Restaurant Brands International QSR

Reference table

1. What they actually sell — and where the revenue really comes from

The FY25 brand disclosure contains a surprise for anyone who thinks of this as a Burger King company:

Two observations, both important.

First, Tim Hortons is 45.0% of revenue — nearly three times Burger King. This is, at the revenue line, primarily a Canadian coffee-and-baked-goods company. That has consequences: Tim Hortons is a mature brand in a mature market, and its 5.1% growth rate is the ceiling on 45% of the business. Any thesis about international quick-service expansion has to contend with the fact that the largest single revenue contributor is domestically anchored in Canada.

Note the disclosure basis changed: FY22 shows Burger King at $1.897B and FY21 at $1.813B, versus $1.514B in FY25 — a decline that almost certainly reflects a reclassification (the International segment was broken out separately in FY23 and FY24) rather than an actual 20% revenue fall. Do not read a Burger King decline from this series.

Second, $2.641B of FY25 revenue — 28% of the total — is not attributed to a brand. The FY24 disclosure separately identified an "International Segment" ($935M) and "Advertising" ($1.116B); those lines are absent from the FY25 breakdown. This is a real disclosure gap and the brand shares above should be treated as approximate.

Geography (FY25) — this disclosure foots exactly:

Canada is 40.8% of revenue. Combined with Tim Hortons at 45.0%, the picture is clear: this is a company whose largest single geography and largest single brand are both mature and growing 4-5%. The United States grew 20.5% in FY25 (the fastest, and likely reflecting the same reclassification noted above), and "Others" — the genuine international growth vector — is only 10.9% of the total.

The structural conclusion: for the consolidated revenue line to grow meaningfully, either Canada/Tim Hortons must reaccelerate or the international 10.9% must become much larger. Consensus assumes neither happens, which is why it models a flat top line.

2. The financial record

Revenue has compounded 13.2% annually over four years — genuinely strong, and directly contradicting the flat forward consensus. But operating margin has compressed every year, from 32.7% to 23.7%, and FY25 operating income, net income and EBITDA all declined despite 12.2% revenue growth. That combination — growing revenue with falling profits — indicates the revenue growth is coming from lower-margin activity (most plausibly consolidated restaurant operations and advertising-fund revenue, which are pass-through in nature, rather than high-margin royalty).

This matters more than it first appears. If the recent revenue growth is largely low-margin pass-through, then the flat forward consensus may be modelling the royalty business correctly while simply assuming the pass-through growth stops. And it means the headline 12.2% growth rate overstates the economic growth of the franchise.

Free cash flow is the reliable series: $1.390B, $1.203B, $1.302B, $1.449B — steady, growing modestly, and highly predictable. That is what a franchise annuity looks like, and it is the correct basis on which to value this company.

The dividend and the payout constraint. Dividends paid rose from $971M (FY22) to $1.108B (FY25), consuming 76.5% of FY25 free cash flow. With $17.579B of debt outstanding, that leaves $341M annually — 2.0% of the debt load — for deleveraging, buybacks and acquisitions combined. There were no buybacks in FY24 or FY25. The dividend is covered 1.31 times by free cash flow, which is adequate but leaves no room for a downturn and very little for balance-sheet repair.

3. The share-count structure — the point most screens miss

This is the single most important valuation mechanic in the dive and it is easy to get wrong.

• Market capitalisation of $25,297,334,660 divided by a price of $72.91 implies 347.0M shares.
• Q1 2026 reports weighted average basic shares of 347M — matching exactly.
• Q1 2026 reports weighted average diluted shares of 459M — 32% higher.

The explanation is the company's partnership structure: exchangeable units held outside the common share count, which convert into common stock and are therefore included in diluted earnings per share but excluded from the quoted market capitalisation.

The consequence for valuation is material:

The vendor's enterpriseValueTTM of $39.97B is lower still, implying $14.67B of net debt against the $16.416B on the FY25 balance sheet — a further discrepancy we note and do not use.

On a fully-exchanged basis the business trades at 20.6x EBITDA rather than the 15.6x the vendor reports — a 32% difference. Similarly, free cash flow per fully-exchanged share is $3.16 (not $4.18), giving a free-cash-flow yield of 4.33% rather than 5.73%. Any screen using the quoted market capitalisation understates how expensive this company is by roughly a quarter. We flag it prominently because it is exactly the kind of structural detail that a mechanical comparison misses.

