Published: · Last reviewed:
Return on equity is useful only when shareholder equity is positive and the denominator has not been hollowed out by debt or buybacks. Damodaran’s January 2026 data shows 14.44% aggregate R&D-adjusted ROE for the U.S. market without financials, but that figure does not tell us how many individual companies clear the common cutoff. Before trusting a high ROE, check the equity base, leverage, sector benchmark, and earnings quality.
A stock screener flashes an 18% return on equity and gives the company a green quality badge. Then you open the balance sheet and find that shareholder equity fell because the company borrowed money and repurchased shares. Earnings barely moved. The ratio improved anyway.
That is the practical problem with return on equity. The numerator can be healthy while the denominator is shrinking, stale, or negative. A single cutoff cannot tell those stories apart.
The useful question is not whether a round-number cutoff is magic. It is whether the company produced that ROE through durable operating performance or through a denominator that became easier to divide by.
What Does a 15% Return on Equity Actually Tell You?
Return on equity equals net income divided by average shareholders’ equity. A 15% ROE means the company generated roughly $15 of net income for every $100 of average book equity during the measurement period.
The ratio compresses three different forces into one number:
- Profit margin: how much profit the company keeps from each dollar of sales.
- Asset efficiency: how much revenue the company produces from its asset base.
- Financial leverage: how much of the asset base is financed with equity rather than liabilities.
This is why two companies can report the same ROE while carrying very different business risk. One may earn strong margins with little debt. The other may reach the same result because its equity base is thin.
A widely repeated Buffett-style screen treats 15% as a quality threshold. Buffett’s own 1979 Berkshire Hathaway letter framed the managerial test more carefully: a high rate of return on equity should come without undue leverage or accounting gimmickry. That exact line is better treated as a practitioner screen than as a universal law.
Primary source: Berkshire Hathaway 1979 shareholder letter
Damodaran’s January 2026 U.S. dataset reports 14.44% R&D-adjusted ROE for the total market without financials. That is an aggregate ratio across 4,822 firms. It is not the median company, and it does not reveal the percentage of firms above the screen.
Do not convert the small difference between the screen and the aggregate benchmark into an expected stock-return difference. ROE is an accounting profitability ratio. It is not an annual portfolio return. Compounding that gap as if it were an investment-return premium creates a precise dollar result from two measurements that are not interchangeable.
The original version of this article made that exact mistake. The corrected analysis removes the unsupported 25-year wealth projection and keeps the part that survives scrutiny: raw ROE needs a denominator check before it can support an investment judgment.
Four Checks Before You Trust Return on Equity
A good ROE review does not require a complicated model. Four checks catch most of the ways the ratio misleads.
Check 1: Is Shareholder Equity Positive and Stable?
Start with beginning and ending shareholders’ equity. Use average equity for the ROE denominator, not only the latest balance-sheet figure.
- If equity is positive and broadly stable, the ratio remains interpretable.
- If equity fell sharply, ask whether losses, dividends, acquisitions, or repurchases caused the decline.
- If equity is close to zero, a small denominator can create an extreme ROE.
- If equity is negative, the arithmetic may produce a negative percentage, but the result is not a meaningful operating-quality comparison.
This same denominator problem can also break the debt-to-equity ratio. Once book equity turns negative, both ratios stop behaving like normal peer-comparison tools.
Check 2: Are Leverage or Buybacks Doing the Work?
Compare ROE with return on invested capital, debt trends, and share repurchases. ROIC is not perfectly standardized, and company-adjusted versions require care, but it is less directly driven by a shrinking book-equity denominator.
A widening ROE-to-ROIC spread is a prompt to investigate, not a universal sell signal. Check whether debt rose, treasury stock expanded, or retained earnings fell. S&P 500 companies spent $942.5 billion on buybacks in 2024, so repurchases are large enough to change equity denominators across the market even when operating performance is unchanged.
Check 3: Is the Sector Benchmark Comparable?
Capital structure, asset age, regulation, and accounting treatment vary by industry. A water utility and a software company should not be judged against the same raw cutoff without context.
The benchmark must also match the reported metric. Damodaran publishes both unadjusted ROE and ROE adjusted for research and development. Mixing a company’s unadjusted filing number with an R&D-adjusted sector number can create a false comparison.
Check 4: Do Cash Flow and Earnings Quality Support the Ratio?
ROE uses net income, so it inherits the limits of accrual accounting. Compare net income with operating cash flow, inspect unusual gains and charges, and look for repeated working-capital adjustments.
A high ROE supported by recurring cash generation deserves more confidence than the same ratio produced by a one-time gain. The cash flow statement is where that distinction usually becomes visible.
| Check | What to inspect | Warning sign | Practical response |
|---|---|---|---|
| Equity base | Beginning, ending, and average equity | Near-zero, sharply falling, or negative equity | Stop using raw ROE as a peer-quality score |
| Capital structure | Debt, buybacks, ROIC, treasury stock | ROE rises while equity falls or debt rises | Separate operating performance from financing |
| Sector context | Comparable industry and same adjustment basis | Universal cutoff applied across unlike sectors | Use a dated, like-for-like sector benchmark |
| Earnings quality | Operating cash flow and unusual items | Net income rises without cash support | Reconcile the income statement with cash flow |
Fast screen: pull five numbers from the latest filing: net income, beginning equity, ending equity, total debt, and operating cash flow. Those figures usually tell you whether the ROE deserves a closer look.
