the solid proof that large-scale CDR deployment is unlikely and dangerous.

Many scenarios integrate unrealistic levels of hydrogen production, which might not be beneficial for the climate

Hydrogen is another key technology of contemporary climate change mitigation scenarios, because many industrial processes are hard to electrify, and thus need some sort of fuel. If hydrogen was produced with electricity from solar panels or wind turbines, its carbon footprint could be close to 0. Many EU scenarios thus include megatons of "green hydrogen" as soon as 2030.

However, this industry still does not exist anywhere in the world. In 2024, 99.9% of hydrogen was produced with fossil fuels (Corporate Europe Observatory 2024). Some hope that the emissions of this process can be captured, but this "blue hydrogen" still emits much more CO2 than simply burning natural gas (Howarth and Jacobson 2021). Quest, one of the largest and most advanced CCS project on a fossil hydrogen plant in Canada, only captures 39% of its emissions, despite the CCS system costing 1 billion (Global Witness 2022).

What about "green hydrogen" then? The ESABCC studied 63 recent EU scenarios that they deemed plausible and compatible with 1.5°C of warming, and found that 54 of them (86%) exceeded 50 GW of green hydrogen in 2030 (ESABCC and EEA 2023). Assuming that this industry could grow as fast as wind and solar power, Odenweller et al. (2022) find that the EU has a 5% chance of having 40 GW of hydrogen in 2030. This means that most EU scenarios rely on completely unrealistic hydrogen production levels to reach climate targets.15 And yet this is an understatement, as wind and solar power are some of the fastest deploying technologies in history. Green hydrogen is the opposite of these technologies: it's extremely inefficient and costly, which explains why despite billions in subsidies, the industry is struggling to materialize (Nippert 2024). Moreover, many of these scenarios do not account for potential leaks in the green hydrogen supply chain, which have a warming potential. If global hydrogen was 70% "green" and the leaks only 5%, then a hydrogen-based economy would actually be worse for the climate than just using natural gas until 2050 (Hauglustaine et al. 2022).

Therefore, most EU scenarios do not seem to consider that a large scale green hydrogen industry is unlikely to materialize and be beneficial for the climate. Similarly to CCS, I should thus investigate what leads modellers and analysts to ignore the evidence.

The optimization and equilibrium principles do not reflect how real market economies work

Trutnevyte (2016) compares the evolution of the British electric system between 1990 and 2014 with a cost-effective scenario, and a large number of near cost-effective scenarios. They find that the cost-effective scenario is not a good approximation of the real evolution of the system, whether in terms of costs, investment costs, deployment of technologies or GHG emissions. The cost difference is estimated to be between 9% and 23%. Some near-optimal scenarios can describe the real evolution of the energy system, but pathways that are between 9% and 23% more expensive are very numerous. The author thus argues for generalizing the use of numerous near-optimal scenarios, or even abandoning the use of cost as the main decision criterion for realism purposes.

Regarding the general equilibrium framework, a very large body of works have showed how its formulation following the works of Walras could not account for the functioning of real-world market economies (Boyer 2009, Souffron et Jacques 2024). Other model types can predict their evolution more accurately: agent-based models, stock-flow consistent models, etc. (Souffron et Jacques 2024).

This lack of realism often leads to underestimate the real "cost" of an energy transition, and overestimate its feasibility within current economic institutions.

Social sciences are not commonly integrated in scenario building, which leads to naturalize economic institutions

Many researchers highlight how rare it is for social scientists (except economists) and their knowledge to be integrated in scenario building processes (Saujot et al. 2022, Süsser et al. 2022, Dioha et al. 2023, Hirt and de Pryck 2023, Krumm et al. 2022). In general, social sciences have only received between 2.5% and 4.6% of the funds dedicated to research on climate change mitigation (Overland and Sovacool 2020). Even when they are present, social scientists are often restricted to auxiliary roles (Silvast and Foulds 2022). Royston and Foulds (2021) find that in the european research programme Horizon 2020, only 10% of the budget has been attributed to social scientists. They show that what is expected of energy social scientists is mostly fostering acceptance of technologies, while conforming to methods and ideas about human nature from neoclassical economics.

In the articles cited above, social scientists are framed as able to enchance the realism of scenarios, by studying the acceptability of infrastructures, energy practices, energy communities, urban and spatial dynamics, etc. Sometimes, social scientists can be said more able to challenge the organization of society (see for example Hirt and de Pryck 2023), but the exact institutions that should be subjected to their inquiry are never really described. These researchers often recommend more interdisciplinary projects, more collaboration in the design of models, and recognition of the value of qualitative knowledge.

My opinion is that studying socio-political feasibility is useless if scenarios are completely unsustainable, as is the case with most climate mitigation scenarios. Thus, I would tend to believe most of this strand of the literature is mistaken on the interest of social sciences for scenario building, which to me is bringing knowledge on alternative institutions, especially when models are insufficient.

In conclusion, social scientists and their knowledge are undoubtedly excluded from scenario building, and that certainly fosters the reproduction of an unsustainable economic organization.

Research on degrowth and alternative economic institutions is underfunded

Finally, it could be hypothesised that degrowth research and education do not receive sufficient funds to be able to challenge the scenarios of EU administrations. This would explain why there are so few degrowth scenarios, why degrowth policies are still too vaguely formulated (Fitzpatrick et al. 2022), and why models to represent alternative economic institutions are lacking (Durand et al. 2024). In fact, Royston and Foulds (2021) highlight that research on energy sufficiency has almost not been funded by the Horizon 2020 program.

There are very good reasons to believe this is the case, although I have not found specific studies on that matter. This hypothesis could be assessed by an analysis of public research funds databases.

Summary of the hypotheses

I summarize here all the hypothesis discussed above. Based on the arguments I developed, I:

  • Rate my level of confidence in each hypothesis.
  • Mention the additional research questions that they might raise.
  • List the methods that would be best suited to assess them.

