Five Reactors, One Winter: What Ontario’s Nuclear Refurbishment Means for Your Bottom Line
For most of the last two decades, Ontario businesses could take one thing for granted: nuclear power supplied a large, steady block of baseload generation. It ran quietly in the background and rarely showed up as something worth discussing.
That is about to become much more visible.
Darlington Unit 4 returned to service in March, followed by Bruce Unit 3 in June. Bruce Unit 4 is already offline for refurbishment. By the start of October, five more units move in the other direction: Pickering Units 5 through 8 will be taken offline and prepared for refurbishment (with full project execution following in January 2027, pending CNSC approval), while Bruce Unit 5 enters its own refurbishment outage. The net effect is a meaningful reduction in Ontario’s available nuclear generation as the province heads into a high-demand winter period.
That’s not a rhetorical problem. It shows up as a number.
Power Advisory’s forecast for the Ontario Energy Board has the average wholesale price climbing from $55.13/MWh in the November 2025–January 2026 period to $94.53/MWh in November 2026–January 2027 : a 71.5 percent increase. On-peak prices follow the same curve: from $66.22/MWh to $108.83/MWh, up 64.4 percent. Power Advisory identifies nuclear availability directly as the main driver of that jump, tying it to the five-unit outage stack arriving in October.
Worth noting: the forecast’s own track record has been mixed. Through most of 2026, realized wholesale prices have generally run above what Power Advisory projected, only the last two months have come in closer to forecast. The fall outage assumption has held up well, but the return-to-service dates didn’t land exactly as modeled, and Power Advisory itself flagged “significant uncertainty” in pricing a market that hadn’t yet lived through a full winter under this supply mix. Take the direction of the forecast with more confidence than the precise number.
Why nuclear isn’t the whole story, but it’s the main one
When low-marginal-cost nuclear generation comes offline, the system is expected to lean more heavily on gas-fired generation to fill the gap. Power Advisory’s forecast also assumed higher fuel costs: its Dawn gas price assumption rises from about C$4.17/MMBtu in November 2025 to C$5.61/MMBtu by February 2027, roughly 35 percent. Separately, Ontario’s Emissions Performance Standards apply a compliance price to gas-plant emissions above the applicable benchmark, not to every tonne emitted, rising from $95/tonne in 2025 to $110 in 2026 and $125 in 2027. Together, these assumptions push up the marginal cost of the gas generation increasingly called on when nuclear availability tightens.
In practice, actual gas prices haven’t risen as sharply this year as that forecast assumed. That’s worth acknowledging rather than glossing over. But winter is a different market than the shoulder seasons : a cold snap that spikes heating demand can move gas prices quickly, and that risk doesn’t disappear just because it hasn’t shown up yet.
It’s not just your commodity line, either
For Class A customers, there’s a second channel worth watching. OPG has applied for a 2027 nuclear base payment of $206.76/MWh, up from $111.33/MWh in 2026. Lower expected nuclear output during the refurbishment period is one contributor, but OPG’s application also points to Pickering refurbishment and Darlington New Nuclear rate-base additions, deferral and variance account recovery, and other rate-setting adjustments : it isn’t simply a story of less output spread across the same fixed costs.
Those regulated nuclear costs are recovered through Global Adjustment, which means refurbishment can touch both the wholesale market and GA. But the two aren’t simply additive: as wholesale prices rise, contracted and regulated generators earn more market revenue, which generally reduces the amount that needs to be recovered through GA. For a Class A customer, the actual net effect depends on both where wholesale prices land and your own Peak Demand Factor, not on treating these as two costs that stack on top of each other.
A reliability story, not a reliability crisis
Here’s a distinction worth holding onto, because it’s easy to blur: adequacy and price are not the same question. The IESO’s most recent full Reliability Outlook, from June 2026 (with an interim update in August), found Ontario’s system adequately supplied and specifically addressed preparations for Pickering’s October shutdown as part of that assessment.
The lights are expected to stay on. What the Reliability Outlook doesn’t promise is that keeping them on stays cheap. Those are two different jobs, and right now Ontario is handling them differently: reliability through procurement and storage, cost through whatever’s left on the margin, which at the moment is gas.
This is the shape of the next decade, not just one hard winter
It’s tempting to treat this stretch as a one-off to wait out. The IESO’s 2026 Annual Planning Outlook suggests otherwise. In its reference scenario, Ontario’s electricity demand grows from about 152 TWh in 2026 to roughly 250 TWh by 2050, an increase of about 65 percent, driven in large part by data centres (expected to reach about 8.6 percent of demand by 2050, some 60 percent higher than previously forecast) and electric vehicles (projected at 15 percent of total consumption and more than half of the growth margin). The IESO expects a roughly 8.5 terawatt-hour energy gap to emerge by 2032, and it says plainly that “on-time delivery of new and refurbished nuclear” is one of the main tools for closing it.
In other words: this refurbishment cycle isn’t a detour. It’s the opening chapter of a longer rebuild. A Long Lead-Time RFP for up to 1 TWh per year of hydroelectric generation and 800 MW of long-duration storage launched in May. A second Long-Term RFP is already underway and bringing wind and solar back into Ontario’s procurement mix for the first time in a decade: its first window closed with 13 contracts totaling 1,115 MW and 2.37 TWh, with draft documents for the second window published in August. None of that supply arrives before the current squeeze does.
The volatility isn’t a blip. It’s the environment for the next several years.
What this means for how you buy
Two businesses can carry the exact same load profile and land in very different places this winter. One reviewed its contract with a clear view of where the November 2026–January 2027 window sits inside it. The other is still on an uncapped variable rate, resetting against whatever the wholesale market does next.
Neither approach is wrong on its face. Only one of them is a choice made with the forecast already in hand.
If you haven’t looked at how your current contract, or your Class A Global Adjustment position, handles a winter wholesale market that Power Advisory forecast roughly 70 percent above the comparable period a year earlier, that’s worth reviewing before the window arrives.
