Every company is, in a certain light, a kind of time machine, and every new product is a missive from the past. When a group of people get together to launch a startup, they’re making a bet that in a few months or years, people are going to want what they’re selling.
In the software industry, the past isn’t too long ago. Because it is possible to code and distribute an app somewhat quickly, a new software product might have only been conceived earlier that year or a year or two earlier.
In a mature consumer-product field — like, say, the car industry — the timeline is longer. A model year 2024 car might have first been conceived of in 2022, and it probably relies on a deeper engineering structure — a “platform” — that might date back to 2018 or earlier. Every new car contains, in essence, two-year-old technology.
But in the “hard tech” industry, the delay can be even longer. It can take more than a decade to get a new type of airplane or power plant to market. These types of technology are the biggest bet of all — because by the time the missive reaches its destination, the world may have changed.
So it was with NuScale, an Oregon-based company developing a small, modular nuclear reactor. Last week, NuScale announced that it was pulling out of a Department of Energy-backed, first-of-a-kind project in Utah.
The company had once planned to build six small, modular nuclear reactors in Utah in conjunction with the Idaho National Laboratory. But despite receiving more than $1 billion in Department of Energy subsidies, NuScale could not make the economics of its project work.
The main problem was that NuScale’s electricity was too expensive. Over the past two years, the estimated price of its project surged, rising by more than 75%. Because electricity projects have to recoup their costs from selling power, those high construction costs helped increase the estimated cost of the project’s electricity by 53%.
By the end, NuScale estimated that power from the project would cost $89 per megawatt-hour. (The average cost of residential electricity in Utah is about $20 per megawatt hour.) Of course, nuclear energy can provide benefits beyond what is captured by price — it is one of the few energy sources that can provide 24/7, zero-carbon electricity — but some costs are too high. NuScale struggled to sell its electrons to nearby towns: It simply could not compete with cheaper electricity from natural gas, solar, or other fuels.
It wasn’t supposed to be like this: NuScale’s smaller size and modular design were supposed to result in lower costs. In essence, NuScale hoped that cost savings would emerge from learning-by-doing and economies of scale — as it got better at making small, modular reactors, it would figure out how to bring down their costs.
That wasn’t a ludicrous idea. Economies of scale have brought down the cost of solar, wind, batteries, and electric vehicles over the past decade. And that idea — that as people do something more, they figure out how to do it more cheaply and efficiently — underpins American and Chinese climate policy.
But the Utah project was the first project of its kind, so NuScale hadn’t yet had the opportunity to take advantage of those economies of scale.
NuScale “shows how much customer matters for a first-of-a-kind deployment. NuScale went down a road that would have proven to be a really interesting model if successful, but it was a lot of legwork,” Ryan Norman, a nuclear analyst at the think tank Third Way, told me. Other advanced nuclear startups have more reliable customer relationships, he added.
Even worse for NuScale, the company found itself building the project amid the worst inflation in a generation. What might have once seemed like a “boring” part of a reactor’s design could create new and spiraling costs.
For instance, NuScale’s design required a lot of concrete, Farah Benahmed, a nuclear policy analyst at Breakthrough Energy, a set of climate investment and advocacy organizations founded by Bill Gates, told me. But concrete costs have risen dramatically, increasing by more than 9% over the past two years and helping to drive the company’s spiraling costs. Other advanced reactor designs don’t rely on concrete to the same degree as NuScale, Benahmed said. (Gates has invested in Terrapower, an advanced nuclear company that competes with NuScale.)
Other key inputs into NuScale’s reactors have also surged in price. From 2021 to 2023, the cost of carbon steel piping more than doubled, according to producer price index data. The cost of fabricated steel plates rose by more than 50%, and the cost of copper wiring rose by 30%.
More broadly, NuScale was founded in 2007 — which means, almost inevitably, that the company was responding to a very different energy moment than the one we have now. At the time, the world was undergoing the first wave of widespread public concern about climate change, driven by Hurricane Katrina, An Inconvenient Truth, and the Intergovernmental Panel on Climate Change’s fourth assessment report. It seemed plausible that Congress might pass a bipartisan cap-and-trade law, which would benefit zero-carbon nuclear power.
Most importantly, U.S. electricity costs were rising, and experts feared they would continue to increase in the 2010s. America’s natural gas supplies seemed to be running out, and the country was preparing to import liquified natural gas in large quantities.
Then came the fracking boom. Cheap natural gas flooded the market, reshaping the domestic energy system and moderating the rise in power prices. The United States never passed a carbon price or a cap-and-trade law. And the economics of building lots of NuScale reactors to provide zero-carbon, 24/7 electricity now look seriously different.
NuScale is not the only clean energy company to run into inflation-driven problems. The offshore-wind company Orsted recently canceled two projects on the Jersey shore due to cost and supply-chain problems. Other offshore projects are also at risk.
Nuclear advocates said that despite its issues, NuScale has accomplished something that no other nuclear startup has. It is the sole nuclear startup to receive approval from the Nuclear Regulatory Commission, the federal agency that must approve nuclear reactors before they can be used. “NuScale has paved the way for how to move through the NRC process. They’re a great example and paved the way for the industry,” Benahmed, the Breakthrough analyst, said.
That approval process took more than four years. It shows another way that it can take years or even decades for “hard tech” companies to get to market — to send their missive from the past to the present.
But despite that long timeline, advocates remain upbeat about the larger industry. “The investor base will do its due diligence to assess what business decisions went wrong with NuScale, but ultimately I think this development is less detrimental to the wave of support we've seen for advanced nuclear from that group,” Norman, the Third Way analyst, said. Because NuScale uses a small version of a light-water reactor — a conventional reactor technology that other advanced-nuclear startups have eschewed — investors probably won’t lose faith in the sector itself.
But they agreed that the make-or-break moment for nuclear is coming up. “The key decision point we need to wrestle with as we continue along the innovation path is: Who is going to lead?” Norma said. “Our allies are waiting. Our competitors are watching. Like it or not, now is the time for the U.S. and industry to prove itself. We've gotta have moxy.”
Editor's note: The original version of this article misidentified one of NuScale’s investors. We regret the error.
Sightline Institute
Penguin Random House
