Could you elaborate on the technical foundations of your methodology, particularly what makes it novel compared to existing approaches, and which data sources were used to validate your approach?
Théo Wittersheim: Our methodology is based on two core principles: analysing dependencies from the perspective of physical flows (for instance, barrels of oil, tonnes of steel, etc.) and monitoring these flows across entire value chains. In this way, we do not stop at the final product crossing a border or reaching a consumer; rather, we map every step in the value chain—from extraction, transformation and transport to manufacturing, trade and final use.
In terms of energy dependencies, this methodology enables us to capture the total energy mobilised globally to support both European economies and consumption. This is how we built our metric, “energy exposure”, estimated at around 14,500 TWh for the European Union in 2022 (the most recent year available in data). This approach allows us to better understand the true level of dependencies and the associated risks, particularly of a geopolitical nature. In public debates on dependencies, the focus often remains limited to monetary flows, direct imports or domestic production (mostly electricity), which accounts for only around 20% of the EU’s final energy consumption. This overlooks a substantial share of upstream dependencies, where many critical vulnerabilities lie.
Recent geopolitical developments illustrate this clearly. Even when Europe is not directly importing energy from a given region, it often remains indirectly dependent through the goods and services it imports. In that case, the dependency is displaced upstream in the value chain: for instance, European industrial products may rely on raw materials imported from Asia, which in turn rely on oil or gas transiting through strategic chokepoints such as the Strait of Hormuz. Without this value-chain perspective, both the scale and the nature of our exposure to risk are systematically underestimated.
Our metric, energy exposure, is similar to a carbon footprint, but it differs from it in two major ways. First, it looks at value chains from an energy perspective rather than a carbon perspective, which provides a more relevant assessment of energy risks and better informs decision-making on strategic autonomy. Second, energy exposure is calculated using a novel hybrid energy model which not only relies on input–output tables (EXIOBASE, GLORIA), but also on energy balance tables (IEA) and trade data (Eurostat, UN Comtrade, EIA, Energy Institute). This provides more detailed information on the energy embodied in the goods and services we consume. For instance, if China consumes electricity to produce manufactured goods exported to Europe, what share of that electricity was generated using coal or natural gas, and where were the coal and gas extracted? This allows us to create a macro-level map of global energy production and consumption, showing which energy suppliers we are connected to, which sectors of our economy rely on them, and which strategic uses are most exposed—such as mobility and household heating, or firms’ production and exports.
We break down the EU’s 14,500 TWh of energy exposure into three components: domestically produced energy (23%, of which 24% is fossil fuels), directly imported energy (54%, of which 98% is fossil fuels), and energy embedded in imported goods and services (23%, of which 81% is fossil fuels). This highlights how fossil dependence extends along the value chain.
This perspective has direct policy implications. A purely final-product view fails to capture systemic risk. For example, while measures such as the 2022 French tariff shield protected households from part of the direct energy price shock, inflation continued to be imported indirectly through goods and services produced outside France’s borders. By identifying where our exposure truly lies, this methodology helps to better target vulnerabilities, anticipate second-order effects, and avoid policies that merely shift risk rather than reduce it—particularly in a context where Europe remains highly dependent on mature oil and gas producers (the EU imports around 97% of its oil and 90% of its gas), and where control over access to energy is increasingly used as a lever in geopolitical power dynamics. The decline of the North Sea has significantly increased Europe’s vulnerability.
If further developed as a foresight tool, energy exposure could go beyond describing dependencies to actively improve policy steering—helping to prioritise vulnerabilities across strategic value chains and assessing the extent to which different policy options under debate would affect the EU’s strategic autonomy.
The EU’s economic resilience depends on the ability of its Member States to reduce their own vulnerabilities. Energy dependency is unequal among the Member States, and if some European countries remain highly dependent on oil or gas imports, this will compromise the decarbonization of other countries and of the European Union as a whole. The European market will remain exposed to significant energy risks, which could lead to imported inflation or supply chain disruptions for all member countries. This is why measuring energy exposure at EU level alone is not sufficient to fully understand dependencies: the exercise needs to be systematically replicated at national level. In this respect, the Governance Regulation could be a relevant vehicle to integrate an energy exposure assessment at the core of National Energy and Climate Plans, thereby improving how these dependencies are understood and addressed.
