Couche-Tard

IThe business

Couche-Tard sells petrol, diesel, cigarettes, drinks, snacks and hot food to people who are in a hurry, and it does it at scale. On 26 April 2026 it had 9,356 stores in North America, 4,820 in Europe and 387 in Hong Kong, 14,563 in all, plus 2,704 more that carry the Circle K name under licence in 12 other countries.⁴ About 145,000 people work in the network.¹⁰ Alain Bouchard opened his first store in Québec in 1980. The head office is in Laval.¹⁰

All company figures on this page are in US dollars, marked US$, because that is how Couche-Tard reports, under IFRS, and 57% of its revenue is American.³ The shares trade in Toronto in Canadian dollars, so the headline price and my values are in C$, and I give the US$ figure beside them. The exchange rate is 1.396, the Bank of Canada’s average on the price date.² The fiscal year ends in late April. Fiscal 2026, written FY2026, is the 52 weeks to 26 April 2026, and the first quarter of fiscal 2027 ended on 19 July 2026.⁴⁵

The key fact about the earnings is in the next chart. In FY2026 the company made US$6.91B of gross profit from merchandise and US$7.30B from road fuel, and spent US$7.75B running the stores, before a US$261M one-off legal recovery C (calculated).⁴³ Merchandise alone does not cover the cost of running the stores. Fuel does the work. Fuel was 73% of revenue and 50% of gross profit.⁴

Three regions, one playbook

The United States is 63% of gross profit: US$9.18B of US$14.46B. Europe, which includes Hong Kong, is 26%, and Canada is 10% C (calculated).³ In the US the company sold 9,675M gallons of fuel in FY2026 at a gross margin of 47.49 cents a gallon. In Europe it sold 17,521M litres at 11.73 cents a litre, and in Canada 5,799M litres at C$0.1559 a litre.⁴ Europe is bigger in volume and smaller in profit per unit. Its fuel margin moves more, because the company describes its European model as more integrated, with a longer supply chain.⁴⁵

Stores come in four kinds. Couche-Tard operates 10,730 of them itself. Another 1,354 are on property it controls but run by independent operators, 1,369 are run by independents on their own property, and 1,110 are franchised.⁴ That mix matters because the first group carries all the store costs and keeps the store profit, and the others pay rent or buy fuel under supply contracts.

What it sells

Merchandise and services were US$19.6B of revenue in FY2026 at a gross margin of 35.2%.⁴ Cigarettes and other nicotine were the largest category: US$7.1B of revenue and US$1.3B of gross profit, 36% and 19% of the merchandise totals.⁴ That is the product whose decline the filing lists as a risk, and it explains the effort in food. The company’s “Core + More” strategy, set out in February 2026, keeps fuel, nicotine and drinks as the core and puts food, private label, car washes, mobility and about 750 new stores by 2030 into the “More.”¹¹⁴ I read this as a plan to keep the traffic while the tobacco profit slowly fades.

The deal machine

The business in its current shape is the product of takeovers. A short list of the largest ones: CST Brands in June 2017 for about US$4.4B including debt; Holiday Stationstores in December 2017 for about US$1.6B; 2,175 European sites from TotalEnergies in December 2023 for about €3.4B (US$3.8B); the 112 MAPCO stores in November 2023 for US$469M; and 270 GetGo sites from Giant Eagle in June 2025 for US$1.6B.¹²¹⁰⁴ The company says it has done 75 deals in 20 years.¹³ The approach in each case has been the same: buy a network of sites that already have fuel and a shop, fit it into Couche-Tard’s purchasing, supply and store formats, and cut costs. I look at whether the returns justify that record in Management.

Two more deals are open. On 31 July 2026 the company launched a cash tender for all of Żabka, Poland’s largest convenience chain, at PLN 32.00 a share, an equity value of PLN 32.6B or about US$8.6B. It would be the largest deal in the company’s history, and I treat it separately in Valuation.⁷⁵ On 9 October it agreed to buy 50 Québec stores it already runs with Irving Oil, plus 2 in Ontario and fuel supply to 71 of its own stores; the price was not disclosed, so I include nothing for it.¹⁴

The state of play now

Results for the first quarter of FY2027, to 19 July 2026, were good. Adjusted earnings a share were US$0.90 against US$0.78, adjusted EBITDA rose 10.5%, and the US fuel margin was 52.61 cents a gallon against 44.00 a year earlier.⁵ Merchandise margin slipped 0.3 points to 35.1%, same-store merchandise sales grew 1.6%, and same-store fuel volumes in the US fell 1.6%.⁵ So the quarter was a fuel quarter. The next results come on 24 November 2026.¹⁵

IIThe moat

The honest test of a durable business is what happened to its profit when conditions were bad. Couche-Tard has not had a bad decade, so I use the one hard test it did have. Revenue fell 8% in FY2020 and another 15% in FY2021, as fuel prices and driving collapsed. Operating income did not fall at all: it was US$2.49B in FY2019, US$3.16B in FY2020 and US$3.68B in FY2021 C (calculated).⁶¹⁶¹⁷ Part of that is lease accounting, which lifted FY2020 reported operating income. But net earnings also rose, from US$1.83B to US$2.35B and US$2.71B.⁶¹⁶¹⁷ People who still drove bought fuel, snacks and cigarettes from the same stops, and the company found margin where it could.

