For a year or two, I’ve been asserting that the 2011 Fukushima nuclear disaster would cost closer to a trillion US dollars when all costs were accounted for. Recently, I was challenged on this point and decided to document and argue my thesis.
Let’s start with principle accountability. Yes, if the tsunami had not occurred, the Fukushima reactor would not have a problem. But that’s facile. If a wind or solar farm had been where the reactor is, it would have been destroyed and the loss would have been trivial in comparison.
It’s also important to point out that the earthquake and tsunami killed thousands directly and slightly fewer indirectly. The reactor meltdown is expected to statistically shorten the lives of a few people. This isn’t a cost comparison of lethality, or a diminishment of the lethality of the earthquake and tsunami.
However, it’s important to understand that the nuclear generator is core to a series of follow-on costs.
First, there’s the basic cost of clean up of the reactor site itself.
Completely cleaning up and taking apart the plant could take a generation or more, and comes with a hefty price tag. In 2016 the government increased its cost estimate to about $75.7 billion, part of the overall Fukushima disaster price tag of $202.5 billion. The Japan Center for Economic Research, a private think tank, said the cleanup costs could mount to some $470 billion to $660 billion, however.
TEPCO indicates that it will take 30–40 years. That’s 3–4 decades of expensive management, execution, and governance. The solutions to the problem have been somewhat intractable so far. The site is very small. If it had anything else on it, the costs would have been orders of magnitude smaller. Site cleanup costs over 3–4 decades of this magnitude are very localized, while the earthquake and tsunami were much less localized.
It’s also important to note that the nuclear technology was comparatively ancient and had insufficient passive safety features, and that it was built in an area known to be inundated with tsunamis. If it had been located uphill a bit, none of these costs would exist. If the flood management system had been designed with a tsunami in mind, most of these issues would have been avoided. But the reactor was sited poorly and protected poorly from understood risks. These are nuclear generation issues, not tsunami issues.
My experience includes billion-dollar multiyear programs in various parts of the world. They never get cheaper or shorter unless requirements fundamentally change. They usually get more expensive and longer. My gut tends to support the JCER’s numbers, and more likely higher in the range. Let’s call it $600 billion for now. And remember, this is just the cleanup of the reactor site, nothing else.
The next lever to question is exports. I’ve tended to lump the Japanese automotive industry’s post-Fukushima challenges in with the reactor rather than the earthquake and tsunami. In researching this answer, of course the reality is more nuanced than either-or.
This graphic is of Honda Group’s automotive exports. There’s a very clear signal in the data that something related to Fukushima caused a significant drop in exports after the disaster.
However, this is the entire Japanese auto industry’s perspective.
That’s right, 2012 was up appreciably in terms of exports. But then it dropped again in 2013 and 2014 as Honda continued to increase. What’s going on?
There were two separate things happening. The first was direct disruption of automotive manufacturing and distribution facilities by the earthquake and tsunami. That was a big impact and relatively quickly improved. Toyota was hard hit, for example.
Toyota hasn’t been immune to the after-effects of the natural disaster that hit Japan in March. Toyota and its affiliates have three factories in the Tohoku region, a center for auto making in Japan, that suffered greatly because of the tsunami. According to Japan’s Cabinet Office figures, the Tohoku region accounts for about 8% of the country’s gross domestic product.
But then we get the radioactive cars concern.
- Radiation concern bars Japanese car shipment
- Mythbuster: Can cars imported from Japan be radioactive?
- Radioactive Used Cars Being Sold in Japan
- Radioactive cars from Japan keep turning up in Central Asia
This is the mostly irrational part of the equation. The cars in question were too far from Fukushima to have been irradiated and were so lightly ‘radioactive’ that flying in a passenger jet is worse, but still, people are weird about radiation after decades of Hollywood and — ironically — Japanese movies demonizing radiation.
The combination has had a toll on Japanese automotive (and other) exports. Barring Fukushima’s impacts, one would assume a return to pre-2008 fiscal meltdown exports by now. But basically they’re static. That’s in the range of $200 billion in lost exports just for the automotive industry.
It’s likely fair to attribute $20 to $50 billion of that to irrational fear of radiation. And once again, that’s solely due to the nuclear plant’s technology, siting, and safety systems, not the tsunami itself.
No nuclear plant, no fear of radiation.
