One of the more vexing and heated debates in climate policy has to do with the role of nuclear power in decarbonization. Is it helpful or not? Necessary or not?
Almost every Democratic candidate was asked the question during the CNN climate forum on Wednesday. There was seemingly little consensus, with answers ranging from the tech-boosterism of Andrew Yang and Cory Booker to the skepticism of Bernie Sanders and Elizabeth Warren.
“What’s the deal with nuclear power?” is consistently the most common question I receive. So, for those joining this conversation concerned about climate change but uncertain of the correct take on nuclear, this post is an introduction.
One thing I hope to convey is that “pro-nuclear” and “anti-nuclear” are not considered policy positions. They are identities, ways of signaling membership in a tribe. You sign up for one team and then scold the other team on social media (you will have lots of company).
If you approach nuclear power as a policy question, on the merits, you will find that, like most things, it’s complicated; there are multiple, overlapping issues involved, and the answers cannot be captured in a single binary.
A side note: Nuclear obviously has a long and contentious history in the US. The fight against nuclear power, bound up with the nuclear-freeze movement, was arguably the seed of the US environmental movement. Until relatively recently, being an environmentalist in the US more or less meant being hostile to nuclear power. But as climate change has become a bigger concern, it has become clear that the environmentalist lens on nuclear and the climate-hawk lens on nuclear are different and can lead to different conclusions.
I’m not going to review all that history (it could fill a book); instead, I’ll approach nuclear power purely through the lens of climate change. From a climate perspective, there are three separate issues to understand: existing power plants, the prospect of building new power plants now, and new, yet-to-be deployed technology in development. Let’s walk through them.
Existing nuclear power plants
According to a US Energy Information Administration (EIA) report from June, “there are 59 commercially operating nuclear power plants with 97 nuclear reactors in 29 U.S. states.” Those power plants cumulatively provide about 20 percent of US electricity — and half of its carbon-free electricity, which is to say, as much carbon-free electricity as all renewables and hydropower combined.
(Some people contest the notion that nuclear is carbon-free. They note that greenhouse gases are released by uranium mining and during both plant construction and decommissioning. Without getting into that tangled argument, we’ll just say that nuclear is extremely low-carbon, certainly relative to any fossil fuel.)
America’s nuclear plants are struggling to compete in wholesale power markets against cheaper natural gas and renewables. And they are shutting down: five have retired in the past five years and 12 reactors at nine plants have announced plans to retire ahead of schedule. If current trends in the power sector continue, more will shut down, either because their licenses are up or for economic reasons, in coming years.
To some climate hawks, this is cause for alarm. Each closed nuclear plant represents hundreds of megawatts, sometimes gigawatts, of carbon-free power lost. And while that power could theoretically be replaced by renewables, storage, and efficiency, in practice it’s mostly replaced by natural gas.
For more background on existing plants and the simple argument for keeping them open as long as possible, see this post.
Most Democratic candidates, to my knowledge, have remained silent on the issue of plant closures. Only Bernie Sanders has come out explicitly against keeping the plants open, saying in his plan that he will not renew their licenses. He reiterated his stance at the CNN town hall, citing the problem of nuclear waste. (Whatever you think of nuclear waste, how does it stack up to what every candidate says is an “existential problem”?)
Warren was somewhat more vague, saying, “We’re not going to build any nuclear power plants and we’re going to start weaning ourselves off of nuclear energy and replacing it with renewable fuels — we’re going to get it all done by 2035.”
Warren may have somewhat misstated her own goals, which she recently adopted from Jay Inslee’s plan. Those goals are 100 percent “carbon-neutral” electricity by 2030 and 100 percent “clean and zero-emission electricity” by 2035.
Those quoted words were carefully chosen by Inslee’s team. “Carbon-neutral” allows for some offsets, i.e., emission reductions from other sectors. “Clean and zero-emission” specifically uses “and” to make room for both renewables and nuclear power, or even, say, biomass electricity with carbon capture and sequestration.
The point of that “and” is to allow for an ongoing role for existing nuclear plants, without which getting to 100 percent clean electricity will be much more difficult. (For a longer discussion of this question — exactly what should count as “clean” electricity — see this post.)
A national carbon price would definitely boost nuclear’s fortunes, but most of the policy that stands to directly help these struggling plants is done at the state level. This post has a few examples of states (New York, New Jersey, Connecticut) that have figured out how to compensate nuclear plants for their carbon benefits and keep them open. This post is about a state (Ohio) that has done the same thing in the most corrupt, backward-looking way possible.
Final note: The question of what to do about existing plants is separate from other questions around nuclear. It’s possible to believe nuclear is doomed, that renewables will stomp it out eventually, and still think existing plants should be run as long as possible.
New nuclear power plants
The current generation of new nuclear plants is not doing so well in the US, as anyone following the ill-fated story of the Vogtle nuclear plant in Georgia is painfully aware. Initial permits were filed to add two new reactors back in 2006. Construction began in 2013. The main contractor behind the plants went bankrupt in 2017. After an endless series of delays (still ongoing) and cost overruns, the total price tag now looks like it may hit an eye-popping $25 billion.
It is extremely expensive to build new nuclear plants in the US. Advocates insist this has as much to do with byzantine bureaucracy and inefficient, bespoke construction costs as it does with technology. They believe that if reactor designs were standardized and rules were streamlined, nuclear could compete.
