What it does · how it got here · where it operates
Let me be plain about what this company does, because it is both very simple and very uncomfortable. British American Tobacco sells cigarettes and the nicotine products meant to replace them. It buys tobacco leaf for pennies, rolls it into a branded product that a addicted customer will buy every single day, and sells it for a price the customer barely questions. It is one of the most profitable business models human commerce has ever devised — and it kills its best customers. Both of those things are true, and an honest owner must hold them in the same hand.
Founded in 1902 and based in London, BAT is one of the two or three largest tobacco companies on earth outside China. It owns Dunhill, Kent, Lucky Strike, Rothmans and Pall Mall worldwide, and — since paying £42 billion for the rest of Reynolds American in 2017 — Newport, Camel and Natural American Spirit in the United States, now its single largest market. In the year to December 2025 it turned over £25.6 billion and earned about £7.8 billion in profit, at operating margins near 40%. The London-listed shares trade as BATS; most of the dividend-hungry world owns it through the BTI American deppositary receipt, at about $60.56 and roughly a $98 billion market value.
"I'll tell you why I like the cigarette business. It costs a penny to make. Sell it for a dollar. It's addictive. And there's fantastic brand loyalty." — a tobacco executive's boast, which Buffett quoted precisely because it is true
That sentence explains the numbers and the discomfort at once. The economics are close to perfect: negligible capital, a customer who returns daily by compulsion, and pricing power that lets the company raise prices every year to more than offset the volumes it steadily loses. And yet Buffett, who understood those economics better than anyone, chose not to own it — and I will come back to why, because it is the whole point of the letter at the end.
| Founded | 1902 · London, United Kingdom |
| Sector / Industry | Consumer Defensive · Tobacco & Nicotine |
| Listings | LSE: BATS (£) · NYSE: BTI (ADR, $) |
| Key brands | Dunhill, Kent, Lucky Strike, Newport, Camel · Vuse, glo, Velo |
| Revenue (FY2025) | £25.6 B |
| Market capitalisation | ~$98 B (ADR) |
Why the economics are extraordinary — and whether you'd understand it
I can explain this business to a ten-year-old, though I would not enjoy doing so. Here is the whole machine, end to end:
The honesty this business demands runs the other way from Coca-Cola's. Coke is easy to understand and its future is bright; BAT is easy to understand and its future is shrinking. Cigarette volumes in the rich world fall a few percent every year, and that decline is permanent — it will not reverse. The entire investment case rests on a single question: can price increases and new nicotine products offset falling volumes for long enough to keep the cash fountain running while you are paid an ~8% cash return (dividend plus buyback) to wait?
Combustibles cash cow · New Categories · geographic mix
Almost all of the profit still comes from the old, doomed, wonderful product — and a growing sliver from the things meant to replace it.
This is the rare pond that is draining. Humanity is, slowly and rightly, smoking less. BAT's task is not to grow the pond but to take a larger, richer share of it while steering its customers toward the smokeless products it also sells — Vuse (the world's leading vapour brand), glo (heated tobacco) and Velo (nicotine pouches). The company's stated ambition is to earn half its revenue from non-combustibles by 2035. Whether it gets there before the cigarette business fades is the entire game.
A wide fortress built on a melting glacier
BAT has a genuine, wide moat — and it sits on ground that is slowly melting. Both halves matter.
The brand & the habit. A smoker does not shop on price between Newport and a no-name; the brand and the addiction together are one of the stickiest franchises in commerce. This is the intangible-asset moat in its purest, and least comfortable, form.
The advertising ban — the strangest moat of all. Because governments forbid tobacco advertising, no new brand can ever be built. That freezes the market in favour of the incumbents forever. Regulation, meant to hurt the industry, has handed its survivors a permanent barrier to entry. It is a moat carved by the state.
Distribution & scale. Getting a product into millions of corner shops across 180 markets, under the world's tightest regulatory regime, is a barrier a newcomer could not cross at any price.
