"Economic moat" is the most important concept Warren Buffett borrowed from corporate strategy and made famous in investing. It is also one of the most over-claimed terms in public-company investor presentations. Most companies that claim a moat do not have one. This post walks through the five canonical moat types in plain English, with one real public-company example for each, and the quantitative anchor that distinguishes a real moat from a marketing claim.
We have a longer treatment with 25 examples at /blog/economic-moat-types-with-examples-2026/. This post is the shorter, beginner-focused version with one canonical example per moat type.
What a moat actually is
A moat is whatever protects a business's above-normal returns from being competed away. In corporate strategy, profits above the cost of capital should attract competition that pushes returns back to normal within 2 to 5 years. With a moat, the above-normal returns persist for 10, 20, sometimes 50 years.
Quantitatively, a real moat shows up as sustained ROIC above 15 percent for 5+ years. Below 15 percent is commodity economics. Above 25 percent sustained for 10+ years is wide-moat territory.
The five moat types described below are the framework Morningstar's Pat Dorsey codified in "The Five Rules for Successful Stock Investing" (2003) and "The Little Book That Builds Wealth" (2008). Buffett's verbal use predates the formal taxonomy by 30 years; the underlying ideas are the same.
Moat 1: Network effects (example: Visa)
The mechanic: the product becomes more valuable to each user as more users join. A telephone network with one user is useless; a network with a billion users is the most valuable utility on earth. The same logic applies to payment networks, social platforms, marketplaces, and exchanges.
Visa (V) is the canonical example. Two-sided network: more cardholders attract more merchants because merchants want to accept payments from as many customers as possible; more merchants attract more cardholders because cardholders want to use their card everywhere. The loop has compounded for 60+ years. Visa now processes roughly 200 billion transactions per year across over 4 billion cards in 200 countries.
Why competitors cannot break in: a new payment network would need to recruit billions of cardholders AND millions of merchants simultaneously. Recruiting either side without the other is worthless. The bootstrap problem is structural.
Quantitative validation: Visa has had operating margins above 60 percent for 15+ years. ROIC has been above 25 percent every year since the 2008 IPO. The market share has been stable to growing. Those are the fingerprints of a real moat.
Other public network-effect examples: Mastercard (MA), Meta Platforms (META), Booking Holdings (BKNG), CME Group (CME), the London Stock Exchange (LSE.L), Airbnb (ABNB).
See /buffett/v/ for the full Buffett-Fit breakdown on Visa.
Moat 2: Switching costs (example: SAP)
The mechanic: once a customer adopts the product, the cost (in money, time, training, or risk) of switching to a competitor is prohibitive. The customer is locked in by the friction.
SAP (SAP.DE / SAP) is the canonical example. SAP is the dominant ERP (enterprise resource planning) software for large companies. An ERP implementation at a Fortune 500 company costs 10 to 100 million USD and takes 18 to 36 months. Once SAP is running the back office, the company's accounting, HR, procurement, supply chain, and finance all flow through it.
Switching to a competitor (Oracle, Workday, Microsoft Dynamics) is not a question of buying different software. It is a question of re-engineering every business process and retraining every employee, with operational risk for the duration. Most large enterprises never switch. SAP's customer base is the same set of large enterprises it has had for 30 years.
Quantitative validation: SAP's net revenue retention is consistently above 100 percent. Customer churn at the large-enterprise tier is in low single digits per year. Operating margin is above 30 percent and has been stable.
Other public switching-cost examples: Microsoft (MSFT, Office and Azure lock-in), Adobe (ADBE, creative-suite skill investment), Salesforce (CRM, custom workflow lock-in), Oracle (ORCL, database lock-in), Sartorius (SRT.DE, validated bioprocess equipment).
See /buffett/sap/ for the full breakdown.
Moat 3: Intangible assets (example: Hermès)
The mechanic: brand, patent, or regulatory licence creates a structural advantage that is difficult to replicate, regardless of how much capital a competitor deploys. The asset that creates value is not on the balance sheet (it is intangible) but is enormously valuable.
Hermès (RMS.PA) is the canonical brand example. Hermès sells handbags at price points starting around 10,000 USD, with multi-year waitlists for the Birkin and Kelly models. The product is leather and stitching; the value is the brand. Hermès deliberately restrains supply (it cannot meet demand by design) because expanding supply would dilute the brand.
The result: Hermès has operating margins above 40 percent (compared to 10-15 percent for most luxury brands and 5-10 percent for non-luxury). The brand is the asset, and the asset compounds because the multi-generational customer base is loyal across cycles.
Quantitative validation: Hermès gross margin has been above 70 percent for a decade. Operating margin above 40 percent. Revenue growth above 10 percent through both expansionary and recessionary cycles. ROIC above 25 percent. These are all signals that the intangible (the brand) is doing the work.
Other public intangible-asset examples: LVMH (MC.PA, luxury brand portfolio), Coca-Cola (KO, beverage brand), Disney (DIS, character IP), Moody's (MCO, ratings-agency licence), ASML (ASML, EUV patent portfolio), Ferrari (RACE, racing heritage).
See /buffett/rms/ for the full breakdown.
