One environmental group says Australia’s peak science body has made errors in its estimation of the cost of nuclear energy production, noting the figures could be three to six times higher than the actual cost.
The group’s comments come amid an ongoing debate over the annual GenCost report that compared costs between renewables and nuclear in Australia.
The Commonwealth Scientific and Industrial Research Organisation (CSIRO), one of Australia’s largest scientific research bodies, released the consultation draft in December 2023.
The report suggested that nuclear power, in particular small modular reactors (SMRs), is the most expensive form of energy source and could not play a role in Australia. Instead, it proposes solar and wind power as a more economically viable option with the “lowest cost range of any new-build technology.”
SMRs are advanced nuclear reactors that have a power capacity of up to 300 MW(e) per unit, which is about one-third of the generating capacity of traditional nuclear power reactors, according to the International Atomic Energy Agency.
While they are smaller in size, SMRs offer greater flexibility, are faster to build, and more affordable than large nuclear plants.
Opposition Leader Peter Dutton, who has been advocating for nuclear power as a replacement for Australia’s ageing coal-fired power stations, criticised the CSIRO’s report as “discredited.”
But the CSIRO chief executive responded, saying the group would “staunchly defend our scientists and our organisation against unfounded criticism.”
Environmental organisation WePlanet Australia became the latest climate group to weigh in on the debate as it challenged the accuracy of CSIRO’s energy costings.
The Issues With the CSIRO’s Study
WePlanet Australia pointed to an analysis by nuclear energy expert Oscar Archer, which suggested the CSIRO had “made errors that misrepresent the costs of nuclear versus other energy sources.”In this case, this means comparing the full, total cost of building nuclear power plants with the component costs of renewables.
In its analysis, the CSIRO cited the excessive project costs of a U.S-based nuclear plant—called the Carbon Free Power Project—as part of its argument that nuclear SMR is “a very high-cost technology.”
“Despite being cancelled, this project is the first to have provided cost estimates for a commercial project with detailed data.”
The “updated real project costing,” according to the CSIRO’s estimates, was A$31,000/kW. This is six times higher than the estimates put out by NuScale—the U.S. nuclear company behind the Carbon Free project—which was only A$5,100/kW, the CSIRO noted.
However, Mr. Archer criticised this approach, explaining that NuScale’s estimated A$5,100 price tag was the “overnight capital cost,” while the CSIRO’s price tag of A$31,000/kW was calculated using the “total project cost.”
Mr. Archer explained that the total project cost includes the extra costs of financing, interest rates, and construction duration, including overruns, among others. These costs are “invariable in the billions” as they follow different trends in different countries.
On the other hand, he suggested the overnight capital cost, also called overnight construction cost (OCC), is a more accurate metric for energy production comparison because it “provides a necessary baseline for costs without the complexities of timeframes and interest rates.”
“While it’s entirely fair to emphasise the fatal escalation of the total cost of the project, apples must be compared with apples,” he wrote.
Mr. Archer described NuScale’s measuring method as “standardised” and “already well advanced.”
Comparing ‘Apples To Oranges’
The opinion was echoed by Jessica Lovering, American astrophysicist and director of energy at the U.S-based Breakthrough Institute, who supported the use of the overnight construction cost in energy research.Ms. Lovering argued if the factor of interest rates is included—as in the case of the CSIRO report—it would make “any comparison of construction costs cross-nationally … impossible.” This would also “obscure” other factors such as the evolution of technology and labor and materials costs.
“Fossil fuels do not pay for their pollution, hydro does not pay for ecosystem disturbance, wind and solar do not pay for their intermittency.”
Ms. Lovering noted that the debate on energy technology costs has been “characterised by data limitations, arbitrary data sets, apples-to-oranges comparisons.”
“Too often apples-to-oranges comparisons between energy technologies are made, by comparing component costs with system costs, and comparing power purchase agreements for solar and wind (including subsidies) with the unsubsidised costs of nuclear and fossil energy technologies,” she said, noting that this was a “serious problem.”
Despite being hailed as a cheaper option, wind and power energy carries a significant price tag when it comes to storage, transmission, and waste. The average wind turbine and solar panel have a lifespan of 25 years before they must be decommissioned and taken apart, while a nuclear reactor has a lifespan of up to 80 years.
Nuclear Costs Also Vary Depending on Country
Ms. Lovering also noted that research on energy costs has put an “overwhelming focus” on cost trends in the United States—“a country where, until recently, no new construction had been initiated for over three decades.”In fact, the United States is the country with the highest cost escalation of nuclear, which Ms. Lovering attributed to the “outsized and anomalous impact” of the Three Mile Island nuclear accident.
Overly depending on U.S. data—as in the case of the CSIRO report—has “obscured the diversity of cost experiences internationally” she said.
Meanwhile, Mr. Archer asked whether Australia could “get a better price for large reactors” than other countries”
