Alphabet Has Entered The Chat
And other thoughts on Berkshire Hathaway's Q3 2025 investment activity...
A few thoughts on Berkshire Hathaway’s investment activity in Q3 2025 — with special attention lavished on a notable newcomer…
Each quarter, all investment managers with more than $100 million under their control must file Form 13F with the SEC to disclose any changes to their portfolios. For us Berkshire Hathaway watchers, this filing offers a priceless peek behind the curtain at the conglomerate’s investment activity during that period.
But any discussion of Berkshire’s 13F comes with a significant caveat. We do not always know — with certainty — who is responsible for a given trade. Media reports have a habit of attributing everything to Warren Buffett, but in recent years much of the buying and selling came courtesy of lieutenants Todd Combs and Ted Weschler. Some brave souls attempt to divine the truth based on position size or industry, but it’s all ultimately just (educated) guesswork.
That little preamble seems especially appropriate this time out because someone at Berkshire opened a sizable new position in Alphabet (formerly Google). This investment was valued at approximately $4.3 billion as of September 30 — and then crossed $5 billion in after-hours trading on Friday afternoon. A stake that size might tip the scales in Buffett’s direction, but is also not so large as to rule out Todd or Ted.
Personally, I’m rooting for Alphabet to be a Buffett buy — even if I don’t feel wholly confident in that wish. If the Oracle did indeed pull the trigger on Google now, just months before his retirement, it would make for a fitting bookend to one of his earlier investment-related regrets. At both the 2017 and 2019 annual shareholders meetings, both Buffett and Munger lamented missing out on money-spinning Google. “I feel like a horse’s ass for not identifying Google better,” said Charlie. The GOATs kicked themselves for not recognizing the enduring strength of Google’s search business.
“We did have some insights on that,” admitted Buffett, “because we were using them at GEICO and we were seeing the results produced. We saw that we were paying $10 a click — or whatever it might have been — for something that had a marginal cost to them of exactly zero.” Charlie summed it up as only he could: “We could see in our own operations how well Google advertising was working — and we just sat there sucking our thumbs.”
Alphabet (née Google) probably doesn’t need much of an introduction, but here’s a quick tour of its sprawling business empire. Basically, Google makes the lion’s share of its money through advertisements on its ubiquitous search engine — but also owns a stable of household names and brands deeply woven into our daily lives. Oh, like the Android mobile operating system, Chrome web browser, a little video platform called YouTube, Google Maps, Gmail, Pixel smartphones, Nest smart home devices, Fitbit wearables, the Google Play app store, Gemini AI, and many others.
Although this new stake in Alphabet caught many of us off-guard, the company ticks a lot of boxes on Berkshire’s investment wish list…
First of all, Berkshire can pour a lot of capital into Alphabet if it so chooses. At the conglomerate’s current size, finding investments large enough to move the needle has proven challenging. That will not be a problem here, with Alphabet the fourth-largest company in the world ranked by market cap ($3.3 trillion).
Of course, I said the same thing about TSMC in 2022 and we all know how that turned out. That’s one elephant I’d love to have back.
Alphabet just posted its first-ever $100 billion revenue quarter ($102.3 billion) — with double-digit growth across every major segment. Google Services (search, YouTube, subscriptions, devices, etc.) generate ~85% of sales, while Google Cloud chips in with the other ~15%. Other Bets, a third bucket home to moonshots like Waymo, contributes negligible revenue and remains mired in red ink.
Operating income rose 9.5% to $31.2 billion in the third quarter, though that includes a $3.5 billion fine from the European Commission. Without that (hopefully) one-time charge, operating income would have increased 22%.
Fears that generative AI chatbots would deliver a knockout punch to Google Search have, so far, proven unfounded. In fact, both overall query volume and commercial queries kept growing — fueled by the very AI Overviews and AI Mode features that Google rolled out in response to its new rivals. Paid clicks and cost-per-click are also each up year over year. The chatbots may have nibbled Google’s market share down below 90% for the first time in years, but the incumbent remains utterly dominant in the continually-growing mobile arena.
In short, Google’s search traffic now exceeds pre-ChatGPT levels with barely a dent in its overall market share. And, as the company integrates its Gemini AI deeper and deeper into the search experience, it seems to be creating an increasingly appealing product.
Alphabet’s formidable war-chest of cash, equivalents, and marketable securities — $98.5 billion at last count — gives it ample firepower to return capital through aggressive buybacks and a growing dividend. Since 2020, the company has shrunk its shares outstanding by ~9% — and still has another $74.8 billion authorized for similar use in the future. Plus, Alphabet launched its first-ever dividend last year and modestly boosted the payout for 2025.
Few brands have infiltrated the English language quite like Google — now a bonafide verb, according to Merriam-Webster, for “search”. This linguistic flex signals not only market leadership, but an unshakeable share of mind among the public. It also brings to mind a comment once attributed to Ted Weschler: “We like verbs in Omaha.” It’s the kind of cultural moat that Berkshire covets.
Google’s co-founders, Larry Page and Sergey Brin, stepped away from daily management in 2019, but remain on the board of directors and —crucially — still hold over 50% of voting power courtesy of their Class B shares. On that subject, Berkshire notably opted for Class A shares of Alphabet (with one vote a piece) as opposed to the typically cheaper, non-voting Class C shares.
And, finally, a little idle speculation: Depending on when exactly these shares were purchased, Berkshire could already be sitting on a big gain. If bought in the first three weeks of July, when Alphabet shares traded in the $180-ish range, it would now be up approximately 50% in the space of just four-plus months.
Beyond the headline-grabbing Alphabet investment, Berkshire also added another 4.3 million shares of Chubb in the third quarter. But, besides that, the other buys can only be considered small-ish (practically irrelevant) nibbles on recent faves like Lamar Advertising, Domino’s Pizza, Sirius XM, and Lennar.
Apple, though, remains squarely on the chopping block. Buffett unloaded another 41.8 million shares from the position, making this the second consecutive quarter of Apple sales. The iPhone maker still sits atop the massive Berkshire portfolio, but the gap over second-place American Express continues to narrow. I have made my peace with Buffett’s decision to pare Apple way back, but I’m more of the Munger mindset that selling any Apple stock usually turns out to be a mistake. (This might also just be the pointless rambling of an inveterate coffee-canner who hates to sell anything.)
Berkshire also sold varying amounts of Bank of America, Davita, Verisign, Nucor, and D.R. Horton in the third quarter. The on-again, off-again romance with D.R. Horton is one of the odder Berkshire-related subplots in recent times. Since 2023, someone at the conglomerate opened a position in the homebuilder, sold all of it, opened a new position, and then fully exited that one as well. In each case, it only took three quarters from start to finish. Why? Your guess is as good as mine.
I think this was Todd Combs. I mean, the guy understands tech companies pretty well.
But does it matter who pulled the trigger?