So we’re up to $650 billion. What else? Well, now we have the nuclear exclusion zone. This is a 2013 map of it.
It’s about 30 kilometers by 5 kilometers for a total of about 150 square kilometers. This was an economically productive region. The region including the zone had a population of close to a million. Its industries included “Information and telecommunications, electronic parts, electrical machinery, chemicals, beverages” as well as agriculture.
Fukushima Prefecture produce became undesirable in and outside Japan, even when much of it was actually safe, as radiation levels vary according to areas and products.
Yeah, back to that irrational fear of radiation.
About a 100,000 people are still not able to return home, and the exclusion zone once again has a decades-long lifespan. If we take Japan’s GDP and divide it by the square kilometers we see that it’s about 13,000 per year per square kilometer. We can adjust that upward quite a bit because this region was obviously not an unpopulated mountainous or island region, but highly productive, industrialized and intensely used industrial land. It’s probably closer to $13 million per square kilometer. 150 square kilometers over 30 years makes that another $60 billion in economic losses.
No nuclear plant, no exclusion zone. That puts us around $710 billion.
Next, let’s look at the electrical generation. Here’s a useful EIA chart.
Prior to the Fukushima disaster, ~30% of Japan’s electricity came from nuclear power plants. There’s an obvious line on the chart that plunges to zero, and that’s nuclear generation. The Japanese government made the call to shut off all nuclear generation until they’d been vetted for safety after Fukushima. Overreaction? Not according to the economic workup so far, regardless of anything else.
But that had direct and indirect costs too. The first is that as nuclear plunged, coal and gas jumped. Japan doesn’t have fossil fuels. Gas jumped from about 280 TWh to almost 400 TWh of generation, a leap of about 43%. Coal jumped from 280 TWh to about 340 TWh, a pop of roughly 21%. Here’s the thing about those jumps. Coal and gas are commodities that have decade-long markets in some cases. Want a lot more on short notice, you pay through the nose. Eventually that stabilizes and prices optimize again, but the 43% extra gas was probably a 100% cost increase and the 21% extra coal was probably a 50% cost increase. Let’s look at a nice, current chart from Lazard (as gas and coal haven’t changed that much).
Gas is $41-$74 per MWh. Coal is $60-$143. They would have already been in the top 50% of those ranges already due to lack of in-nation reserves, but they probably would have been out of those ranges afterward. And given the expectation of the return of nuclear, they would have been paying premium prices the entire time.
What does the math suggest? Pre-Fukushima, let’s call it $60 for gas and $90 for coal because a lot of those costs are fuel. Nuclear was operational, not new, and built when it was cheap due to ignorance and optimism. It was probably cost $70 for the nuclear per MWh.
Post-Fukushima? Gas at $74 and coal at $143 easily due to premium fuel pricing. At 75% averaging factor over the likely 20 years it will take to replace the nuclear with restarted nuclear, wind, and solar, that’s about $100 billion in extra fuel costs.
All due to the nuclear part of the Fukushima disaster. That puts us around $810 billion.
And then there are the negative externalities of coal and gas. Let’s look at this Union of Concerned Scientists of the USA report.
A 2013 study also assessed the economic impacts of fossil fuel use, including illnesses, premature mortality, workdays lost, and direct costs to the healthcare system associated with emissions of particulates, NOx, and SO2. This study found an average economic cost (or “public health added cost”) of 32 cents per kWh for coal, 13 cents per kWh for oil, and 2 cents per kWh for natural gas .
Hmm… that’s another $22 billion for unexpected health costs due to burning extra fossil fuels. That could put us at $833 billion. But it’s tricky. Is this an extra cost now, or an avoided cost earlier? As weird as it sounds, only $22 billion is relatively immaterial in this workup.
So there we are. There’s a reasonable case for an identifiable $800 billion in costs for the nuclear portion of the Fukushima disaster. It’s not hard to see that a 40-year recovery period along with costs excluded from this could add 25% to that without breaking a sweat.
That’s why I say that when all the costs of Fukushima are added in, it’s easy to see that it will be close to a trillion US dollars. Feel free to quibble, but realize that in quibbling, you might get the number down to half a trillion dollars if you are persistent and generous.
I hope this helps answer the question of whether the major costs of the Fukushima disaster — outside of human lives — are heavily skewed by the nuclear reactor’s presence, and that without it the costs would have been much, much lower.