Warren Buffett Turns 96. The Word That Built This Site Turns 31.

Warren Buffett turns 96 on 30 August. It is, in one specific way, unlike any of the ninety-five birthdays before it: for the first time in six decades, he spent this year as Berkshire Hathaway's chairman rather than its chief executive. Greg Abel took over as CEO on 1 January 2026, and on 2 May, Buffett sat in the audience at the annual meeting he had led since the 1960s — an observer at his own company's biggest day, for the first time in most shareholders' living memory.

Most of what gets written on a Buffett birthday reaches for the same three stories — the Coca-Cola stake, See's Candies, the boy who wanted to be rich by 35. They're true, and they're worn smooth from repetition. This piece goes looking instead for four quieter moments from the same eighty-year career — a pinball hustle, a dying map company, a windmill factory, and a scandal nobody trusted Buffett to touch — because each one, in its own small way, is really about the same thing this site spends every day checking: what a moat actually is, and how you tell.

The pinball machines nobody remembers

In 1946, at seventeen, Buffett and a school friend, Don Danley, bought a used pinball machine for twenty-five dollars. To place it, Buffett walked into a barbershop and told the owner, Frank Erico, that he represented a company called Mr. Wilson's Coin-Operated Machine Company — a business that did not exist. Erico liked the pitch enough to take the machine on consignment. It made four dollars its first night.

Within a week they had enough to buy a second machine. Within a year, Buffett and Danley had pinball machines running in barbershops across Washington, D.C., and sold the whole operation to a returning war veteran for $1,200 — roughly fifty times what they'd put in. Decades later, over a meal with Bill Gates in an Omaha candy shop, Buffett described it as "the best business I was ever in. I peaked very early in my business career."

It's a joke, but not only a joke. A seventeen-year-old with an invented company name had already found the shape of a good business before he had a name for it: low capital, a landlord who took a cut instead of a salary, and a machine that kept earning without him standing next to it. Long before "moat" was a word he used, he was already testing for the thing the word describes — does this keep making money whether or not I'm the one running it.

The map company being erased by technology

By 1958, Sanborn Map Company had a problem that no pinball hustle prepares you for: the business itself was disappearing. Sanborn made highly detailed maps used by fire insurers to assess urban risk, block by block — a genuinely dominant, decades-old franchise. Then aerial photography and newer surveying techniques made its door-to-door mapmakers steadily less necessary, and its income had been sliding for most of the decade.

What made it interesting wasn't the maps — it was the arithmetic underneath them. Sanborn's stock traded around $45 a share, but the investment portfolio it held on its own balance sheet was worth roughly $65 a share by itself. The market was pricing the actual mapmaking business — the thing with Sanborn's name on it — at less than zero. Buffett built up roughly a quarter of the company, pushed for change as an active shareholder, and eventually helped split the business in two so the map operation and the investment portfolio could be judged, and priced, separately.

The lesson sits exactly opposite the pinball story: this was a real, long-standing moat that the world had simply moved past. No amount of history or brand recognition put the maps back in demand. Buffett's edge here wasn't finding a hidden fortress — it was refusing to let a dying business's paperwork hide a healthy pile of cash sitting right next to it.

The windmill factory in Beatrice, Nebraska

Three years later came Dempster Mill Manufacturing — a maker of windmills and farm equipment in Beatrice, Nebraska, in an era when windmills were about as unfashionable an industry as existed. Buffett accumulated stock trading at roughly a quarter of its book value: a company priced, in effect, as though it were worth far less than the machinery and inventory it actually owned.

Buying it cheap was the easy part. Turning it around was not — Dempster needed new management and hard operational decisions to actually close the gap between its stock price and its real worth, and Buffett found himself doing the unglamorous, hands-on work of an activist owner rather than a passive investor clipping a discount. It's a role he did the job of, and a role he mostly chose not to repeat once he had the choice. Later Buffett built a career on precisely the opposite temperament — buying businesses he trusted enough to leave alone. Dempster is the quieter, earlier version of him learning that lesson the hard way, at a windmill company in Nebraska nobody talks about anymore.

