Trans Mountain was built to give Canadian oil a second customer. Two and a half years after the expansion opened, it has one: in 2026 Canada is shipping 533,000 b/d of crude outside the United States, more than it sends to the US Gulf Coast. Verified
The world has changed around the pipe. A trade war, a war in the Gulf and Venezuela’s return to the US market have made Canada’s only non-US outlet its most strategic energy asset. But the pipe is full, freight is at records, and the discount on Canadian heavy oil is the widest since 2018. We set out what the pivot earned, who now keeps the value, and our view of the discount to 2030.
In 60 seconds
- The pivot happened. Exports outside the US rose from 131 kb/d in 2023 to 533 kb/d in 2026, while the US Gulf Coast fell from 757 to 439 kb/d. The US share of Canada’s crude exports fell from 97% to 88%.
- It paid. The WCS discount fell from US$18.67 in 2023 to US$11.76 in 2025: about US$8.8bn a year more for Canada’s heavy oil, on our estimate.
- Three forces widened it again. Full pipes, record freight that closed the Gulf Coast re-export route, and Venezuelan heavy back in the Gulf. WCS traded US$25–30 below WTI in late October; Alberta’s budget assumed US$13.
- Venezuela rotated the trade. US imports of Venezuelan crude rose from 0.2 to 0.7 mb/d this year, and Venezuela sent none to China. Venezuelan heavy goes north; Canadian heavy goes west.
- Freight decides who keeps the value. Full Aframax loads from 2027 save about US$1.6 a barrel. Loading VLCCs at Pacific Link’s terminal would save about US$3.2 more, about US$1.2bn a year.
- Our view. US$18–22 in 2027 and US$17–21 in 2028; back to US$12–15 only if Prairie Connector is built. Six dated calls are in Section VIII.
Canada Now Ships More Crude Overseas Than to the US Gulf Coast
The expansion opened in May 2024 and tripled Trans Mountain’s capacity to 890 kb/d. Verified In the fourth quarter of 2025 it delivered a record 807 kb/d: 504 kb/d to the Westridge Marine Terminal, 203 kb/d by pipeline to refineries in Washington State and 100 kb/d within BC. Verified Only the Westridge share can reach Asia.
How to read it → Read left to right: the dark block is total deliveries; each band is one outlet, drawn to scale. The box on the right counts tankers loaded at Westridge since May 2024.
Why it matters → Only the Westridge share can reach Asia, and its dock is Aframax-only. That is the bottleneck behind the rest of this article.
The regulator’s export data show how far the map has moved. Canadian crude exports outside the US averaged 131 kb/d in 2023, 445 kb/d in 2025 and 533 kb/d in January–August 2026. Exports to the US Gulf Coast fell from 757 to 439 kb/d over the same period, while the Midwest rose from 2,447 to 2,780 kb/d and the US West Coast from 286 to 394 kb/d. Verified Trans Mountain did not just add a customer; it took barrels away from the Gulf Coast. Interpretation
How to read it → Each line is Canadian crude exports to one destination, in thousand barrels a day. The shaded bar marks the expansion opening in May 2024; 2026 covers January to August.
Why it matters → Trans Mountain did not just add Asia. It pulled barrels away from the Gulf Coast, so US refiners there now compete with Asian buyers for Canadian heavy oil.
