“Nuclear power is necessary for rich countries to reach net zero by 2050.”
Nobody seriously disputes that nuclear is low-carbon. The fight is over necessity: can a grid run on wind, solar, batteries and other flexible resources reach net zero as cheaply and reliably, or does it need "firm" clean power that runs regardless of weather?
Modelling studies tend to find that firm low-carbon sources cut the cost of deep decarbonisation, but nuclear is only one such source (alongside geothermal, hydro, long-duration storage and gas with carbon capture). Western nuclear projects have run years late and billions over budget, while solar and batteries keep getting cheaper. The answer may differ by country: necessary in some places, unnecessary in others.
Strong reading: net zero is technically infeasible without nuclear. Weak reading: reaching it without nuclear would be substantially costlier or riskier. Most of the real debate is about the weak reading.
Net greenhouse-gas emissions of zero across the whole economy by 2050, as pledged by the EU, UK, US (until 2025) and others. Electricity is the easiest sector, but electrification makes it the backbone.
Systems modelling consistently finds that having firm low-carbon power available lowers the cost of a zero-carbon grid, often dramatically, because wind and solar alone require massive overbuilding and storage to cover long windless, dark periods.
Sepulveda and colleagues modelled 1,000+ scenarios for two US regions: including firm low-carbon resources lowered the cost of zero-emission electricity by 10–62%. Jenkins, Luke and Thernstrom's review of 40 studies reached a similar conclusion.
Sepulveda, Jenkins, de Sisternes & Lester (2018). The Role of Firm Low-Carbon Electricity Resources in Deep Decarbonization. Joule, 2(11), 2403–2420; Jenkins, Luke & Thernstrom (2018). Joule, 2(12), 2498–2510.
Limitation
"Firm low-carbon" includes geothermal, gas with CCS and long-duration storage. The models show the value of firmness, not of nuclear specifically.
Countries that built nuclear at scale decarbonised their electricity decades ago. France and Sweden reached very low-carbon grids by combining nuclear and hydro, something no country has done with wind and solar alone.
French grid electricity typically emits well under 60 gCO2/kWh, with around 65–70% from nuclear; Germany, after phasing out its last reactors in April 2023, remains several times more carbon-intensive per kWh.
RTE (French grid operator) annual electricity reports; Ember, European Electricity Review (2025).
Limitation
France's buildout happened in the 1970s–80s under conditions (state planning, standardised designs) that may not be reproducible. Hydro-rich countries also decarbonised without nuclear (Norway, Iceland).
Major energy agencies' net-zero pathways roughly double global nuclear capacity. Removing it makes the remaining pathways harder and costlier.
The IEA's Net Zero by 2050 roadmap has nuclear output roughly doubling between 2020 and 2050; the IEA estimated that a low-nuclear path would require more investment and raise consumer bills.
IEA (2021). Net Zero by 2050; IEA (2022). Nuclear Power and Secure Energy Transitions.
Limitation
Global pathways are not national requirements, and they rely on assumptions about nuclear costs that recent Western projects have exceeded.
New nuclear in Western countries is extraordinarily expensive and slow. Money spent on reactors that arrive in the late 2030s could cut far more emissions if spent on renewables and storage now.
Lazard's 2025 analysis puts unsubsidised nuclear at $141–220/MWh versus $37–86/MWh for onshore wind. Flamanville 3 connected to the grid in 2024, twelve years late and about four times over budget; Hinkley Point C is expected around 2030 at over £30 billion (2015 prices); Vogtle 3–4 cost over $30 billion.
Lazard (2025). Levelized Cost of Energy+; Cour des comptes (2025) on Flamanville; EDF (Jan. 2024) Hinkley Point C update.
Limitation
Levelised costs ignore the system cost of intermittency (storage, backup, transmission). Costs fall with serial builds, as in South Korea and the UAE's Barakah plant.
Detailed studies show 100% renewable systems can be reliable and affordable with enough transmission, storage and flexible demand. Solar and battery costs have fallen far faster than forecasts, and are likely to keep falling.
Studies by Breyer and colleagues and others model 100% renewable energy systems for many regions at costs comparable to fossil systems. Battery pack prices fell roughly 90% from 2010 to 2024 (BloombergNEF).
Breyer et al. (2022). On the History and Future of 100% Renewable Energy Systems Research. IEEE Access, 10, 78176–78218; BloombergNEF Battery Price Survey (2024).
Limitation
Critics (e.g. Clack et al., 2017, PNAS) argue some 100% renewable studies rely on implausible assumptions about hydro, storage and transmission build rates.
"Necessary" fails on its face: firm clean power can come from other sources. Enhanced geothermal, long-duration storage, hydro, interconnection and gas with carbon capture can all play nuclear's role in different places.
Fervo Energy's enhanced geothermal pilots in Nevada and Utah achieved commercial-scale flow rates in 2023–2024 using techniques from oil and gas drilling, and long-duration storage projects (e.g. iron-air batteries) are entering deployment.
Fervo Energy, Project Red and Cape Station results (2023–2024); Form Energy deployments (2025).
Limitation
These technologies are less proven at scale than nuclear. Betting on them carries its own risk.
Nuclear power has very low lifecycle carbon emissions, comparable to wind.
Recent nuclear projects in Western Europe and the US have suffered severe cost and schedule overruns.
Some firm or long-duration low-carbon capacity lowers the cost of a fully decarbonised grid.