Here is the other side. The profit that rose in that test was mostly fuel margin, and the margin kept rising for five years. A business with a durable edge in selling fuel would see the US margin stay level while volumes changed. Instead it went from 23.60 cents a gallon in FY2019 to 47.49 in FY2026, and Casey’s margin rose 3.9 cents in its latest year, to 42.6 cents before card fees.⁶⁴⁸ Much of that looks like an industry setting, not an edge. Some of it is Couche-Tard’s own doing, such as buying better and optimising prices, and some of it is that retail fuel margins in North America have simply been richer since 2021. I can’t split the two from the outside, and nobody else can either.

What does protect it

• Buying and supply. A network this size buys drinks, snacks and fuel on terms a one-store owner can’t get. Couche-Tard has opened three new distribution centres in FY2026, the first in Otsego, Minnesota in November 2025, to put its own merchandise through its own system.¹⁰⁴
• Sites. A convenience store lives on location. It controls the land under its 10,730 company-operated sites, by ownership or lease, and says it leads convenience retail in Canada, the Nordic and Baltic countries and Belgium.⁴
• The store itself. The merchandise margin rose from 33.5% in FY2021 to 35.2% in FY2026, as food, drinks and private label took share.¹⁹⁴ Same-store merchandise sales were up 1.9% in FY2026 after a 0.4% fall.⁴ This is the piece I would call earned, and it is steady.
• Money. The credit rating is BBB+ from S&P and Baa1 from Moody’s, with US$6.6B of cash and unused credit.⁴ In a downturn that is the capacity to buy sites cheaply.

Where it is thin

Fuel is half the gross profit and the company does not control its price, its cost or its volume. Same-store fuel volumes fell 1.0% in the US and 2.2% in Europe in FY2026.⁴ Electric cars cut into the fuel volume, and fuel gets more competitive at the pump when margins are rich. The company’s own filing says it plainly: results are sensitive to fuel prices and gross margin, and the market’s limited ability to absorb price changes can move them.⁴

Nicotine is the second weak point, at US$1.3B of gross profit, 19% of the merchandise total. The filing lists a sustained fall in demand as a risk, and in Canada the company blames tobacco for a 0.9% fall in same-store sales in the last quarter.⁴ Other nicotine products have grown in the US, but the category is changing quickly.⁴

And returns have been sliding. Return on capital employed, which counts goodwill, was 15.8% in FY2017, 17.5% in FY2023 and 13.7% in FY2026, with a low of 12.2% in FY2025 R (reported).^1820422 A business with a clear edge would see returns rise as it grew. Here the company bought ever-larger networks at full prices, and returns drifted down, which suggests the edge is real but is being paid away in deal prices.

IIIThe competitive advantage

The question here is what Couche-Tard can do that a rival cannot. In the US the comparison is with a company like Casey’s, which is Midwestern, runs mostly its own stores and sells a lot of prepared food.

On the pump there may be an edge. Over the 52 weeks to July 2026 Couche-Tard’s company-operated US stores earned 50.79 cents a gallon before card fees and 44.84 after, which is 8 cents above Casey’s before fees, and the gap a year earlier was 6.7 cents (45.40 against 38.7).⁵⁸ A gap of that size that lasts two years looks like skill, in buying, in supply and in pricing. But the two chains are in different places. Casey’s is mostly small Midwestern towns, and Couche-Tard is in 48 states, with a different mix of places and competitors. I can’t tell which way that cuts, so I can’t separate geography from skill, so in the model I give the company credit for the margin it earned but I do not assume it keeps all of it.⁴

The advantages I can see more clearly are three. The first is integration. The company runs fuel supply, trading and logistics itself in Europe and, in good years, captures more of the fuel move: in the fourth quarter its European margin was 13.44 cents a litre against 9.57 the year before, which it ascribed to a longer supply chain.⁴ That is worth money in volatile times and costs money in calm ones, and the first quarter of FY2027 already shows it easing, at 11.34 cents.⁵ The second is the ability to take over and repair. TotalEnergies’ stations in Germany, Belgium, the Netherlands and Luxembourg were run for a major oil company, and the company says its annual savings reached €61M in about two years, with €120M expected in FY2027 and €170M in FY2029.⁴ The third is financial. Few convenience chains can write a US$1.6B cheque in cash, as it did for GetGo, fund it with commercial paper, and be back to 1.77 times net debt to EBITDA a year later.⁴⁵

