CSIRO’s annual GenCost report sees potential for nuclear after 2030

Australia’s Commonwealth Scientific & Industrial Research Organisation (CSIRO) has included large-scale nuclear for the first time in its annual GenCost report, which has included small modular reactors (SMRs) since its formation in 2018.

CSIRO says GenCost is a leading economic report for business leaders and decision-makers planning reliable and affordable energy solutions to achieve net zero emissions by 2050. It is published annually in collaboration with the Australian Energy Market Operator (AEMO) and “offers accurate, policy and technology-neutral cost estimates for new electricity generation, storage, and hydrogen technologies, through to 2050”. This latest 131-page report is the sixth update since the inaugural report in 2018.

GenCost received the highest volume of feedback to its 2023-2024 consultation in its history “A majority of submissions to the 2023‐24 consultation process requested the inclusion of large‐scale nuclear in addition to nuclear small modular reactors (SMR) that had been included in GenCost since its inception in 2018. In response GenCost re‐examined the appropriateness of large‐scale nuclear and concluded that, although the deployment of large‐scale nuclear would require a significant increase in the reserve margin relative to SMR and existing Australian generation plants, there was no known technical constraint to deploying generation units of this size.”

GenCost also concluded that, due to the current state of the development pipeline in Australia, the earliest deployment would be from 2040. GenCost based its large‐scale nuclear cost on South Korean costs “as the best representation of a continuous building programme consistent with other technologies in the report”. These were then adjusted for differences in Australian and South Korean deployment costs. “Based on this approach the expected capital cost of a large‐scale nuclear plant in 2023 is AUD8,655/kW ($5769/kW). This capital cost can only be achieved if Australia commits to a continuous building GenCost notes that first‐of‐a‐kind (FOAK) premiums of up to 100% cannot be ruled out.

The report says the cost of nuclear SMR “has been a contentious issue in GenCost for many years with conflicting data published by other groups proposing lower costs than those assumed in GenCost”. Calculations were based on scrutiny of the US Utah Associated Municipal Power Systems (UAMPS) regional coalition that develops local government owned electricity generation projects. “Up until the project’s cancellation in November 2023, UAMPS was the developer of a nuclear SMR project called the Carbon Free Power Project (CFPP) with a gross capacity of 462MW. It was planned to be fully operational by 2030.” After conversion to 2023 Australian dollars, project costs were estimated in 2020 to be AUD18,200/kW which is only slightly below the level that GenCost had been applying (AUD19,000kW). “This validated CSIRO’s use of the higher end of the range presented in theoretical studies available at the time.”

However, in late 2022 UAMPS updated their capital cost to AUD28,580/kW citing the global inflationary pressures that have increased the cost of all electricity generation technologies. “The UAMPS estimate implies nuclear SMR has been hit by a 57% cost increase which is much larger than the average 20% observed in other technologies. This data was not previously incorporated in GenCost. Consequently, current capital costs for nuclear SMR in this report have been significantly increased to bring them into line with this more recent estimate. This new data is considered more reliable because all previous data was theoretical whereas the UAMPS project was the first to provide transparent data for a real project.”

The report found the levelised costs of electricity – the total unit costs a generator must recover over its economic life to meet all its costs including a return on investment – used to summarise the relative competitiveness of different generation options were lowest for variable renewables (solar photovoltaic and wind). “If we exclude high emission generation options, the next most competitive generation technologies are solar thermal, gas with carbon capture and storage, large-scale nuclear and coal with carbon capture and storage,” it notes.

Nuclear SMR costs improve significantly by 2030 “but remain significantly higher cost than these other alternatives. For clarity, neither type of nuclear generation can be operational by 2030. Developers will need to purchase the technology in the 2030s sometime after an expected 11 years of pre‐construction tasks are completed. 4 to 6 years of construction would then follow before full operation can be achieved. As such, the inclusion of large‐scale and SMR nuclear in the 2030 cost comparison is only as a point of interest rather than practicality.

According to CSIRO Chief Energy Economist Paul Graham, lead author of the report, GenCost is “flexible” to adjusting assumptions, scope and methodology” in response to feedback received during the formal consultation period and throughout the year. “For example, our approach to the inclusion of large-scale nuclear technology provides a logical, transparent and policy-neutral method of costing a potential deployment scenario in Australia.”