Asia takes most of what reaches the coast. Of 691 tankers loaded at Westridge to 21 October, 455 went to Asia. Reported China has taken nearly half of marine volumes since 2024, according to RBC’s tracking, and Asia took nearly 77% of the crude exported from Vancouver in the first seven months of 2026, up from about 51% in 2024. Reported Korean refiners have stepped up: South Korea aims to import up to 16 million barrels of Canadian crude in 2026, more than triple 2025. Reported In July Japan’s largest refiner bought its first Canadian cargo since 2025. Reported But concentration is real: one Chinese refiner, Zhejiang Petroleum & Chemical, bought about 178 kb/d in the first eight months of 2026, up 36% on 2025. Reported That is roughly a third of Westridge loadings. Estimate
| Measure | 2023 | 2025 | 2026 |
|---|---|---|---|
| Crude exports outside the US, kb/d | 131 | 445 | 533 (Jan–Aug) |
| Crude exports to the US Gulf Coast, kb/d | 757 | 563 | 439 (Jan–Aug) |
| Crude exports to the US Midwest, kb/d | 2,447 | 2,649 | 2,780 (Jan–Aug) |
| WTI minus WCS, US$ per barrel | 18.67 | 11.76 | 15.69 (Jan–Sep); 24.80–29.75 daily in Oct |
| Trans Mountain throughput, kb/d | Legacy line only | 761 (86% full) | 840 in Q2 (94%); 98% in July |
| Heavy toll to Westridge, C$ per barrel | 4.29–4.33 | 10.84–13.18 firm; 14.40 spot | 10.84–13.18 firm; 14.06 spot |
A Dividend of About US$9 Billion a Year
The clearest gain is in the price. The discount of Western Canadian Select to WTI averaged about US$18.70 in the eight months before the expansion and about US$12.00 in the year after it, on the regulator’s figures. Verified On annual averages it went from US$18.67 in 2023 to US$15.56 in 2024 and US$11.76 in 2025. Verified
How to read it → The line is the monthly gap between WTI and Western Canadian Select, in US$ per barrel; higher means cheaper Canadian oil. Coloured bars show the regulator's before and after averages; the red bar spans October's daily settlements.
Why it matters → The discount is the scoreboard for export capacity. When pipes have room it narrows; when they fill, or freight closes other outlets, it widens.
On about 3.5 mb/d of WCS-priced heavy oil, each US$1 a barrel is worth about US$1.28bn a year. Against 2023, the narrower discount was worth about US$4.0bn in 2024 and US$8.8bn in 2025, roughly C$12bn. Estimate Not all of that is Trans Mountain: Mexican decline and Venezuelan sanctions also tightened heavy oil, but the timing is hard to miss. Interpretation
The discount moves fast. Monthly averages ranged from US$11.16 to US$18.99 over the past 14 months, and daily settlements in October ran from US$24.80 to US$29.75, well above the September average of US$16.94. Verified A single day is a poor basis for a year, so we value the discount on annual averages and show recent trading only as full-year scenarios. If October’s levels held for a year, Canadian heavy oil would earn about US$11bn less than in 2023 and about US$20bn less than in 2025. Estimate
How to read it → Solid bars are annual averages of the change in the discount since 2023, multiplied by about 3.5 mb/d of WCS-priced heavy oil. Dashed bars hold one level for a full year: scenarios, not forecasts.
Why it matters → A single day is not a year, but the scenarios show the stakes: each US$1 a barrel is worth about US$1.3bn a year to Canadian producers.
Three Forces Widened the Discount at Once
The pipes are full. Trans Mountain reached 98% of capacity in July and was apportioned by a record 7% in October. Enbridge’s Mainline cleared apportionment in September for the first time this year, then posted 16% on heavy crude and 19% on light for November. Reported
Freight shut the back door. With the Iran war pushing tanker rates to records, shipping Canadian heavy out of the Gulf Coast became uneconomic, and the WCS discount at Houston passed its October 2022 record of US$20.60. Reported At Hardisty, WCS settled US$24.80 below WTI on 5 October and US$29.75 on 30 October, the widest since 2018. Reported
Venezuela is back in the Gulf (Section IV). Commodity Context’s Rory Johnston argues the latest widening has nothing to do with pipeline capacity. Reported We see both: freight and Venezuela explain the timing, and full pipes explain why Alberta has no escape valve. Interpretation
The arithmetic will not ease soon. S&P Global CERA expects Canadian crude output to rise from 5.5 mb/d in 2026 to 5.8 mb/d in 2027 and 6.0 mb/d in 2028. Reported The new pipe that is certain adds about 480 kb/d by the end of 2028; Prairie Connector would add 465 kb/d at the end of 2028, but only if it is sanctioned in mid-2027. Reported Until then, rail carries the marginal barrel. Interpretation
How to read it → Bars stack new export pipe by the year it opens, counted from the end of 2025. The red dashed line is S&P Global CERA's forecast growth in Canadian crude output.
