Warren Buffett has said he would rather own a business earning a high return on equity than one with high earnings in absolute terms. The reason: ROE measures efficiency. It tells you not just how much a business earned, but how much it earned relative to the capital shareholders have committed. A business that earns more with less capital is, structurally, a far better business to own.
"The best business to own is one that over an extended period can employ large amounts of incremental capital at very high rates of return." — Warren Buffett, 1992 Shareholder Letter
Return on equity measures the profit a company generates for each dollar of shareholders' equity:
Net income is the company's after-tax profit from the income statement.
Shareholders' equity — also called book value or net assets — is found on the balance sheet. It represents the total value of what shareholders own: all assets minus all liabilities. It is the accumulated capital that shareholders have put into the business over time, plus retained earnings.
If a company has $1 billion in shareholders' equity and earns $200 million in net income, its ROE is 20%. For every $1 of capital invested, it is generating 20 cents in profit per year.
When a company retains earnings — reinvesting profits rather than paying them out as dividends — those retained earnings become part of shareholders' equity. In future years, the company must earn a return on that expanded equity base to justify the retention.
A company that consistently earns 25% ROE is compounding shareholder capital at 25% annually on the retained portion. Over 10 years, $1 of equity generates approximately $9.31 in cumulative earnings at 25% ROE vs. $2.59 at 10% ROE. This is the compounding effect that Buffett describes as the foundation of long-term wealth creation.
This is also why Buffett prefers retained-earnings businesses to dividend-paying ones (when the ROE justifies retention): if the company can reinvest its profits at a high rate of return, keeping the money inside the business creates more value than distributing it to shareholders who would likely earn less elsewhere.
A single year of high ROE can reflect many things: a one-time windfall, a very low equity base (from recent write-downs), or an unusually strong business environment. What Buffett looks for is ROE that is consistently high across multiple years — because that consistency reveals the business's structural characteristics, not its luck.
A business that earns 25% ROE in one year and 5% the next is unreliable. Its high-ROE year was circumstantial. A business that earns 18–22% ROE across five different years — including a recession — has demonstrated something real: it has structural pricing power, capital efficiency, or a competitive position that persists across economic cycles.
This is the most important nuance in reading ROE: leverage inflates the number artificially, and a high-leverage ROE is often more dangerous than it appears.
Consider two companies that each earn $100 million in operating profit:
Company B shows a much higher ROE — but it is leveraged 5:1. Its equity base is small relative to its obligations. If earnings decline even modestly, Company B may struggle to service its debt. Company A's 9% ROE reflects the same underlying operations with far less risk.
Buffett explicitly adjusts for this: he looks for high ROE achieved without excessive leverage. A company with a low debt-to-equity ratio and consistently high ROE has proven it can generate exceptional returns purely from the quality of its operations — not from financial engineering. That is the combination he prizes.
The DuPont analysis breaks ROE into three component drivers, which helps you understand why a company has a high or low ROE:
This breakdown reveals three distinct paths to high ROE:
The ideal, from Buffett's perspective, is ROE driven by the first two factors, not the third. A company with 40% net margins and efficient capital deployment generating high ROE is a fundamentally different investment from a company achieving the same ROE through 6:1 leverage.
A related metric that eliminates the leverage distortion entirely is Return on Invested Capital (ROIC):
By including debt in the denominator, ROIC measures returns on the total capital committed to the business, regardless of how it was financed. A company with high ROIC and low leverage will show both high ROIC and high ROE, confirming that its returns are genuine. A company with high ROE but low ROIC is almost certainly achieving it through leverage.
Buffett tends to use ROE in public commentary because it is simpler and more widely understood, but his actual analytical framework incorporates the debt context described above.
Some industries structurally generate higher ROE than others due to their economic characteristics:
Industries that structurally struggle with high ROE include capital-intensive manufacturing, utilities (regulated returns), airlines (high fixed costs, commodity pricing), and most commodity producers. In these sectors, high ROE is genuinely unusual and more telling when it does appear.
ROE is one of the six criteria in the Buffett Score, and we evaluate it with two additional considerations beyond the headline number: consistency across multiple years, and the leverage context. A company that scores highly on ROE in our model has demonstrated sustained capital efficiency, not just one good year, and has achieved it without excessive financial risk.
