by Peter Lang
Could the costs of nuclear power have been 10% of what they are if not for the disruption?
A new paper, Nuclear Power Learning and Deployment Rates: Disruption and Global Benefits Forgone , finds that nuclear power could now be around 10% of current cost, and have avoided up to 10 million deaths and 164 Gt CO2 between 1980 and 2015, if not for disruption to progress in the late 1960s and rapidly escalating costs since.
Other global benefits forgone are discussed in the ‘Policy Implications’ section. Figure 3 in the paper (copied below) shows the decreasing costs from 1954 to 1967 (at 32 GW cumulative global capacity of construction starts), and rising costs thereafter for seven countries. The learning rates are derived from the slope of the regression lines.
Figure 3. Regression lines for seven countries: OCC plotted against cumulative global capacity of construction starts.
Regarding the statement that nuclear power could cost around 10% of what it does now if the disruption had not occurred, Note [XII] (in Appendix B in the paper) responds as follows:
“Some readers may question the credibility of the projections of OCC in 2015. This is a counterfactual analysis of what the consequences would have been if the pre-disruption learning and deployment rates had continued. There is no apparent physical or technical reason why these rates could not have persisted. Actual learning rates may have been faster or slower than the pre-disruption rates depending on various socio-economic factors. It is beyond the scope of this paper to speculate on what global economic conditions, electricity demand, public opinion, politics, policy, regulatory responses and a multitude of other influencing factors may or may not have occurred over the past half century if the root causes of the disruption had not occurred.
However, consider the following. A defensible assumption is that if the high level of public support for nuclear power that existed in the 1950s and early 1960s [12,27,28] had continued, the early learning rates may have continued and, therefore, the accelerating global deployment rate may have continued. With cheaper electricity, global electricity consumption may have been higher, thus causing faster development and deployment. In that case, we could have greatly improved designs by now—small, flexible and more advanced than anything we might envisage, with better safety, performance and cost effectiveness.
Rapid learning rates persisted since the 1960s for other technologies and industries, where public support remained high. The aviation industry provides an example of technology and safety improvements, and cost reductions, achieved over the same period in another complex system with high public concern about safety. From 1960 to 2013, US aviation passenger-miles increased by a factor of 19 , while aviation passenger safety (reduction in fatalities per passenger-mile) increased by a factor of 1051 , a learning rate of 87% for passenger safety. The learning rate for the cost of US commercial airline passenger travel during this period was 27% [46,48].
Similarly, the learning rate for solar PV (with persistent strong public support, favourable regulatory environments and high financial incentives) has remained high at 10 to 47% according to Rubin et al. (Figure 8) . Cherp et al.  compare energy transitions of wind, solar and nuclear power in Germany and Japan since the 1970s and find their progress depends on the level of public support, political goals and policies of each country.”
Sixteen notes in the paper address commonly raised concerns relating to statements in the paper (see Appendix B).
Moderation note: As with all guest posts, please keep your comments civil and relevant.