Electric cars, and the one place to look

The electric-car threat is real in fuel. The place to see it first is Norway, where 95.9% of new cars sold in 2025 were battery-electric.²³ Circle K Norway is its operator there. The company says it opened its 1,000th charging point in Norway in 2026 and is the number two charging destination there, and number one in Sweden.²⁴ In all it has about 4,700 charging points in Europe and North America, and charging volumes rose 55% in the year.²⁴ In North America it has 696 points, and in 2026 it agreed with IONNA that IONNA will take over and upgrade most of the existing US chargers.²⁴ It does not publish fuel volumes or profit for Norway, so I can’t say how fast the fuel profit there is shrinking, or whether charging sales replace it. That is a gap in what I know, and it matters because Norway shows where the rest of Europe might be in 10 to 15 years. In my model Europe’s fuel volume falls 1.5% a year, and in the bear 2.5%.

Where rivals can catch up

The part of the formula that a rival can copy is the store. Fresh food, loyalty programmes and good sites are available to any well-run chain, which is why Couche-Tard is spending on them.¹¹ The company’s own plan is to lift same-store merchandise sales 2% to 3% a year to FY2030, and it grew 1.9% in FY2026 and 1.6% in the latest quarter.¹¹⁴⁵ That is close to the target, but with 7.1B of nicotine sales at its core it is being reached against a falling product. Whether the food can carry more of the load than the tobacco loses is the question for the next five years.

IVManagement

Two men run Couche-Tard. Alain Bouchard, the founder, is Executive Chairman and was chief executive for 25 years until 2014. Alex Miller has been chief executive since 6 September 2024, taking over from Brian Hannasch, who retired.¹⁰ The chief financial officer, Filipe Da Silva, has held the post since July 2023.¹⁰ It is a stable team, and the change at the top came without a break in strategy.

Who controls it

Couche-Tard is often described as a company controlled by its founder through shares with extra votes. That was true until September 2022. Then the 10-vote Class A shares were converted into ordinary one-vote shares, after a shareholder vote, and the extra-vote class was repealed.¹⁰ Today there is one class. Mr. Bouchard holds 123,573,993 shares, 13.45% of the company, and 13.45% of the votes, and he is the only holder of more than 10% that the company knows of.¹⁰⁹ At the Q4 2026 price that stake is worth about C$10.0B C (calculated). Directors and executives together own 19%.¹⁰ So the founder has influence from a very large stake, a seat as chairman and a long history, but no lock. Other holders have 86.5% of the votes, and in a large deal they would have a say.

How they are paid, and what that pays them to do

Mr. Miller received US$13.49M in FY2026, up from US$10.01M and US$4.64M in the two years before. Mr. Bouchard received US$8.40M, mostly share units and options.⁹ Total direct pay for the five named officers was US$21.8M, 0.3% of EBITDA.⁹ I don’t think the amount is the issue; what matters is what the pay rewards.

The yearly bonus is 75% on net earnings against a budget, and 25% on personal goals. Nothing is paid below 90% of the budgeted net earnings, and the payout is capped at 250% for results of 130% or more of budget.⁹ That is a plan that rewards growth in profit. It doesn’t reward return on capital, earnings a share or cash flow, and it pays on net earnings that a fuel-margin windfall can lift. Mr. Miller’s bonus went from US$0.22M in FY2025 to US$2.84M in FY2026, the year of record fuel margins and the legal recovery.⁹ I can’t tell from the circular whether the recovery counted toward the bonus. The long-term awards are options (20% a year), performance share units that depend on operating objectives set at the time of grant, and restricted units, with the performance measures not spelled out in the part I read.⁹ All the share units are settled in cash, so they cost the company cash and do not dilute owners.³ Options outstanding are 2.18M, about 0.2% of the shares.³

The two big officers must own five times their salary. Mr. Miller held 45,688 shares and 101,356 deferred share units worth C$11.3M at the April price, and Mr. Bouchard’s holding is far above any requirement.⁹ Executives may take up to half their bonus in deferred share units, with a 25% company match, which is a way to push them to own stock.⁹ The bonus design leans toward growing earnings; the ownership rules lean toward the share price. The first is easier to do with a purchase.

