Warren Buffett's Successor, Greg Abel, Has 69% of Berkshire Hathaway's Portfolio Invested in Just 5 Stocks. Here's My Top Pick for October.

Alphabet rose from 2% of Berkshire Hathaway’s portfolio at year-end 2025 to 9.4% after Q2 2026, becoming its fourth-largest holding; the article expects CEO Greg Abel may add more, but this is an author prediction. Alphabet’s cloud backlog reached $514 billion, up 50% from the prior quarter, with just over 50% expected to be recognized as revenue over the next 24 months. Its forward P/E is 22, while fiscal 2026 capex guidance increased to $195 billion–$205 billion from $180 billion–$190 billion, and management expects capex to increase significantly in 2027, pressuring free cash flow.

Analysis

The investment case turns on whether Alphabet can convert contracted Cloud demand into returns that justify a materially larger infrastructure base—not on Berkshire’s endorsement. Backlog is not near-term free cash flow: verify cancellation/renewal terms, capacity timing, and realized Cloud margins. With capex rising and depreciation following, revenue conversion could coexist with weak FCF and pressure on valuation if returns lag. The reported earnings multiple is especially poor evidence of cheapness given the unusually large investment-related gain; assess normalized operating earnings instead.

Near term, Berkshire’s next 13F may support sentiment, but it is a lagged snapshot and does not establish current buying or intent. Over 1–3 months, Cloud growth, margin commentary, and capex guidance are the meaningful catalysts. Over 6–18 months, successful capacity deployment could strengthen Alphabet against Amazon and other cloud providers; the opposite outcome is underused infrastructure, higher depreciation, and persistently lower cash generation. AI infrastructure suppliers may benefit from spend, but the article does not identify spend allocation or supplier economics.

Contrarian angle: the market may be treating capex as a temporary cost while underestimating the risk that competition forces reinvestment without durable pricing power. Conversely, a rising backlog can be a leading indicator that capacity—not demand—is the binding constraint. Berkshire concentration is a sentiment signal, not an independent valuation floor.

Key Decisions for Investors

• Watch GOOG rather than chase a Berkshire-following headline: consider staged long exposure on weakness, contingent on Cloud revenue growth and margin performance supporting backlog conversion.
• Do not rely on the cited P/E as a valuation anchor. Before sizing, verify normalized earnings excluding investment gains, current consensus estimates, and the implied return on incremental capex.
• For a relative-value expression, monitor GOOG versus AMZN: favor GOOG only if its Cloud growth and margins improve relative to its rising infrastructure burden; otherwise avoid assuming both hyperscalers’ capex will earn similar returns.
• Reassess or exit the long thesis if Cloud growth decelerates while capex guidance rises, or if management indicates weaker backlog conversion or sustained FCF deterioration. Treat a 13F increase as confirmation of sentiment, not proof of fundamentals.