Balance sheet (FY25, 2025-12-31):

• Cash $1.163B; total debt $17.579B ($268M short-term, $15.150B long-term, the balance in leases)
• Net debt $16.416B → 6.8x FY25 EBITDA of $2.424B (my computation; the vendor's 5.72x uses a TTM base and a lower net-debt figure)
• Shareholders' equity $3.634B; goodwill $6.306B on $25.615B of assets — a 25% intangible share, reflecting the acquisition history
• Current ratio 0.99; working capital -$24M — normal for a restaurant business with fast inventory turns and no receivable cycle

6.8x leverage is the defining financial fact. It is serviceable — free cash flow of $1.449B against roughly $700-760M of normalised interest gives about 1.9x coverage after interest — and franchise royalty streams are exactly the kind of contractual cash flow that supports leverage. But it is high, it is not falling quickly (net debt rose from $14.622B to $16.416B in FY25), and there is no revenue growth in the consensus model to grow into it.

4. Data integrity — two significant problems in this file

PROBLEM ONE — the interest expense series contains an almost certainly erroneous quarter, and the FY25 annual figure inherits it.

Reported quarterly interest expense:

The Q3 2025 figure of $18M is a tenth of the surrounding quarters. And the FY25 four quarters sum to exactly $516M — the reported annual figure — so the annual interest expense of $516M carries the anomaly through. FY24's interest expense was $804M on less debt ($15.956B versus $17.579B).

The normalised run-rate, taking the six clean quarters, is roughly $175-190M per quarter, or $700-760M annually. If FY25 interest expense were $730M rather than $516M, pre-tax income would be roughly $214M lower — a material change to the $776M reported net income. We therefore treat the FY25 net income and EPS figures as unreliable and base all valuation work on (a) free cash flow, which is unaffected by the classification, and (b) the consensus adjusted-EPS series, which is validated separately below.

PROBLEM TWO — the gross profit line changes basis mid-FY2025.

Gross margin jumps 14.6 percentage points between Q2 and Q3 2025 and stays there. A step-change of that size in a restaurant franchisor is a reclassification, not an operational event. No gross-margin trend is computed anywhere in this dive, and the same quarter that carries the margin discontinuity carries the interest-expense anomaly — which suggests a single presentational change rather than two independent errors.

PROBLEM THREE — the GAAP EPS fields are not consistent with their own share-count fields. FY25 reports net income of $776M and diluted shares of 457M, which would give $1.70 — but reports diluted EPS of $2.35. The implied share count behind that EPS is roughly 330M, not 457M. The quarterly rows are similarly inconsistent: Q1 2026 reconciles correctly ($338M ÷ 459M = $0.74), but Q3 2025 does not ($315M ÷ 457M = $0.69 against a reported $0.95). The reported EPS series and the reported share-count series are on different bases, most likely because net income includes non-controlling interests that the EPS calculation excludes.

What we did instead: all per-share valuation in this dive uses the consensus adjusted-EPS series, which is internally coherent and independently validated. The check: earnings-calendar actuals for the last four quarters are $0.94, $1.03, $0.96 and $0.86, summing to $3.79 TTM, against a consensus FY25 estimate of $3.679 and an FY26 estimate of $4.050. That series is consistent and usable; the GAAP EPS fields in the income statement are not.

FLAGGED — the consensus EBITDA series is questionable. FY26 estimated EBITDA of $3.087B against FY25 actual of $2.424B is a 27.4% step-up on revenue growth of 4.4%. That would require operating margin to expand roughly six points in a single year. Given the FY25 margin compressed, we treat the estimate EBITDA series as unreliable and compute EV/EBITDA on reported FY25 EBITDA only. This is a milder version of the EBITDA-series failure seen across several files in this batch.

FLAGGED — the far-year revenue estimates are non-monotonic: $9.851B (FY26E) → $9.961B (FY27E) → $9.947B (FY28E) → $10.372B (FY29E) → $10.048B (FY30E). Revenue declines in two of four forward years. FY29 rests on 11 revenue analysts and FY30 on 6. The direction of travel — flat — is the signal; the individual far-year values are not precise.

