Historical Construction Costs of Global Nuclear Power

Arthur Yip
Ph.D. Student
Carnegie Mellon University
Pittsburgh, PA


At the USAEE conference in Pittsburgh, I presented a retrospective analysis of the construction costs of global nuclear power reactors, which was conducted at the Breakthrough Institute and co-authored by Jessica Lovering and Ted Nordhaus. Given the controversial and hazy nature of this subject, the Breakthrough Institute was interested in re-examining common assumptions and assertions.


Analyzing the historical experience has been a common approach to help understand the prospects and challenges on nuclear power. Past studies (Thomas, 1988; Mackerron, 1992; Koomey & Hultman, 2007; Escobar-Rangel & Lévêque, 2015) have documented dramatic cost escalation and recurrent cost surprises, and have even identified the presence of “negative learning-by-doing”, suggesting an “intrinsic” or inevitable increase in costs (Grubler, 2010). These results have played a role in informing integrated assessment modellers and policy makers (Neij, 2008; Junginger et al, 2010; Harris et al, 2013; Azevedo et al, 2013).

Figure 1


We found that the existing literature on the construction cost of nuclear power reactors focuses narrowly on specific time periods or countries, specifically the US and France in the 1970-80’s. The 157 reactors studied by Koomey & Hultman (2007), Grubler (2010), and Escobar-Rangel & Lévêque (2015) in these two countries (out of the 31 countries with nuclear power today) only represent 26% of the total number of completed reactors. This limited subset provides an incomplete picture of the economic evolution of nuclear power construction.

Our study curated historical reactor-specific overnight construction cost (OCC) data that broadened the scope of study, covering earlier, recent, and international experience in nuclear reactor construction. We analyzed and compared the full cost history of nuclear power construction in the US, France, Canada, West Germany, Japan, India, and South Korea, using data from the IAEA and other governmental sources by country. We re-assessed the apparent lack of learning-by-doing in nuclear power construction. We also investigated the role of the Three Mile Island and Chernobyl accidents on OCC and construction durations.

We analyzed data for 349 reactors from 7 major countries with nuclear power reactors, covering a time span of 1954 to 2010 and almost two-thirds of all power reactors ever built globally. This yielded new insight to the historical costs of nuclear power. We found that trends in costs have varied significantly in magnitude and in structure, by era, country, and experience. In addition to rapid cost escalation in countries such as the US, there is evidence of much milder cost escalation in countries such as France, Canada, and Japan, and also cost declines in early construction history such as in the 1960's, and in recent construction history in South Korea.

Figure 2


The empirical comparison of global experiences raises new methodological and practical questions regarding the analysis and application of learning rates and the role of drivers behind technological change other than traditional learning-by-doing. History suggests that other factors such as RD&D, operational experience, regulatory requirements, economies of unit scale, economies of production scale, technology transfer, and international spillover have and will continue to dominate the trends in the cost of nuclear power. We find that the US and French experiences of the 1970's are not necessarily the best or most relevant examples of nuclear cost history, and that it would be ill-advised to draw strong conclusions about future nuclear power costs solely based on those experiences while ignoring trends in other countries and time periods.

The variation in trends dispels the single dominant story of cost escalation. This analysis of historical experience for global nuclear power construction offers a more comprehensive perspective on cost trends than in the literature, and we invite researchers, modellers, and policy makers to reconsider their assumptions regarding the costs of nuclear power construction.

The USAEE session attendees raised excellent points regarding the value of safety and improved performance, the issue of poor budgeting and estimation, the role of technical and regulatory standardization and political economy as a key driver of nuclear power costs and successful deployment.

Click to view a printable version of this article.