All four sources so far explain how a company beats competitors inside a market. Efficient scale explains why, in some markets, competition rationally never shows up. It needs one condition. The market is limited, by geography, by niche demand, by regulation, relative to the minimum efficient size of a competitor. One or two players can serve the whole market at good returns. A third entrant would add capacity the market can't absorb. Prices and returns would crash for everyone, including the entrant itself. A rational, well-funded potential entrant runs that pro-forma and stays out.
So the moat is the entrant's own spreadsheet.
Where does efficient scale show up?
You'll find it in a few classic habitats. Exchanges and market infrastructure come first. Liquidity wants one venue, and the market needs one, maybe two, depositories or clearinghouses. Rating agencies are next. Issuers need two ratings from agencies investors already trust, and a third global generalist adds nothing investors want. Then come pipelines, airports, towers. One pipe serves the route, and a parallel pipe halves both pipes' volumes.
Last are regional cement and aggregates, which have low value-to-weight. Cement travels economically only ~300 to 500 km and crushed stone ~50 km, so each catchment is its own small market. A region that consumes 5 mn tonnes supports two 2.5 mn-tonne plants earning well. A third plant guarantees a price war. So cement returns are decided region by region, and "India cement industry" averages mislead. Watch for stacking here. Most efficient-scale businesses also hold a license (intangible), so the license limits entry legally while efficient scale limits it economically.
Name both. The numbers don't care which you credit.
Advertisement
What does efficient scale look like at CDSL?
Worked example (India): CDSL. India's securities-depository market supports two players, CDSL and NSDL, and has since the 1990s. Every demat account must sit at one of the two. The entire brokerage/registrar/exchange plumbing integrates with both. And the market's total fee pool is modest: annuity-like issuer charges, transaction fees, KYC. A third depository would need years of SEBI process and nine-figure infrastructure to fight over a pool that two incumbents already serve at high returns.
No rational entrant appears, and none has.
Illustratively, at ~FY25 scale (pull the annual report), revenue is ~₹1,000 to 1,100 crore, PAT margins ~40%+, and ROE/ROCE ~30%+ on a business that is mostly cash and receivables. CDSL holds ~75%+ of demat accounts by count, because the retail boom accrued to it. NSDL dominates by value held, which is institutional.
You must always attach a width caveat to regulated infrastructure. SEBI influences tariffs, so the same regulator that blesses the duopoly caps its tolls. IRCTC, back among the licenses, is the same story. India's power-exchange leader IEX is another. With ~90% share of its market, it derated ~30% in days in 2025 when the regulator advanced "market coupling," which would pool orders across exchanges and neutralize its liquidity advantage. Verify the current status.
Efficient-scale moats are wide against competitors and narrow against the state.
What does efficient scale look like at Moody's?
Worked example (US): Moody's. By convention and by many institutional mandates, the bond-rating market needs two ratings. The two the world's investors already price off are S&P and Moody's. Fitch is a distant third, and together the big three take ~95% of the market, stable for decades. An entrant can't sell what investors don't yet trust, and investors won't trust what issuers don't yet use. That's an efficient-scale wall, reinforced by a reputational intangible and the SEC's NRSRO designation. Post-2008 regulation was aimed at the agencies, yet it raised compliance barriers for newcomers. Regulation often concretes an incumbent's moat while punishing it. Issuers pay a toll of a few basis points of each issue, and they recoup it many times over in lower coupons.
A product that costs the customer negative money is very hard to disrupt.
On illustrative ~2024 figures (pull the 10-K), ratings-segment revenue is ~$3.5 to 3.8bn at ~55 to 60% operating margins. Group ROIC is ~20%+ including acquisition goodwill, and far higher on organic capital. Excess returns have persisted across five decades and multiple crises. They even survived the ultimate test, the reputational catastrophe of 2008, with the duopoly's pricing power intact. Buffett's line about the business is that the pricing-power test is whether you can raise prices without losing share. Moody's does, almost annually.
How does efficient scale fail, and how do you test it?
Efficient scale is the moat most often destroyed by growth. Say the niche becomes a boom market because the catchment doubles or the asset class explodes. Then the math that kept entrants out flips, and entry arrives all at once. Watch the denominator.
Evidence trail for efficient scale (where to look). Count the players and the years since last entry, despite visible excess returns. Size the market's total profit pool against a new entrant's minimum efficient scale. That's the pro-forma test, Worked Example 1's method. For physical goods like cement and aggregates, check the freight radius using lead distances in annual reports. Read the regulatory entry requirements and the tariff-setting powers. Look for institutional two-supplier norms in ratings and audit. Then check utilization across the region's capacity.
High utilization with no entry means efficient scale is working. Overcapacity means it never existed.