Moat erosion
How durable advantages die, and the signals that reach the figures before the story.
Every moat is temporary. The question is only whether it outlasts your holding period.
This lesson is about noticing erosion while it is happening rather than after, which is difficult for a specific reason: a moat usually dies in the income statement long before it dies in the story. The narrative stays intact — strong brand, loyal customers, market leader — while the figures quietly stop supporting it.
What erosion looks like in the numbers
None of these is conclusive alone. Together, over several years, they are the shape of an advantage going away.
The last one is worth dwelling on. Companies rarely announce that they are losing pricing power. They announce a strategic decision to invest in competitiveness, which is the same event described from the inside.
How moats actually die
Four patterns cover most of it.
Technology changes the basis of competition. Kodak's moat was manufacturing and distributing film. Digital photography did not out-compete that advantage — it made the advantage irrelevant, which is worse, because there is nothing to respond with.
Regulation removes the protection. A licence, a patent or an exclusive is an advantage held at somebody else's discretion. Patents expire on a date you can look up, which makes this the one form of erosion with a published timetable.
The customer's habit changes. Brands are held in people's heads and heads change slowly — until a generation with different defaults arrives.
The market grows into a second operator. The counter-intuitive one. If a moat is efficient scale, rising demand is a threat: a market that supported one profitable player may support two, and two is a different business entirely.
Re-testing, in practice
Treat a moat as a claim with an expiry date on it.
- Write down the mechanism — the specific thing stopping a competitor.
- Write down what would have to be true for that mechanism to fail.
- Check the second list when the company files, not when the price moves.
That last point is the discipline. A falling share price makes everyone re-examine a thesis; a falling gross margin in an otherwise good year usually does not, and it is the more informative of the two.
This is also why the committee re-grades rather than re-narrates. The five-year series of margins, returns and capital intensity is the same evidence, gathered the same way every time — so a drift shows up as a drift instead of as a change of mind.
Follows Berkshire Hathaway shareholder letter, 1987.