The Case for Paying a Premium for Durable Moats

Costco's price-to-earnings ratio sat in a fairly narrow band for years — roughly 36 to 40 times earnings through fiscal 2020 to fiscal 2023. In fiscal 2024 it broke above 50 times and hasn't meaningfully round-tripped since. Nothing about the warehouses changed. The membership model didn't get structurally stronger in twelve months. What changed is how much the market is now willing to pay for confidence that whatever is protecting Costco's earnings today will still be protecting them a decade out.

That's the real question underneath every "is this stock too expensive" debate, and it's worth stating plainly: what would it take to displace this business, and does the price reflect a reasonable answer to that — or something closer to a story the market is telling itself? A persistent above-market multiple isn't automatically a warning sign. Sometimes it's the correct price for a kind of durability a cheaper, narrower-moat business simply doesn't have on offer, and I think most investors skip straight past that distinction to whether a multiple "feels" high.

MoatScope's Quality × Valuation grid keeps moat strength and price as two separate axes on purpose, because collapsing them into a single buy/don't-buy verdict hides the judgment call that actually matters here. A wide-moat business trading rich isn't automatically overpriced. It might be correctly priced for durable economics that a narrow-moat business can't credibly promise at any price. Sorting out which one you're looking at is what the rest of this is about.

Why Duration Changes What a Dollar of Earnings Is Worth

Every stream of earnings carries an implicit assumption about how long it survives mostly intact. A commodity producer's profits might be trustworthy for the current cycle and genuinely unknowable past it — a new discovery, a demand shock, a substitute technology, and the picture changes fast. A business protected by real switching costs, network effects, or a license nobody else can obtain carries a longer visible runway before the erosion every economic moat eventually faces actually shows up in the numbers.

That visible runway is what a persistent premium is usually pricing — more than it's pricing growth. Costco isn't a fast grower in the way the market usually rewards with a rich multiple; comparable sales growth that's typically mid-single digits, membership fee increases that arrive roughly once every seven years. What it offers instead is confidence: the switching costs built into an annual membership habit, multiplied across tens of millions of paying members, aren't going away next quarter, or the one after that. Not exciting. Reliable, though — and reliability is worth paying for, within limits.

The flip side of the same logic explains why a cheap stock is sometimes cheap for a defensible reason, not a mispriced one. A business with no real switching costs, thin differentiation, and a commodity product is a legitimate two- or three-year forecast at best; anything past that is closer to a guess than an estimate, and the market is right not to pay up for a story it can't actually see past a couple of annual reports. And low multiples on those businesses aren't the market being lazy. They're the market being honest about how little of the future it can responsibly underwrite. Confusing "cheap" with "underpriced" ignores that the multiple is doing real work — it's a rough proxy for how many years of cash flow an investor can reasonably count on.

Three Businesses the Market Has Refused to Let Go Cheap

Three examples make the pattern concrete. Each sits in a different moat category, each has traded at a sustained premium for the better part of a decade or more, and each shows why that premium reflects something structural rather than a crowd chasing a familiar name. None of the three is the fastest grower in its sector. That's the point, not an oversight.

Costco: Loyalty You Can Measure in a Renewal Rate

Costco's fiscal 2025 results put its membership renewal rate at 92.3% in the U.S. and Canada and 89.8% worldwide — figures that barely moved even after the company raised its annual membership fee for the first time in seven years, effective September 2024. That's the durable competitive advantage a membership model is supposed to produce: once a household budgets around a Costco run, switching away means giving up a habit, not just a store. The stock's multiple — north of 45 times earnings for most of the past two years, against a 20-year average closer to 40 times — is the market betting that habit holds. It isn't betting on Costco opening warehouses meaningfully faster than it has for the last decade.

Visa: Infrastructure Nobody Rebuilds From Scratch

Visa doesn't lend money or take credit risk; it operates the rails that move the authorization message between a merchant's bank and a cardholder's bank, and it collects a sliver of nearly every transaction that crosses them. In fiscal 2024, Visa reported net revenue of $35.9 billion against operating expenses of $12.3 billion — an operating margin north of 65%, the kind of number you only see in a business where the marginal cost of processing one more transaction is close to zero and the barrier to a competitor building an equivalent network is close to insurmountable. Visa went public in 2008. The margin structure has barely wavered since, through a financial crisis, a pandemic that shut down in-person commerce for months, and several rate-hiking cycles that were each, in turn, supposed to be the thing that finally cracked consumer spending.