Why it matters → 2027 is the tight year. Until new pipe opens, rail sets the price of the marginal barrel, and the discount reflects it.
Most of the 480 kb/d comes from working existing steel harder. A drag-reducing agent is a polymer injected into the oil that cuts friction, so the same pumps push more barrels; more pumping does the rest. Reported Optimization is the cheapest capacity in the system, and it runs out in 2028. Interpretation
| Project | Adds | When | How |
|---|---|---|---|
| Express-Platte expansion | 30 kb/d | 2026 | More capacity on an existing line |
| Enbridge Mainline optimization, phase 1 | 150 kb/d | 2027 | Upgrades on the existing Mainline system |
| Trans Mountain drag-reducing agent | 90 kb/d | Early 2027 | A polymer added to the oil cuts friction, so the same pumps move more |
| Second Narrows dredging | A third more per tanker | Q1 2027 | A deeper channel lets Aframaxes leave Westridge full, not 75% loaded |
| Trans Mountain mainline optimization | 210 kb/d | Late 2028 | More pumping on the expanded line; system to about 1.19 mb/d |
| Total new pipe | 480 kb/d | By end-2028 | Against about 500 kb/d of new supply (S&P Global CERA) |
Filled with diluted bitumen, 480 kb/d of new pipe needs up to about 145 kb/d of diluent. Estimate Article 52’s diluent forecast counted Trans Mountain’s optimization, up to 420 kb/d including the Puget Sound line, as one step in 2030; the company now plans 300 kb/d of it in 2027–28. Re-running our model with that timing adds about 20 kb/d to Alberta’s diluent shortfall in 2028 and 10 kb/d in 2029: the squeeze arrives on the same timeline, slightly deeper. Estimate
Three Shocks Made One Pipe Strategic
A trade war. Washington put a 10% tariff on most Canadian goods in July and 50% tariffs on others in August, but neither targets crude: goods that meet USMCA rules are exempt, and energy is not among the 50% goods. Reported It also declined to renew USMCA, which now expires in 2036 unless renewed. Reported Crude is exempt because US refiners need it. Trans Mountain lets Canada say the same about Asia, and Ottawa’s goal is to double non-US exports within a decade. Reported
A war in the Gulf. The Iran war cut China’s seaborne crude imports to about 7 mb/d in July, against about 10 mb/d before it. Reported Chinese refiners went looking for heavy sour replacements: Cold Lake traded US$3–4 above Brent delivered to Shandong in early September, and one cargo sold at US$3.70 above Brent for October–November delivery. Reported A route to Asia that passes no chokepoint has become worth paying for (Article 51). Interpretation
Venezuela’s return. After the capture of Nicolás Maduro in January, Washington eased restrictions on Venezuelan oil. US imports of Venezuelan crude rose from about 200 kb/d at the start of 2026 to 701 kb/d in July, the most since 2017, and 655 kb/d in August. Verified In 2025 most Venezuelan crude went to China; in 2026 none has. Reported The heavy oil trade has rotated: Venezuelan barrels go north to Gulf Coast refiners built for them, and Canadian barrels go west to the Chinese refiners Venezuela used to supply. Interpretation For Canada the lesson is blunt: the Gulf Coast is no longer a reliable home for growth barrels, and the Pacific is where the new buyers are. Interpretation
How to read it → The dark centre tile is Trans Mountain today. Each tile around it is one market, with the one number that shows the effect. Red tiles cost Canada, green tiles help, grey tiles are context to watch.
Why it matters → Canada's energy system is one system. The discount, the diluent bill, freight and the next pipe are the same problem seen from different sides.
