As 2024 dawns, the prospects for a nuclear revival in the U.S. look mixed. On one hand, many once-skeptical environmentalists now support the technology. Bipartisan majorities in Congress back funding for nuclear research and deployment. And more than two dozen startups are developing a new generation of small, innovative reactor designs. But momentum slowed in November, when NuScale Power, the Portland, Oregon-based company pioneering small modular reactors (SMRs), announced the cancellation of its showcase project to build a power facility in Idaho. If completed, the project would have been the nation’s first SMR power plant. Backers hoped that the endeavor would prove that this new approach to nuclear technology can deliver affordable, zero-carbon power to our electric grid. Instead, delays and escalating costs forced NuScale to shelve the project before it even broke ground.
News of the NuScale venture’s collapse was met with dismay by many in the nuclear industry. “Not gonna sugar coat it, this is a major blow to the next generation of nuclear power,” Katie Mummah, a leading nuclear proponent and a Ph.D. candidate at the University of Wisconsin-Madison, wrote on X. “This is bad news for the entire nuclear sector, not just the upstart nuclear companies,” energy analyst Robert Bryce told me in an email exchange. But not all experts I surveyed took an apocalyptic view. Adam Stein, director of the Nuclear Energy Innovation program at the eco-modernist Breakthrough Institute, said, “It’s definitely a problem—it’s not great—but it doesn’t have to be a roadblock for the prospects of advanced reactors in general.” In a report cowritten with Breakthrough Institute president Ted Nordhaus, Stein maintains that advocates can’t just “whistle past this graveyard.” Instead, the nuclear industry and its supporters need to confront “the seriousness of the challenges that the technology presently faces and what it will take to overcome them.”
Not all nuclear news was bad in 2023. The United States’s existing reactor fleet now appears secure, after a long period in which power companies, bowing to political and economic pressure, prematurely retired serviceable plants. California’s iconic Diablo Canyon plant, once slated for closure, will keep operating. And a Michigan reactor that closed in 2022 now appears likely to be brought back online. Meantime, the Department of Energy expanded several programs supporting next-generation nuclear research and demonstration projects. Several U.S. companies—including NuScale—are in talks to supply advanced reactors to clean-power ventures in North America and overseas. At the recently completed United Nations COP28 climate summit in Dubai, 22 countries—including the U.S., Great Britain, Japan, Sweden, and South Korea—signed a pledge to triple their nuclear capacity by 2050. While the agreement is largely aspirational, it shows a surprising enthusiasm for nuclear power at an international forum that, until recently, kept nuclear advocates on the sidelines.
Nonetheless, the implications of NuScale’s woes resonated, even in Dubai. Poland and other countries aiming to boost nuclear power production have outlined plans to rely at least partly on the anticipated wave of SMR technology. But if U.S. firms hope to dominate this potential market, they will need to have designs tested and ready for export in the relative near term. As if capitalizing on the NuScale news, China announced that it brought a pair of small, gas-cooled reactors online at its Shidao Bay Nuclear Power Plant on December 6.
The U.S. can’t become a global leader in nuclear technology—or even modestly boost its own nuclear output—if it can’t find a way to start building new plants. Most industry observers saw NuScale’s Idaho project as the flagship for those hopes. My 2020 City Journal article, “Next-Gen Nuclear Power,” featured NuScale prominently. The company’s basic concept makes sense: the conventional power reactors operating in the U.S. today are all cooled with ordinary water—or “light water”—and have an average output of about 1,000 megawatts. The last such project to be completed, two enormous light-water reactors at Georgia’s Vogtle power plant, ran wildly over budget, reaching a total cost of $35 billion.
Not surprisingly, no other utilities seem eager to follow in Vogtle’s footsteps by building such behemoths. Instead of constructing big reactors on-site, NuScale (and most other nuclear startups) propose to build smaller units in factories. These modules could then get shipped by truck or barge to a location where as many as 12 reactors could be combined to form a single power plant. NuScale planned to install six of its 77-megawatt SMRs at the Idaho location.
In theory, building small modular reactors in factories would avoid many of the construction delays that have bedeviled projects built on location. And costs should fall as more units are produced. But someone needs to take the plunge and build the first iteration. As Mummah notes, “Without a first-of-kind reactor, you’ll never get to nth-of-a-kind costs no matter how modular the reactor.” The NuScale project’s collapse so upset nuclear backers in part because the firm had already cleared so many hurdles.
