Berkshire Hathaway's Class A shares climbed roughly 2.3% after the company reported second-quarter results showing something that hadn't happened in nearly four years: Berkshire bought more stock than it sold.
The Numbers
Operating earnings rose 16% year over year, driven by strength in Berkshire's energy and railroad businesses that more than offset weaker results in its insurance underwriting division. Behind the headline number, two capital-allocation figures stood out more. First, Berkshire repurchased approximately $4.5 billion of its own shares during the quarter — a sharp jump from just $235 million in the first three months of the year. Second, and more unusual, Berkshire became a net buyer of equities for the first time in 15 consecutive quarters, purchasing nearly $20 billion more in stocks than it sold.
Why This Streak Mattered
To understand why this is notable, it helps to remember what the streak represented. For roughly four years, Berkshire had been a consistent net seller of equities, a pattern that coincided with the company's cash reserves swelling to record levels — reportedly exceeding $300 billion at points during that stretch. Warren Buffett was repeatedly asked about this cash buildup at shareholder meetings, and his answers generally pointed to one thing: he wasn't finding equities priced attractively enough, relative to their underlying value, to justify deploying that capital at scale. A company sitting on that much cash while consistently selling more than it buys is, in effect, casting a quiet vote that current valuations don't offer enough margin of safety.
What's Actually Driving the Shift
The timing of this reversal — the first full quarter under new CEO Greg Abel, who succeeded Buffett — has drawn attention, though it's worth being careful about over-attributing the shift to leadership alone. Berkshire's capital allocation process has always involved multiple investment managers and a deep bench of analysis, not a single decision-maker acting alone. What can be said with more confidence is that Abel appears willing to put the company's capital pile to work more actively than the pattern of the prior several years suggested, and that the accelerated buyback pace (nearly 20 times the prior quarter's pace) points in the same direction: a management team that currently sees more value in deploying cash than continuing to stockpile it.
Why This Isn't a Market Signal to Copy
It's tempting to read "Berkshire is buying again" as a green light for the broader market, but that interpretation deserves real skepticism. Berkshire's purchasing decisions reflect its own specific analysis of specific businesses at specific prices — a company-by-company judgment, not a top-down call on where the S&P 500 is headed next. A single quarter of net buying, after 15 quarters of net selling, is also a small sample from which to infer a durable trend reversal; it could just as easily reflect a handful of specific opportunities Berkshire's analysts found attractive rather than a broad reassessment of market-wide valuations. Treating any single investor's behavior — however successful that investor has been historically — as a signal to act on your own portfolio is a pattern worth being cautious about generally.
What It Does Tell You
Where this data point is genuinely useful isn't as a market-timing signal, but as a real-world illustration of a discipline worth understanding: patient capital deployment. Berkshire's multi-year willingness to hold cash rather than buy stocks it didn't find attractively priced — even while facing public questioning about "wasting" that cash — and its willingness to reverse course once conditions changed, is a concrete example of valuation-driven decision-making playing out over years, not days. That's a genuinely different mode of operating than reacting to daily headlines, and it's consistent with principles covered in our guide on how to build an investment research checklist — assessing value independently of short-term price action, and being willing to wait when the numbers don't line up.
