Extracting fossil fuels from ever-more-difficult environments is a dangerous business, a truth underlined spectacularly by the explosion at the Massey mine in April that killed 29 miners or the Deepwater Horizon spill that has left the Louisiana coast a blackened brackish mess.
Not in decades has the nuclear option looked more attractive. Earlier this year, the government extended funding to build two new reactors at the Vogtle plant in Georgia, likely the first reactors to go online since 1996, and a lot more may be in the works. Oil and coal disasters like Massey and Deepwater Horizon may be some of the best arguments for nuclear power.
They may also be some of the best arguments against it. Disasters like Deepwater Horizon highlight troubling truths about natural resources. But they also point to some equally troubling truths about accidents and worst-case scenarios.
If you think that the lesson of the Gulf disaster is that we have been racing too fast to drill for oil without adequate equipment or safety protocols, then you may conclude that we should slow down the search for fossil fuels—and perhaps turn toward nuclear. But if you think that the story of the spill is not mainly a story about fossil fuels but about how many levels of “failsafe” mechanisms fail, it is likely to make you more skeptical of the nuclear option.
That last sentence is not what I set out to write when I started thinking about this story. My first impulse was to think that Deepwater Horizon was self-evidently an argument for nuclear power. What, after all, could be a better advertisement for nuclear than the oil-soaked pelicans? Say what you will about it, but nuclear power doesn’t turn the beaches black.
Nuclear does not pollute the air. It does not leave the United States dependent on foreign energy. As global warming has become a more pressing concern, the case for nuclear has gotten even stronger. It’s by far the most viable source of electricity that does not release carbon dioxide into the atmosphere. It has worked well for years. (In France, where nuclear provides about four-fifths of electricity, reactors are still being actively built—the newest started up in 2002.)
With all this in the background, you would think that a giant oil-related accident would be another point against fossil fuels and for nuclear, and in some ways it is. In one very important way, however, it is not. The most visceral concern about nuclear power comes down to, “What if something goes wrong?” Deepwater Horizon is a compelling reminder that things go wrong, in unexpected ways, at unexpected times, to catastrophic effect.
New York Times writer David Leonhardt very aptly and quickly caught what you might think of as the intellectual implications of the disaster in a New YorkTimes Magazine essay about our propensity to underestimate the chances of “low probability, high cost events.” That’s something that has been discussed for a while among economists. The idea that we discount the worst-case scenarios is one of the guiding principles of Nassim Taleb’s business best-seller The Black Swan.
I’ve been wary of jumping on the Black Swan bandwagon because people who go around thinking of worst-case possibilities tend to overcount the cost of worst-case events (when markets collapse, the world doesn’t end). But when it comes to nuclear failure, though, the cost of failure really is incalculable. Bad as the Deepwater Horizon spill is, it is nothing next to the 1986 Chernobyl meltdown, which has left close to 1,100 square miles surrounding the Chernobyl reactor uninhabitable to this day.
One of the news events that has been (depressingly and ironically) overshadowed by Deepwater Horizon was the sentencing, earlier this month, of seven Union Carbide managers for the 1984 gas leak that killed thousands in India. It has been the good fortune of the West that the worst man-made environmental catastrophes—Bhopal, Chernobyl—have happened in other places. They could be chalked up to mismanagement in countries (India, the late-stage Soviet Union) famous for mismanagement.
Deepwater Horizon brings those low probability, high-cost events close. The near meltdown at the Three Mile Island plant did a lot to sour Americans on nuclear power, but Three Mile Island was a disaster that didn’t happen. The safety systems and protocols worked. The plant shut down. The core didn’t melt.
On the Deepwater Horizon well, the safety systems didn’t work. The same high-pressure conditions that made the well fail made the blind shear ram that was supposed to cut the pipe fail as well. This is the problem of the last-resort failsafe plan: It has to work in the least likely and most overwhelming conditions. In other words, it has to anticipate the conditions most likely to lead to disaster.
One of the virtues of the powerful American tort system is that if nothing else, economic self-interest plays a strong role in encouraging companies to take the worst-case scenarios seriously. But the tendency to underestimate their cost tends to trump this as well. Accidents—as British Petroleum is finding out with great pain—turn out not only to be likelier than lay people think, but more expensive and serious than the accountants budget for.
I started off by presenting two views on nuclear power because it is genuinely difficult to choose between them. Duke Energy chief executive Jim Rogers said that, “You can’t be serious about carbon unless you’re serious about nuclear.” Though the pronouncements of energy company chief executives these days are (and should be) taken with a grain of salt, there’s a compelling point to consider in that.
But for all the attractions of nuclear, there remains the looming question of what happens if things go wrong. Nuclear power suffers from what you can think of as a paradox of catastrophe: The worst-case scenario is so terrible that we are actually less able to quantify it and consider its ramifications than we are with other potential disasters. We implicitly recognize this in the laws governing the nuclear industry, which cap the industry’s liability for an accident at $10 billion.
Everybody understands that in the event of a real nuclear catastrophe, that would be a drop in the bucket. The truth is that the costs of that would be so great that we simply put it in the category of those near-inconceivables we don’t want to consider. Which is all the more reason to consider it.