“NuScale is the only company with an SMR design that has been approved by the NRC,” Bryce says. “It had a site for the reactor, and it had customers,” namely a consortium of utilities in the Intermountain West known as the Utah Associated Municipal Power Systems, or UAMPS. NuScale’s design uses the same fuel as existing reactors, Bryce adds, and had staunch backing from the Department of Energy, which had committed roughly $1 billion to the project. “And yet, it still got canceled.”
Though not totally unexpected, the Idaho project’s demise sends a worrisome signal. NuScale had been grappling with delays and rising cost projections for several years. Some of the utilities that signed up to purchase power from the plant later dropped out. Wall Street short-sellers were circling the company. “NuScale had invested half a billion dollars in getting permits from the NRC,” energy analyst Meredith Angwin told me. “If a power plant does not get built after that level of investment, it discourages other groups from investing in similar power plants.” Indeed, other advanced-reactor pioneers are showing signs of stress: X-energy, another company backed by the DOE, recently canceled plans to go public.
Nevertheless, nuclear skeptics’ argument that a new generation of nuclear power plants isn’t essential—and that the country can instead replace carbon-based fuels primarily with wind and solar power—remains shaky. Electricity demand is ramping up in the U.S., especially as federal and state policies encourage electric vehicles and electric home heating. At the same time, as Angwin documents in her book Shorting the Grid, the nation’s electricity supply and distribution networks are growing dangerously unreliable. Regulations intended to reduce carbon emissions are forcing the early retirement of dependable coal-fired power plants, while subsidies boost the rollout of intermittent wind and solar. To date, no large power grid has proved able to run primarily on these sources.
More and more climate advocates now recognize that renewables alone can’t deliver the zero-carbon power grid they desire. Since wind and solar produce power only about one-third of the time, they require massive amounts of “dispatchable generation”—power that can be produced whenever it’s needed—as backup. In the U.S., that backup power increasingly comes from natural gas. The need for all that redundant generation capacity means that trying to run the grid with mostly wind and solar results in both higher costs and higher emissions than backers typically claim.
Some experts argue that the looming crisis in reliability is a more pressing problem than carbon emissions. They warn that relying on a grid powered mostly by wind, solar, and gas could lead to disaster. Thanks to fracking, the U.S. is blessed with ample gas resources. But as Texas learned during the deadly winter blackouts of 2021, natural-gas supplies often fall short during cold snaps. Grid resilience requires robust power sources that don’t depend on the weather—or on just-in-time fuel deliveries. Coal and nuclear plants store fuel on site, which makes them a vital resource during power emergencies. With coal plants closing, and nuclear failing to expand, the U.S. grid faces a dangerous shortfall in reliable capacity.
Despite these risks, the push to eliminate coal power remains politically potent and seems unlikely to be reversed. That leaves nuclear as our only major power source that is at once dependable, zero-carbon, and capable of significant growth in capacity. If nuclear is ever to become a growing source of electricity, it’s vital to understand what went wrong with the first major effort to build a commercially viable, advanced-reactor power plant.
Analysts point to several key factors that derailed the NuScale project. One is the intensely regulated environment in which nuclear power operates. The NRC’s extreme safety requirements put burdens on the nuclear industry that far exceed any detectable health benefits. For example, one study shows that the NRC’s standards for reducing cancer risks related to radiological exposure are 100 times stricter than the Environmental Protection Agency’s rules for air pollution exposure. Ironically, by driving up the costs of nuclear power with such requirements, the NRC has incentivized utilities to rely on dirtier sources of power with far greater health impacts.
Next-generation reactor designs face additional hurdles. The NRC’s byzantine process for licensing new designs was geared toward twentieth-century technology: big, complex, light-water reactors built on-site. New reactor designs are inherently simpler, making them “walk-away safe.” But that doesn’t seem to give them an advantage when seeking NRC approval.
It took NuScale more than six years to navigate the licensing maze. Congress has demanded that the NRC develop a more streamlined approval process for this new generation of reactors. But so far, industry experts say, the agency’s proposed rule changes are even more complex than the regimen they would replace. In 2022, the NRC abruptly rejected an application from the Silicon Valley startup Oklo to build and operate an innovative 1.5–megawatt microreactor. No one knows what unseen approval hurdles other startups might encounter. “It’s clear that the exorbitant cost of getting through the NRC is going to constrain new entrants into the nuclear space,” Stein says.
