When Japan suffered an earthquake and tsunami in 2011, the Fukushima Daiichi nuclear power plant melted down, resulting in one of the worst nuclear accidents in history. In response, the Japanese government shut down all of its nuclear reactors. But subsequent economic research reveals that the unintended consequences of abandoning nuclear energy have been worse than the accident itself.
In this episode of Faster, Please! — The Podcast, I'm joined by Matthew Neidell, an economist in the Department of Health Policy and Management at Columbia University's Mailman School of Public Health. In 2021, Matt coauthored a paper on those unintended consequences called “The unintended effects from halting nuclear power production: Evidence from Fukushima Daiichi accident.” From that paper:
This paper provides novel evidence of the unintended health effects stemming from the halt in nuclear power production after the Fukushima Daiichi nuclear accident. After the accident, nuclear power stations ceased operation and nuclear power was replaced by fossil fuels, causing an increase in electricity prices. We find that this increase led to a reduction in energy consumption, which caused an increase in mortality during very cold temperatures, given the protective role that climate control plays against the elements. Our results contribute to the debate surrounding the use of nuclear as a source of energy by documenting a yet unexplored health benefit from using nuclear power, and more broadly to regulatory policy approaches implemented during periods of scientific uncertainty about potential adverse effects.
In This Episode:
The Fukushima meltdown (1:30)
The consequences of Japan’s shift away from nuclear (7:30)
Japan’s nuclear reversal (17:15)
Public perceptions of nuclear risks (20:43)
Below is an edited transcript of our conversation.
The Fukushima meltdown
James Pethokoukis: The title of the [working] paper is "Be Cautious with the Precautionary Principle: Evidence from Fukushima Daiichi Nuclear Accident." Let's start with a quick explanation. What is the precautionary principle?
Matthew Neidell: One thing I should clarify first: The title of the paper ended up changing. We do talk about the precautionary principle, but it ended up not being the title in the published version. We got a lot of pushback on the use of “precautionary principle” in the title. That said, I'm happy to talk about it, because I think everything in here is relevant to the precautionary principle.
You could teach a whole intro to econ class from this paper. Two things that pop out to me are the precautionary principle and also the idea of trade-offs, because this paper is very much about trade-offs. Starting with that, in what way do you think those principles are illustrated by the Fukushima accident?
I think what's really important here is that—this is almost anytime we think about nuclear but especially when big accidents happen, like Fukushima—we tend to focus on the one thing that happened and we don't think about the alternatives. That's the important thing. We think about nuclear as “nuclear carries risk.” And it does carry risk. There are dangers associated with nuclear. Just about anyone should know that who is following this. But it's “How do the dangers compare to something else, to the alternatives that we can use?”
One of the problems is that we tend to think of nuclear in isolation. Like people are just saying, “Nuclear is bad, therefore we shouldn't do it.” And that's the kind of precautionary principle aspect of things. It says, “Unless we are fully informed about the risks associated with something, there's no uncertainty associated with the risks with something, we shouldn't do it.” And that's hampering because there are so many opportunities that are out there that carry risk. And if we just say, “Let's not engage in these opportunities because there's a chance of risk,” we end up cutting back on so many things that we might otherwise do.
There's a scale of nuclear accident severity. It's a seven-point scale, and so far there have only been two level-seven—the worst—accidents: One was Chernobyl, and the other was Fukushima. In our experience in the nuclear age, Fukushima was one of the absolute two worst accidents that we've had. If you're looking for an example, this would seem to be a fantastic test case about just how dangerous it is, and also just how dangerous the counterfactual is.
Yeah, I think that's right. They are the two most dangerous, the third one being Three Mile Island.
That's actually down the list. I think that’s a five. We'll start with the actual accident: What do we know about the fatalities and the damage from the accident itself?
The biggest thing that people focus on are the radiation deaths. We have the meltdown, there's radiation that's getting out in the environment that's not contained, and how many people are being exposed to that and dying from cancer as a result? That is, I think, the biggest fear to most people. So far we only have estimates of that. We don't know that precisely because people are dying from cancer, unfortunately, all the time. So how we trace those back to this particular accident is hard to know precisely. But we have ways of estimating that, and the estimates out there—and these are not just estimates of how many radiation deaths we've seen so far, but also how many we expect to see over the next 10 or 20 years because of this accident. And the kind of leading number out there that we reference in the paper is 130. That's the number of estimated radiation deaths because of the Fukushima accident.