What they did with the money

The record over a full cycle is the best evidence. In ten years Couche-Tard made US$21.3B of cash that an owner could take out C (calculated). It spent US$14.6B on acquisitions, US$9.4B on buybacks and US$3.2B on dividends, US$27.2B in all, and borrowed the US$5.9B difference C (calculated).^{181261617192021223}

What it got for the money is mixed in a way that is typical for a serial buyer. Diluted earnings a share rose from US$1.06 to US$3.37, 13.7% a year, and the diluted share count fell from about 1,140M to 933M, down 18% C (calculated).¹⁸¹⁶³ That is good. The shares rose less, from C$34.06 at the end of October 2016 to C$80.98, about 9% a year before dividends C (calculated).¹ Return on capital fell from 15.8% to 13.7% over the period, which suggests the deals cost close to what they were worth.¹⁸⁴ Buybacks were large: US$1.6B in FY2026 alone, 30.0M shares, an average of about US$52, or about C$72.⁴ Most were done at prices below today’s C$80.98, which is to the company’s credit, and the biggest years were US$2.4B in FY2023 and US$1.8B in FY2022.¹⁹²⁰ I would call the record fair.

The Seven & i episode is the best single test of discipline. The company offered ¥2,600 a share in cash, a 47.6% premium, for the owner of 7-Eleven, a bid the press put at about US$47B.¹³²⁵ After almost a year of being refused, it said that its due diligence had received 14 files on the US business in 10 weeks, and withdrew on 16 July 2025.¹³ The shares had fallen from C$83.57 before the approach became public to C$68.32. On the day it withdrew they rose 8.3% to C$74.00 C (calculated).¹ The market preferred no deal. The management that made the offer also walked away, and that is the pattern I want to see.

The test now is Żabka. At US$8.6B for the equity it is nearly twice CST Brands, the largest earlier deal at about US$4.4B with debt, and it is in a country where Couche-Tard already operates. The company expects leverage of about 3.0 times net debt to EBITDA at closing, against 1.77 times now, and a return to its target range in the second year.⁷⁵ It states three hurdles: US$250M of savings by year three, higher earnings a share by year two, and a double-digit return on capital in year three before lease accounting. Holders of 57% of Żabka have signed to tender, so the sale is likely, and the price was set by negotiation with them. Those are the right hurdles. I come back to them in Valuation and in the table in section VII.

VFinancials

Three things in Couche-Tard’s statements don’t mean what they first seem to. Revenue follows fuel prices, so it more than doubled from US$37.9B to US$76.5B in ten years without telling us much. Lease accounting changed in FY2020, when IFRS 16 moved rent out of operating costs, so EBITDA and debt before and after that year can’t be compared. And the company’s adjusted figures take out items it calls one-offs, which in FY2026 were a US$261M legal recovery and a US$66M gain on a sale that regulators required R (reported).⁴¹⁶ Net earnings of US$3.14B for FY2026 were US$2.89B with those removed.⁴

With those cautions, the picture is of a business that has grown steadily and with few bad years. Earnings before interest, tax and depreciation went from US$2.37B in FY2017 to US$7.00B in FY2026, and net earnings from US$1.21B to US$3.14B, 11.2% a year C (calculated).¹⁸³ Most of the rise came in two steps: the acquisitions of CST and Holiday in FY2018, and the fuel margin from FY2021. The only real fall was in FY2024 and FY2025, when net earnings slipped to US$2.73B and US$2.58B despite the TotalEnergies deal. In FY2025 expenses rose 9.5% against 7.6% for gross profit.²²⁴ In FY2026 the margin returned.

What an owner can really take out

Net earnings is an accounting number. The cash an owner could take is operating cash flow, which is already after interest and tax and, here, after stock pay, less what has to be spent on the stores, less the principal paid on leases. For FY2026 that was US$5.36B, less US$1.77B of capital spending net of sales of property, less US$0.58B of lease principal: US$3.01B C (calculated).³ That is 96% of net earnings. For ten years the figures are US$21.3B of owner cash against US$24.0B of net earnings, 89% C (calculated).^{181261617192021223} This is a business where reported earnings turn into cash fairly well.

Three adjustments matter for FY2026. The US$261M legal recovery came into cash and I count it: stripped of tax it is about US$200M, which takes owner cash to about US$2.8B E (estimate). A working-capital outflow of US$549M took cash the other way; the company puts it down to higher working capital needs, and it is unlikely to repeat if fuel prices are steady.³⁴ And US$1.99B of capital spending includes growth spending: depreciation on owned property and equipment was US$1.55B, which is my guide to what it costs to stand still, so about US$0.4B was spent on growth, such as new stores and distribution centres E (estimate).³ Put together, FY2026 owner cash was about US$2.8B to US$3.3B, or 5.3% to 6.2% of the market value, and I use the middle.

