Economic Moat: How to Spot a Competitive Advantage in Any Business

An economic moat is a durable competitive advantage that stops rivals from competing away a company’s high returns on capital. According to Pat Dorsey, real moats come from just five sources: intangible assets, switching costs, network effects, cost advantages and efficient scale. You confirm one exists when high returns hold up for years, including through a downturn.

Plenty of companies earn great returns for a year or two. Very few keep earning them for fifteen. The difference is almost always a moat, and learning to spot one changes the question you ask of every business from “is it profitable now?”

to “will it still be profitable in 2040?” Where this fits: this is the hub for Criterion 2, “Does it have a durable moat?” , in the Five Criteria framework.

It builds directly on Criterion 1 and return on invested capital.

What Is an Economic Moat?

Warren Buffett popularised the image.

The business is a castle. The moat is whatever keeps attackers, meaning competitors, away from the profits inside. In a 1999 Fortune article he explained that the key to investing is “not assessing how much an industry is going to affect society, or how much it will grow, but rather determining the competitive advantage of any given company and, above all, the durability of that advantage.”

Note the word durability. An advantage that lasts two years is a head start, not a moat. In his 2007 shareholder letter Buffett was explicit: “A truly great business must have an enduring ‘moat’ that protects excellent returns on invested capital.”

Why moats matter so much to returns

High returns attract competition. If a business earns 25% on its capital and nothing stops others copying it, money will pour in until returns fall to average. This is basic economics, and it happens in almost every industry.

A moat interrupts that process. It lets a company keep earning more than its cost of capital for much longer than competition would normally allow. Since your long-run return tends to track the return the business earns on its capital, the length of that period is worth a great deal.

The Five Sources of a Moat (Pat Dorsey’s Framework)

Pat Dorsey, who led equity research at Morningstar, set out the most practical moat framework in The Little Book That Builds Wealth (2008). He argued that almost every genuine moat comes from one of five structural sources.

Moat source	How it protects returns	Classic example
Intangible assets	Brands, patents or licences rivals cannot copy	Coca-Cola’s brand
Switching costs	Leaving is costly, risky or disruptive	Microsoft Office in businesses
Network effects	Each new user makes it more valuable	Visa and Mastercard
Cost advantage	Structurally lower costs than rivals	Costco’s scale; local quarries
Efficient scale	Market only supports one or two players	Pipelines, regional airports

Intangible assets Intangible assets include brands, patents and regulatory licences. They count only if they let the company charge more, or keep rivals out of the market entirely. A brand is a moat only when it creates pricing power.

Coca-Cola is the textbook case: people pay more for Coke than for a store brand made from very similar ingredients. Buffett began buying Coca-Cola shares in 1988 precisely because of that durable brand. Patents protect pharmaceutical companies for a set period, but they expire, so investors must look at the whole pipeline.

Regulatory licences, such as those needed to run a stock exchange or certain waste sites, are often overlooked. A rival cannot simply buy government approval. The test: could a competitor with deep pockets offer the same product at the same price and win customers?

If a brand does not pass that test, it is only a well-known name.

2. Switching costs

Switching costs are the money, time, risk and hassle a customer faces when moving to a rival.

The best ones feel like convenience, not a trap. Microsoft Office in large organisations is the classic example. Years of files, templates, macros, training and habits are built around Excel and Word.

A rival does not need to be worse to lose; it only needs to be not dramatically better. Switching costs are common in business software, banking relationships, specialist equipment with long service contracts, and outsourced payroll or data processing. They are often strongest where the product is a small share of the customer’s costs but a large share of the customer’s risk.

The test: high customer retention and the ability to raise prices without losing accounts.

3. Network effects

A network effect exists when each additional user makes the product more valuable to every other user.

Merchants accept Visa and Mastercard because consumers carry the cards. Consumers carry the cards because merchants accept them. Hamilton Helmer, in 7 Powers, makes a useful point: network effects can strengthen as the network grows, unlike a patent, which weakens as it nears expiry.

Stock exchanges, online marketplaces and professional networks can all show this pattern. Be careful, though. Many businesses with lots of users do not have network effects.

A streaming service with 100 million subscribers is not more useful to me because you subscribe. The test: does the product get better for existing users as new users join? Is the market leader pulling further ahead?

4. Cost advantages

A structural cost advantage lets a company charge less than rivals and still earn good returns, or charge the same and earn more. The key word is structural.

Being efficient this year is not enough. Dorsey identifies three main sources: cheaper processes, better locations and unique access to resources. Scale is closely related.

Costco’s buying volume, combined with a limited range of products, lets it negotiate prices few rivals can match. Geography can be powerful where goods are heavy and cheap. A gravel quarry near a growing city has a lasting advantage over one 300 kilometres away, because transport costs make the distant rock uncompetitive.

The test: are its costs per unit consistently lower than peers, and would a new entrant need years or huge sums to match them?

5. Efficient scale

Efficient scale is the least intuitive source.

Some markets are only big enough to support one or two profitable players. A newcomer would drive returns for everyone below the cost of capital, so rational competitors stay out. Pipelines, regional airports, some railway lines and niche industrial suppliers often fit this pattern.

Their protection comes not from brilliance but from arithmetic. There is only so much gas flowing between two cities, so a second pipeline would leave both half empty. The test: would a well-funded new entrant be likely to earn a poor return even if it succeeded in taking share?

What Is Not a Moat

Dorsey spends real time on what he calls mistaken moats. These are features that look like competitive advantages but usually are not: A great product. Great products get copied unless something else protects them.

High market share. Being big today does not stop a better rival. Many former market leaders prove this.