You're right, that's what people would mostly focus on. What was the impact of just evacuating because of the meltdown?
With the evacuation, interestingly, that's where more deaths were. Some of the exact numbers are kind of fuzzy what exactly they are, but estimates put it around probably 1000, 1200 deaths or so because of the evacuation process. What's interesting about that is, like you said, that's not the first line of effects that we're expecting to see. Certainly, when it came to the evacuation, there was mayhem when this is happening, but maybe a little bit too much mayhem that led to extra accidents from the evacuation that we shouldn't have seen.
Is that because you're moving sick people from hospitals or there are auto accidents? Why do so many people die during evacuations?
That's a good question, because I think it was a little bit mysterious why the numbers were so high. I think it's the kind of things that you could imagine. We're shuffling people away who maybe weren't the most mobile people to begin with, and were moving them away and that's wreaking havoc as a result. Or you have this mass evacuation and there are accidents along the way. All of that could be contributing to it.
The consequences of Japan’s shift away from nuclear
Those are the deaths that we can calculate from radiation, from the evacuation. But then the Japanese government responded. It responded with a change in energy policy and that had consequences. What did the Japanese government do? And then what did you find about the consequences of their actions?
What the government did—I'd say it was in response to a lot of protests that were happening at the time, at least that was one important contributing factor—was they halted the use of nuclear power as a source of energy in the country of Japan. Until Fukushima happened, about 35 to 40 percent of their energy was coming from nuclear power. And then after Fukushima, it went to zero percent. They weren't using nuclear power for any of their energy. And a couple of other countries that weren't directly involved took similar paths. Germany started phasing out nuclear power, is one big example as well.
Talk about unintended consequences from Germany.
Yes. And that leads to another can of worms that opens up with Germany
Your paper is not even about geopolitics. We could go into that, too. In Japan, no nuclear, which I assume was the cheap energy source for Japan at the time—and the cleanest.
It was. The important thing is that it was a cheap, reliable, and clean source of energy. We don't focus so much on the clean aspect of it in this paper, but I think that's important too. What we really focus on was that it was a relatively inexpensive source of energy. There are all kinds of questions about providing nuclear power. It's very expensive to build these plants, but once they're up and running, they provide energy at pretty low costs. They had a really reliable source of low-cost energy, and all of a sudden they said, “We're going to get rid of 40 percent of our country's energy supply.”
They’ve still got the energy demands that they need to meet. People still need to heat their homes. They need to cool their homes in the summer. Not to mention all of the other basic functions in your home—keeping the lights on and the fridge running. So they had to figure out how to meet those energy demands, and what they did in meeting those energy demands was they started importing fossil fuels. Mainly from mainland China is where they started importing those fuels.
I assume that's coal, mostly.
Coal, gas — those were the two main things that were coming in. And when they were coming in, they were now providing the energy to fill that gap in demand. But they were now more expensive than using nuclear. People's energy bills were going up. Depending on the area of Japan, we're talking about some places were seeing energy bills going up 30 to 40 percent. If you think about during times of the year when you're using energy the most, the middle of winter when you're heating your home to try to keep a nice comfortable environment at home. And now your bill has gone up by 30 to 40 percent—a lot of people ended up cutting back on their energy use.
And this is another economics 101 principle: As the price of a good goes up, people are going to consume less of it. So we saw people cutting back on their energy use, and they're cutting back on their energy use during the coldest time of the year. What's important about that is that's when energy use is really important for your health. One of the things that a climate-controlled environment is doing is it's protecting you from the elements. It's protecting you from freezing temperatures that, most people can relate to, don’t feel good on a regular basis. But also if you have people who might be experiencing heart disease or other kinds of more frail states to begin with, if they're now experiencing colder temperatures for a prolonged period of time, this could be pretty harmful to their health.
