Co-owners,

We've spun the wheel for the North America bucket. For twenty editions we have mostly brought you businesses hiding in plain sight. This week the ticker landed on the opposite: a company everyone has heard of, and almost nobody has looked inside. We also drew it at a historic moment. For the first full year in sixty, it is no longer run by Warren Buffett, who stepped down on the last day of 2025 with the shares near a record high and the balance sheet never stronger. He went out on top.

Before we dive in, a reminder of what the index itself has delivered over time.

Every week, we pull one company at random from the FTSE All-World with ~4,200 companies representing 90% of global stock market wealth. We share the index's long-term returns since inception in 1986 as a reminder of why we're here: the long game. New to the newsletter? Start here.

Traces back to a failing New England textile mill. Warren Buffett took control in 1965. Headquartered in Omaha, Nebraska. Listed on the New York Stock Exchange (NYSE: BRK.A / BRK.B).

Think about your day so far. If there is a car insured in the driveway, an American reading this may have insured it through GEICO. The freight that restocked the supermarket probably crossed the country on a BNSF train. The paint on the wall could be Benjamin Moore, the battery in the remote a Duracell, last weekend's treat a Dairy Queen. Six decades of patient buying turned a dying textile company into the owner of all of them, and roughly sixty more businesses.

A few key facts:

We'll come back to that pile of cash, because it tells you where the company is headed.

Most people picture a stock picker. The reality is a holding company with four engines.

The first, and the heart, is insurance. GEICO, the third-largest car insurer in America, plus a wall of reinsurance and specialist insurers. The clever part is not the premiums, it is the timing. Customers pay now and claim later, so Berkshire sits on ~$176 billion of other people's money in the meantime, the "float", and invests it. It is borrowing that has mostly cost less than nothing. The insurance itself runs hot and cold, but the float only grows, and it funds everything else.

The second engine is BNSF, one of the six great freight railroads of North America, bought outright in 2010. The third is Berkshire Hathaway Energy, the utilities and pipelines that power ~5.4 million American homes, plus grids in Britain and Canada, now racing to feed AI data centres.

The fourth engine is everything else: around fifty wholly owned businesses. See's Candies, Duracell, Benjamin Moore, Fruit of the Loom, Brooks running shoes, NetJets, Clayton Homes, Acme Brick, and Precision Castparts, which forges the parts inside the jet engines you fly behind. We met HEICO a few editions ago, which makes the cheaper replacement version of those very parts.

Here is the part most index investors miss. Almost none of those businesses are listed. You cannot buy shares in GEICO, or See's, or BNSF. They are private, and the only way to own a slice is to own Berkshire. So a share is really a ticket into a giant private portfolio, closer to a permanent private equity fund than an ordinary stock. The difference is that Berkshire buys to keep forever and then leaves the managers alone. It is the opposite of the acquisition machines we met in DSV and JBS, which buy rivals to absorb and reshape them. Berkshire buys good businesses and gets out of the way.

It is still buying. In January 2026 it paid ~$9.7 billion for OxyChem, a chemicals maker whose chlorine ends up in water treatment and construction. In July it took the homebuilder Taylor Morrison private for ~$6.8 billion. It even picked up a family-owned rodent control company, simply because the returns were good and the product is always needed.

On top of the private businesses sits a portfolio of public shares worth more than $320 billion, unusually concentrated. Five names do most of the work: Apple, American Express, Bank of America, Coca-Cola and Alphabet, a newcomer to the top five under Abel. There are also large stakes in Japan's five great trading houses. Which means if you hold a world index fund, you own Apple and Coca-Cola once directly, and a second time through Berkshire.

On 1 January 2026, Warren Buffett handed the chief executive's job to Greg Abel after sixty years and stayed on as chairman. Abel spent his first shareholder letter promising that nothing essential changes: the fortress balance sheet, the decentralised model, the patience.

The cash is the question hanging over him, and he has started to answer it. The pile, long above $370 billion, edged down in the first half of 2026 as Abel put money to work: the OxyChem and Taylor Morrison takeovers, more stock bought than sold for the first time in years, and Berkshire's first real share buybacks in over a year. After a long stretch of Buffett sitting on his hands while prices climbed, that is a quiet change of tempo, though at $365 billion the hoard is still enormous. Berkshire pays no dividend, having managed that exactly once, 10 cents a share in 1967, which Buffett jokes must have been declared while he was in the bathroom.

One of the greatest runs in market history is getting harder to sustain. At Berkshire's size, elephants are scarce and the numbers grow slowly. In 2025 the shares rose ~11%, behind the S&P 500's ~18%, and the long-held stake in Kraft Heinz has been a real disappointment. The energy arm has a thornier problem: its Oregon utility, PacifiCorp, is mired in litigation over the 2020 wildfires. A jury found it grossly negligent, a verdict an Oregon appeals court overturned in early 2026, with the state Supreme Court due to weigh in this November. The bill has already run to billions, and could climb further. A change to US energy tax credits has also made its renewables push less rewarding. None of it threatens the whole, but even this company owns hard problems.

The compounding, though, is almost hard to believe. $100 invested in Berkshire's A shares in 1964 would be worth more than $6 million by the end of 2025. The same $100 in the S&P 500, with dividends reinvested, would be worth about $46,000. The gap is so wide that, as the analyst Chris Bloomstran points out, BRK.A could fall 99% from here and its sixty-year record would still be ahead of the index. When he mentioned it to Charlie Munger, Munger just shrugged: simple compound interest.

You did not have to be that early, or that brave. The index bought Berkshire for you, and with it handed you a railroad, a car insurer, a paint brand, an ice cream chain and a hundred billion dollars of other people's businesses, all run out of one modest office in Omaha. Of everything hiding in your portfolio, this may be the one that hides the most.

If you want to dig deeper into the ins and outs of Berkshire, we cannot recommend Chris Bloomstran's 2025 letter to clients highly enough. It is long and wonderfully detailed, and we have not found a better analysis of Berkshire Hathaway anywhere.

Data and images sourced from Berkshire Hathaway's 2025 Annual Report, Berkshire Hathaway’s Q1 2026 Filings, Berkshire Hathaway’s Q2 2026 Filings, and Semper Augustus's 2025 annual letter. Market value approximate and as of writing.

We're building 90 Percent out in the open. Follow us on LinkedIn where we tease each week's company before it lands here.

Next week, we'll be looking at a company from Europe.