Lost in other headlines from President Trump’s recent trade trip to Asia was the announcement that “the U.S. government will arrange financing and help secure permits and approvals for $80 billion worth of” domestic nuclear reactors built by Pennsylvania-based Westinghouse Electric. As part of the deal, the US government could receive a “20% share of future [Westinghouse] profits and a potential 20% stake in the company if its value surpasses $30 billion by 2029.”
Don’t expect Uncle Sam to see any of that money.
Accounts of the arrangement are murky. Over at The Bulwark, Jonathan V. Last tries to decipher it as part of a broader criticism of the Trump administration’s expanding state-enterprise economy (e.g., Intel, US Steel, MP Materials, Nvidia and AMD, quantum computing). According to JVL:
The government has picked the winner for nuclear reactor construction—Westinghouse—by committing to infuse it with $80 billion in funding while also promising to alter regulatory structures to its benefit.
In return, the government will take 20 percent of “cash distributions” after the first $17.5 billion Westinghouse receives and then, in 2029, if the value of Westinghouse’s business created by the government’s investment is over $30 billion, the government will require it to go public via an IPO and the government will then have a 20 percent ownership stake in the resulting company.
In a footnote, he adds that Uncle Sam will likely get his equity stake because “by 2029, Westinghouse will almost certainly be worth more than $30 billion. The very fact of the government’s $80 billion ‘partnership’ assures it.”
Of this, I’m skeptical. Very skeptical.
Nuclear energy is a lousy business. On paper, plants take a long time and a mountain of money to build. In reality, they take twice as long and cost twice as much as what the paper says. The result is, despite two decades of advocates predicting we’re on the brink of a nuclear “revival” led by next-generation plants, these grand public-private deals mostly yield abandoned projects, lost sunk costs, and corporate bankruptcies.
In the current issue of Regulation, nuclear power expert Steve Thomas takes readers through the recent history of the industry. Writes Thomas:
In 2002, President Bush announced his Nuclear 2010 program, so-called because it was expected the first reactor under the program would come online in 2010. …
Utilities quickly responded and more than 30 reactors were proposed…. Some were to be built in states where they would be exposed to a competitive electricity wholesale market, and analysts soon determined this made them too economically risky even with substantial federal subsidies, resulting in the projects quickly being abandoned. In states with regulated electricity markets, utilities were concerned that regulators might not allow them to recover their costs from consumers if there were time and cost overruns. Most of the other projects were abandoned on these grounds, leaving only two to enter the construction stage: a two-reactor project to join an existing reactor at the V.C. Summer plant in South Carolina, and a two-reactor project to join two existing reactors at the A.W. Vogtle plant in Georgia.
… By 2015, both projects were in bad shape, way over time and budget. … In August 2017, the V.C. Summer project was abandoned. The A.W. Vogtle project continued, and the first reactor was completed in July 2023 with the second unit following in April 2024, six or seven years behind schedule and at more than double the forecasted cost. There are now no proposals for additional large reactor projects in the United States.
Interestingly, the reactor for the Summer and Vogtle projects was the Westinghouse AP1000, which is the centerpiece of the new Trump proposal. How did that work out? Thomas writes that, once Summer and Vogtle’s delays and costs started spiking,
Westinghouse, then owned by Toshiba of Japan, was required to offer fixed-price terms to complete the projects. Those prices soon proved far too low, and in March 2017 Westinghouse filed for Chapter 11 bankruptcy protection.
Toshiba, which bought Westinghouse in 2006 for $5.4 billion, sold it in bankruptcy for $4.6 billion in 2018 to a division of Canada-based investment firm Brookfield Corp. Brookfield later sold 49 percent of Westinghouse to Canada-based global uranium firm Cameco (which operates two of the largest uranium mines in the United States, though they’re largely idle because of the low world price for the element). The Trump nuke deal is with Brookfield, Cameco, and Westinghouse.
That must be a relief for Brookfield, given its returns on the Westinghouse purchase so far. Supposedly, Westinghouse will start construction on some units in Poland beginning in 2028, with first power in 2036—though see the twice the time/twice the price note above. And Westinghouse has licensed the AP1000 design to a Chinese firm (more on this in a moment). Other than that, Westinghouse, Brookfield, and Cameco are beating the bushes, hoping for contracts.
But, nuke supporters say, Trump isn’t just offering taxpayer-backed financing for future US projects; he’s also going to prune back the regulatory thicket to help them go forward. Supposedly, onerous safety and environmental regulations impede safe, economical plants. But I wouldn’t bet on this proving true (and not just because Trump has proven to be a weak deregulator). It’s tough to imagine a government more able and inclined to clear the regulatory path for an energy project than China’s, yet the four Chinese AP1000s took ten years to complete, at an uncertain—but almost certainly high—cost. (In fairness, the Chinese licensee has ten other plants in either the design or construction phase; we’ll see how they pan out.)
Nuke supporters might respond that the Trump deal isn’t just about the AP1000 but also Westinghouse’s Small Modular Reactor (SMR), the AP300. SMRs are said to be the future of nuke; they are small-scale plants that supposedly are easier and cheaper to site, build, and operate. The problem is, they haven’t been shown to be any more economical than their big brothers, because no commercial SMRs have been built. From Thomas:
No current SMR design is under construction, much less operating, so these claims—notably those on cost and construction time—are unproven and no more than marketing hype. … No SMR design that is expected to be offered as a commercial reactor has completed a full safety review by an experienced and credible regulator. … Until a comprehensive safety review is successfully completed, it will not be known if the design is licensable or what the costs will be.
Why, Thomas asks, should we expect SMRs to be more economical than their big brothers, whose costs keep climbing? He writes:
[T]here is no clear evidence on why the real cost of large reactors has continuously increased over the history of nuclear power. Is it because of their size or because of how complex the designs have become? If it is complexity, why would SMRs be less complex than large reactors?
Nuke proponents may try one more argument: What about climate change? Nuclear plants operate with basically zero greenhouse gas emissions, so new plants will be an important part of America’s future energy portfolio. The problem is, the Trump administration has no interest in constraining emissions. Even if it did, my Cato colleague Peter Van Doren and energy analyst David Kemp studied whether a reasonable carbon tax would make nuke competitive with other generation, and they report—basically—nope.
The bottom line is there’s little reason to expect Uncle Sam’s deal with Westinghouse et al. will yield a nuclear revival. Given nuke’s track record, Trump would need to be a four-term president before he cuts the ribbon on a new AP1000—assuming any project reaches completion. The odds are even longer that a financially healthy Westinghouse will pay off in both revenue-sharing and equity for Uncle Sam. Instead of the federal government successfully picking a nuclear power winner, it’s far more likely the deal will be a cross between Solyndra and the Trump Taj Mahal.
That’s not to say that Brookfield and Cameco won’t take Westinghouse public in the next few years. I can imagine Trump convincing some Red state governors (a future Governor Paxton in Texas or Mace in South Carolina, maybe?) to add state taxpayers’ and electricity customers’ money to the federal pile. Amidst all the government largesse and PR, Brookfield and Cameco could then sell off their Westinghouse holdings to credulous investors, including the Trump-led US government.
That certainly would make the two Canadian firms great again. But Americans? Not so much.