And what you did is you looked at a section of large cities to try to figure out those health consequences?
That's what we did. After we found the areas where we saw the decrease in energy usage, we also asked, “Did we see increases in mortality in those areas as well?” And that's precisely what we found. We found that when it was cold out, in the winter—temperatures hovering around 30, 40 degrees, think zero degrees Celsius, right around freezing—when temperatures were falling into that range and energy prices were higher, we saw increases in mortality compared to times where we had similar temperatures but before the increases in energy prices.
You looked at about the 20 or so largest cities over the early 2010s, from 2011 to 2014, and you calculate about 1300 additional deaths.
Because of the higher energy prices, we estimate there are an extra 1300 deaths in the cities where we were able to estimate the effects here.
(1) I assume there would be more if you looked at the entire country; (2) to what extent do you believe this has been an issue beyond 2014?
Both of those are reasons that we think the number is even higher than the 1300. Our study only focused on about 30 percent of the Japanese population, just because that's where the data was available to do the analysis. But if we project our estimates onto other areas, we have every reason to think that there are effects there as well. We think those numbers can be even larger than the 1300. And then also our analysis ended in 2014. There's always a lag on when you can get data and it takes some time to do the analysis. But those effects very likely persisted beyond 2014 as well. So we think the number of deaths is certainly greater than the 1300. It’s hard to put exactly what the number is on it, but we'd say easily it should be several thousand more.
I don't know if you examined this, but do you know if that potential consequence was part of the shutdown debate? Or was the debate just about “This is too dangerous of a technology, we need to shut it down”?
Unfortunately, I think the debate is mostly around just the risks. This is getting back to what we were talking about earlier with the precautionary principle and just focusing on one aspect and losing sight of the whole picture. It was really just saying “This is the risk that we face from using nuclear, so we shouldn't do it,” instead of thinking “These are also the benefits that nuclear brings.” And we haven't even touched on some of the other benefits that nuclear brings. We've just talked about the price benefits right now, that it brings lower energy costs. And we talked a little bit about the potential air quality and greenhouse gas impacts as well. But that's mostly lost at least in the protests that are against it. They're really focusing on the risks and not thinking about the benefits from it.
Matt, I don't know how much time you spend on Twitter. I spend too much. Sometimes I get the impression that there's a certain segment on Twitter who thinks economists are too involved in public policy. But this seems like an absolutely perfect example of where you actually need somebody talking about economics, about trade-offs, about other potential consequences and counterfactuals.
Yeah, I think that's right. Obviously as an economist I'm going to promote what I do as being useful. But it gets back to, I think it was Truman who said, “Give me a one-handed economist,” because all the economists say, “On one hand this, but on the other hand that.” And that's really thinking about the trade-offs here. But it's hard to imagine any situation where there isn't a trade-off, where any decision we're making, there's not also an alternative decision that we can make that has impacts as well. I know that is maybe a bit self-serving and that's what economists do. That's how we approach every problem. We think about not any one thing in isolation, but try to think about the whole problem and everything that's happening in the problem, the costs and the benefits.
Japan’s nuclear reversal
Just to stick with Japan for another moment, they have now reversed course. Japan is now re-embracing nuclear. To what extent have you followed that policy reversal? Part of it seems to be about climate change. I don't know how much also has to do with the kind of findings that you have in the paper.
I don't know the details behind what's driving them to make that decision. But yes, they're planning on bringing back—what is it, seven or nine plants that they're going bring back online? I think a lot of that has to do with the Russian invasion of Ukraine. I think that's a big thing because right now we see gas prices going up everywhere. Whether you're directly relying on Russian gas or not, it's still world markets that are determining prices there. And we see energy prices are going up, and we're especially fearing this winter. The Russian invasion started at the tail end of the winter last year so we had some time to prepare. The hope was things wouldn't go on as long as they are. But they are going on as long as they are, and now we have to think about this winter ahead where we're going have to think about higher energy prices for a lot of people, and how are we going to deal with those higher energy prices?