Smaller reactors should be easier to finance, faster to build, and simpler to install where the power is needed. But, as the Breakthrough Institute report notes, “those advantages evaporate if SMR developers are not able to benefit from the safety advantages of smaller and simpler designs.” In the cases of both NuScale and Oklo, the report continues, “the NRC consistently rejected the safety benefits of smaller reactors, instead reverting to regulatory methodologies, standards, and design requirements established for large light-water reactors.” These requirements raise costs at every point in the nuclear-power supply chain.
Other, less obvious, costs are imposed by the webs of federal and state regulations and subsidies that aim to boost renewable energy. For example, electricity markets in many regions let power producers sell electricity on an almost minute-by-minute basis. The idea of a more market-based electric grid sounds appealing. Unfortunately, these auction-based systems effectively subsidize on-again, off-again wind and solar producers at the expense of more reliable power providers. Wind and solar facilities sit idle most of the time. But when they do produce electricity, they can undercut the prices that dependable “baseload” power producers can charge. That puts capital-intensive nuclear power at a disadvantage. Modern life depends on 24/7 electricity; wind and solar providers pay no penalty for their intermittency.
Economic headwinds pose another challenge for new nuclear companies. Interest rates have roughly doubled since NuScale first struck a deal with the UAMPS consortium of utilities in 2016. And inflation has hammered the materials that NuScale needed to construct its plant. An ominous UAMPS report in January 2023 warned that, over the previous two years, prices for fabricated steel plate had risen 54 percent, carbon steel piping 106 percent, and so on. Overall, those swelling prices helped drive up the project’s construction costs by 75 percent, from $5.3 to $9.3 billion.
Shifting economic tides haven’t just affected nuclear power. “High interest rates and higher commodity prices are hurting every player in the energy sector,” Bryce says. “But those factors are having a particularly large impact on industries with high upfront costs and slow payouts.” Offshore wind and other capital-intensive forms of low-carbon energy have been hit hard, too. The Danish wind-power giant Orsted recently pulled out of two enormous projects off the New Jersey coast. The projects had enjoyed lavish federal and state subsidies, but it wasn’t enough. Orsted’s CEO Mads Nipper told reporters that “the world has . . . turned upside down.”
Biden administration officials often express the goal that the U.S. will “lead the world in nuclear innovation.” Executives in the advanced nuclear sector echo that sentiment, stressing that nuclear power technology could again become a major export business for American companies. The failure of NuScale’s Idaho project dims, but doesn’t extinguish, those hopes.
“It further confirms my belief that new nuclear will succeed in Europe and China before it succeeds in the U.S.,” Bryce said. There’s no question that the industry’s outlook is brighter overseas. The COP28 pledge to triple nuclear output might be aspirational, but it shows that political opposition to nuclear power has faded in many parts of the world. In 2022, the EU voted to include nuclear power in its “taxonomy” of climate-friendly investments, removing a key barrier to investment.
Notwithstanding China’s head start, U.S. companies remain in the running to market SMRs internationally. Last summer, Canada’s Ontario Power Generation company signed a contract to purchase four 300-megawatt SMRs built by the Wilmington, North Carolina, joint venture called GE-Hitachi Nuclear Energy. Those are slated to be installed at Ontario’s existing Darlington nuclear plant. At COP28, Poland announced that it has chosen six locations where it plans to install 24 GE-Hitachi SMRs. The United Arab Emirates has also signed a deal to explore purchasing the GE-Hitachi SMR, as well as a “memorandum of understanding” with the Bill Gates-backed TerraPower to deploy that company’s molten-salt-cooled Natrium reactor. Despite its Idaho setback, NuScale has signed agreements to explore building power plants in Romania and Poland.
In the U.S., a number of advanced reactor projects show potential. One promising scenario is to build SMRs on or near the sites of recently shuttered coal or nuclear plants. The DOE is helping fund a demonstration project to install a Terrapower reactor near the site of a retiring coal plant in Kemmerer, Wyoming. Holtec International, a company that specializes in decommissioning nuclear plants, has been developing its own 300-megawatt SMR for more than a decade. The company recently announced a plan to install two of the units on the grounds of the recently shuttered Palisades nuclear plant in Michigan. The company is also exploring installing SMRs on the site of New Jersey’s retired Oyster Creek nuclear plant, which Holtec is currently decommissioning.