The company’s own free cash flow, which it defines as EBITDA less net capital spending, interest, lease principal, tax and dividends, was US$1.80B in FY2025 and, by my rebuild of the same sum, about US$3.1B in FY2026, against a target of more than US$2.5B C (calculated).¹¹³ That is after dividends and before the buybacks and deals it chooses to make.

A note on the fuel margin and card fees

The fuel margin the company headlines is before card fees, which sit in expenses. For company-operated US stores in the first quarter the margin was 53.87 cents before fees and 47.30 after, a fee of 6.57 cents, and over the 52 weeks to July it was 50.79 before and 44.84 after, up from 45.40 and 39.58 a year earlier.⁵ Fees rise with the pump price, so a spike in prices lifts them. I keep fees in expenses, as the company does, and I put every margin on this page on the headline measure.

Where the cash went

I covered the totals under Management. The point for the statements is that the balance sheet was the funding source. Net interest-bearing debt, including leases, was US$13.3B at the end of FY2026 and US$12.2B in July, from US$6.7B at the end of FY2020 R (reported).⁴⁵¹⁹ Against EBITDA that is 1.99 times, then 1.77 times, which is conservative for a retailer, and the credit ratings are BBB+ and Baa1.⁴⁵ Debt payments, principal and interest, come to US$1.3B in FY2027, US$1.4B in FY2028 and US$1.5B in FY2029, against US$6.6B of cash and unused credit.⁴ The company had repaid a €750M note on 6 May 2026 and issued €750M, US$1.2B and C$500M in the year.⁴ Interest on the debt was US$521M and on leases US$189M in FY2026.³ A debt-funded Żabka deal, taking leverage to about 3.0 times, would move this table, and it is the one real change to the balance sheet ahead.

The price in numbers

On those numbers the shares trade at 18 times adjusted earnings and 9.5 times EBITDA with leases counted, and the owner cash yield is between 5% and 6%. The fall from the June high of C$93.76 to C$80.98 has taken the price back to roughly where it was in early March 2026.¹ It rose 11.7% on 23 June 2026, the day after the FY2026 results, and has since given up all that gain and a little more C (calculated).¹²⁶ I don’t know why, and the filings do not say. The timing coincides with the Żabka tender, announced on 31 July, and the first-quarter results in early September.

VIValuation

First, how this goes badly wrong

Before any numbers, I ask what would hurt an owner most, and I count three ways.

The fuel margin gives back its gains while costs keep rising. The US margin was 23.60 cents a gallon in FY2019 and 47.49 in FY2026.⁶⁴ Each cent is about US$97M of gross profit, 1.4% of EBITDA C (calculated).⁴ If it settles near 42 cents, EBITDA falls by about US$0.5B, and the stores can’t easily make that up while expenses grow 3% a year. This is the one that matters most, and the company has no control over it.

Takeovers at full prices, paid with debt. Return on capital has slid from 17.5% in FY2023 to 13.7%, and Żabka would take leverage from 1.77 to about 3.0 times EBITDA.⁴⁵⁷ A deal that costs more than it earns is easy to hide inside a growing group. If Żabka turned out to be worth 17% less than the price, that alone is US$1.5B, or C$2.3 a share.

The store engine stalls. Merchandise gross profit is US$6.9B, and 19% of it is nicotine.⁴ Costs have to grow slower than that profit, or earnings fall even with steady fuel. Same-store sales grew 1.9% in FY2026, and the company’s target of 2% to 3% sits just above it.¹¹⁴

What an owner can really take out

The cash an owner could take in FY2026 was about US$2.8B to US$3.3B, after stock pay (which is paid in cash and so already deducted), after interest and tax, after lease principal and after the capital needed to keep the stores going and some growth.³ I use US$3.05B as the middle, US$3.3 a share, or C$4.6 E (estimate). Adjusted earnings a share were US$3.10. The two are close, which is good news about quality. The bad news is that both rest on a fuel margin that is at a record. In the model, owner cash before interest is US$3.3B in FY2027 and US$3.1B in FY2028 in the base case, because I let the margin fall back.

What I can’t judge

I can’t say where the US fuel margin goes. Nobody can. It has been at least 45 cents for four years, and that is long enough to be a new level, but short enough to be a cycle. So I show a ladder and not one number. Every cent of margin sustained from FY2028 is worth C$1.79 a share. From 40 cents to 52 cents is a range of C$62 to C$83.

I also can’t judge how fast electric cars take the fuel pools in Europe, because the company does not publish Norway’s fuel volume. And I can’t see inside Żabka beyond its public accounts. Those are the main reasons the range is wide.

