Warren Buffett’s Portfolio Has Half Its Stock Money in Just 3 Names. Here Is What They Are.

Warren Buffett’s Berkshire Hathaway (NYSE:BRK.B) filed its latest 13F for the quarter ended June 30, 2026, disclosed on August 14. The most striking feature of the disclosure was three names carry roughly half of the entire reported equity book by weight.

Those three are Apple (NASDAQ:AAPL | AAPL Price Prediction), American Express (NYSE:AXP), and Coca-Cola (NYSE:KO). All three are long-standing Buffett anchors (we sorted Berkshire’s holdings by valuation and pulled the seven cheapest dividend payers into a free report here: 7 Warren Buffett Stocks to Buy Now). All three are consumer-facing franchises with pricing power. And all three sit inside a disclosed portfolio that gets more concentrated the closer you look.

One critical framing point before the numbers: a 13F covers US-listed long equity only. It excludes Berkshire’s cash and Treasury holdings, its wholly owned operating businesses like BNSF, GEICO, and Berkshire Hathaway Energy, and any non-US-listed exposure. So these three names are roughly half of the disclosed stock portfolio, not half of Berkshire’s money, net worth, or fortune. Berkshire is a holding company, not a fund. Positions are shown as of quarter end and may have shifted since.

Apple: The Anchor Position

Berkshire disclosed 227,917,808 shares of Apple at quarter end, representing 22.04% of the disclosed portfolio. Apple designs the iPhone, Mac, iPad, Wearables, and the fast-growing Services business that layers a high-margin subscription annuity on top of the installed base.

Buffett has publicly framed Apple less as a technology bet and more as a consumer franchise with switching costs, and the fundamentals support the read. Apple trades at a P/E of 42 with a ROE of 171.4% and ROIC of 53.3%. The June quarter delivered revenue of $109.42 billion, up 16.4% year over year, with EPS of $2.02 versus a $1.89 estimate, and Tim Cook called it the company’s “strongest June quarter ever, with double-digit revenue growth across iPhone, Mac and Services, and in every geographic segment.” Apple bought back $62.09 billion of stock in the first nine months of FY26, which mechanically lifts Berkshire’s ownership stake without a single share being traded.

Our 24/7 Wall St. model sees upside of 15.31% at high confidence (0.9), with a base one-year target of $368.95 from a current $319.97. Wall Street’s consensus target is more measured at $323.86, with 6 strong buy, 19 buy, 14 hold, 3 sell, and 2 strong sell ratings. Our model is meaningfully more constructive than the Street here, driven by sector momentum and earnings acceleration; the analyst community is closer to fair value. Predictions are as of publication; the 13F snapshot is as of quarter end.

American Express: The Longest-Running Bet

Berkshire’s disclosed American Express stake stood at 151,610,700 shares, or 17.14% of the disclosed portfolio. American Express operates a closed-loop payments network and card business skewed to premium, high-spend customers.

This is the oldest of Buffett’s blue-chip anchors, and it keeps compounding. Q2 revenue reached $19.64 billion with EPS of $4.53 versus $4.40 expected, and CEO Stephen Squeri highlighted “another excellent quarter, with 10 percent revenue growth, EPS of $4.53, and Card Member spending growth of 9 percent, the highest rate we’ve seen in three years on an FX-adjusted basis.” Management raised full-year revenue growth guidance to 10% and maintained EPS guidance of $17.30 to $17.90. The quarterly dividend has climbed from $0.60 in 2023 to $0.95 in 2026, and diluted share count is running down.

Our model projects upside of 9.05% at high confidence (0.9), with a base target of $355.67 from $326.16. Interestingly, the Street is more optimistic than we are: consensus target is $375.96, with 5 strong buy, 10 buy, 14 hold, 1 sell, and 0 strong sell ratings. The disagreement is worth noting given AXP has fallen 11.12% year to date against a 33.94% run in Apple.

Coca-Cola: The Dividend Compounder

The Coca-Cola position was disclosed at 400,000,000 shares, or 10.86% of the disclosed portfolio. That share count is a well-known constant of the Berkshire book, unchanged for many years, and it means Buffett’s original 1988 cost basis produces an enormous yield on cost as the dividend keeps climbing, from $0.16 per quarter in 1999 to $0.53 per quarter in 2026.

The business is executing. Q2 delivered revenue of $13.38 billion, up 6.74% year over year, EPS of $0.97 versus $0.93 expected, and global unit case volume growth of 5%. New CEO Henrique Braun described “a strong first half of the year” and said the company was “well positioned to deliver on our RAISED 2026 guidance”, which now calls for organic revenue growth of about 5% and comparable currency-neutral EPS growth of 7% to 8%. Trademark Coca-Cola volume grew 5% during the quarter, described as its strongest volume growth in 17 years excluding COVID recovery, helped by the FIFA World Cup activation across more than 180 markets.

Our model flags upside of 10.07% at high confidence (0.9), with a base target of $96.94 from $88.07. Bull and bear cases run to $101.34 and $85.15. Consensus is closely aligned at $94.70, with 7 strong buy, 12 buy, 4 hold, 0 sell, and 1 strong sell ratings. KO trades at a P/E of 29 with a 2.32% dividend yield. Shares are up 27.67% year to date.

What the Top Three Says About Buffett’s Approach

Concentration is the story. Three tickers carrying 22.04%, 17.14%, and 10.86% of a disclosed equity book is the opposite of diversification for its own sake. The sector tilt is unmistakable: one consumer technology franchise, one premium payments network, and one global beverage brand. All three sell products with brand pricing power that survives inflation, recessions, and management changes. None of them are speculative; all three throw off cash and buy back stock. On holding period, this is the essence of the Buffett approach: the KO share count has not changed in decades, AXP has been core since the 1990s, and even Apple, added in 2016, is treated like a legacy holding rather than a trade. The absence of any hot theme, no AI pure-play, no crypto exposure, no highly cyclical bet, is itself the tell.

What to Watch Next

Studying this book, the takeaway for a reader at or near retirement centers on the discipline behind them: fewer tickers to copy, more focus on process: fewer names, higher-quality businesses, and a willingness to sit still. The next 13F, disclosed roughly 45 days after the September quarter closes, will show whether these anchors moved at all, and the next earnings reports from all three names are the near-term catalysts. 13F disclosures are backward looking. Price predictions are projections, not guarantees. And none of this is investment advice.

Contact [email protected] for any questions or corrections.