ENERGY · THE TRAP · COLLISION COURSE 2030

Canada's Electrification Plan That Runs on Gas

Ontario's grid carbon intensity rose in 2024 as gas generation increased. The nuclear supply meant to change that hasn't been built. The province is contracting for more.

By Charles Jesse, Editor · Published June 2026 · 22 sources · 12 min read

A natural gas power plant exhaust stack in soft focus behind a row of residential electric vehicle charging stations, golden hour light casting long shadows across an empty parking lot

THE RECEIPTS Energy — The Trap

8.5 TWh

The electrification plan runs on gas

Ontario's grid carbon intensity rose 25% in 2024. The Toronto Atmospheric Fund attributed the increase to the province's growing reliance on gas-fired electricity generation. IESO's 2026 reference scenario projects an incremental energy gap of 8.5 TWh by 2032. The province's most recent long-term procurement secured roughly 411 MW of non-storage capacity, almost all natural gas, including a 405 MW gas expansion with an LT1 contract expiry of April 30, 2040. The sole nuclear reactor under construction, a design that has never been completed anywhere in the world, would produce about 2.36 TWh per year at full capacity, covering less than a third of the 2032 gap. The consumer fuel charge was set to zero on April 1, 2025. The federal industrial carbon price, which applies to covered gas-fired generators above their output-based standard, remains in effect. By 2030 that standard for new gas-fired electricity falls to zero, meaning covered generators pay the full $170 per tonne on all emissions.

Grid carbon intensity (2024) 73.8 g CO₂eq/kWh (+25%)

Energy gap by 2032 (reference scenario) 8.5 TWh

Non-storage capacity (LT1), almost all gas ~411 MW

Napanee gas expansion (LT1 contract expiry) 405 MW · Apr 30, 2040

SMR Unit 1 share of 2032 gap 28%

Industrial carbon price (2030) $170/tonne

OBPS standard, new gas generation (2030) 0 t/GWh

That's the receipt.

Ontario's grid planner says gas is required for reliability through the late 2030s. The province is contracting for it. The nuclear supply that's supposed to replace it has never been built anywhere. The electrification strategy needs consumers to switch from the pump to the plug, but the plug runs partly on gas and carries a carbon price. Full analysis below.

Full analysis · 22 sources · 12 min

You are being asked to switch. Gas furnace to heat pump. Gas car to electric vehicle. Fossil fuel to plug. The federal government's National Electricity Strategy calls it "a nation-building exercise not seen for generations." The plan depends on the electricity being clean. But the grid that's supposed to deliver that electricity saw its carbon intensity rise in 2024, and the documents that show what comes next are the government's own.

This is Part 2 of The Receipts' Collision Course 2030 series, which documented three federal demand drivers targeting the same electricity grid in the same decade. The original piece asked whether provincial supply plans could deliver on those timelines. This piece examines Ontario, Canada's largest electricity market and the site of the country's sole nuclear reactor under construction, as the primary case study. The structural pattern it reveals, federal policy creating demand while provincial supply falls short, is documented in multiple provinces.

📊 How to read this article ▾

Green boxes contain documented facts with source citations you can verify.

Gold boxes contain our interpretation — clearly labeled analysis you can agree or disagree with.

Grey boxes provide context that both sides of the debate tend to leave out.

Red boxes list specific conditions that would change our assessment — our built-in accountability.

The narrative between the boxes is designed to be read on its own. The boxes add depth if you want it.