Buying trust when everyone else was selling it

The clearest "moat" test of the four came a couple of years later, and it wasn't at a company that made anything at all. In 1963, an employee at an American Express subsidiary was found to have certified warehouse tanks as full of salad oil that were, in large part, seawater with a thin layer of oil floating on top — the collateral behind loans made to a commodities trader who then vanished with the money. American Express was on the hook for the losses, and the stock fell more than 50%, from the mid-60s to the high-30s, inside a few months.

What American Express actually sold wasn't oil-inspection accuracy — it was trust: in its traveler's cheques, in its charge card, in its name on a piece of paper being worth what it said. That is about as intangible a moat as exists, and the market, watching a genuine scandal, priced it as though the trust itself might be gone. Buffett built a roughly 5% stake for about $20 million, without asking for a board seat or a change in management — and even suggested the company use its resources to help compensate the parties defrauded in the scandal. He was betting that customers would keep using a card they'd trusted for years, scandal or no scandal, and that a company mostly stayed a company faster than it stayed a headline. Within a decade, the stock had gone up roughly tenfold from where he bought it.

That's the sharpest version of the same question running through all four stories: is this moat real, or does it just look real because it isn't being tested this month? Sanborn's moat had genuinely eroded and no sentiment could fix it. American Express's had barely been touched at all — it just looked broken because everyone was watching the same bad headline at once. Telling those two apart, cheaply-priced business from cheaply-priced headline, is most of what this job still is.

The sentence, verbatim

All four of those moments came before the word existed for Buffett to use. "Moat" itself traces to one specific sentence, in one specific letter, about one specific company — not a manifesto, a throwaway line about an insurer's cost ratio. From Berkshire Hathaway's 1995 shareholder letter, in a passage about GEICO's underwriting expenses:

"In business, I look for economic castles protected by unbreachable 'moats.' … Thanks to Tony [Nicely] and his management team, GEICO's moat widened in 1995."

Buffett used the word in Berkshire's letters more than twenty times over the following years. By the time anyone else picked it up, it had already done its work: it gave a fuzzy intuition — this company is somehow protected — a name plain enough to argue about.

Why a name is not a measurement

That plainness is also the trap. "Moat" is such a good word that it invites people to use it as a conclusion — a company is admired, therefore it has one — instead of as a question that needs an answer. Buffett's own sentence already contains the discipline that gets lost in the retelling: he didn't say GEICO had a moat in some permanent sense. He said it widened in 1995 — a specific year, a specific number, evidence of the moat doing something, not just existing. That's the same discipline running quietly underneath Sanborn, Dempster, and American Express years before the word existed to name it.

A moat that isn't shown doing something — protecting a margin under attack, making a customer stay when leaving would be easy, letting a price hold when a competitor discounts — isn't a moat yet. It's a reputation. This site exists because that distinction is worth checking, filing by filing, rather than assuming. Every piece we publish is, in some sense, still asking Buffett's 1995 question about a company thirty years later: is the castle real, and is the moat actually holding this quarter — or is "moat" just the word being used because the company is liked?

Happy birthday

Ninety-six years, sixty-plus of them spent running one company, a pinball hustle that taught him more than the eventual thousand-page filings ever did, and a single sentence about an insurance company's cost ratio that gave a whole discipline its name. Whatever else changed this year — a new name on the CEO letter, an audience seat at his own AGM — the word is still doing the same job it did in 1995, and the same job it was quietly doing in a barbershop in 1946: separating the castles that are actually defended from the ones that just look that way from a distance.

Happy 96th, Mr. Buffett.

Educational content, not investment advice. The 1995 moat quote above is verbatim from Berkshire Hathaway's own shareholder letter (berkshirehathaway.com/letters/1995.html), verified against the primary source; the Buffett-to-Gates pinball quote is drawn from converging secondary accounts of that conversation. The Sanborn Map, Dempster Mill, and American Express accounts are drawn from published financial history, not a single primary filing, and are presented as the historical record rather than verbatim quotation. This piece makes no claim, forecast, or recommendation about Berkshire Hathaway, American Express, or any security. Moat & Margin is not a SEBI-registered Research Analyst or Investment Adviser.