Here a csv file containing this table if for whatever reason you want to play with it:

Hypotheses.csv

Hypothesis Domain Confidence Additional question Methods for assessment
European environmental objectives are insufficient, and used in most EU climate mitigation scenarios Legal constraints High Are these targets used outside of the Commission? Why? Interviews with researchers and NGO workers
Existing laws and plans in the EU prevent economic institutions to be challenged Legal constraints High - -
Political and administrative elites that have an interest in preserving the current organization of the economy can constrain the process and outcome of scenarios Organizational constraints High How do imperatives of economic elites conflict with those of the Commission in scenarios? How can they shape scenario-building processes? Observations in scenario-building processes, interviews with lobbyists and ENTSO analysts.
EU administrations filter knowledge and policy ideas that question the current organization of the economy Organizational constraints Low - Observations in stakeholder workshops, interviews with people analysing position papers
Analysts are pressured to be productive, which excludes spending time to question the basic assumptions of their work Organizational constraints Low - Interviews with EC and ENTSO-E analysts.
The scientific controversy over green growth is not closed, despite the overwhelming evidence showing its impossibility Scientific controversies High Which actors bear the cost of doubt? Why? Scientometry, documentary research.
Studies showing the unsustainability of current economic institutions do not rely on the same proof standards than scenarios using models, and thus are hard to compare Scientific controversies Medium - Observation during modelling meetings, conferences and consultations.
Analysts read very specialized papers and do not know much about alternatives to growth and capitalism Professional cultures of analysts Low - Citation networks, prosopography.
Analysts consider a radical reorganization of the economy to be unrealistic or undesirable Professional cultures of analysts Medium - Observations in scenario-building processes, interviews with analysts and researchers.
Analysts do not take environmental problems seriously Professional cultures of analysts Low - Literature review on the sociology of engineers and policy analysts, interviews.
Analysts favor models that tend to preserve the current organization of the economy, because they know how to use them Professional cultures of analysts Low - Prosopography.
Energy and climate policies have been constituted as a specific field of expertise, which is not intended to reorganize the entire economy Professional cultures of analysts Low - Observations in conferences and workshops.
External experts anticipate what will be deemed realistic and acceptable, and thus often exclude radical change External actors involved in the production of scenarios High How do these anticipations happen in the current EU context? Interviews and participant observation.
Researchers who work in research institutes and think tanks depend on external and conditional funding External actors involved in the production of scenarios High How are university researchers and NGOs funded? Documentary research, interviews with grant managers, scientometry of funding.
Most actors participating in consultations and workshops do not question the current organization of the economy External actors involved in the production of scenarios Medium - Analyses of contributions to consultations, observations in workshops, interviews with lobbyists and NGO workers.
Critical actors do not have the same resources as the others to defend their point of view External actors involved in the production of scenarios Low - Interviews with critical actors, analysis of their contributions.
Degrowth and sufficiency communities have not built coalitions with powerful actors External actors involved in the production of scenarios Low - Interviews with critical actors.
Most energy system models are designed to minimize the cost of an energy system, which makes it difficult to use other valuation principles Models and knowledge infrastructures High Do the ESM used by the EC and ENTSOs rely on the equilibrium or optimization paradigm? Model documentation, interviews with analysts.
Most macroeconomic models and IAMs presuppose a market economy and are designed to maximise consumption. Models and knowledge infrastructures High What are the implications of trying to model non-capitalist economies? Literature review.
Most models used to design scenarios cannot explore alternative economic systems Models and knowledge infrastructures High What are the implications of trying to model non-capitalist economies? Literature review.
Redesigning models to represent alternative economic systems requires significant time and funding Models and knowledge infrastructures High - -
In the models used to design scenarios, economic growth has 'positive' consequences Models and knowledge infrastructures Medium What are the consequences of economic growth in the models used for EU climate mitigation scenarios? Model documentation, observations in modelling meetings.
Models used to design scenarios produce insufficient environmental and social indicators to assess sustainability Models and knowledge infrastructures High Are other types of knowledge used to complement and correct model outputs? Documentary research, observations in modelling meetings.
Models are complex and need a lot of time to learn, which reduces the propensity of analysts to adopt new ones Models and knowledge infrastructures High What is the relative importance of complexity and learning time in comparison to other factors? Archive work, interviews with analysts of the EC and ENTSOs.
Using knowledge infrastructures to compare models and scenarios prevents or slows down innovations Models and knowledge infrastructures Medium - Observation in scientific conferences, interviews with coordinators of EU modelling networks.
Methods useful to explore alternative economic institutions are rarely used in EU climate mitigation scenarios Methods and modelling practices High How are scenarios defined? Observations in scenario building processes, interviews with analysts and researchers.
IAMs and macroeconomic models are often calibrated with data that leads to posit economic growth Methods and modelling practices High Are these practices generalized? Interviews with analysts and researchers.
The models used by the Commission and the ENTSOs are not open source, which prevents other actors from contesting them Methods and modelling practices High What is the relative importance of openness in comparison to other factors? Observations in stakeholder workshops, analysis of contributions to consultations.
In IAMs and macroeconomic models, economic output is often poorly modelled, which leads to systematically posit economic growth Bad science and undone science Medium - Model documentation, literature review.
Energy efficiency assumptions and modelling in scenarios is unrealistic, which leads to show green growth is possible Bad science and undone science High - Model documentation.
Most models do not take into account the availability of raw materials Bad science and undone science High - -
Many scenarios integrate unrealistic and dangerous levels of carbon dioxyde removals Bad science and undone science High What incites modellers to ignore the proof that large-scale CDR deployment is unlikely and dangerous? Interviews with researchers and analysts, observations in scenario-building processes.
Many scenarios integrate unrealistic levels of hydrogen production, which might not be beneficial for the climate Bad science and undone science High What incites modellers to ignore the proof that large-scale green hydrogen deployment is unlikely and not necessarily beneficial for the climate? Interviews with researchers and analysts, observations in scenario-building processes.
The optimization and equilibrium principles do not reflect how real market economies work Bad science and undone science High - -
Social sciences are not commonly integrated in scenario building, which leads to naturalize economic institutions Bad science and undone science High - -

Bibliography

Ala-Mantila, S., Heinonen, J., Clarke, J., & Ottelin, J. (2023). Consumption-based view on national and regional per capita carbon footprint pathways and planetary pressures-adjusted human development. Environmental Research Letters, 18(2), 024035. https://doi.org/10.1088/1748-9326/acabd8

Anderson, K., & Jewell, J. (2019). Debating the bedrock of climate-change mitigation scenarios. Nature, 573(7774), 348–349. https://doi.org/10.1038/d41586-019-02744-9

Andersson, J. (2018). The future of the world: Futurology, futurists, and the struggle for the post cold war imagination. Oxford University Press.