One way to deal with those higher energy prices is to increase the supply of energy, and that will hopefully bring down the price of energy. I think that's a big part of bringing the nuclear plants back. I think the same thing is happening in Germany as well. They were slowly decommissioning all of their nuclear plants. They hadn't fully phased them out. I think it was actually supposed to happen this year, 2022. And I think maybe now they're putting the breaks on that and saying “We need to keep some of those plants alive,” especially because Germany in particular is very dependent on Russian natural gas.
There was a somewhat similar study done looking at excess deaths in Germany from switching from nuclear to more coal-fired plants, looking at air pollution, dirtier air causes deaths. That is not something you looked at, though I imagine it would be a factor.
That's not something we looked at directly, but it's absolutely another factor. And this is one that we just looked at the price differential between nuclear versus coal or gas. Another super important benefit from nuclear is that when it's actually producing the energy for people to use, it's not emitting any pollutants locally. So it doesn't have an effect on air quality. When the alternative is coal or gas, that's leading to emissions that contributes to particulate matter, which is this really small fine pollutant that gets deep in our lungs, gets into our circulatory system and causes significant number of deaths. I believe particulate matter is the leading environmental cause of mortality around the globe.
Public perceptions of nuclear risks
Do you have a sense that people overestimate the fatalities from nuclear accidents? I don't know if you watched the Chernobyl miniseries on HBO. It would be very easy for someone kind of watching that and maybe half on their phone to think that hundreds of thousands of people died. And I recently watched a documentary, I think it was on Netflix, about Three Mile Island. They also tried to give the sense that many, many people died, even though the evidence seems pretty anecdotal, what they did give. They didn't bring a lot of experts and economists talking about deaths. In political decision-making, it's easy for politicians to focus on highly visible costs and highly visible benefits. But it seems like with nuclear, they're also looking at costs that perhaps don't even exist, that there's just the sense that it's a lot more dangerous than what it is.
I think that's right. Part of it is about salience. When we think about a nuclear meltdown, it's a shot heard around the world. Everyone knows about Fukushima and everybody knows about Chernobyl. Even if we have the opportunity to forget about Chernobyl—not that we should forget about Chernobyl, we should still remember and learn from Chernobyl—we still have reminders about Chernobyl and the miniseries coming out and telling us how bad it was. If we had a Chernobyl-like incident every five years, we'd be having a different conversation. Then we could say, “Maybe nuclear should be off the table. Maybe it's not worth it.”
But we have very different calculations. We've had one Chernobyl, the worst incident we've had. And then the second-worst incident we've had is Fukushima, where we have 1200 or so deaths. It's orders of magnitude better, and that's only the second-biggest accident that we've had. And if that's what things are like going forward, that's much safer, clearly much safer than Chernobyl. I think the interesting thing embedded in the question you're asking me is about the salience, that we hear about the deaths from these accidents. But the deaths from coal and the deaths from gas, you don't hear about them. They're just in the background. Every day there are people dying from particulate matter, but we're not saying, “It's because of the burning of coal we now have this person dying from particulate matter.” It's just that we know there are some people that are dying and we can later statistically attribute it to the burning of fossil fuels. But it's not reaching salience the same way that a nuclear accident is reaching salience.
An analogy that comes to mind is thinking about flying versus driving. Flying, statistically speaking, is so much safer than driving. The amount of airplane-related deaths in a given year is very small, but the amount of deaths from car accidents is, again, orders of magnitude higher. Car accidents are happening every day, and the only way you really hear about car accidents, other than if it's a really famous person who's involved, is your local news. They're not making national news. But when a plane crash happens, it's national news—it's international news. But when was the last big plane crash that happened? I don't know. I couldn't tell you. It has to have been at least several years away. I know COVID put a lot of pause on a lot of travel, but it has to be years away from when that last accident happened. Yet I think a lot of people are more scared of flying than they are of driving, even though the risks of death from driving are much higher than the risks of death from flying.
Where did the precautionary principle come from? Are there philosophers or economists that have been pushing this idea since the 1970s? Or is this a much old idea that has found new salience?