Last year, a group of scientists — scientists largely sympathetic to nuclear power — took a comprehensive look at this question in a study published in the Proceedings of the National Academy of Sciences. You can read this post to learn more about the study, but here is their blunt conclusion about existing nuclear plant designs:
There is no reason to believe that any utility in the United States will build a new large reactor in the foreseeable future. These reactors have proven unaffordable and economically uncompetitive. In the few markets with the will to build them, they have proven to be unconstructible. The combination of political instruments and market developments that would render them attractive, such as investment and production credits, robust carbon pricing, and high natural gas costs, is unlikely to materialize soon.
This is not to say that it would be impossible to make nuclear competitive. It would depend entirely on policy, but then, expanding renewables also depends on policy. Any rapid, large-scale change in the energy system depends on policy!
With enough subsidies and incentives, anything is possible, in theory. But given that nuclear has been steadily subsidized throughout its life in the US, the industry has acted in a consistently corrupt and incompetent fashion, recent new plants have been financial and political disasters, and renewables continue to defy even the most optimistic cost projections, it’s not clear that the sweeping reforms necessary to revive a moribund industry will gain much political purchase.
A final note: The question of whether the currently existing nuclear industry can or should be revived and new plants should be built is complicated, but again, it is separate from the two other policy questions at issue. The answers need not be linked.
New kinds of nuclear power plants
A variety of new and advanced forms of nuclear power are under development. None are currently commercially viable, but some are getting close. The small modular reactor (SMR) being developed by a company call NuScale has already passed the early stages of Nuclear Regulatory Commission (NRC) review (though review is ongoing). The company hopes to build a working reactor in Idaho by 2026. It promises smaller, faster, cheaper, meltdown-proof reactors.
Andrew Yang is extremely enthusiastic about the prospects for thorium reactors and wants to invest $50 billion in them. (Thorium is a more common element than uranium and produces less waste, which stays radioactive for a much shorter time.) There are molten salt reactors, fast reactors, micro-reactors, and, way out at the edge of science, always just over the horizon, fusion reactors.
All of these technologies are in various stages of research and hype. All of them promise cheaper, faster, and safer nuclear power, with a smaller footprint and less waste.
For the record, that PNAS study concluded that no advanced nuclear technology is even close to market viability and none is receiving anything like the support that might make it viable by mid-century (when total decarbonization is needed).
That’s the relevant policy question here: whether government should support the accelerated development and commercialization of these technologies.
There might be an argument against this kind of R&D, but if there is, I’m not aware of it. If climate change is an existential crisis, we need to investigate every possible tool for addressing it. Government-funded research is extraordinarily effective and government support is absolutely crucial to the development of new clean technologies.
The problem here is not so much that anyone opposes R&D so much as the fact that too few people really support it enough to make it a priority. Consequently, the US energy research budget remains abysmally low and advanced nuclear, like almost every promising clean technology, receives only a fraction of the support it should.
Different Democratic climate plans differ in the amount they emphasize and propose to spend on clean energy research. Buttigieg has called to “at least quadruple” the federal energy research budget. Booker would spend $400 billion over 10 years on research Moonshot Hubs in each state. Warren’s “Green Apollo Program” would also put $400 billion toward R&D. So would Joe Biden’s plan. Sanders would spend $800 billion. But there are few details about how much any of that would go to advanced nuclear research.
A final note: The question of whether or how much to support next-gen nuclear technologies through R&D spending is separate from the previous two questions. The answers need not be linked.
The consequences of disagreements over nuclear are smaller than they appear
People love to fight about nuclear power, but the tangible implications of the fights are somewhat less than the rhetorical heat might lead one to believe.
There are people who have taken on pro- or anti-nuclear as an identity, and their positions on the three issues above might be consistently pro or con. But there’s no reason for ordinary people, who aren’t committed to that identity battle, to dogmatically take one side or the other. As I keep saying, they are separate questions.
My general sense is that the emerging consensus is a mix: It’s worth keeping existing nuclear plants open; it’s unlikely the current nuclear industry can be revived; and it’s worth investing in next-gen nuclear technologies.
That’s also my general sense of what’s going to happen, regardless of how people feel about it. States are rallying to save existing nuclear plants; any kind of carbon price, which any Democratic administration would try to pass, would help them as well. Despite nuclear’s considerable cheering section, though, it’s difficult to imagine the political will amassing to plow sufficient subsidies into a dumpster-fire industry to produce a wave of new nuclear plants built with current designs. However, concern over climate change is rising and every Democrat is proposing to boost the research budget; advanced nuclear is likely to benefit from the boost, if it happens.
Meanwhile, Democrats, despite their obsessive disagreements over nuclear, mostly fall somewhere within that consensus. Some of the presidential candidates (Sanders and Warren) emphasize the transition to renewables, some (Yang and Booker) emphasize exciting research into next-gen nuclear technologies, but in practice, I doubt their effects on the nuclear power industry would differ much one way or the other. Ultimately, nuclear’s fate is likely to be determined by economic forces and policy levers that lie outside the president’s control.
Anyway, that is the shape of the nuclear debate, such as it is. The important lesson for most people, those willing to engage in this debate with an open mind and an eye on evidence, is that you don’t have to choose a team, pro or con. There are multiple, distinct policy questions around nuclear power and it is worth engaging each of those questions on their own terms.