Matthew Kahn has this interesting question:
If the optimal amount of nuclear energy is positive but only very few nuclear plants get build, then potentially no students will become nuclear engineers and subsequently human capital in this field dries up. As a result, nuclear power plants become less safe leading to a higher probability of disasters.
My thoughts on this are:
He is absolutely right, but let’s think this further. Quite interestingly, if there is less human capital in the nuclear energy sector and this increases the probability of nuclear disasters, then this also has an additional feedback effect on the optimal amount of nuclear energy: It should decrease then! And that is interesting, since that means that fewer nuclear plants –> fewer researchers/ nuclear engineers –> higher probability of disasters –> lower optimal amount of nuclear plants –> fewer nuclear plants, etc… So the question is: since there is this vicious circle here and we can indirectly affect the optimal number of power plants by affecting the probability of a disaster, then really what IS the optimal amount of nuclear energy?

Assuming this human capital channel is very large/important, then the following argument should apply: The more plants we build now the more students will become nuclear engineers, the lower the probability of a nuclear disaster and the higher again the optimal amount of nuclear energy. This feedback loop leads eventually to a very large amount of nuclear energy. Or, the other equilibrium will be that fewer plants are expected to be build, leading to fewer nuclear engineers and a higher probability of disasters and potentially a zero optimal amount of nuclear energy in the future. Thus, two equilibria are possible in this scenario, one with and one without nuclear power.
This could make a nice research question indeed. Anyone interested in joining in? 😀

However, I have to say that it is a strong assumption that fewer nuclear engineers implies a higher probability of disasters in the future. That should only happen if a) knowledge is lost or b) other engineers cannot quickly learn the existing knowledge needed for a nuclear plant. Since I am not an engineer, I don’t know how large the synergies or overlaps are with other fields.

Finally, one way to break the vicious circle that leads to an optimal amount of zero nuclear energy is if the governments invest in (nuclear) human capital. Thus, if a potential nuclear engineer student knows that he will be paid (or get a job) by the government whether a plant will be built or not, then this should give enough incentive to keep that kind of (nuclear) human capital intact.