I don't know enough of the history of it going back in time. It definitely gained prominence in the last 50 years. With a lot of the environmental movement that was growing out of the ‘60s and ‘70s, the precautionary principle took more kind of formal definitions and became, in some cases, an official part of policy. That said, the definition of it has changed over time, the precise way that the precautionary principal is defined. Before then, I'm sure it's been used at least unofficially. This is where the precautionary principle gains traction. You can almost think of it as “better safe than sorry.” This is how it gets pushed: “Let's be safe first rather than learn we made a mistake and be sorry later.” It's probably a good way to think about your daily life as an individual, thinking about “better safe than sorry” when you make decisions.
But it's not clear that suddenly becomes a good tool for making decisions for millions of people. It now starts to take a different flavor to it, because as an individual, when you're making this better-safe-than-sorry decision, you're thinking about one thing, one step at a time. In front of you, you have a cup of water, and you say, “I don't know what's in this cup of water. I don't know how long it's been sitting there. Maybe it has some bacteria because it's been sitting there for too long and I could get sick.” So then you say, “Better safe than sorry. I'm not going to drink this cup of water and later I'm going to do something else that's going to quench my thirst.” That's fine, that's perfectly reasonable. I make decisions like that for myself and for my family.
But when you're sitting there making that decision when you have lots of choices in front of you, it becomes different, because then it gets back to like what we talked about before. You have to think about the trade-offs associated with any decision that you make, that if you're not choosing this, you're now choosing something else. And we’ve got to think about the benefits and risks with all the alternatives that we face. And the precautionary principle just gets us away from thinking about those alternatives.
If you were to rank the papers I've mentioned in my writings over the past five years, this paper probably ranks pretty high. It embodies these principles. It provides, I think, a very easy way for people to understand some important principles and concerns a pretty important topic: energy. How much publicity have you gotten from mainstream media about your paper? What kind of response have you gotten, either from environmentalists or economists? I think it is a pretty important piece of research.
Thanks for the kind words. I'd say we've gotten some media exposure from this, but not a tremendous amount. I'm doing this podcast and I did another podcast recently and it's gotten written up in a couple of places, but I'd say it hasn't really hit mainstream media. Some of that, to be fair, is I don't do much self-promotion of my work. You said you spend too much time on Twitter. I spend probably too little time on Twitter. More promotion might have helped with that, so some of that is a result of our choices too.
It seems like, at least as applied to energy because of climate change and more immediately because of the Russian invasion, people are thinking a lot about past decisions and about trade-offs. Do you think views about nuclear energy are changing, as far as the riskiness and more importantly the riskiness compared to other decisions like becoming dependent on a potential enemy for your energy?
I do think it's changing a little bit. And it's interesting how other factors make you change your viewpoint on this. The risk from nuclear hasn't really changed at all, but it's the risk from the other thing has changed. We have to be thinking about the alternative options. We talked about Japan and Germany now changing their tune on nuclear, we also see in the Inflation Reduction Act that was signed recently that now nuclear is being promoted in that as well, encouraging the development of nuclear power. We see Diablo Canyon, the nuclear plant in California, which they were going to decommission. It now looks like they're reversing course on that. I think there is a change.
The question always becomes: How long does this last? If there's another nuclear accident, that could quickly change things. There's a lot of uncertainty when it comes to nuclear and that uncertainty is different than the uncertainty when it comes to other energy sources. If another accident happens, it could be that all of a sudden we change course on nuclear. I hope that doesn't happen, because I feel like we learn from the mistakes. We know with Chernobyl there were just a lot of missteps that happened that led to that accident. And I just heard something recently that they actually had a safety inspection the night before the actual meltdown happened, which is kind of amazing that that could happen. A lot of the information there is behind lock and key with the Russian government, so we don't know a lot of the details. But we've learned over time how to make it safer and safer. Our ability to detect problems is just greater and greater. Hopefully what's happening is that we're learning how low the risk is from nuclear.
And what's important is for everyone to think of it in context to the alternative. That is where we're at a hard point with—we probably don't want to go too far down this path of information and misinformation—what do people know about the risks and how big they are? I don't know that number offhand, what people think about nuclear. But a lot of people think that the risks are worse than they are, and that doesn't help public discourse if people are misinformed about the dangers.