Moreover, supplying electric power to the commercial grid is not the only potential market for advanced reactors. Small reactors generate prodigious amounts of heat, which can be used to produce electricity but also to drive high-temperature industrial processes, such as manufacturing chemicals or cement. Earlier this year, X-energy signed a DOE-backed joint development deal to place four of the company’s gas-cooled reactors at a Dow plant on the U.S. Gulf Coast. And in mid-December, the NRC offered an early Christmas present to Kairos Power, issuing a permit that will allow the startup to begin construction on a 35-megawatt, molten-salt-cooled reactor at Oak Ridge, Tennessee. The project is intended to demonstrate the reactor’s ability to produce industrial heat, rather than electricity.
The growth in power-hungry data centers (due to skyrocket as AI demand grows) could provide another attractive market. NuScale and other companies are exploring deals to build SMR power plants that could pair directly with large data centers. Microsoft and Google have both expressed interest in such an approach. Digital giants hoping to burnish their green credentials might well pay a premium for exclusive access to electricity that is both reliable and 100 percent carbon-free.
If advanced nuclear power boasts many green shoots, it has nothing yet that could be considered a sturdy young tree. That leaves nuclear advocates searching for new policy approaches.
Until recently, many nuclear backers, including in Congress and the White House, have focused on providing federal subsidies to promising projects. But NuScale’s retreat shows that subsidies—while perhaps useful in the short term—aren’t enough to build the industry. It is more important to lower the barriers that impede getting new plants built. That will include reforming the NRC, which the Breakthrough Institute calls “a notoriously capricious regulator.” When it comes to human health, the U.S. nuclear industry has a remarkable safety record; it should be liberated from today’s scientifically dubious regulations. The U.S. also needs to reform the byzantine permitting requirements that hinder construction of most major infrastructure projects.
More broadly, the confusing web of incentives, subsidies, and regulations that governs our power grid needs an overhaul. It makes no sense to prioritize expensive, intermittent energy sources while undermining nuclear power, our best source of clean, dependable power. On a level playing field—especially one that puts a market value on reliability—nuclear power still proves economical. “Advanced reactors are going to be a key component of a resilient, cost-effective energy system in the future,” Stein says. In fact, it might be premature to rule out the possibility that even full-size light-water reactors could make a comeback. One MIT study shows that large reactors of the type that ran over budget at Georgia’s Vogtle plant, would be economically competitive if built today. After all, roughly 60 reactors are under construction around the world—in China, India, South Korea, and other countries—and virtually all of those are full-size designs.
Finally, the anti-nuclear tides that pushed U.S. power companies to close hardworking nuclear plants prematurely appear to have finally crested. In 2022, Michigan’s single-reactor Palisades nuclear plant shut down, despite last-ditch efforts by Governor Gretchen Whitmer and others to keep it open. In a planned transition, the plant passed into the hands of Holtec International for decommissioning. But Holtec did not rush in crews to start dismantling the equipment, instead announcing its interest in refurbishing and reopening the 800-megawatt facility. No retired nuclear reactor had ever been brought back into service in the U.S. Later that year, I published a Manhattan Institute issue brief outlining the many obstacles to reopening closed nuclear plants.
But Holtec persisted and, this past October, announced that it will submit an application to the NRC to upgrade and reopen the plant. The project will get financial support from the State of Michigan and a DOE program intended to help threatened nuclear plants survive. If all goes as planned, the reactor could be back in service by 2025. With plenty of work and a little luck, the existing reactor will be joined by two 300-megawatt Holtec SMRs sometime around the end of the decade. A Holtec spokesperson says that the company wants to turn the half-century-old facility into “a mega-clean energy provider to the region.”
If Holtec’s plans bear fruit, the resurrected Palisades plant could offer a replicable model for the future: a single facility that shows it is possible to keep America’s durable existing nuclear infrastructure humming, while adding extra capacity with innovative new reactor designs. Getting there, however, will require profound reforms in how the U.S. approaches the economics and regulation of our electric grid.