Andrés, J., Boscá, J. E., Doménech, R., & Ferri, J. (2024). The welfare effects of degrowth as a decarbonization strategy. Applied Economic Analysis, 32(95), 112–140. https://doi.org/10.1108/AEA-01-2024-0021

Aykut, S. (2019). Reassembling Energy Policy: Models, Forecasts, and Policy Change in Germany and France. Science & Technology Studies, 32(4), Article 4. https://doi.org/10.23987/sts.65324

Azar, C. (2007). Les émissions optimales de CO2 le sont-elles vraiment ? In Les modèles du futur (pp. 165–183). La Découverte. https://doi.org/10.3917/dec.dahan.2007.01.0165

Baumgartner, T., & Midttun, A. (1987). The Socio-Political Context of Energy Forecasting. In T. Baumgartner & A. Midttun, The politics of energy forecasting. A comparative Study of Energy Forecasting in Western Europe and North America. Oxford University Press.

Beck, S., & Oomen, J. (2021). Imagining the corridor of climate mitigation – What is at stake in IPCC’s politics of anticipation? Environmental Science & Policy, 123, 169–178. https://doi.org/10.1016/j.envsci.2021.05.011

Beder, S. (2001). Research note. Neoliberal think tanks and free market environmentalism. Environmental Politics, 10(2).

Bersalli, G., Tröndle, T., & Lilliestam, J. (2023). Most industrialised countries have peaked carbon dioxide emissions during economic crises through strengthened structural change. Communications Earth & Environment, 4(1), 1–11. https://doi.org/10.1038/s43247-023-00687-8

Bonds, E. (2011). The Knowledge-Shaping Process: Elite Mobilization and Environmental Policy. Critical Sociology, 37(4), 429–446. https://doi.org/10.1177/0896920510379440

Bowker, G. C., & Star, S. L. (2023). Arranger les choses. Des conséquences de la classification. EHESS.

Boyd, E. (2016). Power Sector Modelling 101. https://www.energy.gov/sites/prod/files/2016/02/f30/EPSA_Power_Sector_Modelling_FINAL_021816_0.pdf

Boyer, R. (2009). Marché, état et capitalismes. In P. Steiner & F. Vatin, Traité de sociologie économique. PUF.

Bickhardt, M., Hutteau, F.-X., & Jacquin, U. (2023, December 1). Half Earth Socialism: Planification écosocialiste contre technocratie verte ? Terrestres. https://www.terrestres.org/2023/12/01/half-earth-socialism-planification-ecosocialiste-contre-technocratie-verte/

Brockway, P. E., Sorrell, S., Semieniuk, G., Heun, M. K., & Court, V. (2021). Energy efficiency and economy-wide rebound effects: A review of the evidence and its implications. Renewable and Sustainable Energy Reviews, 141, 110781. https://doi.org/10.1016/j.rser.2021.110781

Capros, P., Van, R. D., Paroussos, L., Karkatsoulis, P., Fragkiadakis, C., Tsani, S., Charalampidis, I., & Revesz, T. (2013). GEM-E3 Model Documentation. Joint Research Center. https://doi.org/10.2788/47872

Cherrier, B. (2023). The Price of Virtue: Some Hypotheses on How Tractability Has Shaped Economic Models. Œconomia. History, Methodology, Philosophy, 13–1, Article 13–1. https://doi.org/10.4000/oeconomia.14116

Child, M., Koskinen, O., Linnanen, L., & Breyer, C. (2018). Sustainability guardrails for energy scenarios of the global energy transition. Renewable and Sustainable Energy Reviews, 91, 321–334. https://doi.org/10.1016/j.rser.2018.03.079

Cointe, B., & Pottier, A. (2023). Understanding why degrowth is absent from mitigation scenarios. Revue de La Régulation. Capitalisme, Institutions, Pouvoirs, 35, Article 35. https://doi.org/10.4000/regulation.23034

Corporate Europe Observatory. (2024). The dirty truth about the EU’s hydrogen push. Corporate Europe Observatory. https://corporateeurope.org/en/dirty-truth-about-EU-hydrogen-push

Cuny, J., & Parrique, T. (2024). Can Europe green its growth ? IDH 21.

d’Alessandro, S., Cieplinski, A., Distefano, T., & Dittmer, K. (2020). Feasible alternatives to green growth. Nature Sustainability, 3(4), Article 4. https://doi.org/10.1038/s41893-020-0484-y

Dahan, A. (2007). Le régime climatique, entre science, expertise et politique. In Les modèles du futur (pp. 113–139). La Découverte. https://doi.org/10.3917/dec.dahan.2007.01.0113

Dekker, M. M., Daioglou, V., Pietzcker, R., Rodrigues, R., de Boer, H.-S., Dalla Longa, F., Drouet, L., Emmerling, J., Fattahi, A., Fotiou, T., Fragkos, P., Fricko, O., Gusheva, E., Harmsen, M., Huppmann, D., Kannavou, M., Krey, V., Lombardi, F., Luderer, G., … van Vuuren, D. (2023). Identifying energy model fingerprints in mitigation scenarios. Nature Energy, 1–10. https://doi.org/10.1038/s41560-023-01399-1