How the cases are built

I build each case year by year. Merchandise gross profit comes from sales growth and margin. Fuel gross profit comes from volume and the margin per gallon or litre, in the US, Europe and Canada separately. Expenses grow at a stated rate from FY2026’s US$7.75B, with the legal recovery put back. EBITDA less tax at 23.5%, lease payments, and capital spending is the cash before interest. I discount it at 8.5% and take off net debt of US$7.4B. Leases are a cost in my cash flows, so I do not also count them as debt. The 8.5% is the Government of Canada ten-year yield of 3.84% plus about 4.7 points, and it is the same rate I use for Costco on this site.²⁷ I stop at FY2036 and then grow the final year’s cash at 2.5% in the base case. FY2027 counts only its second half, because the first has been reported. Currency is held at April 2026 averages.

Three comparisons help. My base EBITDA grows from US$6.7B in FY2026 to US$7.1B in FY2030, 1.5% a year, against the company’s target of 6% to 8%.¹¹ The target assumes fuel profit grows with inflation from a record base, plus a US$850M cost programme. My base lets the margin fall back first. My bull reaches US$8.3B, 5.4% a year, a little under the low end of their range. And my base values the equity at US$48B, against a market value of US$53B C (calculated).

Żabka, as a separate line

The tender prices Żabka’s equity at US$8.6B, 7.8 times its last-12-months adjusted EBITDA of US$1.1B.⁷ That is equity over EBITDA, not enterprise value. Żabka’s own second-quarter call put its net financial debt without leases at PLN 2.3B at 30 June, with cash above PLN 1.6B. An aggregator’s balance sheet adds lease liabilities of about PLN 5.7B, which gives debt including leases near PLN 9.6B and net debt near PLN 8.0B, about US$2.1B C (calculated).²⁸ That puts the enterprise value near US$10.7B, or 9.7 times EBITDA with leases counted, about the multiple Couche-Tard itself trades at. So the price is roughly fair against the buyer’s own valuation, and the case for gain rests on US$250M of savings and sales gains by year three.⁷

I do not value Żabka from the ground up. I have not seen its cash flow statement, and I cannot value it better than the owners who negotiated the price. I treat the price as fair and value only what Couche-Tard adds or loses. In the base case the US$250M comes through half in time: about US$190M after tax, capitalized, less one year’s cost to achieve it, is a gain of about US$0.9B. In the bear, Żabka is worth 17% less than the price and no savings arrive: a loss of US$1.5B. In the bull, the full savings and modest growth give US$2.2B. I weight each by a 90% chance that the deal closes, because holders of 57% have signed to tender and the approvals the company listed, from the European Commission or Poland’s regulator and Romania’s investment screen, were pending at the last public update I found. A secondary report says the offer period was extended on 19 October to 2 December and that Poland’s competition office received a filing on 16 October; I could not confirm either against the official notices.⁷²⁹ If it does not close, all three cases lose that line. I leave out the Irving Oil stores because the price is not disclosed.¹⁴ Taken together, Żabka moves my base by about C$1 a share. The leverage and the management attention matter more than the value line.

The price needs one of three things. The US margin has to settle at about 50.7 cents a gallon from FY2028 and then keep growing, which is above any full year yet and below the latest quarter. Or growth after FY2036 has to be 3.3%, nearly a point above the 2.5% I use, for ever. Or I have to accept 7.9% instead of 8.5%. None of these is silly. All three are bets that the present is normal.

The margin of safety, and the price against the range

My margin-of-safety price is C$51, US$37. It is 70% of my base value. At that price an owner would have a 30% cushion against being wrong about the fuel margin, and the price would be 46% above my bear. The shares last closed below C$51 in June 2022.¹ It would also be 11 times adjusted earnings. At C$80.98 there is no margin of safety. The price is 11% above my base, 28% below my bull, and 2.3 times my bear. My weighted value, with 25% on the bear, 50% on the base and 25% on the bull, is C$73.

The price reflects a real debate. One side says US fuel margins have moved up for good: they have been at or above 45 cents for four fiscal years, Casey’s sees the same, and the first quarter of FY2027 reached 52.61. That side would say my base is too low, and my bull is the central case. The other side says margins were 24 cents in FY2019 and that a boom in margin is an invitation to competitors to cut price, with electric cars waiting behind. That side would say my bear is more likely than I give it. I give the first side more of the weight than the second, and I still end up with a base below the price, because costs are not going to wait. The market could be right, and I name the case for it: this is a well-run company, earning a record, with a founder who owns a seventh of it and has just shown he will walk away from a bad deal.

