Big Reactors in Saskatchewan: What could go wrong?

By Mark Winfield and Susan O’Donnell 

The Saskatchewan government’s recent announcement of its intention to build two large nuclear reactors has again raised questions of the role of large nuclear energy projects in relatively small provincial economies.

The last new full-scale nuclear reactor project in Canada, the Darlington nuclear power plant east of Toronto, was completed more than 30 years ago. The enormous cost overruns on that project contributed significantly to the effective bankruptcy of the province’s utility, Ontario Hydro, leading to its eventual break-up.

As the memories of these previous experiences with large nuclear construction projects have faded, new projects are now being proposed. These projects, and experiences with the handful of new-build nuclear projects initiated in Europe and the United States in the last two decades, give us some indication of the large reactor options, and their potential costs.

New Brunswick’s experience with its long- troubled Point Lepreau reactor, should give Saskatchewan further cause for reflection on the wisdom of such a pathway.  The Lepreau plant is Canada’s lone experience with a large nuclear power plant in a relatively small provincial electricity system and economy, New Brunswick Power is still carrying a C$3.6 billion nuclear debt from the reactor’s original 1975-1983 construction, and the costs of a 2008-2012 refurbishment that was a billion dollars over budget. Since the refurbishment, the poor performance of the Lepreau reactor has been the primary reason New Brunswick Power loses money almost every year.

In Ontario and Alberta, two reactor designs, the CANDU MONARK and the Westinghouse Electric AP1000, have been considered for the expansion of the Bruce Nuclear power plant on Lake Huron, a proposed 10,000-MW Ontario Power Generation plant at Wesleyville on Lake Ontario, and the proposed 4,800-MW Peace River Nuclear Project in Alberta.

The 1,000-MW CANDU MONARK, intended as a successor to the existing CANDU reactors in Ontario and New Brunswick. Although it’s being aggressively promoted to potential international customers, the MONARK design remains incomplete. The situation has already led the Alberta project’s proponents to switch their proposal to favour the AP1000 design by Westinghouse Electric.

Cost information is available on the AP1000 reactor, as two units were completed in 2024 at the Vogtle nuclear power plant in Georgia. The total estimated cost of those two 1,100-MW reactors was US$36 billion, or about $26 billion per reactor in 2026 Canadian dollars. The plant has been described as “the most expensive power plant ever built on Earth.” When it went into service, Vogtle resulted in a nearly 24% increase in Georgia Power’s electricity rates, the largest jump in the utility’s history.

AtkinsRéalis (formerly SNC Lavalin) is currently pitching the CANDU MONARK to Poland, with a reported estimated cost of $45 to $50 billion for a three-reactor plant, or about $15 billion per unit.

The implication of these experiences and proposals is that a new 1,000-MW reactor could carry a price tag of $15 to $26 billion. That would put the total cost for a two-reactor project, like Saskatchewan is proposing, in the $30-$50 billion range. Estimates of the costs of electricity needed to cover the capital costs of new nuclear plants, if they’re financed through electricity rates, range from the mid-20¢ to more than 40¢ per kilowatt-hour.

In the case of New Brunswick, another smaller province considering a large new reactor project, it has been estimated that the project would nearly double to even triple current consumer electricity costs. Such increases would undermine energy affordability, economic competitiveness, and any plans for decarbonization through electrification.

A province could also try to finance the costs through its general tax base. That is the approach that Ontario  been taking, at an estimated cost to the provincial treasury of $7 to $8.5 billion per year. Electricity subsidies now account for more than half of Ontario’s deficit, exceeding annual capital expenditures on education and health care by wide margins.

In the case of New Brunswick, the annual costs of that approach, even spread over the decade or more of construction, could exceed the province’s current, record $1.39 billion deficit, and match or exceed its entire annual capital spending plans in all other areas. Although the Saskatchewan economy is almost double the size of New Brunswick, the scale fiscal impacts of trying to finance two reactors out of general provincial tax revenues would be comparable.

Saskatchewan, for its part, has had another idea – asking the federal government to pay three-quarters of the cost of the province’s first nuclear reactor, a contribution of between $11 to $20 billion from federal taxpayers.

In addition to the financial risks a single large reactor in a relatively small electricity system would repeat and magnify a key problem associated with the original Point Lepreau project in New Brunswick —putting very higher portion of a province’s electricity supply eggs in a single, very expensive and high-risk basket.

All of this should suggest a high degree of caution among provinces considering adding large nuclear components for their electricity systems, and a need to think about other ways of decarbonizing their electricity systems while ensuring their affordability and sustainability.

Mark Winfield is a professor at the Faculty of Environmental and Urban Change at York University in Toronto, and co-chair of the faculty’s Sustainable Energy Initiative.  Susan O’Donnell is adjunct research professor and lead researcher on the CEDAR project in Sustainability and Environmental Studies at St. Thomas University.