Fusion energy has the potential to deliver a reliable, zero-emissions electricity source in the future—if companies can successfully construct and operate the facilities. However, a recent study indicates that even if this future materializes, it may not be affordable.
Typically, technologies become less costly over time. Lithium-ion batteries are currently approximately 90% cheaper compared to their prices in 2013. Nevertheless, historically, various technologies tend to follow this trend at different speeds. The expenses associated with fusion may not decrease as rapidly as those for batteries or solar energy.
Making forecasts about the expenses of a technology that is not yet developed is challenging. However, when billions of dollars in public and private investments are at stake, it is important to reflect on the assumptions we hold concerning our future energy portfolio and its expenses.
A vital metric is known as experience rate—the percentage by which the cost of an energy technology decreases every time capacity doubles. A higher rate indicates a quicker drop in pricing and improved economic benefits with scaling.
Traditionally, the experience rates are 12% for onshore wind energy, 20% for lithium-ion batteries, and 23% for solar panels. Other energy technologies have not decreased in price quite as swiftly—fission holds at merely 2%.
In the new study, published in Nature Energy, the researchers sought to refine predictions about fusion’s future costs by estimating the technology’s experience rate. The team analyzed three crucial factors correlated with experience rate: unit size, design intricacy, and customization requirements. The greater the size and complexity of a technology and/or the more it needs to be tailored for various applications, the lower the experience rate.
The study involved interviews with fusion specialists, including public-sector researchers and individuals from private companies. They asked the experts to assess fusion power plants based on those characteristics and utilized that data to forecast the experience rate. (A note here: The study addressed only magnetic confinement and laser inertial confinement, two of the foremost fusion strategies, which collectively are the primary recipients of current funding. Other methods might offer different cost advantages.)
Fusion facilities are likely to be relatively large, akin to other types of plants (such as those for coal and fission power) that depend on generating heat. They may require less customization compared to fission plants—primarily because regulations and safety considerations should be more straightforward—but more than technologies such as solar panels. Regarding complexity, “there was almost unanimous consensus that fusion is exceedingly complex,” remarks Lingxi Tang, a PhD candidate in the energy and technology policy group at ETH Zurich in Switzerland and one of the study’s authors. (Some experts suggested it was literally beyond the scale the researchers provided.)
The final estimate the researchers propose for fusion’s experience rate falls between 2% and 8%, indicating it will experience a quicker price decline than nuclear power but not as dramatic an enhancement as many standard energy technologies currently being utilized.
This suggests that significant deployment—and probably a considerable amount of time—will be needed for the costs of constructing a fusion reactor to decrease substantially, leading to potentially high electricity prices from fusion plants for a while. Additionally, this rate is much slower than the 8% to 20% that numerous modeling studies currently assume.
“Overall, I believe we should question the current levels of investment in fusion,” Tang states. (The US has committed over $1 billion to fusion for the 2024 fiscal year, while private funding reached $2.2 billion between July 2024 and July 2025.) “When considering the decarbonization of the energy system, is this truly the most effective utilization of public funds?”
Nonetheless, some experts argue that examining past trends to make sense of future energy prices might be misleading. “It’s a constructive exercise, but we must remain humble about the extent of our ignorance,” asserts Egemen Kolemen, a professor at the Princeton Plasma Physics Laboratory.
In 2000, many analysts anticipated that solar energy would stay expensive—but then production surged and costs plummeted, largely due to China’s extensive investment, he notes. “People weren’t exactly incorrect then,” he adds. “They were merely projecting what they witnessed into the future.”
How rapidly prices diminish depends on regulations, geopolitical factors, and labor expenses, he states: “We haven’t built the technology yet, so we can’t be certain.”
This article is from The Spark, MIT Technology Review’s weekly climate newsletter. To receive it in your inbox every Wednesday, sign up here.