How durable? A melting moat. Here is the honesty. The fortress walls are thick, but the ground beneath them is receding a few percent a year and will never stop. This is not Coca-Cola, whose moat widens into new drinks. It is closer to the newspapers Buffett once called "impregnable" — a franchise that was real, and lucrative, and quietly dying. The moat is wide. It is also, unmistakably, narrowing. You are being paid a fat dividend to stand inside it while it does.
"A Smokeless World" · Vuse, glo, Velo · can the second act arrive in time?
Every tobacco company now tells the same story — that it is becoming something cleaner — and an owner must weigh it with a cold eye. BAT's version is called "Building A Smokeless World," and unlike a slogan it has real revenue behind it. Three products carry the hope:
The encouraging news for an owner is that these are no longer a cash drain — the New Categories reached profitability, which is the moment a "transition story" stops being an expense and starts being a business. The sober news is the arithmetic of the race: combustibles still throw off the overwhelming majority of profit, and they are declining now, while the smokeless business — though growing smartly — is not yet remotely large enough to carry the company. The whole question is one of timing: can the new act grow up before the old one fades away? Nobody, including management, honestly knows. That uncertainty is why the stock is cheap.
The Reynolds mistake · the impairment · deleveraging, buybacks & the ITC sale
I judge managers above all on what they do with the cash — and BAT's record is a genuinely mixed one, worth telling honestly.
In 2017 BAT paid £42 billion for the 58% of Reynolds American it did not already own — at the very top of the US cigarette market. It made America BAT's biggest market and loaded the balance sheet with debt and goodwill. Six years later, in 2023, the company wrote down those US brands by roughly £25 billion and swung to a £14.4 billion reported loss. In Buffett's language, a good part of that goodwill was "capitalised adrenaline" — the price of an over-eager acquisition, later confessed. It is the single most important fact about this management's capital allocation, and it argues for humility, not swagger.
The verdict on management is competent, chastened, and rational — not visionary. They overpaid once, badly, and have spent years cleaning it up: honest write-downs, falling debt, and buybacks that are genuinely smart when the shares trade this cheaply. That is the right playbook for the hand they hold. I would simply never forget the £25 billion lesson when they next feel the urge to do something big.
Financials in GBP · valuation multiples on the USD ADR · linked to the data pages
| Metric | Value | Read |
|---|---|---|
| Revenue (FY2020 → FY2025) | £25.8B → £25.6B | ◆ essentially flat |
| Reported EPS swing (2023 → 2025) | −£6.49 → £3.51 | ◆ 2023 = the impairment |
| Operating margin | ~40% | ▲ superb |
| Net profit margin | ~30% | ▲ rich |
| Return on equity (ROE) | 16.3% | ▲ solid |
| Return on invested capital (ROIC) | 8.1% | ◆ dragged by £87B goodwill |
| Net debt / EBITDA | 2.5x | ◆ high but falling |
| Interest coverage | 5.9x | ▲ comfortable |
| Free cash flow (typical yr) | ~£8B | ▲ gushes cash |
| Capex / revenue | ~2% | ▲ wonderfully asset-light |
| Goodwill & intangibles / assets | ~80% | ◆ almost all Reynolds |
Two truths sit side by side here. First, the cash generation is magnificent: ~40% operating margins, ~£8–10 billion of operating cash flow on almost no capital, funding a dividend of ~£5 billion a year with room to spare. Second, the accounting returns look mediocre — an 8% ROIC — but that is an illusion created by the £87 billion of goodwill and brand value piled up buying Reynolds. Strip out that acquisition tombstone and the underlying cash return on the tangible business is enormous; this is the textbook case Buffett describes, where accounting goodwill masks a fabulous cash economics. The real caution flags are elsewhere: flat-to-declining revenue, and a balance sheet still carrying £31 billion of net debt and negative tangible equity. Strong cash, shrinking top line, meaningful leverage — read all three together.
