The extraordinary financial costs of nuclear power

By Alan Finkel

Partial extract from an article to be posted in

They’re being touted as the solution to kickstarting a nuclear power industry in Australia.

According to the Opposition’s Minister for Climate Change and Energy, Ted O’Brien, small modular reactors (SMR) could be built within ten-year period if it wins the next election. 

However, it would likely take 20 years to commence commercial operation of any nuclear reactors in Australia from the time in-principle approval was reached.  To reach that starting point and enable detailed consideration of the challenges and costs of nuclear power, the existing legislative ban on nuclear power in Australia will need to be removed.

There are other obstacles.

While there’s plenty of excitement about SMRs, the problem is there just isn’t enough data about them, mainly because there are none operating in any OECD country.

And it’s unknown when any might be. As Allison Macfarlane, former chair of the US Nuclear Regulatory commission, argues in her article,The end of Oppenheimer’s energy dream, the proposal for small modular reactors to help us in the clean energy transition is fanciful. 

The SMR furthest along the US Nuclear Regulatory Commission (NRC) approval process, from the US company NuScale, cancelled its first planned installation in Utah last November when the initial cost blew out to USD$9 billion, corresponding to USD$20 billion per GW.

The only countries with working SMRs are China and Russia.

Cosmos explainer: what is a small modular nuclear reactor

Micro and large reactors

Micro reactors are intended to generate electrical power up to 10 MW per unit.  Although companies such as Rolls Royce are developing these, there do not appear to be any commercial micro modular reactors that have completed their design.

That leaves full-scale reactors, which have also been mentioned as part of a possible Australian nuclear power play.

Korean company KEPCO builds most of the nuclear reactors in Korea and has now built one at Barakah in the United Arab Emirates. This 5.6 GW plant, scheduled to open this year, has taken 16 years to complete and cost  USD$24 billion (AUD$36 billion).  At 5.6 GW, that is AUD$6.4 billion per GW.  Given salaries and skills shortages in Australia, inflation, interest rates and our regulatory requirements, it would cost more and take longer in Australia.

The Hinkley C plant in the UK was supposed to be finished in 2017 but has been delayed again until 2031 – 23 years after approval.  The estimated construction cost ballooned to AUD$89 billion.  At 3.2 GW electrical power, that is AUD$28 billion per GW.

In the US, the most recent nuclear reactors to be built are the Vogtle 3 and 4built at the existing facility that is home to the Vogtle 1 and 2 reactors.  Both were  anticipated to be in service in 2016.  Vogtle 3 began commercial operation in July 2023.  Vogtle 4 is projected to commence operation in the second quarter of 2024 – 15 years after the construction contract was awarded.

Construction  cost USD$34 billion (AUD$52 billion) for the combined 2.2 GW output of the two reactors, or AUD$24 billion per GW.

Construction of nuclear plants in the United States has declined dramatically over the years.  Approximately 130 were built from the mid 1950s to the mid 1990s.  Only four commenced operation in the 30 years from the mid 1990s to now, and at the time of writing there are no nuclear reactors under construction in the United States. 

In France, only one nuclear power plant is under construction.  The 1.65 GW Flamanville EPR reactor is hoped to be completed and begin to supply electricity later this year, 17 years after construction began.  The most recent cost estimate was AUD$22 billion or AUD$13 billion per GW.  No other nuclear power plants are planned in France.

These high costs and long delivery durations for full-scale reactors are the reasons SMRs are proposed as a way forward in Australia.  However, SMRs are a new technology.  There are none in operation or construction in any OECD countries, thus it is not possible to estimate the costs or delivery schedules.  NuScale’s investment to date suggests that the capital cost for the first units to be delivered will be very high. 

The essential value proposition of SMRs is that they will be built using offsite, volume production, and thus in the long-term will benefit from economies of scale.  It might be decades, though, for this benefit to be realised.

Even if the process of serious consideration of nuclear power in Australia is formalised, it will be essential to continue investing in solar and wind power to meet our clean energy transition targets this decade and next.

Dr Alan Finkel is Chair of the ARC Centre of Excellence for Quantum Biotechnology at The University of Queensland. He is Chair of Stile Education.  He is a neuroscientist, engineer, entrepreneur and philanthropist.  He is former Chancellor of Monash University and was Australia’s Chief Scientist from 2016 to 2020, during which time he led the National Electricity Market Review, the development of the National Hydrogen Strategy, the panel that advised the Australian Government on the Low Emissions Technology Roadmap, chaired the National (Covid) Contact Tracing Review, was deputy chair of Innovation and Science Australia, and chaired the National Review of Industry and School Partnerships in STEM Education.

Subscribe to energise from riaus

Are you interested in the energy industry and the technology and scientific developments that power it? Then our email newsletter Energise is for you. Click here to become a subscriber.

Please login to favourite this article.