8.5 TWh

Energy gap by 2032 (IESO reference scenario)

~411 MW

Non-storage capacity procured (LT1), almost all gas

BWRX-300 reactors completed anywhere

$170/t

Industrial carbon price by 2030

Key Facts — Documented

Energy gap: IESO's 2026 APO reference scenario projects incremental energy needs of 8.5 TWh emerging by 2032. Incremental capacity needs begin at less than 100 MW in 2034 and rise to 950 MW by 2035, reaching 37 TWh and 3,800 MW by 2040. Near-term needs are marginally lower than the 2025 APO due to resources recontracted through the Medium-Term 2 RFP and slightly lower demand in the reference scenario. [1]
Demand growth: IESO's 2025 APO forecast net annual energy demand growing 75% to 262 TWh by 2050. The 2026 APO lowered this to 65% growth, reaching 250 TWh. Ontario's Integrated Energy Plan discusses higher-growth cases of up to 90%. The spread matters because supply planning is sensitive to demand assumptions. [2] [1] [14]
Grid carbon intensity: Ontario's grid carbon intensity rose 25% in 2024 to 73.8 g CO₂eq/kWh. The Toronto Atmospheric Fund attributed this to the province's increasing reliance on gas-fired electricity generation. Reduced nuclear availability during the Darlington refurbishment contributed to higher gas dispatch in 2024. [4]
Gas procurement: IESO's LT1 procurement secured approximately 411 MW of non-storage capacity, almost all natural gas, including the 405 MW Napanee expansion. Under the LT1 contract terms, the expiry date for natural-gas facilities is April 30, 2040. Earlier IESO expedited and same-technology procurements secured additional gas capacity. The LT2 procurement was opened to gas by Energy Minister Lecce on August 28, 2024. [6] [7] [8]
Nuclear timeline: The Darlington BWRX-300 Unit 1 was originally described in conditional terms as a reactor that "could be completed as early as 2028." The current target is commercial operation by end of 2030. No BWRX-300 has been completed anywhere. [9] [10]
Carbon pricing: The consumer fuel charge was set to $0 on April 1, 2025. The federal Output-Based Pricing System remains in effect for covered gas-fired generators. By 2030 the output-based standard for new gaseous electricity generation falls to zero, meaning covered generators pay the full $170/tonne on all emissions. [17]

1. The Documented Gap

Ontario's Independent Electricity System Operator publishes an Annual Planning Outlook that forecasts demand, assesses reliability, and identifies where supply falls short. The 2026 edition, released in March, projects incremental energy needs of 8.5 terawatt-hours emerging by 2032 in the reference scenario. [1] By 2035, the capacity shortfall rises to 950 megawatts. By 2040, it reaches 3,800 MW and 37 TWh. [1]

The near-term picture is marginally better than it was a year ago. IESO notes the improvement reflects resources recontracted through its second Medium-Term RFP and slightly lower demand in the reference scenario. [1] The 2025 APO had put the equivalent capacity gap at 2,100 MW by 2034. That figure has come down. But the direction has not changed: demand is growing faster than committed supply can follow, and the gap widens with every year of the outlook.

The demand side keeps getting revised upward. IESO's 2025 APO forecast a 75% increase in net annual energy demand by 2050, reaching 262 TWh, up from the 60% projected just one year earlier. [2] The 2026 APO moderated this to 65% growth, reaching 250 TWh, reflecting a lower EV adoption forecast and economic headwinds from trade disruptions. [1] Ontario's own Integrated Energy Plan, released three months after the 2025 APO, discusses higher-growth cases of up to 90%. [14] The public numbers vary by document and scenario, but all point the same direction. The spread matters because supply planning is highly sensitive to demand assumptions, and the supply gap widens under every version.

The drivers are the same ones identified in Part 1 of this series: electric vehicles, heat pumps, data centres, industrial electrification. EV-related demand alone accounts for 20 TWh of new load by 2035, with industrial growth, including EV battery plants, steel electrification, and hydrogen production, adding another 23 TWh. [3] Data centres represent a separate and fast-moving pressure. At a Toronto Region Board of Trade panel in late 2025, IESO Chief Energy Transition Officer Chuck Farmer disclosed that the connection queue included 8,400 megawatts of data centre demand, roughly one-third the size of Ontario's entire electricity system. [19]

On June 4, 2026, the federal government released its AI strategy, which states that partnerships are being finalized to provide 850 MW of sovereign AI computing capacity by 2030. [22] That target represents additional federal demand on top of the 8,400 MW already in IESO's connection queue. The federal government is actively soliciting new electricity demand while the grid planner says gas is required for reliability through the late 2030s.

