5 Reasons Why the U.S. Won’t Quadruple Its Nuclear Capacity by 2050
5 Reasons Why the U.S. Won’t Quadruple Its Nuclear Capacity by 2050
I’m adamantly pro-nuclear.
But Trump’s pending executive order aiming for a 4X expansion of the domestic fleet isn't realistic.
Robert Bryce
Share
Publication Note: The Fire Time Magazine appreciates the opportunity to republish this article, which was originally released on May 18, 2025. It is reprinted with permission from Robert Bryce. Click here to subscribe to his Substack. Further use, duplication, or distribution is prohibited without the author’s written permission.
Next week, the White House is expected to release several executive orders calling for a massive expansion of nuclear energy. The orders will reportedly set a goal of quadrupling the U.S. nuclear fleet from today’s 100 gigawatts (GW) to 400 GW by 2050.
One of the orders states, “swift and decisive action is required to jump-start America’s nuclear renaissance.” Another calls for reorganizing the Nuclear Regulatory Commission, and another aims to have the Defense Department finance and host new reactors on military bases.
I’m adamantly pro-nuclear. I’ve been saying the same thing for more than 15 years: If we are serious about decarbonization, we should be pursuing N2N, natural gas to nuclear. In my 2008 book, Gusher of Lies: The Dangerous Delusions of “Energy Independence,” I wrote that nuclear is the only source that can “make a significant dent in the overall use of fossil fuels.” I explained that Toshiba, General Electric, and Westinghouse were all working on new reactor designs and that “some of the world’s leading environmentalists have decided that nuclear power is the best option for the future.” I pointed out that Patrick Moore, Stewart Brand, and James Lovelock had become outspoken proponents of nuclear energy.
Again, I’m not changing my position on nuclear. Having ambitious goals can be good. But if the U.S. wants to build hundreds of reactors over the next three decades, we have to be sober about the scale of the challenge. Further, we must look at the history of the global nuclear sector to better understand what may happen in the future.
Here are five reasons why we won’t see a quadrupling of the nuclear fleet over the next 25 years, along with six charts, and my (realistic) prediction about how much nuclear capacity the U.S. should be able to build by 2050.
1. The Trump administration may want 300 GW of new nuclear by 2050, but there’s no historical precedent for building that much capacity that quickly.
The math here is straightforward. Constructing 300 GW of nuclear reactors in 25 years will require commissioning an average of 12 GW of new capacity every year for two and a half decades. Given the recent history of the nuclear sector in America, that goal appears absurd. The newest reactors in the U.S., Vogtle 3 and 4, with a total capacity of about 2,200 megawatts (MW), took 15 years to build. Of course, as more reactors get built, the time needed to construct each succeeding reactor should shrink. But even looking at France, which expanded its reactor fleet rapidly from the 1970s to the 1990s, never came close to building 12 GW per year. (More about France in a moment.)
As seen above, the U.S. roughly doubled its capacity between 1978 and 1997, growing from nearly 50 GW to about 98 GW, which is about the amount we have today. Thus, over those 19 years, the U.S. added 48 GW, or 2.5 GW per year. And yet the Trump administration is saying the U.S. should now grow its nuclear fleet at a rate five times faster than that.
China is building more nuclear capacity than any country on the planet. But it hasn’t come close to building 12 GW of new capacity annually. As seen in the chart above, which uses a screenshot from the World Nuclear Association, since 2010, China has been adding about 2.5 GW of new capacity per year. Thus, China—which doesn’t have to contend with anti-nuclear chuckleheads like Ed Lyman, the Union of Concerned Scientists, or Greenpeace—hasn’t built nuclear capacity any faster than the U.S. did three decades ago.
Finally, the U.S. nuclear sector still lacks momentum. Yes, a handful of companies, including Natura Resources, TerraPower, and Kairos, have construction licenses and are moving dirt, but none have their final licenses from the Nuclear Regulatory Commission. As seen above, the International Atomic Energy Agency does not include the U.S. among the countries building new reactors. (By the way, it’s embarrassing to see Bangladesh and Pakistan are on that list, but the U.S. isn’t!)
2. The costs are still stratospheric.
The two new reactors at Plant Vogtle cost $36 billion, about twice as much as initially projected. That’s nothing new. Cost increases are routine in large construction projects. But they are particularly problematic when it comes to nuclear energy. Last week, the provincial government in Ontario approved a plan to build its first small modular reactor at the Darlington Nuclear Generating Station. The approval of the GE Hitachi BWRX-300 reactor is a significant milestone. If it progresses on schedule, the 300-MW reactor will likely be the first SMR built in North America.