The widest gap between bulls and bears on the whole board
| Yardstick | Today | Read |
|---|---|---|
| P/E — reported earnings | ~13x | priced as a declining franchise |
| Dividend yield | ~5.3% | well-covered (~68% payout) |
| Total cash return (div + buyback) | ~8% | you're paid handsomely to wait |
| Free cash flow yield | ~4.5% | converts cleanly |
| EV / EBITDA | 10.4x | cheap for the cash |
| Price / book | 2.1x | book is ~all intangible — ignore it |
I have not seen a wider disagreement anywhere on this board. A mechanical discounted-cash-flow model, fed BAT's enormous cash flows, spits out ~$184 — nearly triple the price — because it quietly assumes the cash lasts. A lone analyst covering the ADR sits at $40, assuming it doesn't. The market splits the difference at ~13× earnings and a 5.3% yield — the multiple Buffett's own framework assigns to a narrowing franchise. All three can't be right, and the honest answer is that the value depends entirely on how fast cigarettes decline and how well the smokeless pivot lands — which no model can know. What I can say is that the deep-value bargain of 2023–24 is largely gone: the stock has roughly doubled off its lows, and the ~8% yield has compressed to ~5.3%. What was a screaming cigar butt is now a fairly-valued, high-yield melting-ice-cube.
Decline · litigation · regulation · the ESG stain — with sources
This is the most risk-laden name we have examined, and it deserves a clear eye. The secular decline of cigarettes is not a risk but a certainty; the only question is pace. Litigation is real and large — BAT's Canadian arm agreed to a multi-billion industry settlement that drove heavy 2024–25 charges and cash outflows. A US menthol ban would strike directly at Newport, a crown jewel of the Reynolds deal. And beyond the numbers sits the ESG stain: a large and growing pool of investors and funds simply will not — or cannot — own a tobacco company, which caps the buyer base and is part of why the stock stays cheap. None of these is likely to be fatal to the cash flows this decade. All of them are reasons the market pays only ~13× for them.
Dear shareholder — this is the hardest letter I have written to you, not because the business is complicated, but because it is simple in a way that troubles the conscience. So let me separate, as I always try to, the two questions that must never be blurred: is this a wonderful business? and is this a business I want to own?
On the first question, the answer is almost embarrassingly clear. British American Tobacco has some of the finest economics I have ever studied. It takes a cheap leaf, wraps it in a beloved old brand, and sells it to a customer who is compelled to come back tomorrow — at a price he does not argue with, and which the company lifts a little every year. It costs almost nothing in capital to run, so nearly every pound it earns can be mailed straight to owners. An advertising ban that was meant to punish the industry has instead frozen out all new competition forever. On the cold arithmetic of return on capital, this is a franchise most companies would kill for.
And yet I must tell you the second truth, because it is the more important one. The ground under this fortress is melting. Cigarette volumes fall a little every year and will never rise again — this is not a cycle to wait out, it is a tide going out for good. Management's answer, the smokeless pivot into Vuse, glo and Velo, is real and finally profitable, but it is nowhere near large enough yet to carry the company, and the race between the fading old business and the growing new one is one that nobody — not you, not I, not the chief executive — can honestly handicap. Add to that a balance sheet still groaning under thirty billion pounds of debt from an acquisition they plainly overpaid for, and you have a wonderful cash machine bolted to a shrinking foundation.
Then there is the part no spreadsheet captures. Many years ago, when the great tobacco businesses were on offer at a song, I said something I have never taken back: I understood exactly why the economics were superb, and I chose not to own them anyway. There are things a person can do and things a person would rather not, and for me this fell on the wrong side of that line. I tell you this not to lecture — you are a grown investor and this is your decision, not mine — but because the ESG stain is also a hard financial fact: a whole universe of buyers is barred from ever owning this stock, and that is part of why it stays cheap and may always be.
So where does that leave us on price? The screaming bargain of two years ago is gone — the shares have roughly doubled, and the yield that once neared nine per cent has settled to a still-generous five-point-three. At about thirteen times earnings you are being paid an ~8% total cash return to stand inside a narrowing moat — a fair deal, not a steal. For the income investor who has made his peace with what this company sells and accepts that its best days of decline-defiance are priced in, it is a reasonable, well-covered, eyes-open holding. I would want it cheaper — back in the low fifties, where the yield tops six per cent — before I called it a bargain again. And I would size it as what it is: a high-yield melting ice cube, not a compounder to leave to my grandchildren. That, at least, is a distinction Coca-Cola never forced me to make.