Bias against catastrophic nuclear tail events
Published : Mar 22, 2011 - 19:20
Updated : Mar 22, 2011 - 19:20
The worst case. These three words have been at the back of everyone’s mind ever since the Fukushima reactors began malfunctioning after being swamped by a tsunami. Remarkably, these reactors have been at the front of few experts’ mouths.

Many experts have shied away from describing worst-case outcomes, which are terrifying to contemplate and risky to mention. The risk isn’t just panicking the public. Crying wolf can threaten one’s expert status.

The bias toward calm, cool expression has been on full display in this crisis.

The Japanese are particularly good at the stiff upper lip. The authorities have serially indicated that exploding reactor housing is not a big problem, that released radioactive steam is not a big problem, that the significant cracks in containment vessels are not a big problem, that burning spent fuel ponds are not a big problem, and that the contamination of food and water is not a big problem. To top off all this, Tokyo Electric Power Co.’s president, Masataka Shimizu, made a formal apology “for causing such a great concern and nuisance.”

Earth to Shimizu: This isn’t just a nuisance.

What’s already occurred is horrible enough, what with death and severe injury to plant workers and the contamination of local milk, spinach, beans, and, presumably, fish. But the worst-case scenario at Fukushima is far beyond this. It entails six reactors melting down and all 4,277 tons of spent fuel stored at the plant burning out of control. And this at a site just 150 miles from Tokyo’s 14 million inhabitants, whose water is already showing traces of radiation.

Yes, Japan can hope for a strong wind blowing out to sea. But let’s just imagine how we might move from the current situation to this worst case? Another earthquake followed by another tsunami could certainly get us there. Sound crazy? Maybe not. Maybe last week’s earthquake was a foreshock for an even larger one that’s coming.

If this thought makes you queasy, you are feeling the bias we humans have against considering terrible tail events, or outcomes that are at the far end of the probability distribution. If we haven’t seen it, and it hurts to think about it, we ignore it. But equating what has happened with what will happen is folly. When it comes to earthquakes, what we’ve recorded to date is a minute span of geological history.

Nor can we count on future geological processes following past patterns. Four of the nine largest quakes to strike our planet since 1900 have occurred in the last seven years. This list doesn’t include Japan’s 1995 Kobe earthquake, which was, until now, the costliest on record.

The U.S. Geological Survey website states emphatically that no one can predict earthquakes. What it doesn’t say is that neither the Survey nor anyone else knows the statistical distribution governing earthquakes of different magnitudes.

Hence, the probability of extremely powerful earthquakes and their attendant tsunamis may be much higher than we think. Case in point ― the 9.0 magnitude earthquake off the northeast coast of Japan was 10 times bigger than the maximum quake the builders of the Fukushima plant considered possible.

Japan has 55 nuclear reactors. It’s in the process of developing another 11. That’s 66 “nuisances” waiting to happen, if not by earthquake and tsunami, then by terrorist, or by human error ― the cause of the Chernobyl meltdown.

Thanks to Chernobyl, an area the size of Switzerland, or 16,000 square miles (41,000 square kilometers), is uninhabitable for the next 300 years. Japan is about nine times bigger than Switzerland. If Japan has nine Chernobyls, it’s game over.

Is this unthinkable? It depends on your time frame. A lot can happen in 700 million years, which is the half-life of uranium-235, or 24,000 years, which is the half-life of plutonium-239, or even 30 years, which is the half-life of cesium-137. Japan’s thousands of tons of spent fuel are chock full of these ingredients. Yes, new technologies are reprocessing spent fuel, but they carry their own risks, including the production of weapon-grade plutonium.

Since we care about our kids, who will care about their kids, who will care about their kids, we effectively care about all our future descendents. And since our kids, grandkids, great-grandkids will inter-marry, we personally need to worry about all future humanity.

Our “carespan” far exceeds our lifespan and forces us to worry not just about the likely effects of nuclear energy, but the tail events. Since the tail events are potentially so catastrophic to our progeny, economics tells us to place all the weight in our planning on the worst-case scenario.

Were Nobel physicist Enrico Fermi alive, he would be appalled by the nuclear tail risk we are manufacturing for ourselves and bequeathing to our loved ones. Fermi, who played a major role in the Manhattan Project, would have no hesitation in telling Japan to shut the reactors.

Fermi developed the first nuclear reactor and had grave doubts about their net benefit, wondering when mankind will “grow sufficiently adult to make good use of the powers that he acquires over nature.” Fermi wrote these words two years before he died at age 53 of stomach cancer caused by radiation exposure to his reactor.

By Laurence Kotlikoff and Eugene Stanley

Laurence Kotlikoff is professor of economics at Boston University and author of “Jimmy Stewart Is Dead.” Eugene Stanley is professor of physics at Boston University, a member of the National Academy of Sciences, and co-author of “An Introduction to Econophysics.” The opinions expressed are their own. ― Ed.