Deprez, A., Leadley, P., Dooley, K., Williamson, P., Cramer, W., Gattuso, J.-P., Rankovic, A., Carlson, E. L., & Creutzig, F. (2024). Sustainability limits needed for CO2 removal. Science, 383(6682), 484–486. https://doi.org/10.1126/science.adj6171

DG CLIMA, DG ENER, DG MOVE, De Vita, A., Capros, P., Paroussos, L., Fragkiadakis, K., Karkatsoulis, P., Höglund-Isaksson, L., Winiwarter, W., Purohit, P., Gómez-Sanabria, A., Rafaj, P., Warnecke, L., Deppermann, A., Gusti, M., Frank, S., Lauri, P., Fulvio, F. di, … Kalokyris, T. (2021). EU reference scenario 2020: Energy, transport and GHG emissions : trends to 2050. European Commission. https://data.europa.eu/doi/10.2833/35750

Dioha, M. O., Montgomery, M., Almada, R., Dato, P., & Abrahams, L. (2023). Beyond dollars and cents: Why socio-political factors matter in energy system modelling. Environmental Research Letters, 18(12), 121002. https://doi.org/10.1088/1748-9326/ad0a54

Directive (EU) 2016/2284 of the European Parliament and of the Council of 14 December 2016 on the Reduction of National Emissions of Certain Atmospheric Pollutants, Amending Directive 2003/35/EC and Repealing Directive 2001/81/EC (Text with EEA Relevance ), 344 OJ L (2016). http://data.europa.eu/eli/dir/2016/2284/oj/eng

Drews, S., Savin, I., & van den Bergh, J. (2024). A Global Survey of Scientific Consensus and Controversy on Instruments of Climate Policy. Ecological Economics, 218, 108098. https://doi.org/10.1016/j.ecolecon.2023.108098

Durand, C., Hofferberth, E., & Schmelzer, M. (2024). Planning beyond growth: The case for economic democracy within ecological limits. Journal of Cleaner Production, 437, 140351. https://doi.org/10.1016/j.jclepro.2023.140351

Dyke, J., Watson, R., & Knorr, W. (2024, August 20). The overshoot myth: You can’t keep burning fossil fuels and expect scientists of the future to get us back to 1.5°C. The Conversation. http://theconversation.com/the-overshoot-myth-you-cant-keep-burning-fossil-fuels-and-expect-scientists-of-the-future-to-get-us-back-to-1-5-c-230814

ENTSO-E & ENTSOG. (2023). TYNDP 2024 Scenarios Storyline Report, July 2023. ENTSOG & ENTSO-E. https://2024.entsos-tyndp-scenarios.eu/wp-content/uploads/2023/12/ENTSOs_TYNDP_2024_Scenarios_Storyline_Report_2023-12-04.pdf

EC. (2023a, August 4). European Commission Response on Beyond Growth Letter. https://friendsoftheearth.eu/wp-content/uploads/2023/10/EC-response-on-Beyond-Growth-letter.pdf

EC. (2023b). European Critical Raw Materials Act. https://ec.europa.eu/commission/presscorner/detail/en/ip_23_1661

EC. (2023c). Better Regulation toolbox—July 2023. European Commission. https://commission.europa.eu/system/files/2023-09/BR%20toolbox%20-%20Jul%202023%20-%20FINAL.pdf

EC. (2024a). Securing our future. Europe’s 2040 climate target and path to climate neutrality by 2050 building a sustainable, just and prosperous society. European Commission. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=COM%3A2024%3A63%3AFIN

EC. (2024b). Carbon capture, storage and utilisation. European Commission. https://energy.ec.europa.eu/topics/oil-gas-and-coal/carbon-capture-storage-and-utilisation_en

Engler, J.-O., Kretschmer, M.-F., Rathgens, J., Ament, J. A., Huth, T., & von Wehrden, H. (2024). 15 years of degrowth research: A systematic review. Ecological Economics, 218, 108101. https://doi.org/10.1016/j.ecolecon.2023.108101

ESABCC & EEA. (2023). Scientific advice for the determination of an EU-wide 2040 climate target and a greenhouse gas budget for 2030–2050. ESABCC. https://data.europa.eu/doi/10.2800/609405

European Parliament. Directorate General for Parliamentary Research Services. (2023). Beyond growth: Pathways towards sustainable prosperity in the EU. European Parliament. https://data.europa.eu/doi/10.2861/602232

EU. (2023). Strategic foresight report 2023. European Commission. https://commission.europa.eu/document/download/ca1c61b7-e413-4877-970b-8ef619fc6b6c_en?filename=SFR-23-beautified-version_en_0.pdf

Fitzpatrick, N., Parrique, T., & Cosme, I. (2022). Exploring degrowth policy proposals: A systematic mapping with thematic synthesis. Journal of Cleaner Production, 365, 132764. https://doi.org/10.1016/j.jclepro.2022.132764

Floyd, J., Alexander, S., Lenzen, M., Moriarty, P., Palmer, G., Chandra-Shekeran, S., Foran, B., & Keyßer, L. (2020). Energy descent as a post-carbon transition scenario: How ‘knowledge humility’ reshapes energy futures for post-normal times. Futures, 122, 102565. https://doi.org/10.1016/j.futures.2020.102565

Fremstad, A., & Paul, M. (2022). Neoliberalism and climate change: How the free-market myth has prevented climate action. Ecological Economics, 197, 107353. https://doi.org/10.1016/j.ecolecon.2022.107353

Fressoz, J.-B. (2024). Sans transition: Une nouvelle histoire de l’énergie. Seuil.

Global Witness. (2022). Hydrogen’s hidden emissions. Shell’s misleading climate claims for its Canadian fossil hydrogen project. Global Witness. https://www.globalwitness.org/documents/20314/Hydrogens_Hidden_Emissions_January2022.pdf

Greenberger, M., & Hogan, W. H. (1987). Energy-Policy Modelling in the US: Competing Social Alternatives. In T. Baumgartner & A. Midttun, The politics of energy forecasting. A comparative Study of Energy Forecasting in Western Europe and North America. Oxford University Press.

