The 63% Figure Is Wrong — And the Real Story Behind Berkshire's Concentrated Portfolio

The recent headline claiming Greg Abel has 63% of Berkshire Hathaway's portfolio in just five stocks gets the concentration direction right but the number wrong. The Q1 2026 13F filing, reported to the SEC on May 15, shows those five names — Apple, American Express, Coca-Cola, Bank of America, and Chevron — actually account for 68% of the portfolio of the $263.1 billion public equity portfolio. The top ten holdings stretch that to roughly 91%. That's one of the most concentrated institutional portfolios in existence.

But that 67.1% figure is already stale. It was taken on March 31. Since then, Abel has deployed $10 billion into Alphabet, $6.8 billion into Taylor Morrison, and $2.6 billion into Delta Air Lines. The five-name list that the media keeps recycling is no longer the portfolio's top five. The concentration number is mechanical, not philosophical, and it's shifting fast. The real question isn't whether 63% or 67% is accurate. It's whether the concentration reflects deep conviction or simply the mechanical cleanup of a departed portfolio manager, and whether the businesses inside it are worth the risk.

About a third of that compression wasn't Abel choosing five names over thirty — it was 16 names disappearing at once. Todd Combs, one of Berkshire's two longtime portfolio managers, left for JPMorgan in late 2025. Abel liquidated the stocks Combs had managed. The exits included Visa, Mastercard, Amazon, UnitedHealth Group, Domino's Pizza, and Charter Communications. Those positions are gone.

The portfolio dropped from 42 positions to 29 securities in one quarter. When you eliminate 16 names and don't replace them with an equal number of new ones, concentration rises mathematically. That doesn't mean the remaining five don't deserve the weight — they do, as I'll get to. But it's worth separating intentional conviction from forced consolidation.

The selling streak is even longer than the Combs exits. Berkshire has been a net seller of publicly traded stocks for 14 consecutive quarters, offloading roughly $8.2 billion in Q1 alone. Apple's position has been reduced from a peak of 915 million shares to 228 million. Bank of America has been cut by more than 75% since 2024. This isn't panic; it's discipline. But a fourteen-quarter selling run with 29 names left to show for it means every dollar of remaining capital carries more weight.

The Five Names: An Asset Check

Stripping away the concentration headline, the five names themselves pass the hard-to-replace test.

Apple holds the largest position at $57.8 billion, or 22% of the portfolio. The company generates $136.7 billion in trailing free cash flow with a 71% return on invested capital. Revenue grew 14% year-over-year, and its 48% gross margin and 33% operating margin reflect an ecosystem that customers don't leave. Apple trades at 35 times trailing earnings and 40 times forward earnings. That's expensive on a multiple basis. But Berkshire isn't pricing this position against the S&P 500; it's pricing it against the $136.7 billion in annual free cash flow that flows back to the balance sheet. Even at that price, the dividend yield is negligible at 0.34%, so the income case doesn't apply here. This is a cash-flow engine, not an income play.

American Express at $45.9 billion (17.4%) is the portfolio's closed-loop network play. Its proprietary card architecture — where it controls both the issuer and acquirer sides of a transaction — is a moat that can't be rebuilt by a new entrant. Revenue growth of 11%, an ROE of 34%, and a 19% free cash flow margin show the engine is still compounding. The stock trades at 20 times trailing earnings, a reasonable premium for a business with this kind of structural advantage. The 1.05% dividend yield is modest but growing, with a 20.5% payout ratio leaving ample room to reinvest.

Coca-Cola at $30.4 billion (11.6%) is the dividend anchor. A 2.4% yield, 24 consecutive years of dividend growth, and 62% gross margins from a brand portfolio that commands shelf space in every country on earth. The stock trades at 26 times earnings — above its historical average, but justified by pricing power and 30% operating margins. This is the kind of holding that serves a retirement portfolio: predictable cash flows, durable demand, and a dividend that's been growing for more than two decades.

Bank of America at $25.0 billion (9.5%) is the cheapest name in the top five at 14 times trailing earnings and 1.5 times book value. The bank earns 1.8% on its dividend yield, and its 11.6% ROE is in line with large-cap banking peers. BAC has been the most heavily reduced position, so this is less a conviction buy and more a residual holding that hasn't yet reached its sell-off threshold.

Chevron at $17.5 billion (6.6%) is the only top-five name that Abel is actively trimming, having sold 46 million shares in Q1 — a 35% reduction. The 3.6% dividend yield is attractive, and the company generates solid cash flow, but the 18x trailing P/E with a 31x forward multiple suggests the market has priced in oil-price optimism. Abel appears to believe the energy transition and valuation compression justify profit-taking.