But the cost is Total Bonkers CrazytownTM.
As the Globe & Mail explained, Ontario Power Generation plans to build four of the SMRs at an estimated cost of $20.9 billion (Canadian dollars) or about U.S. $15 billion. Thus, the SMRs will cost about U.S. $12.5 million per megawatt of capacity. The Globe & Mail continued, saying the price tag is “far higher than what independent observers argue are necessary for widespread adoption of SMRs.” The projected SMR cost at Darlington is more than what the Tennessee Valley Authority expects to spend on the BWRX-300 it plans to build at the Clinch River Nuclear Site. Last September, the TVA estimated it would spend U.S. $5.4 billion on the SMR, or about U.S. $18 million per megawatt!
Those costs are particularly galling given that in 2021, when OPG announced plans to build the BWRX-300 at Darlington (with an expected completion date of 2028), the cost for four of the SMRs was expected to be U.S. $3.4 billion, or U.S. $2.8 million per megawatt. Now, just four years later, the costs have increased more than fourfold.
The Globe & Mail also provided a key cost comparison, saying that “a recently-completed 377-MW natural gas-fired power station in Saskatchewan cost” about U.S. $591 million. That works out to about U.S. $1.6 million per megawatt of new capacity. Here’s another key comparison: Last month, China’s State Council approved the construction of 10 new nuclear reactors with about 12,000 MW of capacity. The projected cost is U.S. $27.4 billion, or about U.S. $2.3 million per megawatt of capacity.
Thus, China is now building nuclear plants at costs comparable to those of new gas-fired power plants in North America. Furthermore, China is approving about 10 new reactors per year. Unless or until the cost of building new reactors comes down—and I mean way down—it’s hard to see how the U.S. will build large amounts of new capacity.
3. Without robust financial support from the federal government, we won’t see a rapid nuclear buildout. In fact, we might not see one at all.
Federal support was key to constructing America’s newest reactors, Vogtle 3 and 4. Those reactors were backed by $12 billion in loan guarantees from the Department of Energy’s Loan Programs Office. As I noted here back in January, the LPO under Jigar Shah’s leadership, rushed through a spate of questionable deals during the last few weeks of the Biden administration. Last week, the DOE announced it is investigating (and may terminate) some of those deals, including about $15 billion worth of projects.
However, the LPO will be needed to finance new reactors. Emmet Penney, a senior fellow at the Foundation for American Innovation, is an ardent supporter of new nuclear capacity in the U.S. I spoke to him by phone on Thursday. He was blunt in his appraisal of the need for the LPO. “The utilities and the banks won’t take the risk on new nuclear without federal backing,” he said. “And the state utility commissions won’t approve these projects because of the risk to ratepayers unless the federal government provides loan guarantees or other financial support.”
It’s too early to know what will happen to the LPO. It’s also too early to know what will happen to the nuclear-friendly tax credits passed under the Inflation Reduction Act. Congress is considering legislation that could dramatically cut the corporate welfare going to Big Solar and Big Wind. Let’s hope that happens. But the hard reality is that new nuclear capacity in the U.S. will need federal dollars—and lots of them. If those dollars aren’t available, the deployment of new nuclear reactors could slow to a trickle.
4. There are too many reactor designs.
As I noted a few weeks ago, dozens of companies want to break into the commercial nuclear power business. The tally of SMR purveyors now stands at 50. (Ultra Safe Nuclear is highlighted because it declared bankruptcy last year.) On the surface, having a lot of designs can be seen as a good thing. However, too many designs in the market can hinder adoption because all those reactor designs are vying for regulatory approval.
Standardization would help speed deployment. Between 1978 and 1997, France increased the size of its nuclear fleet nearly 10-fold, going from about 6,400 MW to almost 63,000 MW. Thus, over 19 years, it deployed about three gigawatts annually.
The quick expansion was aided by France’s use of a standardized system of pressurized water reactors designed by Framatome. The International Atomic Energy Agency explains that France has the “highest degree of standardization among countries with large nuclear fleets. This also translates into a standardized approach when dealing with the back end of the nuclear fuel cycle, which involves spent fuel and waste management, decommissioning, and environmental remediation.” (About 34 of France’s 57 operating reactors have 900 MW of capacity. The remaining reactors are rated at either 1,300 MW or 1,450 MW. France’s newest reactor, Flamanville 3, which came online in late 2024, is rated at 1,600 MW. Like Vogtle, Flamanville 3 was plagued by delays and cost overruns. Flamanville 3 took 12 years to build and cost about $14 billion, four times its initial estimated cost.)
