DL 340 · Moats in Depth · Lesson 1 of 5

Measuring a moat: ROIC that persists

A moat is a claim about the future. The best evidence for it is in the past: a return on capital that stayed above the cost of that capital for years, while rivals had every reason to copy it.

Level:Advanced
◆
Read time:6 min
◆
Course:DL 340
The idea

What a moat leaves in the accounts

Nobody can audit the future. What you can audit is whether a business has already done what a moat is supposed to let it do: earn more on its capital than that capital costs, and keep doing so while competitors watched and tried to take the profit away.

Buffett made the point in a 1999 Fortune article. What matters, he said, is a company's competitive advantage "and, above all, the durability of that advantage." This lesson is about measuring the durability. The Economic Moat lesson covers the five sources and the franchise idea, so we won't repeat them.

We'll use ROIC throughout, because it doesn't care how a company is financed. Our ROIC lesson shows how to calculate it, and Margins & Returns explains why it beats ROE for this job.

Economic profit
(ROIC − Cost of capital) × Invested capital
The spread in brackets is the moat's footprint. A company that earns exactly its cost of capital creates no value, however fast it grows.
The evidence

Returns drift toward the middle

Competition pulls returns toward the average. High returns attract new entrants and copycats, prices get cut, and the returns come down. Low returns push capacity out of an industry, and the survivors recover. Economists call it mean reversion. Every time you say a company has a moat, you're betting against it.

Michael Mauboussin and Dan Callahan have measured the pull for years, in Measuring the Moat and later in The Base Rate Book. They sort companies by ROIC and follow each group forward. The top group's returns fall and the bottom group's rise, yet the gap usually narrows without closing: companies that start out earning high returns tend to stay above average for a long time. How fast the spread fades varies a lot from one industry to another.

Mauboussin's name for the stretch during which a company earns more than its cost of capital is the competitive advantage period. A wide moat is simply a long one. Your job is to estimate its length, and the past record is the first clue.

Worked example

Same average, opposite stories

Two illustrative companies, ten years of ROIC each. Haverford Seals makes seals for industrial pumps. Moreton Solar had a panel design nobody else could make, until rivals worked it out. Take a 9% cost of capital for both.

YearHaverford SealsMoreton Solar
118%35%
219%30%
317%26%
416%22%
518%15%
620%11%
719%9%
818%10%
917%8%
1019%7%
Ten-year average18.1%17.3%
Lowest year16%7%
Years above 9%10 of 107 of 10
Last three years, average18.0%8.3%

On a ten-year average they're almost twins. Haverford has never earned less than 16%. Moreton's spread has been shrinking for a decade and turned negative in year 9.

Worked example
Economic profit in year 10 (illustrative)
Haverford: invested capital $800M(19% − 9%) × 800 = $80M
Moreton: invested capital $1,500M(7% − 9%) × 1,500 = −$30M

Moreton is the bigger company, with a ten-year record that looks almost as good. It's now losing about $30 million of value a year for its owners.

So look past the average. Check the worst year, especially if it fell in a recession. Count the years above the hurdle. And look at the slope of the last five years, which is where a fade shows first: Haverford's last five average 18.6%, Moreton's 9%.

Where it breaks

When ROIC misreads the moat

  • Investment that never reaches the balance sheet. Brands, software and drug pipelines are built with marketing and research, which are expensed. Invested capital looks small, so ROIC looks huge. The persistence still tells you something, but the level is overstated. The incremental ROIC lesson touches on capitalising R&D.
  • A bought moat. After a big acquisition, goodwill swells invested capital, and ROIC can look ordinary even when the businesses inside it are excellent.
  • High returns on almost no capital. A staffing agency or a consultancy can show 30% ROIC because it barely needs any capital. That alone doesn't prove anyone is kept out. Check whether margins are high too.
  • The industry, not the company. If every firm in a sector earns 20%, the advantage may belong to the industry's structure, regulation or efficient scale. Ask whether the company beats its peers as well as its cost of capital.
  • Survivors. Long ROIC histories come from companies that lasted. Many of the ones whose moats collapsed fast aren't in the data any more.
On Dividend Line

Persistence on Dividend Line

Every company page has a Moat tab. Its verdict rests on eight numbers calculated from up to ten years of annual accounts, and the first of them is ROIC persistence: the ten-year average of after-tax operating profit divided by equity plus long-term debt, marked down if any single year fell below 6%. Gross margin level and consistency, earnings predictability, cash conversion and share count discipline sit beside it. Until the written analysis exists for a company, the tab shows these numbers directly, under "The numbers behind the verdict".

For the shape of the series, use the ROIC & ROE chart in the Chartroom tab, which plots each year against a dashed 15% hurdle. That's where you'd spot a Moreton.

The Moat Screener works on one year. Its ROIC column is the latest annual figure, and the switch "Earns more than its cost of capital" keeps companies whose latest ROIC is above a flat 8%. Moreton would have passed in year 8 and failed in year 10, long after its moat had gone. Use the screener to find candidates, then check the record.

Wide and narrow moats with a latest ROIC of at least 15%. Open each company's Moat tab for the ten-year picture. You'll need a free account.Open the Moat Screener →
Remember

Takeaways

“
Key takeaway
  • A moat's footprint is a spread: ROIC above the cost of capital, sustained while competitors try to close it.
  • Returns tend to fade toward the average. The length of the fade, the competitive advantage period, is what you're estimating.
  • Haverford and Moreton had nearly the same ten-year ROIC. The lowest year, the count of years above the hurdle and the recent slope told them apart.
  • Expensed investment, goodwill and tiny capital bases all distort ROIC. Read the level with care and the persistence with more confidence.
Sources

Sources for this lesson

  • Michael Mauboussin and Dan Callahan, "Measuring the Moat", Credit Suisse (2013)
  • Michael Mauboussin, Dan Callahan and Darius Majd, The Base Rate Book, Credit Suisse (2016)
  • Michael Mauboussin and Paul Johnson, "Competitive Advantage Period: The Neglected Value Driver", Financial Management (1997)
  • Warren Buffett with Carol Loomis, "Mr. Buffett on the Stock Market", Fortune (November 1999)
Next lesson · DL 340 · 2 of 5 ◆

Five pricing-power tests

Every management team says it has pricing power. Five tests you can run from public documents to check, starting with the one most analysts stop at and going well past it.
Continue
Educational disclaimer · This content is for educational and informational purposes only. It does not constitute investment advice, tax advice, legal advice, or a recommendation to buy or sell any security. Always conduct your own research before making investment decisions.