Moody's and S&P Global: A Duopoly Regulation Helped Build

Moody's and S&P Global don't publish a combined market-share number — they're competitors, not partners — but independent trackers of the ratings industry have consistently put their joint share of global credit ratings output at around 80%, a concentration that gets even tighter in sovereign debt, where the two firms' combined share tops 99%. Part of that is reputation built over a century. Part of it is structural: both are designated Nationally Recognized Statistical Rating Organizations by the SEC, a status new entrants can seek but rarely secure at meaningful scale, and bond covenants and investment mandates routinely reference their ratings by name. Replacing either firm would mean rewriting contracts across the fixed-income market, not just building a better rating model — which is a much higher bar than out-innovating a competitor on price or service.

What the Premium Is Actually Pricing In

Line these three up and the pattern isn't growth. None of them is compounding revenue at a rate that would normally justify a rich multiple on its own — and that's the point. What the market is pricing is duration: the probability that whatever generates today's cash flow is still generating it, largely unchanged, five or ten years out. A moat doesn't have to be dramatic to be valuable. It has to be durable enough that a long-duration forecast isn't a leap of faith.

So a persistent premium is, in effect, a duration bet. Not a growth bet. That distinction matters because it changes what should actually worry you about owning one of these businesses. It's not a disappointing quarter of revenue growth; the market was never really paying for growth here in the first place. It's evidence that the thing protecting the earnings — the renewal habit, the network, the license — is quietly weakening, which is a slower and much easier signal to miss.

A persistent premium is a bet on duration, not a referendum on how much you like the business.

It also explains why the market sometimes gets duration badly wrong in the other direction — paying a premium multiple for a story of durability that turns out not to be real. A brand that looks unassailable can lose pricing power in a handful of quarters once a cheaper substitute becomes socially acceptable. A platform that looks locked-in can watch switching costs evaporate once a regulator or a new interface removes the friction that was holding customers in place. The three businesses above have each been tested by at least one real threat to their position and come through with the underlying metrics intact. That track record is worth something. It isn't a guarantee of the next one.

Where a Justified Premium Turns Into Overpaying

None of this is a license to pay any price for a wide moat. The argument gets shakier from here: there's no clean formula for exactly how much duration is worth, and I'm less confident than I'd like to be about how this reasoning holds up the next time one of these businesses gets tested by a real, multi-year downturn rather than a sharp-but-short one. Our companion piece on judging moat quality separately from a valuation model makes a related point: the moat question and the price question are different jobs, and mixing them up produces bad answers to both.

A wide-moat classification tells you a business is likely still standing in a decade. It says nothing on its own about whether today's price has already baked in a few more decades of dominance than any moat, however wide, can plausibly promise. Overpaying for a real moat is still overpaying — the moat just makes the mistake slower to show up in the numbers, which is arguably worse. You can hold a wonderful business for years at too rich a multiple and simply earn a mediocre return while everyone around you keeps calling it a great investment.

A useful practical check, short of building a full model: ask what has to remain true for today's multiple to make sense five years from now, and be specific about it. If the answer is "the renewal rate stays roughly where it is," that's a testable, monitorable claim — you'll know within a year or two if it's wrong. If the answer requires the business entering entirely new categories at high returns, or a competitor that hasn't yet been named failing to show up, the premium is leaning on hope rather than duration. The first kind of premium is the kind worth paying. The second kind is speculation wearing a quality label.

Key Takeaways

• A persistent above-market multiple can be the correct price for genuine durability. Costco, Visa, and the Moody's/S&P Global duopoly have each traded rich for the better part of a decade because their moats extend the visible life of their cash flows, not because they're fast growers.
• The premium is usually a duration bet, not a growth bet — watch for the moat itself weakening (falling renewal rates, eroding network effects, a regulatory door cracking open), not for an ordinary soft quarter of revenue growth.
• A wide moat justifies some premium, not any premium. Judge moat quality and price as separate questions before combining them into a verdict.
• Treat the historical evidence with some humility. The sample of moats actually tested by a prolonged, multi-year downturn is smaller than the case studies above make it look.