“It will be difficult to find a company to build a power plant in Switzerland”

The Federal Council is in favour of a return to nuclear power in Switzerland. How can we explain the resurgence of this energy source, which was thought to have been shelved after Fukushima?

Switzerland’s electricity supply is facing a challenge: consumption is increasing, particularly due to society’s electrification (mobility, heating, industry, data centres). At the same time, environmental strategy requires a reduction in greenhouse gas emissions. In this context, nuclear power appears to be a solution. It produces a lot of electricity while emitting very little CO₂. Another advantage is that nuclear reactors generate energy continuously, which can help balance the system, whereas renewable energies (wind and solar) only produce electricity at certain times.

Is the main obstacle to its revival the Swiss population’s reluctance towards nuclear energy?

This is indeed a key issue, as the question will have to be put to a vote. It should be noted that the rejection of nuclear energy is different from the (partial) rejection of renewable energies. In the case of wind power, it is mainly local opposition – people who do not want to live near wind turbines and want to preserve the landscape – while at cantonal and national level, the population tends to be more in favour of wind turbines. With nuclear power, it is somewhat the opposite. Residents of municipalities with nuclear power plants in Switzerland and who benefit from advantages such as lower taxes would be more willing to accept a new reactor near their homes. At the cantonal and federal levels, however, the population is much more divided. The reason is simple: in the event of a nuclear accident, it is not only the population in the immediate vicinity of the reactor – that have accepted this risk – that would be affected, but a large part of Switzerland.

Are Swiss nuclear power plants safe, given that our country has the world’s oldest nuclear plant, Beznau, which was commissioned in 1969?

Switzerland has special regulations in this area: reactors do not have a predefined service lifespan, but must be constantly modernised to meet the latest safety requirements. Our power plants are therefore safer today than when they were built. If we decide to build new reactors, they will theoretically be even safer. The risk of an accident is therefore minimal. However, it is not zero, and the consequences of an accident would be dramatic: ‘Very low risk, very high impact’. The assessment and acceptance of such a risk are highly subjective. They depend on each individual’s assessment. Are we prepared to accept an extremely low risk that could have enormous consequences? It is up to the population to decide.

According to a study by the Association des entreprises électriques suisses (AES), the umbrella organisation for the Swiss electricity sector, electricity consumption is likely to rise to 90 terawatt hours per year by 2050, compared with 57 terawatt hours in 2024. Can this demand be met without nuclear energy?

In principle, yes. It all depends on how quickly expansion takes place and how well the energy system is structured and managed. Smart grids, for example, make it possible to adjust consumption to production, and, in conjunction with battery storage, to balance out daily fluctuations in electricity production and consumption or to compensate for peaks in production and consumption. The real problem with renewables is storing the photovoltaic electricity generated in summer until winter. Solar energy is much more abundant in summer than in winter. This requires incentives in the electricity market to ensure that the right technologies are further developed and, above all, that existing ones are also used.

Nuclear power can help solve this problem, but it is not a panacea: it generates continuously even in winter and provides significant winter electricity, but it is – mainly for economic reasons – poorly suited to smoothing out fluctuations in production and, in particular, consumption. Band energy must also be distributed across fluctuating consumption using storage systems. Grid expansion is needed not only to absorb production peaks, but also to cope with increasing consumption peaks caused by heat pumps and the charging of electric cars. There is no clear answer to the question of whether an energy system with or without nuclear power is most economical.

Would building a new power plant not be as profitable as is claimed?

Building a new power plant represents a considerable financial risk for investors and operators. Most of the costs are related to construction, while operating costs remain relatively low. Profitability therefore depends heavily on construction costs and the utilisation rate of the plant. Today, the amortisation period for a reactor is generally estimated at 60 years. However, in a liberalised energy market dominated by renewables, the capacity utilisation is uncertain, especially in summer. This is because electricity surpluses from solar power plants are to be expected at that time. If you have to shut down your power plant in summer because electricity prices are very low or even negative, the average electricity costs per kilowatt hour increase accordingly.

