isnaini azizah
The allure of fusion power is undeniable. It promises a virtually inexhaustible, zero-emissions source of electricity, a holy grail in the global fight against climate change. However, the immense engineering and scientific hurdles involved in harnessing the power of the stars on Earth translate into significant upfront costs and a complex development process. The question of how quickly these costs will recede as fusion plants are built and deployed is critical for policymakers, investors, and the public alike, shaping decisions about where to allocate vast sums of money in the transition to a sustainable energy future.
Historically, technological progress is often characterized by a phenomenon known as the "experience curve," or learning curve. This concept quantifies the percentage by which a technology’s cost declines each time its cumulative production capacity doubles. A higher experience rate signifies a more rapid cost reduction, leading to faster market penetration and greater economic benefits as the technology scales. For context, lithium-ion batteries, which have revolutionized portable electronics and electric vehicles, boast an impressive experience rate of 20%. Solar modules have also seen remarkable cost declines, with an experience rate of 23%. Onshore wind power, a mature renewable technology, has an experience rate of 12%. In stark contrast, established energy technologies like nuclear fission have a significantly lower experience rate of just 2%, illustrating the inherent challenges in cost reduction for complex, capital-intensive industries.
The new study, conducted by researchers at ETH Zurich and other institutions, sought to provide a more realistic estimate of fusion power’s experience rate by examining factors that historically correlate with cost decline. The team identified three key characteristics: unit size, design complexity, and the degree of customization required. The underlying principle is that larger, more complex, and highly customized technologies tend to have slower learning curves and therefore experience less dramatic cost reductions as production scales up.
To gather data, the researchers conducted in-depth interviews with a diverse group of fusion experts. This included academics involved in public research endeavors and professionals working within the burgeoning private fusion sector. These experts were tasked with evaluating fusion power plants based on the three identified characteristics. It is important to note that the study primarily focused on two leading fusion approaches: magnetic confinement (like tokamaks and stellarators) and laser inertial confinement. These two methods currently attract the vast majority of global funding and research efforts, though other, less-funded approaches may exhibit different cost dynamics.
The consensus among the interviewed experts painted a picture of fusion power plants as likely being relatively large installations, comparable in scale to existing thermal power generation facilities such as coal and fission plants. While fusion plants are expected to require less customization than fission plants, largely due to anticipated simpler regulatory frameworks and safety protocols, they are still predicted to need more bespoke engineering than mass-produced technologies like solar panels.
Perhaps the most striking finding relates to the perceived complexity of fusion. Lingxi Tang, a PhD candidate in energy and technology policy at ETH Zurich and a co-author of the study, highlighted the near-unanimous agreement among experts that fusion is "incredibly complex." Some respondents even indicated that the complexity of fusion systems exceeded the scales provided by the researchers, underscoring the unprecedented scientific and engineering challenges involved.
Based on these expert assessments of unit size, design complexity, and customization needs, the researchers projected an experience rate for fusion power in the range of 2% to 8%. This places fusion on a learning curve that is steeper than nuclear fission but significantly less pronounced than that of most renewable energy technologies currently being deployed.
The implications of this lower experience rate are substantial. It suggests that achieving a significant reduction in the cost of building a fusion reactor will necessitate a very large number of deployed plants and likely a considerable amount of time. Consequently, electricity generated by fusion power plants could remain expensive for an extended period. This projected cost trajectory diverges from the more optimistic assumptions of 8% to 20% learning rates often used in current fusion modeling studies, which can lead to an overestimation of future cost competitiveness.
The study’s findings have prompted critical questions about the current scale of investment in fusion power. Tang voiced concerns, stating, "On the whole, I think questions should be raised about current investment levels in fusion." She further questioned the allocation of public funds, asking, "If you’re talking about decarbonization of the energy system, is this really the best use of public money?" This sentiment is particularly relevant given the substantial financial commitments being made, with the US allocating over $1 billion to fusion in fiscal year 2024 and private sector funding reaching $2.2 billion between July 2024 and July 2025.
However, the study’s conclusions are not universally accepted, and some experts caution against relying too heavily on historical cost trends to predict the future of energy technologies. Egemen Kolemen, a professor at the Princeton Plasma Physics Laboratory, acknowledges the value of such exercises but emphasizes the inherent uncertainties. "It’s a good exercise, but we have to be humble about how much we don’t know," he stated.
Kolemen offered a compelling historical parallel: the dramatic price collapse of solar power. He recalled that in the year 2000, many analysts predicted solar would remain prohibitively expensive. However, a massive surge in production, largely driven by China’s strategic investment, led to an unprecedented drop in prices. "People weren’t exactly wrong then," Kolemen explained, "They were just extrapolating what they saw into the future." He argues that future cost reductions for fusion will be influenced by a complex interplay of factors, including evolving regulations, geopolitical dynamics, and labor costs. "We haven’t built the thing yet, so we don’t know," he concluded, underscoring the speculative nature of any precise cost predictions at this early stage.
While the Nature Energy study provides a valuable, data-driven perspective that tempers expectations of rapid cost declines for fusion power, the future remains open to innovation and unforeseen breakthroughs. The inherent complexity and novelty of fusion technology mean that its economic journey may well defy easy categorization based on past technological learning curves. The ongoing debate highlights the critical need for continued research, transparent assessment of progress, and careful consideration of investment priorities as the world strives to build a sustainable energy future.