Looking beyond 2030, how do you anticipate declining fossil fuel use and increasing reliance on green energy imports will reshape Europe’s energy dependency profile?
Théo Wittersheim: Fossil fuel extraction in Europe has been steadily declining since the early 2000s. The decline of the North Sea is irreversible and will keep increasing Europe’s vulnerability as long as we continue to rely heavily on fossil fuels—which represented 74% of the EU’s energy exposure in 2022.
Our analyses of oil and gas supplies point to a risk horizon for Europe as early as the 2030s for oil, and around the 2040s for gas. The combined effects of resource depletion and supply concentration create significant geopolitical dependency risks. If Europe does not decarbonise for climate reasons, it will ultimately be forced to do so for physical reasons. If this transition is not anticipated and managed, it is likely to occur under severe constraints, with serious geopolitical, economic and social consequences.
If the EU manages to decarbonise, it can regain control over a significant share of its energy exposure. However, its 74% reliance on fossil energy cannot be easily replaced by low-carbon energy sources. Detailed analysis is needed to estimate Europe’s low-carbon production potential by 2050, and the implications in terms of constraints on consumption. In the case of France, electrification—of buildings, transport and industry—is one of the most essential levers for reducing dependence on fossil fuels, alongside efficiency and sufficiency measures. This is expected to be important for most, if not all, EU Member States. Furthermore, efficiency will be required to reduce energy demand, and sufficiency may also be needed in order to match demand with supply.
Do your findings suggest that Europe is shifting from fossil fuel dependency towards new forms of dependency, for example on critical materials or external clean energy supply chains?
Théo Wittersheim: The main difficulty in producing such a mapping is access to sufficiently detailed, high-quality data. We also need to compare orders of magnitude when it comes to dependencies. No energy system is completely free of vulnerabilities.
At first order, the key point is to compare the timescales of our dependence on fossil fuels—IEA member states have around three months of strategic stocks of oil and gas—with the timescales of our material dependencies: solar panels or wind turbines have a lifespan of more than 20 years, and electric vehicle batteries typically last more than 10 years. This means that, if lithium were no longer available on the market to produce new EV batteries, existing batteries could still operate for years. The same is true for renewable technologies. This should be contrasted with disruptions to fossil fuel trade, which could affect us within weeks to months. This is not the same level of dependency.
Decarbonisation requires increasing quantities of copper, whether for electricity generation or for the electrification of end uses. This is expected to be true for many critical materials, with different stakes for each, depending on the technologies they are required for. For example, in France, road transport electrification is by far the largest driver of copper demand in decarbonisation, followed—though to a lesser extent—by the expansion of electricity generation capacity. Key electrification technologies—such as electric vehicles, batteries and renewable energy infrastructure—are all highly copper-intensive. Their rapid deployment over the next decade is therefore expected to significantly increase copper consumption. A more systematic view is required to understand the role of each material.
Therefore, the transition raises broader questions linked to industrial realities and the extent to which a circular economy can improve our control over dependencies. Demand-side measures—such as modal shifts away from cars, controlling vehicle size, or managing the expansion of data centres—also need to be considered, as they could play a key role in mitigating these pressures and securing the transition. In that sense, the challenge is not simply a shift from fossil to material dependency, but the need to anticipate and manage a new set of physical constraints across both energy and material systems.
Do your findings suggest any consistencies or inconsistencies between EU climate policies and trade policies, particularly when considering indirect fossil energy dependencies embedded in global value chains?
Théo Wittersheim: Our findings point to several risks of tension between EU climate and trade policies, when viewed through the lens of physical energy dependencies.
First, diversifying suppliers of oil and gas does not necessarily mean securing supply, as we rely mostly on mature producers, which creates a twofold pressure: geopolitical and geological. One of our previous analyses—carried out under the auspices of France’s Ministry of Defence and based on Rystad Energy data—shows that 14 of the EU’s 16 main oil and gas suppliers are mature producers, with the cumulative oil production of these 16 suppliers potentially declining by up to 50% by 2050. This indicates a structural risk that cannot be overlooked. If the oil and gas production of most of our suppliers were to decrease sharply in the future, then decarbonisation should not be seen only as a climate imperative, but also as a strategic necessity to reduce exposure to declining and increasingly constrained resources.
