Expeditors International: A Fine Business Priced Like an Indestructible One
What the business is
Some businesses own the ships. This one owns the paperwork. Expeditors International is an asset-light freight forwarder and customs broker: it does not own aircraft or vessels. Instead, it buys cargo space wholesale from airlines and ocean carriers, fills that space with customer freight, and keeps the spread — the net revenue. On top of that sit fee-based services: customs brokerage, warehousing, distribution. In 2025, revenue broke down as airfreight 36%, ocean freight 25%, and customs and other services 39%.
The moat is a network. More than 250 offices, concentrated on the Asia–North America lanes, aggregate volume from shippers. That volume buys negotiating power with carriers. That negotiating power wins more shippers. The loop feeds itself, and Morningstar has identified exactly this self-reinforcing network effect as the basis for a wide moat. Add the switching costs of large shippers whose compliance and multi-lane logistics are embedded in Expeditors’ processes, plus a portfolio of customs licenses and regulatory know-how that no algorithm can wish away, and you have a genuine — if not impregnable — barrier.
The capital economics are the quiet star. Because the model owns almost nothing that rusts or depreciates, maintenance capital spending is negligible, and owner earnings sit close to reported earnings. The five-year average return on equity is 37.3%, remarkably consistent. The direction of the moat, however, is my concern. Digital forwarders are taking share in standardized lanes and routine freight, and unmanned customs clearance has already been demonstrated in the industry. AI is collapsing the marginal cost of the quote-book-track layer of this business. What remains defensible — regulatory liability, deep integration with large shippers, volume-based buying power — is real, but the low-complexity territory the moat protects is being nibbled at the edges. My judgment: the moat is narrowing, gently. Not collapsing. Narrowing.
What the numbers say
The arithmetic here is unfussy, which is one of the few kindnesses this business offers an analyst. Owner earnings are reported at $814 million for the year — an annual figure that, given the asset-light model, differs very little from net income. This was a mid-cycle year: ocean rates have normalized, and the result sits between neither the 2022 peak nor a trough, but at a reasonable baseline.
The capital allocation record deserves its own sentence of applause. Essentially no long-term financial debt. No M&A empire-building, ever. Decades of cumulative buyback programs completed, a fresh $3 billion authorization approved in February 2026, $875 million returned to shareholders in 2025, and a 5% interim dividend raise in 2026. A simple principle, repeated: invest organically, return the surplus, repeat. Management also refuses quarterly guidance and reports bad quarters — a 15% drop in Q4 2025 net income, for instance — without cosmetics. One gray area: the thesis notes the recent buyback price may sit above intrinsic value, and “repurchase below intrinsic value” is the standard I prefer.
Now the price. Here are the figures that matter, adjusted to the latest close:
- Market cap (price-adjusted, 2026-09-02 close): $24.42 billion
- Owner-earnings yield (price-adjusted): 3.33%
- Share price (2026-09-02 close): $187.43
- 10-year Treasury: 4.80%
An owner-earnings yield of 3.33% against a risk-free 4.80% is a spread of minus 1.47 percentage points. The market is paying a rich multiple — the thesis notes a PE near 28.8 — for a business whose earnings swing with freight rates and whose moat is, in my assessment, slowly eroding. Applying conservative multiples of ten to twelve times owner earnings produces an intrinsic range that the current market cap exceeds by roughly two and a half times at its top; even a generous fifteen-times case, per the thesis, falls well short of today’s price. There is no margin of safety here. There is a premium.
What Wall Street believes vs what I see
The current headline mood is a study in ambivalence. One contributor recommends a Hold after strong Q2 results tempered concerns. Another worries openly about visible weakness in the ocean segment. A third asks about “the AI elephant in the room” — which at least means the right question is being asked, even if nobody in the headlines has answered it.
My answer is the arithmetic above. The consensus is arguing about quarters; the valuation is pricing in decades. A 3.33% owner-earnings yield means the market has already paid for consistent 37% returns on equity, flawless capital discipline, and a moat that holds its ground against software. It has paid, in other words, for everything going right — while the evidence says the moat’s least complicated territory is already contested. Paying a premium multiple for a brokerage business whose moat is an open question is not one mistake; it is two mistakes stacked neatly on top of each other.
Verdict
Verdict: pass.
The business quality is real — debt-free, disciplined, consistently high returns on capital — but the rubric is unambiguous: an owner-earnings yield of 3.33% does not clear the 4.80% Treasury, and the price sits far above any conservative estimate of intrinsic value. Quality at the wrong price is still the wrong price. The verdict changes if the market cap falls toward owner earnings at twelve times — the thesis puts that somewhere in the $70s per share — or if management demonstrates three consecutive years of defending net revenue margins and share against digital erosion, evidence that the AI elephant is a working employee rather than a houseguest.
This analysis is AI-generated, educational, and not investment advice. Figures may contain errors or be delayed. Disclaimer