2. The Nuclear Bet

Ontario's plan to close the gap depends overwhelmingly on nuclear power. The Integrated Energy Plan envisions nuclear supplying more than 70% of Ontario's electricity by 2050, requiring up to 17,800 MW of new nuclear capacity on top of the existing fleet. [14] The plan begins with a single reactor.

The Darlington BWRX-300, developed by GE Vernova Hitachi Nuclear Energy, is a 300-megawatt small modular reactor under construction at the Darlington site on the north shore of Lake Ontario. The Canadian Nuclear Safety Commission issued a Licence to Construct on April 4, 2025. [11] The reactor building foundation was set on May 1, 2026. The target is commercial operation by the end of 2030. [10]

The original public framing was conditional. When OPG and GE Hitachi announced the project in December 2021, the reactor was described as one that "could be completed as early as 2028." [9] The current commitment is end of 2030. The public target moved from a conditional early framing to a firm date, and the firm date is later.

Even if Unit 1 delivers on time, the arithmetic is clear. A 300 MW reactor operating at a 90% capacity factor produces roughly 2.36 TWh per year. IESO's 2032 energy gap, in the reference scenario, is 8.5 TWh. Unit 1 covers approximately 28% of it. Three follow-on units are planned for the 2033 to 2036 window, which would bring the Darlington SMR fleet to roughly 1,200 MW and 9.4 TWh. Beyond that, the next tranche of nuclear, including a potential Bruce C expansion of up to 4,800 MW, has no committed in-service date. A pre-development agreement was announced in May 2026. Earliest commissioning on standard new-build timelines would be late 2030s at best. [14]

The construction timeline is where the documented record runs hardest against the schedule. The World Nuclear Industry Status Report found that the average construction time for the 18 reactors connected to grids globally between 2021 and 2023 was approximately 10 years. [12] The World Nuclear Association's own data corroborates the underlying grid-connection figures, confirming seven reactors were connected in 2024 across China, France, India, the UAE, and the United States. [21] WNISR's 2025 edition reported that fewer than a third of expected grid connections materialized that year, with four reactors connected against a beginning-of-year expectation list of 13. [13] Western projects fare worse. Vogtle Units 3 and 4, the only reactors completed in the United States in over three decades, each took more than a decade and cost more than double their original estimates. Flamanville-3 in France took 17 years.

No BWRX-300 has been completed anywhere in the world. There is no reference plant for any regulator, builder, or investor to point to. The Darlington project is not just Ontario's first new reactor in a generation. It is the first of its kind globally. [13]

OPG's record on its most recent nuclear project, the Darlington refurbishment, is more encouraging. The company reported the four-unit refurbishment came in approximately $150 million below the $12.8 billion budget and four months early. [15] The Ontario Auditor General's 2018 report found that 18 prerequisite projects ran more than $725 million above their initial estimates, though about $345 million of that overrun had been incorporated into the $12.8 billion project estimate. [16] Separate OPG filings identify additional turbine and generator lifecycle work outside the 2016 refurbishment scope. Those costs do not mean the refurbishment was over budget, but they complicate the headline framing. [15] None of this makes OPG unusual among nuclear operators. It does make the C$20.9 billion SMR fleet estimate, announced by the Minister of Energy and Mines in May 2025 without an Ontario Energy Board review comparable to the refurbishment disclosure, worth independent scrutiny. [10]

3. What Actually Fills the Gap

The gap exists now. The nuclear supply arrives later, if it arrives on schedule. Something has to generate the electricity in the interim. The documented record shows what that something is.

IESO's Long-Term 1 procurement, completed in May 2024, secured approximately 411 MW of non-storage capacity, almost all of it natural gas, alongside 1,784 MW of new battery storage. [6] The largest gas award was the 405 MW Napanee Generating Station expansion, a simple-cycle gas turbine. Under the LT1 contract terms, the expiry date for natural-gas facilities is April 30, 2040, meaning the Napanee expansion is contracted to operate through the end of that decade. [7] Earlier IESO expedited and same-technology procurements secured additional gas capacity beyond the LT1 totals. The battery storage provides peak shifting, not net new energy. Gas provides both.

