Since the nuclear accident at Fukushima, there has been a lot of discussion about nuclear power, its safety, and practicality. Fortunately, most of it has been very level headed, and several diarists and commenters have asked that we foster a rational discussion -- neither alarmism nor excessive optimism about nuclear power as a panacea for our need to reduce our dependence on fossil fuels.
I should get my biases out of the way: I think that nuclear power is an inevitable part of the solution to reducing carbon emissions, but I'm not optimistic or enthusiastic about it. As I've tried to explain in other's diaries, I think the solution will look something like an aggressively developed mix of wind, solar, hydro, geo-thermal, nuclear and a continued reliance on carbon in the near term that declines over time, bound together in a smart grid. There is no one solution to our energy problems, no one single technology or form of generation.
In this diary, I'd like to explain why. My background is in finance, economics and law, especially in developing countries, and their intersection with environmental issues. I know a bit about the electrical grid and power generation, having worked in the 1990s, on project finance for an electrical power plant, although I haven't worked in the private sector in over a decade.
When we talk about power generation, carbon emissions and nuclear power, we should talk about the economics of these problems and economics invariably involves law.
The environmental economics of power are mostly about the problems of externalities. Externalities occur when an enterprise does not have to internalize the cost of what it's doing, making and selling. This could be as minor as mushroom hunters gathering "free" mushrooms from a national forest to sell to high end restaurants, or the service that bees provide to farmers in pollinating their crops. But the biggest externalities problems involve pollution -- for example, when a factory or power plant emits pollution, causing costs to others that are not reflected in the price of their products that consumers pay.
Law plays the role of trying to internalize externalities. When nuisance law says that a slaughter house that allows blood and gore to run across the company's property lines onto another's property must pay that other property owner or stop allowing the stuff to run, the law is internalizing that externality; it's forcing the slaughter house to pay the cost of its gory run-off.
Obviously carbon emissions have been a huge externality for electrical generation. If it causes massive climate change, that would be the worst case of externalities in human history.
Laws like carbon trading, carbon taxes or classifying CO2 as a pollutant are ways of internalizing the externality. It is one of many steps we need to take in reducing carbon emissions.
If we carry this process far enough, it could be economically feasible to capture carbon at levels that will not destroy the environment.
My biggest concern with nuclear power is that it also does not internalize its externalities, and I've rarely come across a proponent of nuclear energy willing to discuss it or face it.
In fact, the law relentlessly externalizes the costs of nuclear power.
The reason the law does this is because of lobbying by the nuclear power industry and the pleading of well meaning experts.
But the law externalizes the cost of nuclear power for a more fundamental reason -- which is that, if we took all the costs together, nuclear power is almost infinitely expensive.
It is only competitive in a purely financial, present day sense because most of its costs are externalized by statute and because it is heavily subsidized, such that its costs are carried by governments or consumers. This may be something we have to do to reduce carbon emissions, but it doesn't argue well for nuclear power being a long term, viable, sustainable solution to our energy problems.
By infinitely expensive, I don't mean that the cost of mining and refining uranium is infinitely expensive or that operating a nuclear power plant is.
But the two areas that cause nuclear power to be infinitely expensive without externalities are the problem of nuclear waste and the problem of risk of nuclear accidents.
Some components of nuclear waste will have to be stored for tens of thousands of years. Even if this storage were cheap, using simple financial calculations, discounting the future costs to a present day internalizing cost, would make nuclear power prohibitively expensive.
But storage isn't even cheap, even though the volume of nuclear waste isn't that great; in fact, it's fantastically expensive. Discounting to present value these high current costs of storing and monitoring nuclear waste for, say, 50,000 years makes the costs extremely high. The only way we can economically generate nuclear power is simply to ignore the problem (and cost) and tell ourselves that we have no choice because carbon emissions are worse and push the costs off to future generations.
I realize we did the same thing with carbon fuels -- convinced ourselves that we could ignore the externalities, so I'm not saying that nuclear is worse than coal at this. But I am saying that the ultimate sustainable solution will not be one that externalizes costs just as badly as the coal generated power industry did.
A lot of discussion of nuclear waste just assumes that we will be a more scientifically advanced civilization some time in the future and will "solve" the problem. Of course, we don't know if that's true or will come to pass. We don't even know what kind of civilization we are pushing these costs onto. For all we know some time in the future a pirate civilization ethos will take over that will cherish using nuclear waste to obliterate their enemies. Or maybe we won't even be here and we will be pushing these costs onto other species.
The other reason that nuclear power is too costly -- unless costs are externalized is the probabilistic risk of accidents. Very roughly, the internalized cost of possible future accidents is the probability of the accident happening multiplied by the cost of the accident.
As Chernobyl, Three Mile Island and Fukushima have demonstrated, the risk of catastrophic accidents is non-zero. Most proponents of nuclear power on this board consistently argue that the probabilistic risk of accidents is zero or near zero. They argue that every former accident was special or caused by design flaws of older nuclear plant designs, or human error, or a unique natural catastrophe that cannot happen again. This "reasoning" would not be tolerated as an argument in any other field. Three Mile Island, Chernobyl and Fukushima were each different. There is no reason to believe that even with advanced new designs, fate or the flow of events will not find some design flaw or combine to produce new, unanticipated accidents.
I would agree that the risk of catastrophic accidents is very low, but it is non-zero. To say it is zero today is to argue from a blatantly non-reality based perspective.
Moreover, the cost of the accidents are phenomenally large. The Chernobyl exclusion zone is about 20 miles. The potential cost can mean abandoning entire urban-industrial areas -- all the businesses, farms, factories, homes, infrastructure, etc. Multiply the low but non-zero risk times the cost yields a result that cannot be incorporated into the price of nuclear generated power.
In other words, even if the annual probability of risk is extremely low, you are multiplying that very low number by a very high number -- the actual cost of an accident.
If you think this is fear mongering, then look at the signals the economy sends. Private insurers will not insure nuclear power plants without statutory intervention. I know that insurance companies are not popular, but at least we can admit they like to make money. If they like to make money and if they could make money insuring nuclear power, they would do so. They can't. That's not fear mongering; that's cold eyed capitalist calculation. So the insurance costs are socialized to governments.
If lots of people are convinced to oppose nuclear power because of Fukushima, please don't call that irrational, or fear mongering, or anything else. It's that they have made a different cost benefit calculation from the one you would make.
I think that some proponents of nuclear power are somewhat mezmerized by the very low fuel requirements of nuclear power. Back in the 1960s, we all heard the public service announcements that nuclear power seemed to offer almost unlimited energy at no cost -- meaning at almost no fuel cost. The low fuel cost is true. But fuel cost is not the main cost of generating electricity. The electrical company still to build a very large, complex machine, hire very skilled engineers and technicians to run the place and hold the waste. The "cheapness" of nuclear power is an illusion based on the cheapness and very small volume and energy density of the fuel. But that cheapness in legal-economic terms is large based on externalizing costs, subsidies and refusing to address certain long term issues.
Lastly, I would say let's continue having this discussion on a rational basis.