FLAGGED — the FY25 brand disclosure covers only 72% of revenue (section 1), and the profile and technical blocks report slightly different 52-week ranges ($61.33-$81.96 versus $61.86-$81.67).

CHECKED — the most recent quarter. inc_q[0] (Q1 2026) has internally consistent net income, revenue and diluted EPS ($338M ÷ 459M = $0.736 versus the reported $0.74), share counts that do not duplicate the prior quarter, and an interest expense of $171M consistent with the normalised run-rate. The quarter itself is clean; it is the FY25 annual figures and the mid-2025 quarters that carry the problems.

5. Knowledge base — a genuine, recent, high-conviction endorsement

Five claims match, of which three are substantive and two tangential.

The two claims that matter most, both from the same source on the same date, and both recent:

• invest_like_the_best, 2026-02-10 (bullish, conviction 82, entities: Restaurant Brands International, Burger King, Tim Hortons): "RBI has compounded to roughly a 30x multiple on invested capital and is still growing, validating the concentrated operator-investor model."
• invest_like_the_best, 2026-02-10 (bullish, conviction 90, entities: Burger King, Tim Hortons, Hunter Douglas, Popeyes, Firehouse Subs): "Prefer simple, long-standing, well-moated businesses with strong brands that own the direct customer relationship and are hard to disintermediate."

This is the strongest single-name endorsement in the batch, and it deserves careful reading. The first claim is a specific, quantified statement of realised historical return on invested capital, made in February 2026 at conviction 82, naming the company. The second is a principle claim that explicitly lists all four of the company's brands as exemplars of the quality characteristics the source seeks. Together they constitute a considered, third-party, non-management endorsement of the business model.

The third substantive claim, from a separate source:

• lex_fridman, 2024-02-20 (bullish, conviction 70, entities: Restaurant Brands, QSR, Burger King): "Owns Restaurant Brands (Burger King etc.); franchise QSR wins on systems that let a stranger operate a successful store at scale." — a disclosed personal holding plus a clear articulation of why the franchise model works. Revealed preference, not commentary.

Two tangential claims:

• business_breakdowns, 2023-05-05 (neutral, conviction 75, entities: Cinnabon, Burger King, Taco Bell, Pillsbury): "Great license/collab deals combine each partner's distinct strength so one plus one equals three; both must bring credibility." — a brand-licensing principle that mentions Burger King as an example.
• no_priors, 2025-09-11 (bullish, conviction 60, entity: Mubadala): "Brazilian food-service franchises (Burger King, Starbucks, Subway) bought out of bankruptcy with Mubadala are a good long-term investment." — this concerns a Brazilian franchisee operating company, not Restaurant Brands itself. Counted for completeness, weighted at close to zero.

Read: Moderate-High conviction, and it is the right kind of conviction for this name. Note what the endorsement actually says: it is a quality and durability thesis, not a growth thesis. The conviction-90 claim explicitly prefers "simple, long-standing" businesses — the source is buying an annuity, not an exponent. Our verdict is entirely consistent with that lane: this is a good business worth owning, and the disagreement is about the price and the expected return, not about the quality. A Hold that agrees with the panel about the business and differs about the entry point is a coherent position, and we would rather state it than manufacture a Buy from a quality claim.

6. Valuation — priced in or room?

At $72.91 (common market cap $25.30B, fully-exchanged equity $33.47B, net debt $16.416B):

The consensus EBITDA series is rejected (section 4), so forward EV/EBITDA is not computed. The estimate revenue series is non-monotonic beyond FY27 and should be read as "flat" rather than as precise values.