The next best use of the same dollar

The closest comparable on this site is Costco. Its base return, from today’s price to 2036 with owner cash counted, is 0.7% a year in my work, and Walmart’s is negative 0.9%. Couche-Tard’s base return is 7.5% E (estimate). That beats both and the 3.84% on a ten-year Canadian bond by about 3.7 points, and it falls short of the 8.5% I ask for. It differs from the 7.9% above because that is the rate at which my base cash flows equal the price, while this return assumes the value I put on the business in October 2036. Costco may well be the better business, and its price leaves a lower return. Couche-Tard is a good business at a fair-to-full price. Those are separate questions, and I answer them separately: the quality is high and the price is a little above what I think it is worth. I give neither a grade.

If the bull case needs many things at once, I multiply them out. For the bull I need the US margin to hold near 49 cents (about 45% likely, by my guess), expenses to grow no more than 2.2% a year (40%), merchandise sales to grow 4.8% a year against 1.9% in the last year (35%), and Żabka to deliver its savings (50%). Together that is 3% if they are independent. They are not independent, since a strong consumer helps three of them, so I call it about 10%, not 25% E (estimate). My weighting of the bull at 25% is generous on that arithmetic.

Where this could be in 5 and 10 years

For the base to be right at 5 years, the US margin has to be near 47 cents in FY2030 and 48 in FY2031, and merchandise sales have to grow about 4% a year. EBITDA is then US$7.3B in FY2031, and the business is worth C$82 a share, C$9 more than today, because five years of cash have gone out and the value of the rest has grown E (estimate). For it to be right at 10 years, EBITDA reaches US$8.1B and owner cash before interest US$3.9B in FY2036, and the value is C$94. Bear and bull differ on the fuel margin and on how fast costs grow: in the bear EBITDA stays near US$5.8B, in the bull it reaches US$10.3B by FY2036. The four numbers that drive it are fuel margin, expense growth, the share of that growth that needs capital, and the 8.5% rate.

The most likely way my base is wrong is the fuel margin. Four cents more from FY2028 lifts year 10 to C$103, and four cents less cuts it to C$85 E (estimate). These are estimates of value, not forecasts of the share price. The market price can sit above or below value for years.

VIIThe case for the price

What most people miss

Three things cut in Couche-Tard’s favour and get little attention. First, the stock pay is paid in cash and is already inside the cash flow I use. The share units are settled in cash, options are 0.2% of the shares, and the diluted share count is down 18% in nine years, so owners were not diluted to pay for the earnings.³¹⁸ Second, leverage has come down while it spent: net debt to EBITDA, with leases, was 1.77 times in July against 2.25 times at April 2024, after paying US$1.6B for GetGo and buying back US$1.6B of shares in FY2026.⁴⁵ Third, the cash is being earned. Free cash flow on the company’s own definition was about US$3.1B in FY2026 against a target of more than US$2.5B, TotalEnergies savings are on plan at €61M a year with €120M expected in FY2027, and adjusted earnings a share rose 15% in the first quarter while the shares were 14% below the June high.¹¹⁴⁵¹

The bull case

In plain words: fuel margins have reset higher for good, the stores are getting better at food and loyalty, costs are under control, and the company has a track record of buying networks and cutting costs. Management put numbers on it in February 2026. From FY2026 to FY2030 it aims for same-store merchandise growth of 2% to 3% a year, merchandise revenue growth of 4% to 5%, fuel gross profit growing with inflation, expense growth at or below inflation, adjusted EBITDA growth of 6% to 8%, and adjusted earnings a share growth of 10% or more.¹¹ It points to a cost programme worth about US$850M of EBITDA, 750 new stores by 2030, and the Żabka deal as accretive by year two.¹¹⁴⁷ Mr. Miller called it the next stage of the growth journey and Mr. Da Silva said the targets were calibrated and measurable.¹¹

I found no named analyst or fund manager whose published view on this company I could verify, so the bull case on this page is management’s own. The one named holder I can document is CDPQ, which sold 8.7M shares back to the company in July 2024 at C$80.50, a 3% discount, and kept about 3.5% at the time; the sale was described as a portfolio rebalance.¹⁰ That tells me about a holder’s timing in 2024 and nothing about today. Mr. Bouchard’s 13.45% is a holding, not a stated reason.¹⁰

Against reality

If Couche-Tard hits the low end of its target, EBITDA is US$8.5B in FY2030, and my model then gives about C$116 a share, a little above my bull and 43% above the price. So the price is possible, and it needs less than the full plan. The distance from today is the point. In the 12 weeks to July, adjusted EBITDA was up 10.5%, but last year’s base was weak, and over the 52 weeks it is only 2.5% above FY2026.⁵ To reach US$8.5B it has to grow 5.7% a year from the last 12 months, for 3.75 years C (calculated). That works only if fuel profit keeps growing with inflation from a record, merchandise grows over 4% against 1.9% now, and costs grow less than 2.5% against 2.7% to 3.1% now.

