The Breakthrough Institute, a US-based pro-nuclear lobby group, has been in the middle of a debate about the economics of nuclear power. The origin of this debate was an article by the Institute's Jessica Lovering, Arthur Yip and Ted Nordhaus, published in the Energy Policy journal last year.1
They compiled overnight construction cost data from a number of countries and concluded:1
"In contrast to the rapid cost escalation that characterized nuclear construction in the United States, we find evidence of much milder cost escalation in many countries, including absolute cost declines in some countries and specific eras. Our new findings suggest that there is no inherent cost escalation trend associated with nuclear technology."
The article attracted scepticism even from nuclear advocates, with former World Nuclear Association executive Steve Kidd writing:2
"The article tries too hard to refute the general contention ‒ based on US and French experience ‒ that the costs of nuclear power stations have risen substantially over time. It incorporates new evidence on the costs of early demonstration reactors in the US and France, and data from a wider number of countries (including Canada, Germany, Japan, India and South Korea), that show a variety of more favourable trends. With the exception of South Korea, these apply only in particular time periods.
"Full data from the UK was conveniently unavailable ‒ the cost escalation record of the 14 AGRs [Advanced Gas-cooled Reactors] was even worse than the US experience in the 1980s ‒ while the escalating costs of the two EPRs under construction in Europe and the four AP1000s in the US are also ignored."
Energy Policy has recently published two detailed responses3,4 to the Lovering et al. article and a rejoinder by Lovering et al.5 US academics Jonathan Koomey, Nathan Hultman, Arnulf Grubler argue that Lovering et al. "use analytical methods that mask nuclear power's real construction costs, cherry pick data, and include misleading data on early experimental and demonstration reactors."3
Koomey et al. take issue with the use of overnight costs ‒ omitting financing and other time-related costs ‒ by Lovering et al.:
"While overnight costs do have a long history, there is simply no economic basis for comparing the costs of reactors without including the cost of capital and the construction duration. A key aspect of nuclear reactors that makes them such high-risk investments are that they are large scale, complex, and predominantly site-built. Hence construction takes years (even in the best case) and can extend over a decade or more. Almost all modern reactor programs analyzed in detail to date have experienced significantly lengthened construction times, which is ignored in the use of overnight construction costs by Lovering et al.
"Given that financing constitutes a significant part of nuclear costs in the real world, and that the very nature of nuclear power as a large scale, capital-intensive technology makes it particularly sensitive to financial risks, a study that ignores return on capital cannot give a true picture of the costs of nuclear power."
Koomey et al. note the overnight cost data presented by Lovering et al. do not even support their conclusions:3
"Analyzing the costs of electricity generation technologies is an exercise fraught with pitfalls. Unfortunately, Lovering et al.'s assessment of nuclear costs made several consequential errors. Their analysis incorrectly omits interest during construction, and thus substantially underestimates the effect of cost escalation over time. ...
"We note that the authors cherry pick data to suit their conclusions. Nevertheless, the presented data itself don't even support their stated conclusions, which is deeply puzzling. While Lovering et al. claim that their data show a more nuanced picture, suggesting that cost escalation for nuclear reactors is not a real problem, their own data for the modern era show the contrary. …
"The article presents graphs for nuclear construction costs in the US, France, Canada, West Germany, Japan, India, and South Korea, but the only country where overnight costs appear to decline over time in the modern era is South Korea. In that case the data do not come from an independent source but from the country's nuclear utility, have not been independently audited, and are not disclosed (and of course do not include interest during construction, as discussed above). As a result, they do not meet the critical scientific criteria of reproducibility and thus utmost caution is advisable in drawing strong conclusions from those numbers.
"Lovering et al.'s results suggest one example of overnight costs decreasing in the modern era, but the most sensible interpretation of their data is that almost all countries showed cost escalation from the 1970s onwards. This effect would be even more dramatic if the authors had included the costs of financing for a full accounting of nuclear construction costs and their historical evolution."
In a separate response, Alexander Gilbert, Benjamin K. Sovacool, Phil Johnstone, and Andy Stirling accuse Lovering et al. of being selective with their choice of data, unbalanced analysis, and biased interpretation.4 They write:
"In conclusion, several methodological decisions limit the applicability of Lovering et al.'s analysis to overall nuclear construction costs. Difficulties concerning the impact of interest rates on total installed costs, the role of time overruns, accounting for independent cost variables, the normalizing of global data, and comparisons with existing energy sources all serve to blunt Lovering et al.'s implied critique of earlier studies. Indeed, several
conclusions in the existing literature remain unrefuted:
- Nuclear energy displays serious cost escalations both in the form of rising capital costs over time and in cost overruns at individual plants;
- There are regional and temporal variations in these trends, but similar patterns nonetheless persist across countries and timeframes;
- Compared to other technologies, the intensity of these cost escalations is highly distinctive of nuclear reactors;
- Policymakers and energy modelers addressing nuclear energy need to be aware of elevated capital costs, the critical role of interest rates, and the near certainty of cost and time overruns."
Gilbert et al. also write:
"Our own work on the role of cost overruns in nuclear economics yields several points that deserve highlighting. One of them is that almost all nuclear reactors suffer from cost overruns. Another is that nuclear cost overruns occur in all countries. Yet another is that cost overruns are much greater for nuclear than for other energy sources. A final one is that nuclear cost overruns are heavily influenced by interest costs and time overruns. Lovering, et al. do not challenge this picture from the existing literature. Indeed, by failing to address
the roles of interest costs or construction delays, their study effectively ignores some of the most important issues in understanding historical nuclear construction cost trends."
Gilbert et al. also take issue with the public statements made by Lovering et al., which have been even more inaccurate than their quasi-academic article in Energy Policy:
"Lovering and colleagues have repeatedly referred to their data or analysis publicly as reflecting the "real costs of nuclear power", as offering a "complete construction cost history" of the industry, or proving that "nuclear plants can be built quickly, safely, and cheaply". In light of both Lovering et al.'s actual results and our previous criticisms, these characterizations of their study are misleading and inaccurate."4
References:
1. Jessica R. Lovering, Arthur Yip, and Ted Nordhaus, April 2016, 'Historical construction costs of global nuclear power reactors', Energy Policy 91, pp.371–382, www.sciencedirect.com/science/article/pii/S0301421516300106
2. Steve Kidd, 27 March 2016, 'Achieving better nuclear economics – new designs and industry structure?', www.neimagazine.com/opinion/opinionachieving-better-nuclear-economics-ne...
3. Jonathan Koomey, Nathan E. Hultman, Arnulf Grubler, March 2017, 'A reply to "Historical construction costs of global nuclear power reactors"', Energy Policy, Vol. 102, pp.640–643, www.sciencedirect.com/science/article/pii/S0301421516301549
4. Alexander Gilbert, Benjamin K. Sovacool, Phil Johnstone, and Andy Stirling, March 2017, 'Cost overruns and financial risk in the construction of nuclear power reactors: A critical appraisal', Energy Policy, Vol. 102, pp.644-649, www.sciencedirect.com/science/article/pii/S0301421516301690
5. Jessica R. Lovering, Ted Nordhaus, Arthur Yip, March 2017, 'Apples and oranges: Comparing nuclear construction costs across nations, time periods, and technologies', Energy Policy, Vol. 102, pp.650–654, www.sciencedirect.com/science/article/pii/S0301421516306000