6a. What today's price assumes (the inversion)

At $72.91 — 18.0x FY26E and 16.5x FY27E with a 3.48% yield — the price embeds the following falsifiable claims:

• Revenue is essentially flat through FY30 (consensus: $9.851B → $10.048B, a 0.5% compound rate, declining in two of four forward years). The market is not paying for growth, and this is the most important thing to understand about the valuation.
• Adjusted EPS nonetheless reaches ~$4.83 by FY28 (consensus, 6 analysts) — a 19% increase from $4.05 with no revenue growth, requiring operating margin expansion, interest reduction on the $17.6B debt load, and/or share-count reduction. This is the most fragile assumption in the price: margin has compressed in each of the last four years (32.7% → 23.7%), and there were no buybacks in FY24 or FY25. Two of the three levers are currently running backwards.
• The dividend is maintained and grows. At $1.108B against $1.449B of free cash flow, the payout is 76% and coverage is 1.31x. Any material free-cash-flow decline puts the dividend — and therefore half the expected return — at risk.
• Leverage does not force action. 6.8x net-debt/EBITDA with $341M of annual free cash flow after the dividend means deleveraging takes decades on current policy. The price assumes refinancing remains available on reasonable terms.
• The market keeps paying ~16-18x forward earnings for a no-growth, 6.8x-levered franchisor. That is a defensible multiple for a stable annuity; it is not obviously a cheap one.

6b. The return bridge (why the multiple moves)

Expected return ≈ EPS growth + multiple drift + shareholder yield. This is the clearest return bridge in the batch and it is worth setting out precisely, because it produces the verdict directly.

• EPS growth: consensus +9.0% (FY27E) and +9.4% (FY28E). But note the source of it — with revenue flat, all of it is margin, interest and share count, and margin has compressed for four straight years while buybacks have been zero for two. We haircut the consensus growth in our base case, which is the principal difference between our $75 and the street's $83.25.
• Multiple drift: our base case assumes the multiple HOLDS, applying 17x to FY27E against the 16.5x the market pays today — essentially flat. We do not assume expansion (there is no growth to justify it) and we do not assume compression (a 3.5% covered dividend and a 0.53 beta provide support). The roll from 18.0x FY26E to 16.5x FY27E at a constant price is mechanical, not re-rating, and we do not count it.
• Shareholder yield: +3.48%, and it is the largest single component of the expected return. No buybacks. The dividend is 76% of free cash flow.

The bridge reads: modest earnings growth, a flat multiple, and a 3.5% dividend — netting to roughly 3% price appreciation plus 3.5% income, or about 6.5% total return. For a low-beta, high-quality annuity that is an adequate outcome. It is not a compelling enough return to justify committing new capital, and that single sentence is the verdict. Note what is absent from this bridge: there is no growth term and no re-rating term. Everything comes from the coupon and from slow margin work on a levered balance sheet.

Scenario bands (stated arithmetic, no DCF):

• Bear ~$57 — 13x FY27E adjusted EPS of $4.416. Consumer weakness pressures franchisee economics, Tim Hortons (45% of revenue) stalls, free cash flow declines and the market re-rates a 6.8x-levered no-growth franchisor. -22% from spot, and roughly the 52-week low. Note the dividend would be under pressure in this scenario, removing half the total return.
• Base ~$75 — 17x FY27E adjusted EPS of $4.416. Revenue is flat as modelled, margins stabilise, the dividend is maintained, and the multiple holds. +3% from spot, plus a 3.48% yield ≈ 6.5% total return.
• Bull ~$92 — 19x FY28E adjusted EPS of $4.832. International unit growth reaccelerates the top line above the flat consensus, operating margin recovers toward 26-28%, buybacks resume, and leverage falls below 6x. +26% from spot. This matches the $92 street high.

Base sits 3% above spot and 10% below the $83.25 street consensus. The gap is entirely our haircut to consensus EPS growth, for the specific reason that two of the three levers producing it — margin and buyback — are currently moving the wrong way.

6c. Variant perception (where we differ, what would surprise)

