How governments are funding SMRs — and whether it's working
Billions of dollars, bold promises, and one very embarrassing cancellation — the state of public money in the small modular reactor race.
Governments have never exactly been shy about betting big on energy. Coal subsidies, oil depletion allowances, decades of loan guarantees for conventional nuclear — the public purse has a long history of picking sides. Now it’s picking SMRs. Across the U.S., Canada, and the UK, federal and provincial treasuries are pouring billions into small modular reactors with a confidence that is, frankly, striking given how few of these machines have actually produced a single commercial kilowatt-hour. The question worth asking — clearly, without flinching — is whether the money is doing what money is supposed to do, or whether it’s papering over problems that subsidies can’t fix.
That’s not a rhetorical dismissal. The case for government involvement in SMR development is real and defensible. These are first-of-a-kind technologies with long lead times, enormous upfront costs, and supply chain gaps that private capital alone won’t bridge. The International Energy Agency now describes nuclear as experiencing a “strong comeback”, and the International Atomic Energy Agency has raised its global nuclear projections for five consecutive years. Something is definitely happening. Whether what’s happening is a genuine industrial renaissance or an enormously expensive dress rehearsal is the real question, and the answer is probably both.
The U.S. playbook: milestone grants and first-mover bets
The American approach to SMR funding has settled into a familiar structure: cost-shared grants, phased milestones, and a bet that getting a couple of reactors built first will unlock an orderbook that drives down costs for everyone who follows. In December 2025, the U.S. Department of Energy selected Tennessee Valley Authority and Holtec Government Services to each receive $400 million in federal cost-shared funding to support early deployments of advanced light-water SMRs. 🔬
The two projects couldn’t be more different in character. TVA plans to advance deployment of a GE Vernova Hitachi BWRX-300 at the Clinch River Nuclear site in Tennessee, alongside accelerated deployment of additional units with Indiana Michigan Power and Elementl. Meanwhile, Holtec plans to deploy two SMR-300 reactors — named Pioneer 1 and 2 — at the Palisades Nuclear Generating Station site in Michigan, demonstrating viability for additional orders both domestically and abroad. Both projects are aimed at bringing new nuclear generation online in the early 2030s, which sounds reassuringly specific until you remember that nuclear timelines have a long tradition of gentle optimism.
The money itself comes from a program with some bureaucratic history. The $900 million for SMR support was appropriated in the Consolidated Appropriations Act of 2024 using funds from the Bipartisan Infrastructure Law, first announced by the DOE in June 2024. When the Trump administration took over, the solicitation was reissued with modifications — the community benefits requirements were stripped out, and the framing was reworded to align with an “energy dominance” agenda. ⚡ The structure stayed essentially the same. The $800 million for two first-mover teams, the additional $100 million for fast followers, the focus on Gen III+ light-water designs — all unchanged. What changed was the rhetoric, not the architecture of the support.
What makes this program worth watching closely:
TVA’s BWRX-300 is the only SMR design currently under construction anywhere in the Western world, in Canada
Holtec’s SMR-300 is pursuing a “one-stop-shop” model — vendor, constructor, and electricity merchant rolled into one
The program requires multi-reactor orderbooks, meaning the DOE isn’t just funding single plants, it’s trying to seed an industry
Applicants must include a U.S. utility, a reactor vendor, and an EPC company — a consortium structure designed to share risk
The Clinch River site is one of the only federally-approved, undeveloped nuclear sites in the U.S., giving TVA a meaningful head start on permitting
The honest concern here is that confidence in the program depends heavily on whether these are genuinely replicable designs or just expensive one-offs with aspirational price tags. The DOE is clearly betting on replication. Whether the market obliges is a different story.
Canada’s C$3 billion wager on Darlington
If the U.S. approach is “grants and milestones,” Canada’s is “equity investment and provincial partnership.” The investment includes $2 billion from the Canada Growth Fund and $1 billion from Ontario’s Building Ontario Fund, financing four GE Hitachi BWRX-300 reactors at the Darlington site east of Toronto, with the first reactor scheduled to start operating by late 2029. 🌱
That timeline would make the Darlington SMR the first commercial SMR to operate in North America — and, remarkably, the first in any G7 country. Russia and China already have SMRs running, but they’re state-built, state-run machines in a different category altogether. The Darlington SMR received its license to construct in April 2025 from the Canadian Nuclear Safety Commission, a significant step toward commercial deployment.
The Conference Board of Canada estimates that the deployment and operation of the four SMR units will increase Ontario’s GDP by CAD $35.1 billion over 65 years and Canada’s GDP by $38.5 billion, sustaining 18,000 jobs during construction and 2,500 jobs over the projected 60 years of operation. Those numbers are large enough to be exciting and speculative enough to treat with appropriate skepticism. GDP multiplier estimates for big infrastructure projects tend toward the optimistic. But even if the reality comes in at 70 cents on the dollar, that’s still a serious economic argument.
What Canada has done smarter than most is build a genuine federal-provincial coalition. Ontario, Saskatchewan, New Brunswick, Alberta — multiple provinces have signed onto SMR strategies, creating the conditions for a fleet deployment rather than a single demonstration project. 💡 A fleet is the whole point. The economics of SMRs don’t work with one reactor. They need serial production to drive costs down, the same logic that made Boeing and Airbus viable versus building custom aircraft for each airline. Canada seems to understand this more viscerally than most governments, probably because it’s already building.