Great execution. Operating well is valuable, but a competitor can hire good managers too. A brilliant CEO.

Managers leave. A moat should be a feature of the business, not a person. This is why Criterion 2 asks you to name the specific source.

“It is the market leader” does not count.

How to Test a Moat with Numbers

A story about a moat is a hypothesis. The numbers are the evidence.

Here is how to test it.

Test 1: ROIC above 15% for five years or more

If a company has a moat, it should already show up as high returns on capital. A company claiming a wide moat while earning 7% on capital is telling you the moat is not working.

Start with the Criterion 1 work in our ROIC guide.

Test 2: ROIC held up through a downturn

This is the most important test. Anyone can earn high returns in a boom.

Look at how ROIC behaved in the worst recent year for that industry, such as 2008–09 or 2020. A dip is normal. A collapse to single digits, followed by a slow recovery, suggests customers had alternatives when money got tight.

Test 3: Stable or rising gross margins

Pricing power shows up at the gross margin line. If a company has passed on cost inflation for a decade while keeping its gross margin steady, customers are not shopping around.

Test 4: Market share and retention

Look for stable or rising market share, high customer retention or renewal rates, and long contracts.

Companies with switching costs often disclose retention figures in their annual reports.

Test 5: Price increases without volume loss

Read how revenue grew. Growth driven by price with steady volumes is a strong sign of pricing power.

Growth driven by discounts and promotions is the opposite.

Worked Example: Two Companies Through a Recession

Here are two hypothetical companies. Both looked excellent before a recession in Year 3.

Year	Company A ROIC	Company A gross margin	Company B ROIC	Company B gross margin
Year 1	21%	58%	19%	34%
Year 2	22%	58%	20%	35%
Year 3 (recession)	17%	57%	6%	27%
Year 4	20%	58%	9%	29%
Year 5	22%	59%	12%	30%

Before the recession they looked almost identical on ROIC. The downturn told the truth. Company A’s gross margin barely moved, and ROIC stayed well above 15% even in the bad year.

Its customers kept buying at similar prices. That is what a switching-cost or brand moat looks like in data. Company B had to cut prices to keep volumes.

Its ROIC collapsed and had not recovered two years later. Its earlier returns were probably the result of a strong cycle, not a moat. Company A passes Criterion 2.

Company B does not.

How Wide and How Durable Is the Moat?

Spotting a moat is half the job.

The other half is judging how long it will last. Morningstar’s ratings are a helpful reference: a wide moat is one expected to protect excess returns for at least 20 years, a narrow moat for at least 10. Ask these questions: Can you imagine the advantage still intact in 15 years?

If it needs heroic assumptions, be sceptical. Is the moat getting wider or narrower? Growing networks and deepening integrations widen moats.

Expiring patents and cheaper substitutes narrow them. Is it inside your circle of competence? If you cannot explain why customers stay, you cannot judge whether they will keep staying.

Signs a Moat Is Eroding

Moats rarely vanish overnight. They leak. These are the early warning signs I watch for: Gross margin slipping for three or more years, especially while competitors hold steady.

ROIC trending down from its historical range, even if it is still above 15%. Rising marketing or promotion spending just to keep sales flat. Market share loss to a new type of competitor, not just a familiar rival.

A technology or regulation change that lowers switching costs or opens a protected market. Management talking about “investing to defend our position” repeatedly, without explaining what has changed. Acquisitions to replace fading organic growth.

One sign is worth investigating. Several at once usually mean the castle is under attack. Our full list of warnings is in red flags in a business.

Where Moats Come From: Industry, Model and Boredom

Moats do not appear at random. They are more common in some settings than others. Industry structure.

A moat is easier to keep in an industry with few rational competitors, weak buyers and no close substitutes. Porter’s five forces are the standard tool, covered in why industry analysis is non-negotiable. Business model.

Subscriptions tend to build switching costs. Marketplaces build network effects. Franchisors lean on brands.

See different business models explained. Dull businesses. Many durable moats sit in unexciting niches that attract little new capital.

That is one reason Peter Lynch liked them, as explained in why boring businesses often make the best investments.

How We Use This in the Five Criteria

Criterion 2 asks: what stops competitors eroding this company’s returns? A company passes our house default when you can do two things: Name one of Dorsey’s five sources of moat (intangible assets, switching costs, network effects, cost advantage, efficient scale) and explain it in two sentences.

Show evidence that ROIC held up through a downturn, staying at or near the 15% Criterion 1 level in the industry’s worst recent year. If you cannot name the source, the company fails, however good recent returns look. If you can name it but the numbers never proved it, treat the moat as unconfirmed and demand a bigger margin of safety at Criterion 5.

Having worked in investor relations, I have heard many companies describe their “unique competitive position.” Very few could point to a downturn where the numbers proved it. That second half is what separates a moat from a slogan.

Common Mistakes When Analysing Moats

Confusing a popular product with a moat. Popularity fades. Ask what stops a copy.

Assuming size equals protection. Scale helps only when it creates a structural cost gap. Ignoring the downturn test.

Boom-time returns prove very little. Paying any price for a moat. A wide moat does not justify an unlimited valuation.

Price still matters. Anchoring on the past. Moats can erode.

Review the evidence every year. Counting “network effects” that are not there. More users is not the same as more value per user.

Your Next Step

Pick one company you own or follow. Name its moat source in two sentences, then check its ROIC and gross margin in the worst year of the last recession. Start with the numbers in our ROIC guide, see how the moat test fits the full process in the Five Criteria framework, and record your answers with the Five Criteria Checklist.