• Our variant perception is on the composition of the forecast EPS growth, not on the business. Consensus models EPS rising 9% a year on flat revenue. That requires margin expansion, interest reduction and buybacks. In the last two reported years, operating margin compressed from 28.8% to 23.7%, buybacks were zero, and net debt rose from $14.622B to $16.416B. All three levers ran backwards while consensus models them running forwards. We think the street is under-examining where the earnings growth is supposed to come from, and that is why our base is 10% below theirs.
• We differ from the street on the effective valuation, because of the share structure. On a fully-exchanged basis the business trades at 20.6x EBITDA, not the 15.6x the vendor's own key-metrics block reports, and free cash flow per fully-exchanged share is $3.16 rather than $4.18. Any peer comparison built on quoted market capitalisation makes this company look roughly 25% cheaper than it is. We have not seen this adjustment discussed and we think it is material.
• We agree with the panel about the business and differ about the price, and we state that explicitly rather than manufacturing disagreement. The conviction-82 claim about a ~30x return on invested capital is a statement about realised historical returns on capital deployed; it is not a statement that the equity is cheap at 16.5x forward earnings with revenue flat for five years. Both things can be true.
• Positive surprise that would force a repricing: the 2026-08-06 print delivering revenue above $2.6B (versus the $2.521B consensus) — decisively above the flat trajectory — with operating margin recovering above 26% and, most powerfully, a resumption of buybacks. Any evidence that the top line is not flat would change the entire return bridge, because the current bridge has no growth term at all.
• Negative surprise: free cash flow below $1.3B (putting the 1.31x dividend coverage under strain); Tim Hortons revenue growth turning negative; net-debt/EBITDA rising above 7.5x; operating margin below 22%; or any commentary on refinancing difficulty.

7. Moat and competitive position

The moat is genuine and it is the reason this is a Hold rather than an Avoid. A franchise system of this scale has three reinforcing advantages, all of which the knowledge base articulates better than we could:

Brand and customer relationship. The conviction-90 claim names the characteristic precisely: businesses "with strong brands that own the direct customer relationship and are hard to disintermediate." Tim Hortons in Canada is a national habit rather than a restaurant chain; Burger King and Popeyes have decades of brand equity.

The system itself. The lex_fridman claim states it well: "franchise QSR wins on systems that let a stranger operate a successful store at scale." The operating manual, supply chain, training and marketing infrastructure are the actual asset — the thing an independent operator pays a royalty to rent.

Capital efficiency. Because franchisees own the restaurants, the company deploys very little capital: capex was $265M in FY25, just 2.8% of revenue. That is what allows $1.449B of free cash flow on $9.434B of revenue and what produced the ~30x return on invested capital the knowledge base cites.

Where the moat is under pressure. Operating margin has compressed from 32.7% to 23.7% in four years — a nine-point decline that the moat did not prevent, and the single most concrete evidence that competitive or mix pressure is real. Delivery aggregators sit between the brand and the customer, directly attacking the "owns the direct customer relationship" characteristic the knowledge base prizes. And Tim Hortons at 45% of revenue is concentrated in a single mature market with limited unit-growth runway.

The vendor's peer list (Domino's, Darden, InterContinental Hotels, Li Auto, NVR, Ralph Lauren, Smurfit Westrock, Ulta, Williams-Sonoma, Yum China) is largely a market-capitalisation cohort. The genuinely comparable names are Domino's ($12.06B), Darden ($23.62B) and Yum China ($16.88B) — all materially smaller, and none directly comparable on capital structure.

8. Insider activity

The record shows eight transactions, all A-Awards (equity grants) at a price of $0, on 2026-07-07: seven tranches to Chief Corporate Officer Duncan Fulton totalling 772.3 shares, and one of 572.6 shares to Thomas Benjamin Curtis, President of Burger King US & Canada. The fractional share quantities indicate dividend-reinvestment or unit-conversion mechanics rather than discretionary transactions.

There are no open-market purchases and no open-market sales in the record.

Read: neutral, and genuinely uninformative. Unlike three other names in this batch — where chief executives and founders made substantial open-market disposals — there is simply no signal here in either direction. We note it and move on.

9. Verdict, kill-criteria and flip conditions

Hold. Restaurant Brands is a high-quality, wide-moat, low-beta franchise annuity with a 23.7% operating margin, $1.449B of free cash flow, a 3.48% dividend yield, a beta of 0.526 and a recent, high-conviction third-party endorsement in our knowledge base — including a February 2026 claim at conviction 82 that the company has compounded to roughly a 30x return on invested capital.

It is a Hold rather than a Buy for reasons of arithmetic, not quality:

1. Consensus models revenue flat from FY26 to FY30 — $9.851B to $10.048B, a 0.5% compound rate, declining in two of four forward years. There is no growth term in the return bridge.

2. The forecast EPS growth depends on three levers, two of which are currently running backwards. Operating margin compressed from 28.8% to 23.7% over two years; buybacks were zero in FY24 and FY25; net debt rose from $14.622B to $16.416B. Consensus models all three improving.

