Fusion Power Probably Won’t Be Cheap. We Should Talk About That.

Fusion power is the dream: steady, zero-emissions electricity, no meltdowns, no radioactive waste hanging around for millennia. But a new study from researchers at ETH Zurich throws a bucket of cold water on the hype. Even if we get fusion plants built and running—which is still a very big if—the electricity they produce might never be cheap.

Technologies usually get cheaper as we build more of them. Lithium-ion batteries are roughly 90% cheaper than they were in 2013. Solar modules follow a similar curve. But not everything follows that path. Nuclear fission, for example, has barely budged in cost over decades. The key metric here is something called the experience rate: the percentage by which costs drop every time installed capacity doubles. For onshore wind, it’s about 12%. For lithium-ion batteries, 20%. For solar modules, 23%. For fission? A paltry 2%.

The ETH team, led by PhD candidate Lingxi Tang, wanted to figure out where fusion might land on that spectrum. Since we don’t have any working fusion plants to measure, they took a different approach. They identified three characteristics that historically correlate with a technology’s experience rate: unit size, design complexity, and how much customization each installation needs. Bigger, more complex, and more customized generally means slower cost declines.

They then interviewed a bunch of fusion experts—both from public research labs and private companies—and asked them to rate fusion plants on those three axes. The results weren’t great.

Fusion plants will likely be large, similar to coal or fission plants that generate heat. They’ll probably need less customization than fission plants, because regulations and safety requirements should be simpler. But on complexity, the experts were nearly unanimous: fusion is incredibly complex. Some literally said it was off the scale the researchers gave them.

The final estimated experience rate for fusion: between 2% and 8%. That’s faster than fission, but far slower than solar, wind, or batteries. And it’s a lot lower than the 8% to 20% that many energy modeling studies currently assume.

What does that mean in practice? It means you’d need a massive amount of deployment—and probably decades—for the cost of building a fusion reactor to drop significantly. Electricity from fusion plants could stay expensive for a long, long time. Tang puts it bluntly: “If you’re talking about decarbonization of the energy system, is this really the best use of public money?” The US allocated over $1 billion to fusion in fiscal year 2024, and private-sector funding hit $2.2 billion between July 2024 and July 2025. That’s a lot of cash for a technology that might never compete on price.

Now, there are caveats. The study only looked at magnetic confinement and laser inertial confinement—the two dominant approaches that get almost all the funding. Other, more exotic fusion concepts could have different cost profiles. And as Egemen Kolemen from Princeton Plasma Physics Lab points out, extrapolating from past trends can be misleading. In 2000, plenty of analysts said solar would stay expensive. Then China went all in, production exploded, and prices crashed. “People weren’t exactly wrong then,” he says. “They were just extrapolating what they saw into the future.”

He’s right. We don’t know how regulations, geopolitics, or labor costs will play out. We haven’t built the thing yet. But that’s exactly the point: we’re betting billions on a technology whose economics are a complete unknown. The ETH study at least gives us a more grounded estimate than the hand-wavy optimism we’ve been running on. It’s not a death sentence for fusion, but it’s a reality check. If you’re putting money into fusion expecting it to be the next solar panel, you might want to reconsider.