There’s a truism in climate circles that says solving global warming doesn’t require anything we don’t already have. Reasonable people can debate that, but the upshot is that electrifying buildings, transportation and industry is the best path forward, and we already have plenty of cheap, carbon-free ways to generate that power. It’s just a matter of will.
Solar and wind usually top the list, but nuclear is often included alongside. All of them are carbon-free energy sources that don’t require any major scientific breakthroughs. Together, the trio generates a third of U.S. electricity, and the majority of that, 19%, is from nuclear.
Nuclear’s continued significance is partly due to inertia. Between the 1950s and the 1990s, the U.S. built 90 nuclear reactors. But over the last 20 years, only three new reactors have been connected to the U.S. grid.
That stagnation has sent the industry searching for answers. What changed? Investors, sensing an opportunity in sclerosis, poured $5 billion into a new generation of fission power startups last year, according to the Nuclear Energy Institute. Many of the new approaches aim to solve the issues that plagued existing reactor designs, like cost overruns, waste and safety concerns.
Yet nuclear fission still faces challenges. Even with the new designs, it probably won’t be enough to make a difference when it comes to the climate because nuclear’s biggest hurdles aren’t technical at all.
Dollars and cents
Part of nuclear’s problem is cost. The U.S.’s newest reactors, Vogtle 3 and 4 in Georgia, are expected to cost $30 billion when the latter is completed (Vogtle 3 was hooked up in August). They’re already seven years behind schedule and more than $16 billion over budget.
The cost of power produced by those reactors is estimated to be a whopping $141 to $221 per megawatt-hour, based on Lazard’s calculations of the leveled cost of energy (LCOE), a common way to compare different generation technologies. That’s about what it costs to run peaker plants fueled with natural gas. They’re only turned on when demand surges, and they’re among the most expensive to run. But Vogtle 3 and 4 aren’t small peaker plants — they’re massive reactors that are supposed to generate power 24/7 at a reasonable cost.
Pretty much every other technology significantly undercuts the price of electricity produced by the new reactors. Onshore wind ranges between $24 and $75 per megawatt-hour, according to Lazard’s LCOE estimates. Utility-scale solar ranges from $24 to $96 per megawatt-hour and $46 to $102 if it’s linked to a four-hour grid-scale battery.
Nuclear advocates point to the fact that reactors are always generating, which increases their value to the grid, and they’re not wrong. Utilities love “firm” sources of energy. But even when accounting for that value, today’s reactors and proposed advanced designs struggle given their high upfront costs, according to the U.S. Energy Information Administration.
Recently, the EIA attempted to determine the competitiveness of various generating sources that are likely to come online in 2040. It calculated a value-cost ratio: Any value above 1 means it’s advantageous to deploy relative to existing sources; anything below means that it either costs too much or doesn’t provide enough value to the grid or both. Geothermal was the big winner at 1.20. Onshore wind and solar were slightly below at 0.88 and 0.98, respectively. Advanced nuclear came in at 0.47. In other words, nuclear would have to halve its costs or double its value. The latter is a steep challenge and the former an impossibility given that most reactors run with better than 90% uptime.
So far, none of the advanced reactor designs that are closer to being built appear able to solidly undercut solar or wind or so-called hybrid renewables, which include battery storage. Small modular reactor (SMR) company NuScale estimates its LCOE will drop to $86 per megawatt-hour once it’s built several reactors. Oklo suggests a very broad range of $40 to $90 for its SMR in its investor presentation from July, figures that notably don’t include decommissioning costs, which could tack another 5% to 10% onto construction costs. They also don’t account for delays in regulatory approval. On that front, the company has already stumbled, with its application to the Nuclear Regulatory Commission having been denied on the first try in 2022.
Elsewhere, the story is the same. TerraPower, which is building a larger reactor, is shooting for $50 to $60 per megawatt-hour. And Radiant Nuclear, a recent Andreessen Horowitz investment, hasn’t disclosed an LCOE for its small reactors, though its choice of markets — diesel backup generators, which are pricey to run — suggests its on the higher end.
The Inflation Reduction Act does offer plants put in operation after 2025 a 30% investment tax credit or $25 per megawatt-hour production tax credit for the first 10 years. But those credits also apply to solar, wind and other zero-carbon sources.
More than money
But let’s say these new companies can bring their LCOE down to a reasonable level. Then what? Get ready for epic NIMBY battles.
It’s true that the public has been warming to nuclear, with 55% of Americans in favor, according to an April survey from Gallup, and 57% in favor, according to an August survey from Pew Research Center. Yet while people may like nuclear in the abstract, it’s likely that most respondents wouldn’t want a reactor anywhere near their homes.
That’s partly why most nuclear plants are massive projects built in rural areas; companies can focus their time and resources on convincing a handful of people to let the project move forward. Now imagine companies trying to deploy SMRs across the country. They’d have to embark on the same PR campaign hundreds of thousands of times over. Keep in mind that just under half the country is opposed to new nuclear plants; smaller groups have stalled more banal projects like apartment buildings for years, even decades.
After reading all that, you may think I’m opposed to nuclear. Quite the contrary. I’m not at all. It’s a safe, proven technology that has the potential to offer predictable, carbon-free power. I just can’t see a way to make it work on a timeline that’s consistent with keeping global warming below 1.5°C by 2050. The fresh crop of startups may solve all the technical and cost challenges in a decade, but then who knows how long it’ll take to achieve public buy-in and deploy the new reactors in meaningful numbers.
Nuclear’s track record over the last 30 years doesn’t inspire confidence.
What about today’s nuclear plants? I say keep them running as long as it’s safe and practical to do so. They’re producing cheap, dispatchable and carbon-free power today; there’s no reason to leave them by the wayside. But at the same time, we shouldn’t bet the planet (or too many dollars) on breakthrough fission technology. Competing carbon-free technologies are already cost competitive with nuclear, and in the decades it’ll take to get new reactors deployed, they’re only going to get better.