3. Leverage is 6.8x net-debt/EBITDA and the dividend consumes 76% of free cash flow, leaving $341M a year — 2.0% of the debt load — for everything else.

4. The share structure makes the business roughly 25% more expensive than it appears. On a fully-exchanged 459M share basis, enterprise value is $49.9B and EV/EBITDA is 20.6x, not the 15.6x the vendor reports.

5. The expected total return is roughly 6.5% — 3% price plus a 3.48% yield. Adequate for a defensive annuity; not a reason to commit new capital.

Pre-registered KILL criteria (any one converts Hold → Trim/Avoid):

• Free cash flow below $1.3B — dividend coverage falling toward 1.15x, at which point the payout that provides half the expected return comes into question.
• Tim Hortons revenue growth turning negative — 45.0% of revenue in a mature market is the load-bearing brand, and it determines the consolidated result more than the other three combined.
• Net-debt/EBITDA rising above 7.5x, or any indication of refinancing difficulty on the $17.6B debt load.
• Operating margin below 22% — continuing a four-year compression that has already removed nine points.
• A dividend cut or freeze, which would remove the primary reason to own the shares at all.
• The 200-DMA (~$72) decisively lost on company-specific news rather than market beta.

Pre-registered FLIP TO BUY conditions (any two):

• Revenue growth above 5% for two consecutive quarters, breaking the flat trajectory consensus assumes and restoring a growth term to the return bridge.
• Buybacks resuming at scale, taking shareholder yield above 5% combined and demonstrating that free cash flow can fund more than the dividend.
• Operating margin recovering above 26%, reversing the four-year compression.
• A price below ~$62 (roughly 14x FY27E, near the 52-week low), where the dividend yield would exceed 4.1% and the total-return arithmetic would clear a reasonable hurdle.

Where QSR fits in the Synthos Framework Portfolio. The defensive / income sleeve, as a low-beta ballast holding with a covered 3.5% yield. It is the natural counterweight to the high-beta positions elsewhere in this batch — a 0.526 beta against Affirm's 3.674 and AST SpaceMobile's 2.677 — and it is the only name here with a meaningful dividend. But ballast should be bought when it is cheap, and at 16.5x FY27E with revenue flat for five years it is merely fair. Existing holders: hold for the income, with the kill-criteria armed. No new capital. Logged as a tracked Synthos call (Hold) as of 2026-08-04 at $72.91.

Single biggest risk: 6.8x leverage against a flat revenue line. $17.579B of debt is comfortably serviced by $1.449B of free cash flow today, and franchise royalties are exactly the contractual cash flow that supports leverage. But consensus models the top line growing 0.5% a year for five years, so there is no growth to grow into the balance sheet, and with the dividend absorbing 76% of free cash flow there is only $341M a year available to reduce it. That is a slow-moving risk rather than an acute one — which is precisely why it is under-priced.

Most fragile assumption in the price: that EPS grows 9% a year on flat revenue. That requires margin expansion, interest reduction and buybacks. Over the last two reported years, operating margin compressed nine points, buybacks were zero, and net debt rose $1.8B. Consensus is modelling three tailwinds that have recently been headwinds, and there is no revenue growth to fall back on if they do not turn.

