Brendan Frank is Clean Prosperity’s Vice President, Policy
To keep pace with anticipated growth in power demand, Canada needs $1 trillion to $2 trillion of investment in its grids over the next 25 years. Provinces and the federal government are working to steer some of this investment towards large nuclear reactors.
Two large reactor technologies under consideration
The federal government’s June 2026 Nuclear Energy Strategy targets the construction of 10 new large-scale reactors across Canada. The prospectus for September’s Canada Investment Summit highlighted opportunities for investment for Ontario’s proposed 4,800 MW Bruce C project, the proposed 10,000 MW Wesleyville nuclear generating station, and up to 1,500 MW at Point Lepreau in New Brunswick. Saskatchewan subsequently announced intentions to build 2,000 MW of large reactors in the province’s northwest by 2042.
If Canada moves ahead with building new large-scale nuclear power, the choice of reactor technology becomes one of the most consequential energy policy decisions in decades. Last week’s news that the first reactor at the Darlington New Nuclear Project will come in behind schedule and over budget has only upped the stakes for this decision. Electricity policy is the domain of the provinces, but they’ll need federal help in this undertaking. No nuclear fleet in any country has succeeded without significant state backing — political, institutional, and often financial. The federal government plans to lend a hand through both broad and bespoke financial support and regulatory streamlining.
Provincial success depends on federal support and effective coordination
The most successful nuclear buildouts are the most coordinated and repeatable. Central to this is a fleet-based policy that rigorously builds the same reactor design over and over. This maximizes the benefits of learning by doing, creates more predictable supply chains, and ensures that engineering and project management expertise accumulate and spread out. Canada used this approach in its first nuclear buildout that concluded in the 1990s: 22 CANDU reactors across Ontario, Quebec, and New Brunswick.
Provinces are effectively looking at two large reactor designs. Each has its advantages and disadvantages. The latest generation of CANDU technology is AtkinsRéalis’ CANDU Monark reactor, which is still awaiting a final design. Its principal competitor is the Westinghouse AP1000, which already has six units operational in the United States and China with another 20+ under consideration or construction. A Canadian consortium made up of Brookfield Renewable Partners and Cameco acquired Westinghouse in 2023, increasing the appeal of AP1000 deployment in Canada. The table below compares the CANDU Monark and the AP1000 across some of the criteria that governments might consider.
| CANDU Monark | AP1000 | |
|---|---|---|
| Time to market | Final design expected by 2030 | Six reactors in operation worldwide with additional units under construction |
| Geopolitical risk | Largely domestic supply chain; fuel supply is fully secure as the reactor does not require enriched uranium | Largely foreign supply chain; enriched fuel supply depends on U.S. and/or European suppliers; U.S. government seeking 20% equity stake in Westinghouse IPO in exchange for financing initial AP1000 buildout in U.S. |
| Scale | 925 MW | 1,100 MW |
| Capital costs ($/W) | Unknown until final design is complete | Approximately CAD $24/W based on recent builds in North America (or roughly CAD $27 billion per reactor) |
| Fleet potential | Hinges on successful first-of-a-kind (FOAK) deployment in Canada | Would be part of the existing global fleet; U.S. and China have absorbed FOAK costs |
| Supply chain | Canadian supply chain already oriented towards CANDU | Canadian supply chain would need to reorient |
| Export and partnership opportunities | Federal Crown corporation Atomic Energy of Canada Limited owns the intellectual property for the reactor | Canadian-owned Westinghouse holds the intellectual property for the reactor, but export and licensing require the approval of the U.S. government |
Advantage Mixed Disadvantage
Governments should coordinate to take advantage of fleet economics
One of the key benefits of building 10 large reactors, as envisioned by the federal strategy, should be fleet economics. The first-of-a-kind (FOAK) project should be the single most expensive point in the entire program. In the long run, what matters most is sustaining a learning curve across all 10 reactors that leads to a reduction in costs with each successive build.
Economics should compel the federal and provincial governments to coordinate and build a single kind of large reactor design repeatedly. The risk is that regional interests and incentives override sound economic decision-making.
Every deviation from a national fleet-based approach will carry additional costs. If provinces go their own way, ratepayers risk paying first-of-a-kind prices for far longer. If a government switches designs midstream due to project execution problems, sticker shock, or something else, it would need an incredibly strong rationale to justify discarding the accumulated learning to restart at the top of the cost curve.

Transparency is critical to ensure durable and economical reactor choices
The stakes of the reactor technology choice are high. Nuclear is a long-lived generating technology; these decisions will have to outlast the careers of the people making them. The technology choices and ensuing buildout will take decades and must be capable of enduring multiple political cycles and governments. Decision-making transparency is essential to secure and sustain public confidence in Canada’s nuclear buildout.
A transparent, competitive selection process can surface critical information and provide a public yardstick by which to measure success. But most importantly, governments should be clear upfront about their criteria for evaluating reactor technologies. And if provincial governments choose not to coordinate on fleet-based development, they should explain why they are forgoing this approach.
While the choice of reactor to build ultimately rests with the provinces, the federal government wields significant influence. The federal nuclear strategy all but endorsed the CANDU reactor and committed to ensuring that its latest design, the Monark, is available by 2030. Ottawa has an incentive to focus on CANDU. It owns the intellectual property via Atomic Energy of Canada Limited and last year announced a loan of up to $304 million to help AtkinsRéalis finalize the Monark’s design. But like the provinces, Ottawa should also be more transparent and explicit about all the criteria that guide its ongoing support for CANDU.
To succeed, Canada’s next nuclear buildout will require project discipline above all else. As we’re learning at Darlington, discipline is ultimately downstream of technology choice. Governments will need alignment and ongoing coordination over a multi-decade timeframe. As we wade into the early planning stages, they should be making their decisions in the daylight.
To learn more about our work on nuclear power, check out our recent reports:
- Nuclear Nation Building: Policies and pathways for scaling new reactors across Canada
- Chain Reactions: How to address constraints on Canada’s nuclear supply chain
- A Nuclear Grand Bargain: How to grow an affordable and low-emitting electricity grid in Saskatchewan
Photo credits: Darlington Nuclear Generating Station, Robert T Bell, Wikimedia Commons, CC BY-SA 4.; Ali Kazal from Upsplash