Grubler, A., Wilson, C., Bento, N., Boza-Kiss, B., Krey, V., McCollum, D. L., Rao, N. D., Riahi, K., Rogelj, J., De Stercke, S., Cullen, J., Frank, S., Fricko, O., Guo, F., Gidden, M., Havlík, P., Huppmann, D., Kiesewetter, G., Rafaj, P., … Valin, H. (2018). A low energy demand scenario for meeting the 1.5 °C target and sustainable development goals without negative emission technologies. Nature Energy, 3(6), 515–527. https://doi.org/10.1038/s41560-018-0172-6

Guillemot, H. (2007). Les modèles numériques de climat. In Les modèles du futur (pp. 91–112). La Découverte. https://doi.org/10.3917/dec.dahan.2007.01.0091

Hache, E., Simoën, M., & Sokhna, S. G. (2018). Electrification du parc automobile mondial et criticité du Lithium à l’horizon 2050. ADEME. https://librairie.ademe.fr/mobilite-et-transport/3803-electrification-du-parc-automobile-mondial-et-criticite-du-lithium-a-l-horizon-2050.html

Hauglustaine, D., Paulot, F., Collins, W., Derwent, R., Sand, M., & Boucher, O. (2022). Climate benefit of a future hydrogen economy. Communications Earth & Environment, 3(1), 1–14. https://doi.org/10.1038/s43247-022-00626-z

Heymann, F. (2019). Demand forecasting methodology. ENTSO-E. https://eepublicdownloads.entsoe.eu/clean-documents/sdc-documents/MAF/2020/Demand_forecasting_methodology_V1_1.pdf

Hickel, J., Brockway, P., Kallis, G., Keyßer, L., Lenzen, M., Slameršak, A., Steinberger, J., & Ürge-Vorsatz, D. (2021). Urgent need for post-growth climate mitigation scenarios. Nature Energy, 6(8), 766–768. https://doi.org/10.1038/s41560-021-00884-9

Hirt, L. F., & de Pryck, K. (2023). Diversifying knowledge for climate change mitigation: Illuminating the common good and desirable futures. PLOS Climate, 2(12). https://doi.org/10.1371/journal.pclm.0000321

Hodencq, S. (2022). Méthodes et outils pour un processus de modélisation collaboratif et ouvert des systèmes énergétiques [Université Grenoble Alpes]. https://hal.science/tel-03809331

Hourcade, J.-C. (2007). Les modèles dans les débats de politique climatique: Entre le Capitole et la Roche tarpéienne ? In Les modèles du futur (pp. 140–164). La Découverte. https://doi.org/10.3917/dec.dahan.2007.01.0140

Hourcade, J.-C. (2017). 5. Prospective énergétique: Modèles contre futurologie. In C. Jeandel & R. Mosseri (Eds.), L’énergie à découvert (pp. 32–33). CNRS Éditions. http://books.openedition.org/editionscnrs/10909

Howarth, R. W., & Jacobson, M. Z. (2021). How green is blue hydrogen? Energy Science & Engineering, 9(10), 1676–1687. https://doi.org/10.1002/ese3.956

International Energy Agency. (2022). Executive summary – The Role of Critical Minerals in Clean Energy Transitions—Analysis and Key findings. International Energy Agency. https://www.iea.org/reports/the-role-of-critical-minerals-in-clean-energy-transitions/executive-summary

Jarvis, A., & Forster, P. M. (2024). Estimated human-induced warming from a linear temperature and atmospheric CO2 relationship. Nature Geoscience, 1–3. https://doi.org/10.1038/s41561-024-01580-5

Juaied, M. A., & Whitmore, A. (2023). Prospects for Direct Air Carbon Capture and Storage. Costs, Scale and Funding. Belfer Center for Science and International Affairs, Harvard University. https://www.belfercenter.org/sites/default/files/pantheon_files/files/publication/Al%20Juaied%20Whitmore_Prospects%20for%20DACCS_20231129.pdf

Keepin, B., & Wynne, B. (1987). The Role of Models—What Can We Expect from Science? A study of the IIASA World Energy Model. In T. Baumgartner & A. Midttun, The politics of energy forecasting. A comparative Study of Energy Forecasting in Western Europe and North America. Oxford University Press.

Keyßer, L. T., & Lenzen, M. (2021). 1.5 °C degrowth scenarios suggest the need for new mitigation pathways. Nature Communications, 12(1), Article 1. https://doi.org/10.1038/s41467-021-22884-9

Kikstra, J. S., Li, M., Brockway, P. E., Hickel, J., Keysser, L., Malik, A., Rogelj, J., Van Ruijven, B., & Lenzen, M. (2024). Downscaling down under: Towards degrowth in integrated assessment models. Economic Systems Research, 1–31. https://doi.org/10.1080/09535314.2023.2301443

Kinchy, A., & Schaffer, G. (2018). Disclosure Conflicts: Crude Oil Trains, Fracking Chemicals, and the Politics of Transparency. Science, Technology, & Human Values, 43(6), 1011–1038. https://doi.org/10.1177/0162243918768024

Kolkman, D. (2022). The (in)credibility of algorithmic models to non-experts. Information, Communication & Society, 25(1), 93–109. https://doi.org/10.1080/1369118X.2020.1761860

Koppelaar, R. H. E. M., Keirstead, J., Shah, N., & Woods, J. (2016). A review of policy analysis purpose and capabilities of electricity system models. Renewable and Sustainable Energy Reviews, 59, 1531–1544. https://doi.org/10.1016/j.rser.2016.01.090

Krumm, A., Süsser, D., & Blechinger, P. (2022). Modelling social aspects of the energy transition: What is the current representation of social factors in energy models? Energy, 239, 121706. https://doi.org/10.1016/j.energy.2021.121706