Provenance & disclosures

• Traceability: 5 KB claims match (kb_claim_count: 5, breadth 2, net conviction positive-moderate), of which 3 are substantive: invest_like_the_best 2026-02-10 (conviction 82 — "RBI has compounded to roughly a 30x multiple on invested capital and is still growing") and 2026-02-10 (conviction 90 — a principle claim naming all four of the company's brands as exemplars of simple, well-moated businesses that own the direct customer relationship); and lex_fridman 2024-02-20 (conviction 70 — a disclosed personal holding plus the franchise-systems argument). Two are tangential and weighted at close to zero: business_breakdowns 2023-05-05 (a brand-licensing principle mentioning Burger King as an example) and no_priors 2025-09-11 (concerning a Brazilian franchisee operating company, not Restaurant Brands). Note the character of the endorsement: it is a quality-and-durability lane, not a growth lane — which is entirely consistent with a Hold verdict that agrees about the business and differs about the price.
• Data as-of: fundamentals 2026-03-31 (Q1 2026; fiscal year ends December) · estimates 2026-08-04 · prices 2026-08-04 ($72.91, 50-DMA $73.88, 200-DMA $72.23, RSI 43.9, beta 0.526) · KB claims 2026-08-04. Sole market-data source: the Synthos vendor file for this ticker.
• Data REJECTED / treated as unreliable: (1) the FY25 interest-expense figure of $516M and, consequently, FY25 GAAP net income and EPS. Quarterly interest expense runs $200M, $204M, $206M, $194M, $186M, $187M, $18M, $125M, $171M — the Q3 2025 reading of $18M is a tenth of its neighbours, and the four FY25 quarters sum exactly to the reported annual figure, so the annual number inherits the anomaly. Normalised run-rate: $700-760M annually. FY24 interest expense was $804M on less debt. All valuation work uses free cash flow and the consensus adjusted-EPS series instead. (2) The gross-profit line, which changes basis mid-FY2025 — gross margin jumps from 33.8% (Q2 2025) to 48.4% (Q3 2025) and stays there, a reclassification rather than an operational event. No gross-margin trend is computed. (3) The GAAP EPS fields, which are inconsistent with their own share-count fields — FY25 net income of $776M ÷ 457M diluted shares = $1.70 against a reported diluted EPS of $2.35 (implying ~330M shares); Q3 2025 shows the same mismatch. Most likely net income includes non-controlling interests that EPS excludes. (4) The consensus EBITDA series — FY26E of $3.087B against FY25 actual of $2.424B is a 27.4% step-up on 4.4% revenue growth, requiring six points of margin expansion in a year when margin compressed. Forward EV/EBITDA is not computed.
• Validation of the series we DO use: earnings-calendar adjusted-EPS actuals for the last four quarters ($0.94, $1.03, $0.96, $0.86) sum to $3.79 TTM, consistent with the consensus FY25 estimate of $3.679 and the FY26 estimate of $4.050. That series is coherent and is the basis for every per-share figure in this dive.
• SHARE-STRUCTURE ADJUSTMENT (material, and easily missed): market capitalisation reflects ~347M common shares while diluted earnings are struck on ~459M — a 32% difference arising from exchangeable partnership units. On a fully-exchanged basis, equity value is $33.47B, enterprise value is $49.88B, and EV/EBITDA is 20.6x — not the 15.6x the vendor's key-metrics block reports. Free cash flow per fully-exchanged share is $3.16 (a 4.33% yield), not $4.18 (5.73%). The vendor's enterpriseValueTTM of $39.97B implies $14.67B of net debt against $16.416B on the FY25 balance sheet — a further discrepancy, noted and not used. Any peer comparison built on quoted market capitalisation understates this company's valuation by roughly a quarter.
• Data gaps explicitly noted: the FY25 brand disclosure covers only 72% of revenue ($6.793B of $9.434B), with the International and Advertising lines broken out in FY24 but absent from FY25; the Burger King revenue series is not continuous across a segment reclassification (FY22 $1.897B versus FY25 $1.514B is presentational, not a real decline); far-year revenue estimates are non-monotonic (declining in FY28 and FY30) and rest on 11 and 6 analysts respectively — read the direction (flat), not the values; and the profile and technical blocks report slightly different 52-week ranges.
• Data CHECKED and clean: the most recent quarter (inc_q[0], Q1 2026) is internally consistent — net income $338M ÷ 459M diluted shares = $0.736 against a reported $0.74, share counts do not duplicate the prior quarter, and interest expense of $171M is consistent with the normalised run-rate. The geographic disclosure foots exactly to $9.434B.
• Fair-value caveat: the $57 / $75 / $92 anchors are scenario multiples applied to the consensus adjusted-EPS series (not GAAP EPS, which is unreliable in this file), and are not a discounted cash-flow model. The base case assumes the multiple holds at roughly 17x and applies a haircut to consensus EPS growth, for the stated reason that two of the three levers producing that growth — operating margin and buybacks — have run backwards over the last two reported years. Expected total return of ~6.5% (3% price + 3.48% yield) is the arithmetic that produces the Hold.
• Not investment advice. Independent research, educational and informational only, never personalised.
• Version: 2026-08-04-full.