Press release

EU taxonomy will lead to new nuclear projects and more expensive gas

The impact of the new EU taxonomy – the system that classifies which investments are sustainable – is likely to support the growth of nuclear power in some countries but bringing new gas online within the defined emissions limit of 270g CO2e/kWh limit may be unrealistic, according to Rystad Energy research.

The challenge for gas is twofold: abatement technologies are not yet market-ready at scale and their costs, once implemented are likely to boost prices further. The resulting spike in production costs suggests that even existing gas power plants could become economically unviable, according to our research. Hydrogen blending and carbon capture and storage (CCS) are the primary options for gas emissions reductions with the latter unlikely to be a viable option before the mid-2030s. Rystad Energy’s research also shows a 800,000 tonnes potential gap in hydrogen production targets between REPowerEU, an earlier initiative aimed at increasing energy security, and what is implicit in the taxonomy. This discrepancy might make hydrogen blending for power generation a challenge.

For natural gas, the EU taxonomy sets out strict guidelines for any new power plant, clarifying that projects should replace an existing solid or liquid fossil fuel power plant, without exceeding the preexisting capacity by more than 15%. In practice, this means that if a gas power plant would take a minimum of four years to build, the earliest replacement of a coal power plant in the EU would be in 2026. Rystad Energy expects 44.6 gigawatts (GW) of coal capacity to shut down in the region between 2026 and 2031 – which could be replaced by a maximum of 51.3 GW of new gas power capacity.

For nuclear, a gradual phaseout towards 2040 is expected, but unlike gas, its emissions are negligible and new plants have no requirement to replace coal capacity under the taxonomy, instead, being only required to adhere to safety and waste disposal standards.

Europe is between a rock and a hard place – the taxonomy is an attempt to soften the squeeze. Germany, for example, is considering a potential U-turn on its aggressive nuclear phase-out over energy fears, and countries across the continent are ramping up coal use to reduce gas dependency. Despite the taxonomy’s ambitions, European countries are set to prioritize energy security in the short term. In the longer term however, we expect renewable energy and storage to become the staple for the European energy system.

Lars Nitter Havro, senior analyst, Rystad Energy

Find out more in PowerCube

The European power mix – until the war in Ukraine – has been characterized by a growing variable renewable energy presence, a declining share of the least preferred energy sources such as coal, and a view of natural gas as a transition fuel. Moving forward, gas will still play a central role during the transition of the European power mix towards renewables. However, for gas power plants granted a construction permit prior to 31 December 2030, the taxonomy imposes a direct emissions limit of 270g CO2e/kWh. There is also a 100g CO2e/kWh lifetime emissions limit for power plants after 2030, which would spike costs significantly or decrease the plant’s annual operative hours considerably. Therefore, to assess the potential impact that the EU taxonomy on the cost of power from gas, we need to take into account the economic ramifications of abatement.

Hydrogen and CCS are the primary options for emissions reductions

For hydrogen, in light of the recent REPowerEU initiative (which sets targets for exclusively green hydrogen to be used within the EU), we applied our updated cost forecasts for green hydrogen to determine the cost of production. Adequate reduction in emissions using hydrogen equates to 44% by volume mix of hydrogen, and we use a carbon price of $100 per tonne, which would push power generation costs up to around €75-€80 ($75-$80) per MWh – almost double the unabated cost of generation.

For CCS, a new gas power plant replacing coal would require a ready-to-go CO2 transport and storage infrastructure. This does not yet exist in Europe and is unlikely to do so until the early 2030s. We give an indication of the potential costs for CCS on a gas power plant based on Rystad Energy’s CCS levelized cost models (see figure below) for comparison only. We note that post-2035, all power plants will be incentivized to run on zero carbon gases, which may potentially exclude CCS as a viable candidate although this remains to be seen. Additionally, for energy security purposes, CCS does not lower demand for natural gas, whereas hydrogen considerably lowers it.