Three months after the LT1 results, Ontario Energy Minister Stephen Lecce directed IESO to change its second long-term procurement from "non-emitting" to "technology agnostic," explicitly opening the door to additional gas. [8] The directive was issued on August 28, 2024. The rationale was reliability. The effect is that Ontario is contracting for gas-fired generation through 2040 under LT1, and is open to additional gas under LT2 that could extend further.

This is not a contingency plan. It is the plan. IESO's 2025 APO Capacity Expansion Scenario states that gas generation "is modelled to operate at levels far below its capability and can provide more energy as necessary." [5] The 2025 APO summary puts it plainly: gas generation will be required for reliability into the late 2030s and beyond. [20] The system has spare gas capacity. If nuclear is late, gas runs harder.

The measured outcome is already visible. Ontario's grid carbon intensity rose 25% in 2024, reaching 73.8 g CO₂eq/kWh. The Toronto Atmospheric Fund attributed the increase to "the province's increasing reliance on gas-fired electricity generation." [4] It is worth noting that reduced nuclear availability during the Darlington refurbishment contributed to higher gas dispatch in 2024. Whether the increase represents a durable trend depends partly on nuclear refurbishment recovery and future gas dispatch patterns. But the direction in 2024 was unambiguous: the grid got dirtier in the year the electrification strategy was supposed to begin delivering cleaner power.

4. The Carbon Price on the Bridge

On April 1, 2025, the consumer fuel charge was set to zero. The price Canadians paid on gasoline, home heating fuel, and natural gas disappeared from their bills. But the federal Output-Based Pricing System, which applies to large industrial emitters including gas-fired electricity generators, remained in effect. [17]

The OBPS works through a declining output-based standard. For new gaseous electricity generation, the standard started at 206 tonnes per gigawatt-hour in 2025 and falls to zero by 2030. [17] Once the standard reaches zero, covered generators pay the industrial carbon price on all emissions, not just the excess above a benchmark. At $170 per tonne in 2030, a combined-cycle gas plant emitting roughly 0.37 to 0.45 tonnes per MWh would face an additional cost of approximately $63 to $77 per megawatt-hour. A simple-cycle plant, like the Napanee expansion, could face $94 to $111 per MWh. [17]

That cost is not a line item on a consumer's hydro bill, but it is embedded in the electricity rate. Every heat pump that replaces a gas furnace, every EV that replaces a gasoline car, shifts household energy consumption onto the electricity grid. The switch can still reduce household emissions, particularly for heat pumps, which use electricity roughly three times more efficiently than a gas furnace produces heat. The cleaner the grid, the larger the benefit. But Ontario's planning documents show the marginal generation source is gas through the late 2030s, and the industrial carbon price applies to that generation.

The federal National Electricity Strategy, released in May 2026, frames the grid build-out as essential to "affordability, economic growth, and climate change." [18] It commits to doubling Canada's electricity supply by 2050. It does not reconcile this commitment with the grid planner's documented reliance on gas-fired generation for the next two decades. The electrification plan is real. The demand it creates is real. The clean supply it assumes is not here yet, and the documents that tell us what fills the gap in the interim are the government's own.

Plain English: The federal government wants Canadians to replace gas furnaces and gas cars with electric heat pumps and EVs. For that to deliver its full emissions benefit, the electricity has to be clean. Ontario's grid planner says gas-fired generation is needed for reliability through the late 2030s, and the province is actively procuring more of it. The nuclear reactor that's supposed to start changing this has never been built before, and even if it arrives on time, it covers less than a third of the projected shortfall. The switch can still reduce emissions at the household level, especially for heat pumps. But the cleaner the grid is, the larger the benefit, and Ontario's planning documents show the marginal source is gas for the foreseeable future.

