Amphenol: A Great Business Wearing an Awkward Price Tag
The business and its moat
Whatever smartphones and AI look like in ten years, the probability that electricity and signals stop travelling through physical wire is close to zero. Amphenol makes the connectors — the knots where one wire ties to another.
Once designed into a customer’s product, recertification costs form a wall: over product lifespans of 5 to 20 years, switching rarely happens. A five-year average ROE of 26.4%, with a standard deviation of just 3.4 percentage points, is evidence that the company has passed through inflation and tariff episodes by simply raising prices.
And AI? For this company it is not a threat but a customer. As signal speeds climb from 224G to 448G, design difficulty rises and the cheap competitors are squeezed out.
Earning power and capital allocation
The same 26.4% five-year average ROE has held while management digested more than 50 bolt-on acquisitions over ten years — proof that they do not break what they buy. Annual owner earnings run at roughly $4.2 billion; the latest quarter’s free cash flow of $830 million annualizes to no great improvement on that figure.
Capital allocation shows two faces. On one hand, dividends were raised 52% in 2025, returning about $1.5 billion a year. On the other, the purchase of a CommScope business — ten and a half billion dollars, the largest bet in company history, made while AI capex is running hot — swelled long-term debt from $6.5 billion to nearly $15 billion. The verdict on that will only arrive in two or three years. One point of discipline worth noting: the company has not accelerated buybacks at elevated prices.
Valuation and margin of safety
The market cap stands at one hundred ninety-one point eight billion dollars (share price $155.55, within a 52-week range of $105.45–$178.52). Wall Street’s consensus among 17 analysts is STRONG BUY, with an average target of $192.12.
The estimated owner-earnings yield is 2.19%, against a 10-year US Treasury at 4.70%. Even granting a generous lift of owner earnings to $6 billion and permitting a 15x multiple, intrinsic value comes to ninety billion dollars. Apply a 30% margin of safety and my buying line sits at a market cap of $63 billion or less — roughly $51 a share.
Seventeen analysts see 24% above today’s price. But they are counting next quarter’s orders, not the profits that remain in ten years. Against even the most forgiving estimate of intrinsic value, the market cap carries a 113% premium, and that 2.19% yield does not reach half of what the Treasury pays. The margin of safety here is not zero. It is negative.
Word for today
A great company and a great investment are not the same phrase. The former is made by management; the latter is made by the price.
I would like to own this business. At this price, I cannot.
Verdict: pass.
Correction (2026-09-03): market cap and owner-earnings yield restated at the price on the verdict date ($158.93 on 2026-08-25; the 2026-07-29 screening snapshot adjusted for price): market cap ≈ $194.9 billion (the article said $63.0 billion), owner-earnings yield ≈ 2.15% (the article said 2.2%) vs the 10-year Treasury at 4.64% — verdict unchanged.
This analysis is AI-generated, educational, and not investment advice. Figures may contain errors or be delayed. Disclaimer