Unfortunately, this sort of analysis is rarely conducted because the underlying data are private and often unaffordable. Institutional studies on this topic are required to raise awareness and should be systematised. For instance, this could be integrated into EU energy security frameworks—either as part of a revision of energy security criteria—or as a dedicated indicator within National Energy and Climate Plans under the Governance Regulation.
Second, trade relationships are inherently reciprocal: they involve mutual commitments over time. This raises a fundamental question of consistency. Trade agreements with countries heavily dependent on fossil fuels must be designed and negotiated with climate considerations in mind. They should serve to accelerate the decarbonisation of partner countries, rather than a way of externalising our responsibility or masking our exposure to energy risks—which will remain present, even if externalised. Similarly, trade agreements involving fossil fuel imports must be aligned with our decarbonisation targets and should not lock the EU into continued dependence on fossil fuels and on specific suppliers. Finally, trade agreements that lead to increased European exports should be consistent with the resulting energy consumption, to ensure that this does not create new energy dependencies.
97% of the petrol consumed in the EU is imported. Which sectors should the EU focus on to reduce our energy dependence?
Théo Wittersheim: Some sectors may not consume large quantities of oil or gas themselves but may nevertheless be heavily reliant on other sectors that are, in turn, heavily dependent on them. We have to think in terms of value chains. If the first domino falls, which others are likely to follow, given their lack of short-term alternatives?
Petrol is mostly used for two purposes. In industry, it is used as a key input for the chemical sector. For instance, many polymers are derived from petrol (often referred to broadly as “plastics”), and large-scale substitution is not easy to achieve. In that sense, the circular economy will have an important role to play, as will reducing demand.
Outside industry, petrol is mostly used to power vehicles. The transport and agricultural sectors are heavily dependent on refined petroleum products. If an oil shortage makes it difficult for lorries to operate, industry may no longer be able to produce or sell, and cities may no longer be supplied with essential goods.
This leads to a final way of framing the question. It is not only about the sectors that consume the most petrol, but also about the final needs that require petrol to be met. Assessing the energy dependencies linked to the different needs of a population, an economy or a country is essential in order to determine which ones must be secured first against future fossil fuel-related risks, and at what cost.
How can the EU face, and adapt to, a decrease in petrol production and petrol supply?
Théo Wittersheim: The issue must be approached through a physical-flow lens and a genuinely systemic analysis. From that perspective, the priority should be to decarbonise and reindustrialise strategic sectors based on a physical assessment of value chains and uses. “Decarbonise” means mobilising all relevant low-carbon levers for the Member States that have an interest in them—renewables, nuclear—while recognising the constraints of these levers. It may therefore also imply efficiency and sufficiency measures, in order to close the physical equation. The exact proportions remain to be determined through further foresight analysis, to document the choices that need to be made.
This also means improving the physical consistency of scenarios and foresight exercises. Today, many prospective exercises still contain major inconsistencies or weak forms of physical balancing. This is precisely the focus of another stream of work we have launched at The Shift Project, where we are seeking to involve the foresight community and sectoral stakeholders to improve coherence and comparability across scenarios.
In this respect, the ongoing reflection on the EU’s energy security architecture and the Governance Regulation could provide a useful vehicle for such topics. These frameworks could integrate a more explicit physical-flow and below-ground assessment of oil and gas supply under the energy security dimension. Energy exposure could be used as a tool to better assess the decarbonisation dimension within the Governance Regulation, at the heart of National Energy and Climate Plans. Likewise, improving foresight methodologies and their physical consistency could become a core component of the Regulation, both to strengthen the quality of NECPs and to support the European Commission’s evaluation processes.
In short, adapting to declining oil and gas supply is not only about finding substitutes. It is about understanding where these resources truly sit in the economy, identifying which functions are genuinely strategic, and building a physically coherent, sectoral and systemic transition pathway that reduces vulnerability while preserving the essential functions of society.