Moat 4: Cost advantage (example: Costco)
The mechanic: the company's larger scale or structurally different cost model lets it deliver the same product at lower unit cost than smaller competitors. The lower cost is passed through to customers as lower prices, which attracts more volume, which compounds the cost advantage.
Costco (COST) is the canonical example. Costco operates roughly 875 warehouse-club stores globally with hyper-efficient supply chain, limited SKU count (roughly 4,000 SKUs per store versus 100,000 at a traditional grocer), and a membership-fee model that effectively turns customers into shareholders. The membership fee covers most of operating profit, which allows Costco to mark up products at razor-thin margins (around 11 percent gross margin versus 25-30 percent for traditional grocers).
The result: Costco's prices on identical products are 20 to 40 percent lower than at traditional grocery stores. Customers come back, because the price differential is real. Walmart and Sam's Club are the closest competitors, but Costco's cost structure remains structurally advantaged on the warehouse-club model.
Quantitative validation: Costco's operating margin is low in absolute terms (around 3-4 percent) but stable. Membership renewal rate is above 90 percent globally and above 92 percent in the US. Same-store sales are positive every year. Market share continues to grow.
Other public cost-advantage examples: Walmart (WMT, distribution scale), Amazon (AMZN, AWS scale), Aldi (private, but the model is the same), Geico (BRK.B's subsidiary, direct-to-consumer auto insurance), Southwest Airlines (LUV, point-to-point route structure).
See /buffett/cost/ for the full breakdown.
Moat 5: Efficient scale (example: Waste Management)
The mechanic: the market is just big enough to support one or two profitable competitors. A third competitor entering would push everyone's returns below cost of capital. As a result, incumbents are protected by the fact that no rational competitor would enter.
This moat is rare. It typically appears in geographically segmented businesses where the local market is too small to support multiple efficient operators.
Waste Management (WM) is the canonical example. Trash collection in a given metro area is a route-based business: you need trucks driving past every house every week. Below a certain density of customers, the operation is unprofitable. The first operator in a given region captures the dense routes; a second operator enters at lower density and lower profitability; a third operator can rarely enter at all without destroying everyone's returns.
The result: in most US metro areas, two or three operators (WM, Republic Services, Waste Connections) split the market and operate at healthy returns. New entrants do not appear because the math does not work.
Quantitative validation: Waste Management has had ROIC above 12 percent for 15+ years (below the 15 percent ROIC moat threshold but justified by the regulatory and capital-intensity profile). Operating margin around 18 percent and stable. Revenue grows roughly with GDP plus modest pricing.
Other public efficient-scale examples: Republic Services (RSG, also waste), pipelines (KMI, EPD, ENB.TO), rail (UNP, CSX, NSC, CP), some regional utilities, some midstream energy.
See /buffett/wm/ for the full breakdown.
Combined moats are the strongest
The wonderful businesses Buffett actually owns tend to have multiple stacking moat types, not just one. Examples from his current portfolio (see /investors/warren-buffett/ for the current Berkshire holdings):
- Apple (AAPL): intangible asset (brand) + switching costs (App Store, iCloud, ecosystem) + cost advantage (supply-chain scale)
- Coca-Cola (KO): intangible asset (brand) + cost advantage (distribution scale) + sometimes regulatory (bottling partnerships)
- American Express (AXP): network effect (card network) + intangible asset (brand, particularly the premium positioning)
- Moody's (MCO): intangible asset (NRSRO licence) + network effect (issuers use the rating most investors trust most)
When you screen for the multi-moat compounders, the cohort shrinks dramatically. That is also why those names trade at premium multiples: the market understands the durability of the moat and prices the persistence into the valuation.
How to validate a claimed moat in 5 minutes
For any stock claiming a moat, run three quick checks:
- ROIC above 15 percent for 5+ consecutive years. If not, the moat is not yet visible in the numbers.
- Gross margin stability through the last downturn. Pricing-power moats hold gross margins through recessions; commodity businesses cannot.
- Market share trend over 10 years. A moat business holds or grows share; an eroding moat loses share to competitors annually.
If a company passes all three, the moat claim is probably real. If it fails any, ask why.
Where invest-like fits
Each stock on invest-like at /buffett/[ticker]/ surfaces:
- The qualitative moat-source tagging (network, switching, intangible, cost, efficient scale)
- The 5-year ROIC trend with the colour coding for "above 15", "10 to 15", or "below 10"
- The gross-margin stability score
- The market-share trend where available
The Moat pillar of the Buffett-Fit Score combines all of the above into a single Strong Fit / Partial Fit / Avoid verdict for the moat dimension.
For the longer treatment with 25 examples, see /blog/economic-moat-types-with-examples-2026/. For the five pillars of the Buffett framework where the moat is pillar 1, see /blog/buffett-5-pillars-stock-valuation/.
Disclosure
Educational tool. The five moat-type taxonomy is Pat Dorsey's framing, also used by Morningstar's analyst team. The named companies above are illustrative; moats erode and the examples may change over time. Not investment recommendations. Past moat strength does not guarantee future moat strength.
Author: Zaid Ghazal, founder of invest-like, Kiel, Germany. Not a registered investment adviser. Full methodology at /methodology/.