Kuhnhenn, K. (2018). Economic growth in mitigation scenarios: A blind spot in climate science. Heinrich Böll Stiftung, 25. https://www.boell.de/sites/default/files/endf2_kuhnhenn_growth_in_mitigation_scenarios.pdf

Labussière, O., & Nadaï, A. (2015). Visions du futur et scénarios. In L’énergie des sciences sociales (pp. 37–66). Éditions A. Athéna. https://doi.org/10.4000/books.allianceathena.216

Lage, J., Thema, J., Zell-Ziegler, C., Best, B., Cordroch, L., & Wiese, F. (2023). Citizens call for sufficiency and regulation—A comparison of European citizen assemblies and National Energy and Climate Plans. Energy Research & Social Science, 104, 103254. https://doi.org/10.1016/j.erss.2023.103254

Latour, B. (1987). La Science en action. Éditions La Découverte. https://hal-sciencespo.archives-ouvertes.fr/hal-02057134

Latour, B., & Woolgar, S. (1979). La Vie de laboratoire: La production des faits scientifiques. Éditions La Découverte. https://hal-sciencespo.archives-ouvertes.fr/hal-02057284

Lauer, A., Capellán-Pérez, I., & Wergles, N. (2025). A comparative review of de- and post-growth modeling studies. Ecological Economics, 227, 108383. https://doi.org/10.1016/j.ecolecon.2024.108383

Lauer, A., de Castro, C., & Carpintero, Ó. (2024). Between continuous presents and disruptive futures: Identifying the ideological backbones of global environmental scenarios. Futures, 163, 103460. https://doi.org/10.1016/j.futures.2024.103460

Laurens, S. (2015). Les courtiers du capitalisme. Milieux d’affaire et bureaucratie à Bruxelles. Agone.

Lebaron, F. (2009). La formation des économistes et l’ordre symbolique marchand. In P. Steiner & F. Vatin, Traité de sociologie économique. PUF.

Li, M., Keyßer, L., Kikstra, J. S., Hickel, J., Brockway, P. E., Dai, N., Malik, A., & Lenzen, M. (2023). Integrated assessment modelling of degrowth scenarios for Australia. Economic Systems Research, 0(0), 1–31. https://doi.org/10.1080/09535314.2023.2245544

Li, M., Pelz, S., Lamboll, R., Wang, C., & Rogelj, J. (2025). A principle-based framework to determine countries’ fair warming contributions to the Paris Agreement. Nature Communications, 16(1), 1043. https://doi.org/10.1038/s41467-025-56397-6

LOCOMOTION H2020 project. (2019, October 23). Integrated assessment model. LOCOMOTION. https://www.locomotion-h2020.eu/locomotion-models/locomotion-iams/

Lund, H., Arler, F., Østergaard, P. A., Hvelplund, F., Connolly, D., Mathiesen, B. V., & Karnøe, P. (2017). Simulation versus Optimization: Theoretical Positions in Energy System Modelling. Energies, 10(7), Article 7. https://doi.org/10.3390/en10070840

Lunde, T., & Midttun, A. (1987). Electricity Forecasting in Norway: Administrative Centralism. In T. Baumgartner & A. Midttun, The politics of energy forecasting. A comparative Study of Energy Forecasting in Western Europe and North America. Oxford University Press.

Magee, C. L., & Devezas, T. C. (2017). A simple extension of dematerialization theory: Incorporation of technical progress and the rebound effect. Technological Forecasting and Social Change, 117, 196–205. https://doi.org/10.1016/j.techfore.2016.12.001

Matarasso, P. (2007). La construction historique des paradigmes de modélisation intégrée: William Nordhaus, Alan Manne et l’apport de la Cowles Commission. In Les modèles du futur (pp. 44–62). La Découverte. https://doi.org/10.3917/dec.dahan.2007.01.0044

Matti, C., Jensen, K., Bontoux, L., Goran, P., Pistocchi, A., & Salvi, M. (2023). Towards a fair and sustainable Europe 2050: Social and economic choices in sustainability transitions. Joint Research Center. https://doi.org/10.2760/804844

Menezes, T., Pottier, A., & Roth, C. (2023). The Two Sides of the Environmental Kuznets Curve: A Socio-Semantic Analysis. Œconomia. History, Methodology, Philosophy, 13–2, Article 13–2. https://doi.org/10.4000/oeconomia.15729

Mitchell, T. (2013). Carbon Democracy. Le pouvoir politique à l’ère du pétrole. La découverte.

Moglen, R., Chawla, K. P., Levi, P., Sun, Y., Phillips, O., Leibowicz, B. D., Jenkins, J. D., & Grubert, E. A. (2023). The state of macro-energy systems research: Common critiques, current progress, and research priorities. iScience, 26(4), 106325. https://doi.org/10.1016/j.isci.2023.106325

Morena, E. (2023). Fin du monde et petits fours: Les ultra-riches face à la crise climatique. La Découverte.

Morrison, R. (2018). Energy system modeling: Public transparency, scientific reproducibility, and open development. Energy Strategy Reviews, 20, 49–63. https://doi.org/10.1016/j.esr.2017.12.010

Nippert, A. (2024). Hydrogène mania. Enquête sur le totem de la croissance verte. Le passager clandestin.

