An assessment of the costs of the French nuclear PWR program 1970–2000

The paper reviews the history and the economics of the French PWR program, which is arguably the most successful nuclear-scale up experience in an industrialized country. Key to this success was a unique institutional framework that allowed for centralized decision making, a high degree of standardization, and regulatory stability, all epitomized by comparatively short reactor construction times.

Drawing on largely unknown public records, the paper reveals for the first time both absolute as well as specific reactor costs and their evolution over time. Its most significant finding is that even this most successful nuclear scale-up was characterized by a substantial escalation of real-term reactor construction costs. Specific costs per kW installed capacity increased by more than a factor of three between the first and last reactor generations built. Conversely, operating costs have remained remarkably flat, despite lowered load factors resulting from the need for load modulation in a system where base-load nuclear power plants supply three quarters of electricity.

The paper draws a number of cautionary lessons for technology, policy, and modeling studies in a climate-constrained world. First, the inherent technology characteristics of nuclear power: large-scale, complex, and with lumpy investments introduce a significant economic risk of cost overruns in the build-up process. Anticipated economic gains from standardization and ever larger unit scales not only have not materialized, but the corresponding increasing complexity in design and in construction operations have reversed the anticipated learning effects to their contrary: cost escalation. Second, cost projections and policy rationales based on relative economic merits of competing technology options are fraught by persistent uncertainties and biases, suggesting that the real cost of a scale up of a technology as large and complex as nuclear might be in fact unknowable ex ante, severely limiting conventional deterministic economic calculus and decision making (e.g. cost minimization) models. Lastly, the French nuclear case illustrates the perils of the assumption of robust learning effects resulting in lowered costs over time in the scale-up of large-scale, complex new energy supply technologies. The uncertainties in anticipated learning effects of new technologies might be much larger that often assumed, including also cases of "negative learning" in which specific costs increase rather than decrease with accumulated experience.