Context — What Both Sides Omit

Nuclear advocates omit the interim. The case for nuclear is strong on a 30-year horizon: high capacity factor, low emissions, baseload reliability. But the first BWRX-300 is four and a half years from commercial operation under the best-case timeline. The rest of the fleet is a decade or more away. "Nuclear is the answer" is not an answer to what happens between now and 2035.

Anti-nuclear voices omit the alternatives' limitations. The 1,784 MW of battery storage procured under LT1 provides peak shifting, not net new energy. Renewables are intermittent. Quebec's surplus is shrinking as Hydro-Québec commits it to its own electrification under Action Plan 2035. Demand response, projected at 152 MW rising to 183 MW by 2050, is too small to bridge a multi-gigawatt gap. [3] If you remove nuclear from the plan entirely, you are left with gas for longer.

Both sides omit the rate impact. Every year of gas-fired bridge generation adds carbon-priced electricity costs to consumer bills at the precise moment those consumers are being asked to electrify. By 2030, the OBPS cost on new gas-fired generation could add $63 to $111 per MWh depending on plant type. The affordability case for electrification weakens as the gas bridge extends. [17]

Interpretation

This article does not argue that Ontario's nuclear strategy is wrong. It documents a timing gap between the demand that electrification policy creates and the clean supply that electrification policy assumes. That gap is filled by gas. The gas carries a carbon price. The carbon price raises the cost of the electricity that consumers are being told to switch to. This is not a failure of intent. It is a structural consequence of the documented mismatch between demand timelines and supply timelines.

If Darlington Unit 1 slips three years, to 2033, and the follow-on units shift proportionally, the cumulative shortfall over the decade could reach 25 to 30 TWh of expected clean generation that does not arrive. At combined-cycle gas marginal intensity of approximately 400 to 490 g CO₂/kWh, replacing that generation with gas would add roughly 10 to 14 megatonnes of CO₂ cumulatively. This is not a primary-source forecast. IESO has not published an equivalent scenario. It is the structural math their own gap figures and standard capacity-factor assumptions make possible.

Counter-interpretation: Ontario's nuclear bet may pay off. Darlington is not a greenfield nuclear program starting from zero. OPG has an active nuclear workforce, a recently completed multi-unit refurbishment, an existing licensed nuclear site, established supply-chain relationships, and a regulator already engaged with the project. The BWRX-300 is intentionally simpler than the large Western new-build designs that drove decade-long overruns: smaller output, natural circulation, fewer major systems, no large forgings of the kind that delayed Flamanville and Olkiluoto. If Unit 1 establishes the construction method, Units 2 through 4 may benefit from repetition. Gas reliance in the interim, while not ideal, may be a defensible bridge if the destination is a grid that is 70% nuclear by 2050. The fair critique is not that the schedule is impossible. It is that Ontario is treating a first-of-a-kind schedule as a planning dependency before any reference plant exists.

What Would Change This Assessment

If Darlington Unit 1 completes hot functional testing by Q2 2029, as publicly reported by OPG or CNSC, the end-of-2030 target becomes plausibly achievable and the first-of-a-kind delay thesis weakens substantially.
If IESO publishes a delayed-nuclear scenario showing no material increase in gas generation or electricity-sector emissions through 2035, the structural gas dependence documented here would be overstated.
If Ontario's grid carbon intensity declines below the 2023 baseline by 2026, the directional finding that the grid is getting dirtier would need revision.
If federal or provincial policy materially shields ratepayers from OBPS costs on gas-fired electricity generation, the rate-impact finding no longer holds.
If another BWRX-300 reaches major nuclear construction milestones (reactor pressure vessel installation or equivalent) before Darlington, the schedule risk associated with a sole first-of-a-kind project would be materially reduced.