What the Filing Missed: The Alphabet Bet

The Q1 13F doesn't capture what may be the most significant shift in Berkshire's public equity posture. In June, after the filing deadline, Abel approved a $10 billion private placement in Alphabet — $5 billion in Class A shares at $351.81 and $5 billion in Class C shares at $348.20, both at a discount to market price. That purchase more than tripled Berkshire's Alphabet stake from roughly 18 million shares to nearly 58 million.

Abel described the move as a rapid signoff following an inquiry from Goldman Sachs. Buffett, by contrast, had historically passed on Google because he couldn't see far enough into the technology business to estimate its intrinsic value decades out. Abel's willingness to deploy that kind of capital into AI infrastructure represents a genuine philosophical shift. Alphabet is expected to become Berkshire's third- or fourth-largest holding, displacing Chevron or Coca-Cola from the top five.

This matters for the concentration question because it means the five-name thesis was already breaking down before the market noticed. Once Alphabet enters the top five, the concentration metric will look different again.

The Mechanics Behind the Concentration

The Cash Pile: $397 Billion of Optionality

Berkshire ended Q1 with $397 billion in cash and short-term Treasuries. Adjusted for railroad cash and T-bill payables, that's roughly $380 billion. Even after deploying $16.8 billion on Alphabet and Taylor Morrison, the pile remains enormous.

The cash isn't idle baggage. It's optionality against the concentrated equity portfolio. When 91% of your public stocks sit in ten names, you need a balance-sheet buffer that can absorb a shock without forcing fire sales. The Treasuries are earning 4-5%, which is a reasonable holding pattern, though the cash drag has cost Berkshire relative to the S&P 500 over the past decade. That under-earning is the tax on discipline.

Buybacks are the other valve. In Q1, Berkshire repurchased $235 million in shares at roughly 1.4 times book value. The pace was deliberately slow. Management only buys back when it believes the shares trade below intrinsic value. At 1.43x book as of mid-May and 15.6 times trailing earnings, BRK is cheaper than the S&P 500 (which trades around 25x), but the margin of safety isn't wide enough to trigger aggressive repurchases yet.

The Valuation Gap

Here's where the concentration risk meets the value thesis. Berkshire's public equity portfolio, despite holding Apple at a rich multiple, trades at a discount to the broader market as a whole. The conglomerate's 15.6x trailing P/E is well below the S&P 500's 25x. Its 1.5x book value is below the 2.5x+ that large-cap financials like JPMorgan and Goldman Sachs command. That multiple gap exists because the market penalizes Berkshire for size, cash drag, and the uncertainty of what happens after Buffett.

Those are real concerns. The $380 billion cash pile represents a drag on returns if it can't be deployed productively. The concentrated portfolio means that a hit to any of the top five names flows directly to Berkshire's book value. But the portfolio's underlying businesses — measured by aggregate ROE and free cash flow yield — outperform the S&P 500. The discount is a gap, not a trap.

Abel's approach reinforces the value frame. He's not trying to chase growth; he's buying entire businesses (Taylor Morrison), taking private placements at a discount (Alphabet), and trimming positions that have run ahead of their fundamentals (Chevron). The 14-quarter selling streak reflects a refusal to buy expensive stocks, even when idle cash is expensive.

The Portfolio Role

For a retirement portfolio, Berkshire itself remains a compounding engine rather than an income play — it pays no dividend. But the concentration of its public equity holdings in dividend-generating, cash-flow-rich businesses (Coca-Cola at 2.4%, Chevron at 3.6%, Bank of America at 1.8%, American Express at 1.0%) provides an indirect income layer. Berkshire's operating subsidiaries generate roughly $45.5 billion in operating cash flow, which supports the capital allocation machine.

The gate that matters going forward is whether Abel can deploy the cash pile at valuations that preserve the 1.5x book entry point. If he buys back more aggressively or deploys cash into whole-business acquisitions at attractive multiples, the concentration risk in the public portfolio becomes less material — the operating and wholly-owned businesses offset it. If the cash stays parked for years, the drag grows and the multiple gap narrows.

The five-name concentration is real, but it's a snapshot, not a strategy. The real strategy is simpler: buy durable businesses at fair prices, hold them through cycles, and don't rush to deploy capital when nothing meets the bar. Abel inherited that framework. The Alphabet bet and Taylor Morrison acquisition show he's willing to bend it when the numbers work. The concentration number in the headlines is already outdated — and it probably always was.

Rating: Buy. Berkshire trades at a meaningful discount to its component businesses and the broader market, with a balance sheet that provides optionality against concentration risk. The primary invalidation condition is sustained cash drag — if the $380 billion pile earns only Treasury rates for three years or more without meaningful deployment, the valuation gap closes. Until then, the multiple discount represents a compounding opportunity, not a puzzle.