If developed to the maximum limit and abated using exclusively green hydrogen, the 51.3 GW of available capacity for new gas power plants translates into a demand for approximately 0.9 million tonnes of hydrogen per year (assuming a 40% capacity factor). However, the REPowerEU initiative – which sets out a demand target of 20 million tonnes of hydrogen per year by 2030 – has only earmarked 0.105 million tonnes of hydrogen for power generation by 2030. Ergo, there is a potential misalignment between the two, amounting to around 800,000 tonnes of hydrogen demand per year.

Should the REPowerEU be the one validated out of the two statements, we expect two things: firstly, there is likely going to be a supply shortage due to the extremely tall hydrogen order, and secondly, that the amount earmarked for power production would only amount to 5.1 GW of coal capacity actually being replaced (with around 5.9 GW of new gas generation). Given the expected hydrogen supply shortage, we believe the 5.9 GW of new gas power capacity is more probable. As this would push the cost of gas generation up significantly, we expect these power plants would move from baseload providers into peaker-plant territory, with the vast increase of renewables under the REPowerEU plan covering most of the coal shutdown.

Nuclear: emissions are less relevant

In 2021, lifecycle emissions generated from nuclear power plants emitted 99.7% less eqCO2 /TWh than European gas power plant, and 99.8% less than European coal power plants. And this despite making up a substantial proportion of the European power mix. As such, the EU taxonomy outlines which types of nuclear power plants fall within an acceptable range, primarily based on safety. Existing nuclear power plants that have Generation III-type reactors will be included in the taxonomy if the power plants are frequently upgraded to the latest safety and waste disposal standards. While Generation III reactors can be defined in different ways, these are generally understood to have low waste production, high efficiencies, passive safety systems, and have average lifetimes of about 60 years. Built since the mid-1980s, only about 14.5 GW of European reactors are likely to be considered Generation III. And under the taxonomy, after 2040, these plants will no longer fit the bill. The taxonomy includes Generation III+ and Generation IV reactors as well. New Generation III+ reactors have a green light until 2045, whereas Generation IV reactors are included indefinitely. By 2035, the inclusion of these types of nuclear plants would have a small effect on the addition of nuclear power in the EU.

Rystad Energy expects about 48.28 GW of today’s European nuclear fleet to be decommissioned by 2035 – with only 26.51 GW of capacity added over this period – making a net nuclear generation loss of 21.77 GW. Considering most of the plants earmarked for shutdown in the EU are Generation II reactors, the taxonomy will not save these projects. It is more likely that Generation III+ plants that have been proposed will become viable, and we assume that at most, the EU taxonomy could bring about 20-25 GW of new capacity. The EU taxonomy might also save Generation III reactors from being decommissioned, but these projects were likely to be around for a longer time regardless. Meanwhile, Generation IV reactors have been in discussion since the early 2000s but are still only in conceptual stage. Commercialization, planning, and construction is unlikely before 2035. While nuclear has no explicitly stated goal of replacing coal, countries that have favorable nuclear programs could see this as an opportunity to replace coal plants with more reactors rather than natural gas.

Rystad Energy believes this would have a more lasting effect post-2035 in Eastern EU countries, where funding for a new Generation III+ reactor under the green taxonomy could offset the project’s costs in the first few years of operation and replace coal. Also, it is highly likely that the increased funding for nuclear would kickstart Generation IV plants, which could come online in the 2040s. In addition to ramping up new installations, there is approximately 9.4 GW of dormant nuclear power plant capacity that could realistically be restarted in Europe, with a lead time of around 12 months on average. These power plants are distributed across Germany, Sweden, and Spain.


Lars Ivar Nitter Havro
Senior Analyst
Phone: +47 24 00 42 00 ‎

Victor Ponsford
Media Relations Manager
Phone: +47 94 97 49 77

About Rystad Energy
Rystad Energy is an independent energy research and business intelligence company providing data, tools, analytics and consultancy services to the global energy industry. Our products and services cover energy fundamentals and the global and regional upstream, oilfield services and renewable energy industries, tailored to analysts, managers and executives alike.

Rystad Energy’s headquarters are located in Oslo, Norway with offices in London, New York, Houston, Aberdeen, Stavanger, Rio de Janeiro, Singapore, Bangalore, Tokyo, Sydney and Dubai.