The links run both ways. Every barrel of diluted bitumen loaded at Westridge carries about 0.3 barrels of diluent, sold to Asia at a heavy crude price and replaced in Alberta at a light oil price, much of it from the US Midwest (Article 52). Estimate The condensate Alberta does produce comes largely from liquids-rich Montney and Duvernay gas wells, whose pace depends on LNG demand from LNG Canada’s second phase and Ksi Lisims. Interpretation And projects deducted C$29.7 billion of diluent costs in 2025 before royalties, so a wider discount and dearer diluent both cut the Crown’s take. Estimate Oil, gas, condensate and the provincial budget are one balance sheet. Interpretation
What Canada’s Pivot Means South of the Border
The United States is the other side of every number here. Canada supplied 64% of US crude imports in 2025, up from 41% in 2015. Reported Three US regions feel the pivot in different ways.
The Midwest is more dependent than ever. Canadian exports to the Midwest averaged 2,780 kb/d in January–August 2026, 334 kb/d more than in 2023, and its refiners processed about 2.92 mb/d of Canadian crude in the first half. Verified They buy at prices tied to Hardisty, so a wider discount is largely a transfer from Alberta to them, and to whoever holds rail and pipe space. On averages to September, that transfer is already about US$5bn a year more than in 2025; at our 2027 central view it would be about US$11bn. Estimate
The Gulf Coast has Venezuela. Exports to the Gulf Coast fell 318 kb/d between 2023 and 2026; Gulf refiners processed about 337 kb/d of Canadian crude in the first half of 2026, against 526 kb/d in 2024. Reported Enbridge’s Houston Oil Terminal, opened in July, gives Canadian barrels on the Gulf an export route, but at today’s freight rates few can use it. Interpretation
The West Coast is buying more. Exports to the US West Coast rose from 286 to 394 kb/d. Verified About 200 kb/d goes by pipe to Washington State refineries; on our arithmetic, most of the rest goes by tanker to California, which closed 284 kb/d of refining, 17% of its capacity, with Phillips 66’s Los Angeles refinery in October 2025 and Valero’s Benicia plant in April 2026. Estimate For Washington, a full Trans Mountain is mostly good news: Midwest refiners get cheaper heavy oil, Gulf refiners have Venezuela, and Canada still sends 88% of its crude exports to the United States. Interpretation
Freight Decides Who Keeps the Asian Premium
Westridge sits behind the Second Narrows in Burrard Inlet and loads only Aframaxes, about 75% full: Japan’s July cargo carried 559,070 barrels against a full load of about 750,000. Reported In normal markets it paid to sail those Aframaxes to the lightering zone off California and move the oil onto very large crude carriers (VLCCs), which carry about 2 million barrels. In June 2024 Argus put that route at US$4.36 a barrel to China, against US$6.32 for a direct Aframax, and 78% of Asian cargoes used it. Reported
The Iran war flipped that. VLCC rates hit records, a direct Aframax voyage fell to about 54% of the cost of the transfer route, and by August no Asian cargo was transferred. Reported Platts assessed a Vancouver–China Aframax at about US$3.1 million a voyage in August, about US$5.70 a barrel, before a record US$7.5 million on 10 September. Reported
How to read it → Each bar is the freight cost per barrel from Vancouver to China by route. The top panel uses June 2024 rates, the bottom August 2026 rates. Solid bars are reported; dashed bars are our estimates.
Why it matters → Full Aframax loads from 2027 save about US$1.6 a barrel; VLCCs at Pacific Link's Roberts Bank terminal would save about US$3.2 more. Freight is the core of Pacific Link's case.