If the U.S. is going to deploy new reactors quickly, the industry will have to standardize around a handful of designs, and by that, I mean maybe five different reactor types. It will also need to narrow the market to a handful of reactor sizes, just as France has done.
5. The U.S. has to solve the fuel supply and enrichment problem.
In 2023, in “No U,” I explained that the U.S. will need “dozens of tons” of nuclear fuel that it doesn’t have. I went on, writing:
Four decades ago, the U.S. nuclear sector was largely self-sufficient in uranium and nuclear fuel supplies. In 1980, the U.S. produced a record 43 million pounds of uranium oxide. Today, it isn’t producing any uranium oxide. Over the past four decades or so, the U.S. went from being the world’s biggest exporter of nuclear fuel to its biggest importer.
Five months ago, in a piece about SMRs, I followed up on the fuel supply issue, noting that the U.S. will need to roughly triple its domestic enrichment capacity to accommodate the expected growth in domestic nuclear capacity. The looming ban on imports of Russian-made nuclear fuel has made the need for domestic production even more urgent. Last year, President Biden signed into law the Prohibiting Russian Uranium Imports Act, which, as the EIA noted recently, bans imports of “uranium products from Russia beginning in August, although companies may apply for waivers through January 1, 2028.”
In short, if the U.S. is going to expand its nuclear capacity, it has to fund and develop domestic nuclear fuel production capacity, and it has to make that happen right now.
Given these challenges, how much new nuclear capacity should the U.S. be able to build by 2050?
I’ve listed five reasons why we need to be sober about expanding nuclear energy in the U.S. I could have added other challenges, including the need to reform the Nuclear Regulatory Commission, the nuclear waste issue, supply chain challenges, the uncertainties created by Trump’s trade war, and the availability (or rather, the growing shortage) of skilled labor.
For all those reasons, it will be nearly impossible for the U.S. to quadruple its nuclear capacity by 2050. So what is a reasonable estimate?
As shown in the charts above, it’s realistic to believe that with bipartisan political support at the federal level—and by that I mean unwavering, decades-long, bipartisan support from Congress and the White House—the U.S. could match the productivity of the 1980s, when it was building 2.6 GW of new capacity per year. At that pace, and assuming no existing reactors get shuttered in the interim, the U.S. could have 163 GW of reactors pumping out oodles of sweet, zero-carbon juice by 2050.
But I also believe the U.S. should—repeat, should—be able to double its nuclear capacity by 2050. Doing so would require building an average of four gigawatts per year for 25 years. That’s more capacity than the French built during the 1980s, but I believe it’s an achievable target.
Last month, Matthew Wald and Deric Tilson wrote a sharp piece called “How to Define Nuclear Success” for the Breakthrough Journal. After a disquisition on the market dynamics of the U.S. electric grid, they concluded that the nuclear renaissance “will dawn when a rational actor will calculate that a reactor—large, small or in between—is better than any of the alternatives. That will move the technology beyond the demonstration projects and back into the commercial mainstream.”
The U.S. has to move its nascent nuclear power sector out of demonstration mode and into commercial mode for many reasons, including energy security, land sparing, grid stability, and technological leadership. As I have noted before, making the nuclear renaissance happen won’t be quick, cheap, or easy, but we must make it happen. Trump’s pending executive orders may be overly ambitious, but they are an overdue and much-needed step in the right direction.
Author’s Note: This piece has been corrected. As Rod Adams pointed out in the comments of the original article, I did not convert the cost figures for the SMRs at Darlington from Canadian dollars to U.S. dollars. I have made the correction. Thanks to Rod for noting the error.
Want to get more content like this from Robert Bryce? Click below to subscribe to his Substack.
Robert Bryce
Robert Bryce is a Texas-based author, journalist, podcaster, film producer, and public speaker. Over the past three decades, his articles have appeared in numerous publications including the Wall Street Journal, New York Times, National Review, Field & Stream, and Austin Chronicle. He has given nearly 400 invited or keynote lectures to dozens of groups including the Marine Corps War College, Sydney Institute, Jadavpur University, Northwestern University, and a wide variety of professional associations and corporations. He has also appeared on dozens of TV and radio shows, including NPR, BBC, MSNBC, Fox, Al Jazeera, CNN, and PBS.
