There are no solutions, only tradeoffs [Thomas Sowell]

Figure 1. Hypothetical Nuclear Tradeoff Curve

Here's a simple multiple choice question. Would the planet be better off with

a) 2 cent/kWh nuclear power and a Three Mile Island(TMI) or larger release every 2500 reactor years, or

b) 4 cent/kWh nuclear power and a TMI or larger release every 17,000 reactor years, or

c) 8 cent/kWh nuclear power and a TMI or larger release every 50,000 reactor years, or

d) 16 cent/kWh nuclear power and a TMI or larger release every 100,000 reactor years?

Stop and make a choice. Which did you pick?

With two cent per kilowatt-hour nuclear power, we could lift billions out of poverty, almost automatically decarbonize the electricity grid, convert most rail and light transportation to electricity, afford widespread desalination, produce CO2-free nitrogen based fertilizer, make solid inroads into decarbonizing steelmaking and have a shot at synfuels and CO2 capture. Humanity would be wealthier and healthier. Wealth is health, especially for poor people. We'd probably end up with more than 10,000 gigawatt scale reactors and 4 Three Mile Island or larger releases a year.

With 4 cent/kWh nuclear power, nuclear power could compete successfully with fossil fuel for baseload electricity generation, in most parts of the world. We'd see most coal plants replaced by nuclear when the coal plants were retired. But humanity would not be a great deal richer. Fossil fuel would continue to dominate the non-electricity markets. We'd end up with maybe 2500 gigawatt scale reactors and about one TMI or larger release every 10 years.

At 8 cents per/kWh, nuclear power is a meaningless sideshow. Well connected parasites will extract taxpayer money to play with toy reactors and conduct interminable studies; but nuclear will make no significant contribution to humanity's wealth and health. Releases of radiation from nuclear power plants would be very rare.

At 16 cents per/kWh, nuclear is dead. There will be no plants and no nuclear power plant releases.

Figure 1 sketches the trade off. It shows a hypothetical power cost versus release frequency curve. This is not a real curve. I made it up to match my made up question. But it is qualitatively correct and it does make a couple of very important points.

1) The only way we can have zero releases is to have zero plants. Cost rises exponentially as the release frequency approaches zero.

2) Where you want to be on that curve depends on how you value electricity cost versus release frequency. If you value cheap power much more than the cost of a release, then you want to be in the lower right portion of the curve. If you regard a release to be an intolerable catastrophe, the upper left is where you should be.

I believe the societal optimum is in the 2 to 3 cents/kWh portion of the curve; somewhere around the green oval in Figure 1. The planet would be far better off with 3 cents/kWh nuclear and several releases a year, than with more expensive nuclear and fewer releases. Provided we react intelligently to the releases, the health impacts from the lack of pollution and the increase in wealth will far outweigh the health impacts from the releases, CO2 emissions would drop drastically without the need for intrusive, wealth and health destroying regulation.

However, the NRC and other similar regulatory bodies take a very different view. They see no benefit from cheap, dispatchable, very low pollution and CO2 power, but they own any problems. These are the incentives we have imposed on these bureaucrats; and they respond accordingly, just like you and I would. Their job is to prevent releases, so their optimum is far up the cost curve, somewhere around the red oval in Figure 1 or perhaps higher. Safety is the overriding priority. NRC Chairman Hendrie put it best when he said the NRC's responsibility was safety “without regard to economic and social costs". We have a total disconnect between society’s optimum and the regulator’s optimum. On most of this planet, it is the regulator's optimum that rules.

To make matters worse, the regulatory process is far from perfect. A great deal of the regulatory cost does not decrease release frequency. In some cases, such as imposing the Double Ended Guillotine Break criteria, the regulation probably increases the number of releases per reactor-year. As a result we end up well to the right of the tradeoff curve, in the vicinity of the dashed red oval or worse. We end up in a situation where we could have the same release frequency at a great deal less cost. Economists call this “inefficient". A better description would be “tragically wasteful". As a practical matter, we end up with little or no nuclear power.

An alternative to NRC-style regulation is a market based system. If we can impose real competition on the suppliers of electricity, we will end up somewhere around the dashed green oval. This requires that the private costs to investors and operators of a release to be consistent with the public costs. The GKG has proposed a regime for ensuring that this is the case. This underwriter based system would be far closer to the societal optimum than the omnipotent regulator optimum. By imposing a well-designed radiation release compensation system, we can nudge the underwriter based optimum still closer to the societal optimum.

Sowell's statement is both an important insight and an obvious fact of life. But for nuclear, it misses the point. The real question is: who is making the tradeoff? Is it a process that actually balances cost versus risk? Or is it an all powerful bureaucracy for which one side of the tradeoff is unimportant?