Two changes alter the arithmetic. Dredging at the Second Narrows, due in early 2027, lets Aframaxes leave full; on our estimate that cuts about US$1.6–1.7 a barrel from a direct voyage, about US$0.3 billion a year on 500 kb/d. Estimate When VLCC rates normalise, the California transfer still saves about US$1 a barrel even with full Aframaxes, so expect some cargoes to return to it; the direct route keeps the time advantage. Estimate Full loads also stretch the dock: at about 34 Aframax calls a month, Westridge can load about 625 kb/d at today’s cargo sizes and about 840 kb/d full. Estimate That headroom is what lets most of Trans Mountain’s optimization, to 1.19 mb/d, reach the water. Interpretation
The price ladder shows where the value goes. On 21 October, WCS traded US$28 below WTI at Hardisty and the Pacific grades at Westridge US$23–25 below Brent; in September, Canadian heavy for delivery to China sold above Brent. Reported The roughly US$27 gap between Westridge and China covers freight, time, risk and margin. Interpretation With the pipe full, that margin goes to whoever holds space on Trans Mountain and the ships, not to producers selling at Hardisty. Interpretation
How to read it → Each bar is a price against its benchmark: WTI for Hardisty, Brent for Westridge and China. Arrows mark the costs between steps.
Why it matters → With the pipe full, the value of Asian demand goes to whoever holds pipe space and ships, not to the barrel sold at Hardisty.
What It Cost, and Who Is Paying
Ottawa bought the pipeline for C$4.5 billion in 2018, and the expansion cost C$34.2 billion, about six times the 2012 estimate. Reported In December 2024 it lent C$20 billion to Trans Mountain’s federal parent to repay an C$18 billion bank loan. Reported In 2025 the pipe earned C$3.0 billion of revenue, C$2.31 billion of adjusted EBITDA and C$556 million of net income, and paid its parent C$1.7 billion in dividends, interest and fees. Verified The company expects about C$11 billion of cash available for distribution over 2025–30. Reported Critics say the profit looks better than it is because most of the debt now sits with the parent, and that long-term contracts cover only about half the cost. Reported
How to read it → Left: what the Crown put in. Right: what the pipe returns, as reported by the company.
Why it matters → The pipe now earns cash, but on a cost base six times the original estimate. The same choices on tolls and contracts now face Pacific Link.
Shippers pay for it through tolls. The heavy toll from Edmonton to Westridge rose from about C$4.30 a barrel before the expansion to C$10.84–13.18 for firm service and C$14.06 for spot. Verified The regulator approved a settlement on 28 September that lets Trans Mountain contract up to 90% of capacity, up from 80%, on 15- and 20-year terms, with new tolls from January 2027. Verified Locking in 90% means Trans Mountain fills first in any squeeze, and the Mainline absorbs the variation. Interpretation
How to read it → Grey is the legacy toll; the blue band is the firm toll range across contract tiers; red is the spot toll. Quarterly ranges, in C$ per barrel.
Why it matters → Shippers pay about three times the old toll, and new tolls from January 2027 lock 90% of capacity into 15- and 20-year contracts.
Ottawa has said it does not intend to own the pipe for good, but the company’s chief executive now calls it “a sovereign pipeline”. Reported And Ottawa is doubling down: Trans Mountain leads the development of Pacific Link, which the federal and Alberta governments would own equally, with at least 10% offered to Indigenous communities. Reported The Crown is becoming the long-term owner of Canada’s Pacific oil corridor. Interpretation
Our View to 2030, and Our Calls
Relief comes in steps: about 90 kb/d from a drag-reducing agent and full tanker loads in early 2027, 210 kb/d of mainline optimization by the end of 2028, and Prairie Connector’s 465 kb/d at the end of 2028 if it is sanctioned in mid-2027. Reported
Our view of the discount follows from that. Pipeline parity, the cost of moving heavy oil to the Gulf Coast plus its quality discount, is about US$13, which is also Alberta’s budget assumption. Reported When pipe space runs short, a congestion premium sits on top: about US$6 in May and June this year, before freight spiked, and up to US$20 in past squeezes. Estimate Assuming the war freight premium fades during 2027 and Venezuelan flows hold, we expect US$18–22 in 2027 and US$17–21 in 2028. With Prairie Connector, the discount returns to US$12–15 from 2029; without it, supply growth keeps it near US$20 until Pacific Link. Estimate The gap between US$13 and US$20 is worth about US$9 billion a year to Canadian producers. Estimate
How to read it → Navy bars are annual averages (2026 is January to September); the red bar is October's daily range. Shaded boxes are our ranges; the gold line is US$13, Alberta's budget assumption and pipeline parity.