The shadow in the picture: NB Power announced a potential move away from SMRs to other energy sources in 2025. New Brunswick was supposed to be part of the coalition, and its drift out is a reminder that provincial politics can reshape a national strategy faster than anyone in Ottawa would like.
The UK’s long competition and the Rolls-Royce selection
Britain’s approach has been the most theatrical — a formal technology competition, shortlisting, final negotiations, all with the gravitas of a procurement process for a new aircraft carrier. The results finally arrived. Rolls-Royce SMR — a joint venture of Rolls-Royce, Qatar’s sovereign wealth fund, BNF Resources of France, Constellation of the U.S., and Czech utility CEZ — was selected by the UK government to deploy the first SMRs at Wylfa in Wales. 🔬
Rolls-Royce SMR has received UK government funding of £210 million as part of Phase 2 of the Low-Cost Nuclear Challenge Project, supplemented by £280 million of private capital. That’s a meaningful public-private split, and the private money coming from a sovereign wealth fund and a major European utility suggests genuine commercial confidence — or at least genuine diversification of sovereign risk by parties who don’t share a budget with British taxpayers.
The UK government’s ambitions extend well beyond Wylfa. The government sees Britain as a potential exporter of nuclear technology, with the Advanced Nuclear Framework providing a pathway for privately-led advanced nuclear technologies, and the National Wealth Fund exploring potential investment opportunities. Prime Minister Starmer’s nuclear steer in November 2025 was unusually direct for a British government document: regulatory reform, streamlined licensing, explicit targets for delivery. For a sector that has watched UK nuclear projects slide from bold announcement to expensive embarrassment for decades, the tone was noticeable.
The honest complication is that the UK’s timeline consistently runs toward the mid-2030s for commercial operation, a fact that sits uncomfortably alongside Canada’s 2029 target. If Darlington delivers, Britain may find itself importing lessons from a country that moved faster, and possibly importing the technology too.
The warning light that won’t go away
None of this confident government spending happens in a vacuum. There is a body of critical analysis on SMR economics that can’t be hand-waved away, and the bluntest statement of it came from the Institute for Energy Economics and Financial Analysis. IEEFA documented that NuScale’s SMR costs more than doubled from 2015 to 2023, rising from $9,964 per kilowatt to $21,561 per kilowatt, prompting the cancellation of its signature six-reactor facility in Idaho in cooperation with the Utah Associated Municipal Power Systems. 📈
The NuScale collapse deserves to be understood precisely, because its misreading has become a rhetorical sport. The project didn’t fail because of engineering problems or regulatory obstruction. It failed because the power was too expensive. The target price rose from $58 per megawatt-hour to $89 per megawatt-hour, and this included an expected $1.4 billion subsidy from the U.S. Department of Energy plus support under the Inflation Reduction Act of around $30 per MWh. Subsidized to the eyeballs and still too expensive for municipal buyers to commit. That’s not a regulatory problem. That’s a cost problem.
IEEFA’s broader argument is that SMR developers consistently underestimate costs before construction begins, a pattern they trace across multiple projects and countries. They raise an additional concern that’s worth taking seriously:
Dollars invested in SMRs are not available for wind, solar, and battery storage
A utility that sinks billions into an SMR has every incentive to run it at maximum capacity, potentially crowding out cheaper renewables
The “learning curve” argument — that costs fall with each successive build — has not been demonstrated at commercial scale for SMRs
Regulators, utilities, and investors are being asked to trust cost projections for technologies that have never been built at commercial scale in the relevant markets
Is IEEFA right that SMRs are simply too expensive and too slow? Probably not as a categorical, permanent verdict. But as a description of the present moment and the near-term risk? They have a real point. The best counter-argument isn’t that the critics are wrong about current costs — it’s that governments and utilities are trying to buy down those costs through deliberate first-mover investment. The question is whether the buying-down is working.
Is the money actually working?
Here’s where it gets complicated, and where honest analysis requires holding two things at once. ⚡
Government funding for SMRs is doing some things well:
It’s creating real projects with real construction timelines — Canada’s Darlington is actually being built
It’s forcing supply chain development, including domestic uranium enrichment and forging capacity
The consortium structures — utilities, vendors, constructors, off-takers — are more sophisticated than previous nuclear funding models
The U.S. and UK are explicitly requiring multi-reactor orderbooks as a condition of support, pushing toward fleet economics
It’s doing some things badly:
Cost transparency remains poor — IEEFA notes that SMR developers “consistently sought to shield their construction cost estimates”
No Western commercial SMR has completed construction yet, so every cost projection is still a projection
The NuScale experience showed that even generous government support doesn’t automatically produce a viable power price
Timelines remain optimistic by historical nuclear standards
The technology the DOE just gave $400 million to TVA to deploy — the BWRX-300 — is under construction right now at Darlington. GE Vernova CEO Scott Strazik described it as “the only commercial SMR technology being built right now in the Western world.” That’s genuinely meaningful. It means the learning curve argument isn’t entirely theoretical — there will be actual data on this design within a few years. If Darlington comes in close to budget and close to schedule, the entire investment thesis for SMR government support strengthens considerably. If it doesn’t, the critics will have earned their told-you-so.
Think about what you believe about the role of government in energy transitions: should public funds routinely take first-mover risk on technologies that markets won’t touch alone, or does that crowd out better alternatives? The SMR funding question is, at its core, that older argument in nuclear clothing. 🔬 The answer probably isn’t binary — some bets are worth taking, some aren’t, and the difference lies in whether the underlying technology has a credible path to commercial viability rather than just a credible path to the next grant application.
What would you want to see from these programs before calling them a success?