In one sentence: The numerical proof of a moat is a return on equity (ROE) of at least 10%, held over at least ten years and not falling, in a company with no debt.
Key ideas
- Dynamic learning stages. Phil uses the four stages of competence (unconscious incompetence, conscious incompetence, conscious competence, unconscious competence) and says Munger and Buffett are at the fourth stage, so they find it hard to explain what they do. Expect the second stage to be unpleasant. [03:00–07:00]
- Investing as habit. Danielle links reading the news "as an owner" to Gretchen Rubin's work on habits. The aim is a routine that needs no daily willpower. [07:00–12:00]
- ROE is the number. ROE is profit divided by shareholders' equity. Phil's rule is that Buffett and Munger treat consistent high ROE as the sign of a wonderful business. [23:00–25:00]
- Equity from the balance sheet. Assets minus liabilities is equity, using a household example (a $500k home and cars less $310k owed leaves $190k). Profit comes from the income statement. The cash-flow statement is the third. You rarely need to compute ROE yourself, because every finance site shows it. [31:00–37:00]
- Why ROE and not ROIC, for now. ROIC (return on invested capital) includes borrowed money, so debt dilutes the number. Phil prefers to start with companies that have no debt. Then ROE and ROIC match and the debt question disappears. Debt is "the second number": the right amount is none. [25:00–29:00]
- The thresholds. ROE of 10% or more, for at least ten years, and not trending down. A rising series (20%, 21%, 22%, 23%) is ideal. A falling one needs an explanation, unless the company is simply getting much bigger. [34:00–36:00, 43:00–47:00]
- Why ten years. Ten years is long enough for competitors to attack, so a steady ROE shows the moat is durable. Younger companies lack that record and belong in a small "risky" bucket of the portfolio. [43:00–45:00]
- Check the source. The Google results for IBM (about 72%) and Walgreens (about 10%) show the number is easy to find. IBM's figure is flattered by debt, so it isn't a starter case. Check more than one website. [37:00–43:00]
- How a mercenary wrecks ROE. An executive who wants a bigger jet buys an overpriced factory. Equity that earned 15–20% now earns 2%, and the drop shows up in ROE. [47:00–48:30]
- Wonderful versus nirvana. A wonderful business has high ROE and no debt. The best also grows. Management then decides whether to reinvest or return cash (buybacks and dividends). [49:00–51:00]
How it maps to RuleOne
- ROE history is a core screen metric. A stock page should show the ten-year ROE, flag any year below 10% and show the debt level.
- "Look for a falling ROE" is a good alert for the Holdings page.
- ROIC is the better number once debt is understood. The course should get there later (module m4).
Buffett, Munger and Graham links
- Buffett's letters repeatedly prefer high returns on equity with little debt. The 1979 letter is the one to read first, and he criticises earnings figures that ignore the capital employed.
- Munger's "show me the incentive" idea fits the mercenary CEO story.
Words to know
- ROE (return on equity): annual profit divided by shareholders' equity.
- Equity: assets minus liabilities, the owners' stake.
- TTM: trailing twelve months.
Try this
Open three stocks on /stocks/ from industries you know. For each, write down the ten-year ROE and the debt level. Mark any that pass 10% every year with no debt, and say whether the ROE is rising or falling.
Check yourself
- What is ROE, and what are Phil's thresholds?
Answer
Profit divided by equity. At least 10%, for at least ten years, not falling, ideally with no debt. - Why start with no-debt companies?
Answer
Debt inflates ROE and makes the business harder to judge. With no debt, ROE and ROIC are the same. - Why is a falling ROE a warning?
Answer
It may mean management is putting your capital into poor projects, or that the moat is eroding.
Short quotes
"Return on equity is the number that you need to know." (Phil, ~23:30, auto-transcribed)