Why it matters → The gap between US$13 and US$20 is worth about US$9bn a year to Canadian producers. Whether Prairie Connector is sanctioned in 2027 decides which path Alberta is on.
Pacific Link is the long answer, and its case rests on freight. The Alberta government estimates it at C$35.2–43.7 billion for 1 mb/d to a deep-water terminal at Roberts Bank; service is targeted for 2032–34. Reported IEEFA estimates its tolls at C$18.70–23.72 a barrel, about 60% above Trans Mountain’s committed tolls at the low cost, and argues shippers’ netbacks would be US$5–8 a barrel lower. Reported On our freight model, loading VLCCs at Roberts Bank would save about US$3.2 a barrel against a full Aframax at normal rates, about US$1.2 billion a year. Estimate That covers much of the toll gap at the low cost, not all of it; the rest has to come from a narrower discount, which is why the open season next spring is the number to watch. Interpretation
| Risk | Why it matters | What to watch |
|---|---|---|
| China concentration | About half of Westridge volumes go to China, and about a third to one refiner | Korean, Japanese and Indian cargoes; any Chinese trade measure |
| The war premium fades | Asian buyers pay up for heavy oil that avoids Hormuz; that eases if Gulf flows return | Chinese seaborne imports back near 10 mb/d; Pacific grade discounts |
| Freight | Record Aframax and VLCC rates take a large share of the Asian premium and close the Gulf Coast re-export route | Vancouver–China Aframax and transpacific VLCC rates |
| Venezuela | More Venezuelan heavy in the Gulf pushes Canadian barrels out; renewed sales to China would soften the Asian bid | US imports of Venezuelan crude (EIA); any cargo to China |
| US policy | Crude is exempt from US tariffs so far; that can change | Any tariff, quota or licence on Canadian crude |
| Diluent | Every 1 mb/d of new pipe needs about 300 kb/d of diluent (Article 52) | Condensate premium at Edmonton; Southern Lights expansion |
| Cost and tolls on Pacific Link | Tolls well above Trans Mountain’s could leave it part-full | Open season terms, spring 2027 |
What Trans Mountain teaches applies to every West Coast export, crude, LNG or LPG: Asia pays for supply that avoids chokepoints, but the dock decides the ship, the ship decides the freight, and contracts decide who fills first. Interpretation
| Our call | By | What would prove us wrong |
|---|---|---|
| The WCS discount averages US$18–22 in 2027, well above Alberta’s US$13 budget assumption | End-2027 | A 2027 average below US$18 |
| Trans Mountain is apportioned again every winter until its mainline optimization is complete | Late 2028 | A winter without apportionment before 2028 |
| Asia keeps more than 65% of Westridge tankers even if Gulf flows recover | 2027 | Asia’s share below 60% for two quarters |
| Some Asian cargoes go back to the California VLCC transfer once VLCC rates normalise, even after dredging | 2027 | No transfers in 2027 despite normal VLCC rates |
| Prairie Connector is sanctioned in 2027, because the discount makes the case for it | End-2027 | No investment decision by the end of 2027 |
| Ottawa keeps Trans Mountain at least until Pacific Link is built | 2030 | A sale agreement before Pacific Link construction starts |
Trans Mountain did what it was built to do, and the world made it more valuable than its planners expected. The task now is the one it exposed: Canada’s oil, diluent and shipping capacity have to grow together, or the discount will keep taking back what the pivot earned.
Insight Stream — LNG/LPG & Waterborne Trade Analysis
Condensate and diluent balances, LNG contract economics, LPG benchmarks and NGL export infrastructure. Independent market intelligence, built only on sources you can check.