Sources (22)

IESO, 2026 Annual Planning Outlook: Ontario's electricity system needs 2027–2050, March 2026. IESO PDF
IESO, 2025 Annual Planning Outlook: Ontario's electricity system needs 2026–2050, April 2025. IESO PDF
IESO, 2025 Annual Planning Outlook: Demand Forecast Module, April 2025. IESO PDF
Toronto Atmospheric Fund, 2024 GTHA Carbon Emissions Inventory: Electricity Grid. TAF
IESO, 2025 Annual Planning Outlook: Capacity Expansion Scenario, Costs, and Emissions Module, April 2025. IESO PDF
IESO, Long-Term 1 RFP procurement results, May 2024. Approximately 411 MW non-storage capacity (410.69 MW maximum contract capacity), almost all natural gas, plus 1,784 MW battery storage. IESO
IESO, Long-Term 1 RFP contract template: expiry date for natural-gas facilities set at April 30, 2040. Napanee Generating Station Expansion: 430 MW nameplate / 405 MW maximum contract capacity, simple-cycle gas turbine, contract awarded May 9, 2024. See also: Atura Power project information sheet. IESO
Ontario Ministry of Energy directive to IESO regarding Long-Term 2 RFP, August 28, 2024; reported by Northern Ontario Business, August 29, 2024.
OPG and GE Hitachi Nuclear Energy, joint press release: "OPG selects GE Hitachi as technology partner for Darlington Small Modular Reactor," December 2, 2021. Described as a reactor that "could be completed as early as 2028." OPG
GE Vernova product website, BWRX-300 specifications; Ontario Ministry of Energy and Mines, Darlington SMR fleet budget of C$20.9 billion (C$7.7B for Unit 1), May 8, 2025.
Canadian Nuclear Safety Commission, Record of Decision: OPG Application for a Licence to Construct One BWRX-300 Reactor at the Darlington New Nuclear Project Site, April 4, 2025. CNSC
Mycle Schneider Consulting, World Nuclear Industry Status Report 2024. Average construction time of 18 reactors connected to grids 2021–2023: approximately 10 years. WNISR
Mycle Schneider Consulting, World Nuclear Industry Status Report 2025. Four reactors connected to grids in 2025 against a beginning-of-year expectation list of 13. No BWRX-300 completed anywhere. WNISR
Government of Ontario, Energy for Generations: Ontario's Integrated Energy Plan, June 12, 2025. Projects up to 17,800 MW of new nuclear by 2050; nuclear supplying 70%+ of electricity; discusses demand growth of up to 90%.
Ontario Power Generation, Q4 2025 Financial Release, March 12, 2026. Darlington Refurbishment completed ~$150M below $12.8B budget and four months early. Separate filings identify additional turbine and generator lifecycle work outside the 2016 refurbishment scope.
Ontario Auditor General, Annual Report 2018, Section 3.02: Darlington Nuclear Generating Station Refurbishment. Documented 18 prerequisite projects running >$725M (>75%) above initial estimates; approximately $345M of that overrun was incorporated into the $12.8B project estimate.
Government of Canada, Output-Based Pricing System. Consumer fuel charge set to $0 effective April 1, 2025; OBPS remains in effect. Covers gas-fired electricity generators above 50,000 t CO₂e/year. Output-based standard for new gaseous electricity generation: 206 t/GWh (2025), declining to 0 t/GWh (2030). ECCC
Government of Canada, Powering Canada Strong: A National Strategy for an Electrified Canadian Economy, May 14, 2026. NRCan
Chuck Farmer, IESO Chief Energy Transition Officer, Toronto Region Board of Trade panel, December 2025: "Our latest queue includes 8,400 megawatts of data centre demand. That's one-third the size of Ontario's entire electricity system today."
IESO, "Seven Graphs and a Map: 2025 Annual Planning Outlook," May 27, 2025. Gas generation "will be required into the late 2030s and beyond for reliability." IESO
World Nuclear Association, Global Nuclear Industry Performance, 2025. Confirms seven reactors connected to grids globally in 2024. WNA
Government of Canada, AI for All: Canada's Artificial Intelligence Strategy, June 4, 2026. States partnerships being finalized to provide 850 MW of sovereign AI computing capacity by 2030.

No corrections have been issued for this article. See our corrections policy.

This article was produced using AI-assisted adversarial verification. Read how and why.