Odenweller, A., Ueckerdt, F., Nemet, G. F., Jensterle, M., & Luderer, G. (2022). Probabilistic feasibility space of scaling up green hydrogen supply. Nature Energy, 7(9), 854–865. https://doi.org/10.1038/s41560-022-01097-4

Overland, I., & Sovacool, B. K. (2020). The misallocation of climate research funding. Energy Research & Social Science, 62, 101349. https://doi.org/10.1016/j.erss.2019.101349

Pahker, A.-K., Keller, M., Karo, E., Vihalemm, T., Solvak, M., Orru, K., Tammiksaar, E., Ukrainski, K., & Noorkõiv, M. (2024). What’s worse, communism or carbon? Using the Transitions Delphi approach to identify viable interventions for the Estonian energy transition. Energy Research & Social Science, 109, 103421. https://doi.org/10.1016/j.erss.2024.103421

Palmer, G. (2018). A Biophysical Perspective of IPCC Integrated Energy Modelling. Energies, 11(4), Article 4. https://doi.org/10.3390/en11040839

Parpan, G., Andrieu, B., Vidal, O., Delannoy, L., Boulzec, H. L., Gervais, M., Jegourel, Y., & Delalande, S. (2025). Examining copper supply feasibility in decarbonization pathways: A mine-level dynamic approach. https://eartharxiv.org/repository/view/8881/

Parrique, T. (2024). A response to Savin and van den Bergh: Ceci n’est pas degrowth. Timothée Parrique. https://timotheeparrique.com/a-response-to-savin-and-van-den-bergh-ceci-nest-pas-degrowth/

Parrique, T., Raworth, K., Liegey, V., Friends of the Earth Europe, European Environmental Bureau, European Youth Forum, & Wellbeing Economy Alliance. (2023a). Post-growth Europe: 400+ experts call for wellbeing economy. https://friendsoftheearth.eu/publication/post-growth-europe-letter/

Parrique, T., Raworth, K., Liegey, V., Friends of the Earth Europe, European Environmental Bureau, European Youth Forum, & Wellbeing Economy Alliance. (2023b). Response to European Commission repeats call for a post-growth Europe. https://friendsoftheearth.eu/publication/response-to-european-commission-repeats-call-for-a-post-growth-europe/

Pearce, F. (2024). Nations Are Undercounting Emissions, Putting UN Goals at Risk. Yale Environment 360. https://e360.yale.edu/features/undercounted-emissions-un-climate-change

Pelz, S., Rogelj, H., & Riahi, K. (2023). Evaluating equity in European climate change mitigation pathways. IIASA. https://pure.iiasa.ac.at/id/eprint/18830/1/report_equity_iiasa_euab%20(1).pdf

Peters, B. G., & Nagel, M. L. (2020). Zombie Ideas: Why Failed Policy Ideas Persist (1st ed.). Cambridge University Press. https://doi.org/10.1017/9781108921312

Plehwe, D. (2022). Reluctant transformers or reconsidering opposition to climate change mitigation? German think tanks between environmentalism and neoliberalism. Globalizations, 1–19. https://doi.org/10.1080/14747731.2022.2038358

Pinto, M. F. (2017). To Know or Better Not to: Agnotology and the Social Construction of Ignorance in Commercially Driven Research. Science & Technology Studies, 30(2), Article 2. https://doi.org/10.23987/sts.61030

Pircher, B. (2023). The EU’s Commission Regulatory Scrutiny Board: Better regulation or biased influence on legislation? AK Wien, Abteilung Europa und Internationales und Lobby Control. https://urn.kb.se/resolve?urn=urn:nbn:se:lnu:diva-120484

Pitron, G. (2018). La guerre des métaux rares. La face cachée de la transition énergétique et numérique. Les liens qui libèrent.

Polanyi, K. (2009). La grande transformation: Aux origines politiques et économiques de notre temps. Gallimard.

Pottier, A. (2016). Comment les économistes réchauffent la planète. Seuil.

Puiseux, L. (1987). The Ups and Downs of Electricity Forecasting in France: Technocratic Élitism. In T. Baumgartner & A. Midttun, The politics of energy forecasting. A comparative Study of Energy Forecasting in Western Europe and North America. Oxford University Press.

Regulation (EU) 2018/842 of the European Parliament and of the Council of 30 May 2018 on Binding Annual Greenhouse Gas Emission Reductions by Member States from 2021 to 2030 Contributing to Climate Action to Meet Commitments under the Paris Agreement and Amending Regulation (EU) No 525/2013 (Text with EEA Relevance), 156 OJ L (2018). http://data.europa.eu/eli/reg/2018/842/oj/eng

Regulation (EU) 2021/1119 of the European Parliament and of the Council of 30 June 2021 Establishing the Framework for Achieving Climate Neutrality and Amending Regulations (EC) No 401/2009 and (EU) 2018/1999 (‘European Climate Law’), 243 OJ L (2021). http://data.europa.eu/eli/reg/2021/1119/oj/eng

Regulation (EU) 2023/857 of the European Parliament and of the Council of 19 April 2023 Amending Regulation (EU) 2018/842 on Binding Annual Greenhouse Gas Emission Reductions by Member States from 2021 to 2030 Contributing to Climate Action to Meet Commitments under the Paris Agreement, and Regulation (EU) 2018/1999 (Text with EEA Relevance), 111 OJ L (2023). http://data.europa.eu/eli/reg/2023/857/oj/eng

Roubelat, F. (1993). La prospective: Méthodes et outils. Téoros: Revue de recherche en tourisme, 12(2), 7. https://doi.org/10.7202/1078019ar

Rouchier, J. (2020). Modéliser l’environnement avec l’économie: Deux approches. In È. Chiapello, A. Missemer, & A. Pottier (Eds.), Faire l’économie de l’environnement (pp. 177–196). Presses des Mines. https://doi.org/10.4000/books.pressesmines.6520

Royston, S., & Foulds, C. (2021). The making of energy evidence: How exclusions of Social Sciences and Humanities are reproduced (and what researchers can do about it). Energy Research & Social Science, 77, 102084. https://doi.org/10.1016/j.erss.2021.102084

Royston, S., Foulds, C., Pasqualino, R., & Jones, A. (2023). Masters of the machinery: The politics of economic modelling within European Union energy policy. Energy Policy, 173, 113386. https://doi.org/10.1016/j.enpol.2022.113386

Sakai, M., Brockway, P. E., Barrett, J. R., & Taylor, P. G. (2019). Thermodynamic Efficiency Gains and their Role as a Key ‘Engine of Economic Growth’. Energies, 12(1), Article 1. https://doi.org/10.3390/en12010110