I multiplied out the odds for the bull case in Valuation: about 3% if the four conditions are independent, perhaps 10% with the overlap E (estimate). The price asks for less than the bull. It needs the fuel margin and the costs, and does not need Żabka’s savings, so it is more likely than the bull, and still not the central case.

What a 10x would take. Ten times C$80.98 is C$810, a market value of about C$745B, or US$534B. At 18 times earnings that needs about US$29.7B of net income, 9.4 times the US$3.14B of FY2026 C (calculated). At today’s net margin of 4.1% that is revenue of about US$720B, more than nine times today’s. In ten years it would take 26% a year. For comparison, the shares rose 9% a year in the last ten years, from C$34.06 to C$80.98, and earnings a share 13.7% a year, with US$14.6B of deals.¹¹⁸³ Couche-Tard sells into fuel and convenience markets that grow a few percent a year, and it would have to own a large part of all of them. Tenfold from here is close to impossible, and I would not build anything on it.

Where I come down

What moved me was cash: owner cash is 96% of net earnings, the balance sheet is strong, and the walk-away from Seven & i was the right call. What did not move me was the record margin; it is a gift from the market, and I treat part of it as temporary. What would raise my values is a US fuel margin at or above 50.7 cents on the reported measure for a full year, with expense growth under 2.5%. Each cent is worth about C$1.8 a share.

VIIIRisks, and what would change my view

The three ways this goes badly wrong are at the start of Valuation. These are the rest.

• Nicotine. US$7.1B of sales and US$1.3B of gross profit, in a category the filing says may be hurt by price rises, taxes, bans and illicit trade.⁴
• Electric cars. Norway’s new cars were 95.9% electric in 2025. The company does not publish Norway’s fuel volumes, so the pace of loss is invisible to me.²³
• Legal items. The US$261M net recovery in FY2026 included payments in a card-fee case where the company was a plaintiff, offset by other legacy matters. It is unlikely to repeat.⁴
• Leverage. About 3.0 times at the Żabka closing. The company says it will be back in range in the second year, with no expected change in its rating.⁷
• The minority put. A US$292M redemption liability, which the filing shows in the FY2027 column of obligations, relates to the Belgian business, and I treat it as debt.⁴
• Currency. The company reports in US dollars and earns about 37% of its gross profit outside the United States, in euros, Nordic and Polish currencies, Hong Kong and Canadian dollars.³ A 5% move in the currencies of its cash balances changes net earnings by about US$41M, partly offset by US$26M on its dollar debt.⁴
• Succession and control. Mr. Bouchard’s 13.45% has no extra votes, but the company is closely tied to him and to the founder-led culture.¹⁰
• Tax and interest. The tax rate was 22.9% and rose from 22.0%, and variable-rate debt is small, so interest rates matter little in the near term.⁴

Geopolitical risk, and what it means for the value

Couche-Tard deconsolidated its Russian subsidiaries in April 2022, has no stores of its own in the Middle East, and has no state owner. Its exposure is indirect. It sells fuel in 48 US states, ten Canadian provinces, the Nordic and Baltic countries, Poland, Germany, Belgium, the Netherlands, Ireland and Hong Kong, and it licenses the Circle K name in 12 other countries, including Saudi Arabia and the Emirates.⁴ The chokepoints are the price of crude and the sea lanes that carry it, trade rules between the US and Canada, and the border between NATO’s eastern members and Russia. The filing names “hostilities and geopolitical tensions in the Middle East” and tariffs, including retaliatory tariffs, as risks.⁴

The effect of a conflict on this company is not one-way. In the last quarter of FY2026 the fuel margin rose sharply in Europe because of price volatility, which the company credited to its supply chain.⁴ A spike is first a gift and then a cost, because customers drive less: European same-store fuel volumes fell 4.4% that quarter.⁴ I rerun the base case for each scenario below. They are not predictions. The price of C$80.98 is above my base, so it already assumes the margins of today persist and treats the volatility windfall as normal. I see nothing in it that allows for any of these cases.

The largest single loss is a long disruption in the Gulf, where I put the odds at about 1 in 8 over five years on my own judgement, since I have no model for it. The most likely is trade friction, at about 1 in 4, because tariffs are already in the filing and same-store sales grew only 1.9%. Weighted by those odds, trade friction costs more than the Gulf, about C$2 a share against about C$1. Hong Kong, with 387 of 14,563 stores, is small, and the filing does not break out its profit. Żabka sits close to a war and its 13,000 stores earn in zloty and euros, and I show it separately for that reason. Odds here are guesses and the values depend on my model.