Furthermore, the costs of a power plant depend not only on the construction costs (ed. note: French auditors estimate the total cost of the EPR reactor in Flamanville, which went into operation in December 2024, at €23.7 billion), but also on interest rates over 60 years. However, it is impossible to predict how these will evolve.

While it is certainly not fundamentally different in relation to a wind farm, firstly, the amortisation and construction periods are much shorter (around 20 years) and, secondly, the amount to be invested that is exposed to the profitability risk is around a thousand times lower. A nuclear power plant probably needs government support not only as investment aid, but also – at least in part – to cover the investment risks. If you look at what is happening elsewhere in the world, all new nuclear power plants being built today are subsidised by the public sector, which assumes part of the investment costs and risks. A new project in Switzerland without public assistance is not realistic.

Switzerland is a rich country. We could afford it...

Considering the federal government’s current austerity measures, that doesn’t seem so clear. If Switzerland decides to provide financial support for construction, the law will have to be amended, which will require a new referendum. The population would therefore have to vote on the issue several times: on whether to accept the ‘Stop the Blackout’ initiative or the Federal Council’s indirect counter-proposal; on a financing law and on the ‘general authorisation’ (framework approval). In addition, a construction and operating licence would be required. And finally, investors will have to decide whether to embark on such a project, given the financial conditions. Each of these political, economic and technical decisions could delay or halt a new construction project, further increasing the financial risk for investors.

Assuming that people are in favour of all of these issues, when could a new power plant be built in Switzerland?

Even in a very optimistic scenario, it would be unlikely before 2050. A construction company would also have to be selected, as we do not have one in Switzerland capable of carrying out such a project. Only a few companies in the world have the necessary expertise. Currently, Rosatom (Russia) and CNNC (China) are the largest power plant builders. However, it is difficult to imagine at this stage that a Russian company would build a power plant in Switzerland, and CNNC builds almost exclusively in China. The construction of a new plant in Switzerland would therefore probably be entrusted to a French (EDF), American (Westinghouse) or South Korean (KEPCO) company. Most of these providers are facing or have faced financial difficulties. EDF recently completed reactors in Finland and France and is currently building two in the United Kingdom. However, all of these projects have experienced delays and cost overruns. The Olkiluoto EPR project, originally scheduled to take four years and cost €3 billion, ultimately took 13 years and cost around €11 billion. EDF also planned to build six new reactors in France, as well as two more in the United Kingdom. However, the French Court of Auditors has questioned the profitability of these activities, and the timetable for the new French facilities has been postponed.

Westinghouse, for its part, had started building two reactors in the United States (Vogtle). However, work was halted in 2017 due to the company’s bankruptcy and completed by other companies with public funds. The Japanese have not built any new reactors for nearly 20 years and have faced technical problems. That leaves the Korean company KEPCO, which delivered six reactors to the United Arab Emirates (UAE) on schedule (after eight years of construction). But in the UAE, there are no financing problems, complex authorisation procedures or referendums. In short, it will be difficult to find a company that is willing and able to build a nuclear power plant in Switzerland. The 2050 deadline is therefore rather optimistic, and it may take longer. A new power plant is not a solution to the sharp rise in electricity demand expected well before 2050; at best, it could only partially replace the gradual phase out of existing power plants – and even then, only if Gösgen and Leibstadt nuclear power plants continue to operate well beyond 60 years.

Could small modular nuclear reactors (SMRs), which are expected to be much cheaper and therefore easier to finance, be a solution for Switzerland?

SMRs are currently still in the development phase. They could be commercially available in the 2030s, but experience with costs and operation will still be very limited at that point. The idea behind SMRs is to build factories that produce large numbers of units in order to benefit from economies of scale and reduce costs. It remains to be proven that this works. But if a plant produces 10 or 20 SMRs per year – or even a hundred, which would be huge – will the cost advantage really be as significant as claimed? At this stage, the International Atomic Energy Agency (IAEA) is talking about cost parity with large power plants by the 2030s at the earliest. We are not in the automotive industry, where factories produce millions of vehicles and thus achieve enormous economies of scale. If the concept works, it could be an option. But I wouldn’t rely on it.