Saltelli, A., Arnald, P., & Di Fiore, M. (2024). Mathematical models: A state of exception. International Review of Applied Economics. https://doi.org/10.1080/02692171.2024.2365727

Saujot, M., Dubuisson-Quellier, S., Martin, S., Barbier, C., Pourouchottaminh, P., Pascal, C., Cécile, D., Hajjar, J., Podesta, G., Thiriot, S., Gallic, T. L., Branche, S. L., Laurent, D., Gersende, C., & Julien, D. (2022). Pour une meilleure intégration des dimensions sociales et des modes de vie dans les exercices de prospective environnementale. Iddri - Sciences Po. https://sciencespo.hal.science/hal-03614943

Saurugger, S. (2002). L’expertise: Un mode de participation des groupes d’intérêt au processus décisionnel communautaire. Revue française de science politique, 52(4), 375–401. https://doi.org/10.3917/rfsp.524.0375

Savin, I., & van den Bergh, J. (2024). Reviewing studies of degrowth: Are claims matched by data, methods and policy analysis? Ecological Economics, 226, 108324. https://doi.org/10.1016/j.ecolecon.2024.108324

Schweitzer, J., & Mix, T. L. (2022). It Is a Tradition in the Nuclear Industry... Secrecy: Political Opportunity Structures and Nuclear Knowledge Production in France. Sociological Research Online, 27(2), 504–524. https://doi.org/10.1177/13607804211025052

Silvast, A., & Foulds, C. (2022). Sociology of Interdisciplinarity: The Dynamics of Energy Research. Springer Nature. https://doi.org/10.1007/978-3-030-88455-0

Souffron, C., & Jacques, P. (2024). A Successful Assessment of the Economic Impacts of Ecological Transition Policies in the EU Requires the European Commission to Broaden the Range of Its Modelling Tools (SSRN Scholarly Paper 4640677). https://doi.org/10.2139/ssrn.4640677

Sovacool, B. K., Axsen, J., & Sorrell, S. (2018). Promoting novelty, rigor, and style in energy social science: Towards codes of practice for appropriate methods and research design. Energy Research & Social Science, 45, 12–42. https://doi.org/10.1016/j.erss.2018.07.007

Spragg Nilsson, T. (2024). 2040 Target: ​​Increased ambition welcomed – but 90% reduction is not enough. CAN Europe. https://caneurope.org/2040-target-reaction/

Süsser, D., Ceglarz, A., Gaschnig, H., Stavrakas, V., Flamos, A., Giannakidis, G., & Lilliestam, J. (2021). Model-based policymaking or policy-based modelling? How energy models and energy policy interact. Energy Research & Social Science, 75, 101984. https://doi.org/10.1016/j.erss.2021.101984

Süsser, D., Martin, N., Stavrakas, V., Gaschnig, H., Talens-Peiró, L., Flamos, A., Madrid-López, C., & Lilliestam, J. (2022). Why energy models should integrate social and environmental factors: Assessing user needs, omission impacts, and real-word accuracy in the European Union. Energy Research & Social Science, 92, 102775. https://doi.org/10.1016/j.erss.2022.102775

The Shift Project. (2019). Scénarios énergie-climat. Évaluation et mode d’emploi. The Shift Project. https://theshiftproject.org/article/scenarios-energie-climat-evaluation-mode-emploi-rapport-shift/

Traité de Lisbonne modifiant le traité sur l’Union européenne et le traité instituant la Communauté européenne, signé à Lisbonne le 13 décembre 2007, 306 OJ C (2007). http://data.europa.eu/eli/treaty/lis/sign/fra

Vadén, T., Lähde, V., Majava, A., Järvensivu, P., Toivanen, T., Hakala, E., & Eronen, J. T. (2020). Decoupling for ecological sustainability: A categorisation and review of research literature. Environmental Science & Policy, 112, 236–244. https://doi.org/10.1016/j.envsci.2020.06.016

Vieille Blanchard, É. (2007). Croissance ou stabilité? L’entreprise du Club de Rome et le débat autour des modèles. In Les modèles du futur (pp. 19–43). La Découverte. https://doi.org/10.3917/dec.dahan.2007.01.0019

  1. In the diagram below, I did not include national administrations. That is mainly because an exhaustive research on climate mitigation scenarios for the 27 member states is no work for a single person. ↩︎

  2. Spoilers: it does. ↩︎

  3. This narrative is widespread in official documents, and is most tangible in the Better Regulation toolbox, a 600-pages document providing guidelines and tools for "evidence-informed policymaking" (EC 2023c). ↩︎

  4. Peters and Nagel (2020) call this conception causal beliefs: the causal links established are to be taken literally. ↩︎

  5. Although it will be impossible to tell before a few years, because global warming is a tendency, not an event. ↩︎

  6. Especially if, as with the EU's net zero target for 2050, the threshold is presented as the last effort required by EU countries to do their fair share of climate change mitigation. ↩︎

  7. More precisely, per my terminology, retroforecasting. ↩︎

  8. Although the validity of the second study has been strongly contested by degrowth researchers (see Parrique 2024). ↩︎

  9. Initially, energy policy was mostly about energy security, and more specifically fossil fuel reserves. ↩︎

  10. IAMs also often include a "land sector" module, which allows to quantify GHG emissions from land use change and forestry. Most other models do not. ↩︎

  11. An extreme example of this can be found in Andres et al. (2024), where the authors assess the "welfare effects" of "degrowth policies", where "welfare" actually means GDP. Of course, they find that degrowth is undesirable because it reduces GDP. ↩︎

  12. For many foresight researchers, originality is considered as necessary for scenarios (The Shift Project 2019). But originality is relative to the observer and their objectives. ↩︎

  13. Which, let's not forget, is an unfair and unsustainable policy target. ↩︎

  14. The GEM-E3 documentation does not mention "rebound effects" (Capros et al. 2013), but it could be modelled without naming it. ↩︎

  15. And let's not forget that EU climate targets are unfair and unsustainable. ↩︎