Dr. Andrew Cottone is a Visiting Fellow for the Center for Energy & Environmental Policy at the Caesar Rodney Institute.
America doesn’t need to reinvent the nuclear reactor every time it builds a power plant. It needs to learn how to build proven designs repeatedly, safely and affordably.
Delaware has an opportunity to learn from decades of costly nuclear construction experience, documented by the U.S. Department of Energy, as it considers its own nuclear energy future. The Delaware Nuclear Energy Feasibility Task Force, on which I served, recently concluded that nuclear power could play an important role in the state’s long-term energy future, but small modular reactors are not yet a near-term solution. Questions remain about construction timelines, costs, suitable locations, regulations and public acceptance.
Rather than rushing into expensive, one-of-a-kind projects, Delaware should use this time to develop clear policies and identify potential sites, while the nuclear industry demonstrates which reactor designs can be built reliably and affordably, again and again.
Standardization is key to affordable nuclear energy
Nuclear energy already produces vast amounts of reliable, low-carbon electricity. Yet its future in the United States may depend less on a scientific breakthrough than on a basic industrial principle: Build the same proven design repeatedly.
France took a more coordinated approach to nuclear power, beginning in the 1970s, standardizing much of its reactor fleet around a limited number of pressurized-water designs. Designs evolved, but the core systems remained similar. Repetition allowed construction teams to gain experience, operators to train on familiar equipment, regulators to review known systems and suppliers to manufacture parts at greater scale. Standardization did not eliminate every cost or performance problem, but it helped France build a large fleet of nuclear reactors more quickly and consistently than a one-off approach would have allowed.
Nuclear power is often criticized as expensive, but the evidence suggests that the true challenge is not the technology itself but how it is deployed. France built a world-leading nuclear fleet by standardizing around a handful of reactor designs, achieving construction costs of roughly €1.25 million per megawatt, measured in 2010 euros, according to France’s national audit office. The United States, meanwhile, built dozens of reactor variants and saw its recently completed Vogtle units 3 and 4 project in Georgia reach approximately $15 million per installed MW in 2024 dollars, excluding financing costs, according to the U.S. Department of Energy. Adjusted for inflation and currency differences, France’s historical cost was approximately $2.4 million per MW in 2024 dollars — less than one-sixth of Vogtle’s cost.
Can new reactor designs deliver on their promises?
SMRs are intended to be smaller, more flexible and more factory-built than traditional nuclear plants. They could serve industrial sites, military installations, remote communities and data centers. However, the field is fragmented. According to the Nuclear Energy Agency, developers are pursuing many competing concepts, including light-water, molten salt, gas-cooled, sodium-cooled and lead-cooled reactors. Only a small number of SMR projects are operating commercially.
That diversity also masks a deeper constraint: Most SMR designs have little or no fleet-scale operating history — experience gained from operating multiple reactors of the same design over time. Unlike conventional light-water reactors, whose performance reflects decades of accumulated experience across hundreds of units, many new SMR technologies have not yet demonstrated long-term reliability, maintenance requirements, fuel performance, component life, staffing needs or actual operating costs under commercial conditions. Engineering models and test facilities are essential, but they cannot fully substitute for years of real-world operation. The industry, therefore, needs disciplined first-of-a-kind demonstrations, followed quickly by repeated deployment of the designs that perform best, so operating evidence and standardized learning can accumulate, rather than being dispersed across dozens of unrelated concepts.
Practical path forward: Standardize proven nuclear designs
The answer is not to stop innovation or have the government choose a winner too early. Developers should continue testing new ideas, and regulators should demand strong evidence of safety and performance. Once designs have demonstrated their safety and reliability, policymakers, utilities and investors should favor repeat deployments. Reusing approved designs and common components can shorten reviews, strengthen domestic supply chains, improve workforce skills and give manufacturers the volume needed to lower costs.
Nuclear power has the potential to provide reliable, around-the-clock electricity for generations. But technical promise alone will not make it affordable. The United States must pair innovation with industrial discipline. In nuclear energy, the breakthrough we need may be the willingness to stop reinventing the plant — and start building proven designs again and again.
Nuclear energy’s biggest obstacle may be our inability to resist redesigning the toaster every time we want another slice of bread.
Reader reactions, pro or con, are welcomed at civiltalk@iniusa.org.