AutoZone: Profitable, but Standard ROE Breaks
AutoZone shows why a negative equity base should not be read as an automatic verdict on business quality.
Its Form 10-Q for the quarter ended May 9, 2026 reported $641.5 million of quarterly net income. The same filing showed a stockholders’ deficit of $2.785 billion, compared with a $3.414 billion deficit at August 30, 2025.
Positive income divided by negative average equity produces a negative ROE. The calculation is possible, but the interpretation is broken. It would make a profitable operator look weak simply because accumulated repurchases and other equity movements pushed book equity below zero.
| Metric | Reported value | What it tells the ROE reader |
|---|---|---|
| Q3 FY2026 net income | $641.5 million | The company remained profitable during the quarter |
| Stockholders’ deficit, May 9, 2026 | $2.785 billion deficit | The standard equity denominator is negative |
| Stockholders’ deficit, August 30, 2025 | $3.414 billion deficit | The denominator was already negative at fiscal year-end |
| Adjusted after-tax ROIC, trailing four quarters | 36.3% | Company-defined non-GAAP evidence of strong capital productivity |
| Repurchases since January 1, 1998 | 156.0 million shares for $39.8 billion | Long-running buybacks materially shaped book equity |
Primary source: AutoZone Form 10-Q, quarter ended May 9, 2026
AutoZone does not prove that buybacks are good or bad. It proves something narrower and more useful: once equity is negative, standard ROE no longer answers the question investors think it answers.
No single ratio replaces ROE. Review operating margins, cash generation, debt capacity, and a clearly defined return-on-capital measure. That broader check is also closer to how an economic moat should be evaluated: through durable economics, not one accounting output.
How Should Investors Compare ROE Across Sectors?
Damodaran’s January 2026 dataset makes the sector problem visible. Selected R&D-adjusted figures range from 3.25% for specialty chemicals to 20.21% for system and application software. Machinery sits near the common cutoff, while water utilities are closer to 10%.
| Industry | Number of firms | R&D-adjusted ROE |
|---|---|---|
| Software (System & Application) | 309 | 20.21% |
| Machinery | 105 | 15.11% |
| Aerospace/Defense | 79 | 11.32% |
| Utility (Water) | 14 | 10.27% |
| Chemical (Specialty) | 59 | 3.25% |
| Total Market without Financials | 4,822 | 14.44% |
Primary source: Damodaran, Return on Equity by Sector (US), January 2026
The table is not a ranking of sector attractiveness. It shows how accounting structure changes the baseline. A 12% ROE may be ordinary in one industry and strong in another. A 20% ROE may still be poor if peers produce more with cleaner balance sheets.
Historical-cost accounting adds another complication. Curtis, Lewis-Western, and Toynbee found that asset turnover can be biased upward when appreciated assets remain recorded at modified historical cost. Firms with older assets may therefore look more efficient partly because their book asset base is lower.
Esplin, Hewitt, Plumlee, and Yohn also found that separating operating and financing activities can improve profitability forecasts, but only under the appropriate component-forecasting approach. The practical takeaway is modest: decomposition can help, but it does not turn every accounting ratio into a clean predictor.
Return on Equity FAQ
What is a good return on equity?
A good ROE is positive, persistent, supported by cash flow, and strong relative to comparable companies with similar leverage. A universal cutoff such as 15% can be a starting screen, but it is not a final quality verdict.
Is 15% return on equity rare?
Damodaran’s 14.44% January 2026 figure for the U.S. market without financials cannot answer that question because it is an aggregate ratio, not a distribution of individual-company ROEs. You would need company-level data and a defined universe to measure how many firms exceed 15%.
Can return on equity be negative?
Yes. ROE can be negative because net income is negative, because average equity is negative, or both. Those cases mean different things. A loss on positive equity signals unprofitability, while positive income divided by negative equity usually signals a denominator that is no longer useful for normal peer comparison.
How do share buybacks affect ROE?
Buybacks reduce shareholders’ equity through treasury stock. If net income stays unchanged while equity falls, reported ROE rises. That increase may reflect capital allocation, leverage, or a thinner denominator rather than better operating performance.
Should I use ROIC instead of ROE?
Use both when the definitions are clear. ROIC reduces the direct effect of a shrinking equity denominator, but issuers calculate adjusted ROIC differently. Reconcile the formula, compare it with cash flow and debt, and avoid treating any single ratio as a complete quality score.
The Bottom Line on Return on Equity
A high ROE can be useful. It can also be a clean-looking answer to the wrong question.
Before trusting the ratio, verify that average equity is positive, identify what changed the denominator, compare the company with a like-for-like sector benchmark, and confirm that cash flow supports reported earnings. Those checks separate operating quality from leverage, repurchases, and accounting measurement effects.
One boundary matters most: do not turn an ROE difference into a stock-return forecast. The accounting ratio can guide investigation. It cannot supply a portfolio return premium by itself.
Try the four checks on one holding
Open its latest filing and compare beginning equity, ending equity, debt, operating cash flow, and repurchases. The denominator usually explains more than the green screener badge.
- Damodaran, NYU Stern: January 2026 U.S. sector ROE table. This article uses the R&D-adjusted column and labels the 14.44% nonfinancial total-market figure as an aggregate ratio.
- AutoZone Form 10-Q: Quarter ended May 9, 2026. Filing values are shown in millions or billions after rounding from the SEC table’s thousands.
- S&P Dow Jones Indices: 2024 S&P 500 repurchase expenditure of $942.5 billion, used only to establish the scale of buybacks.
- Esplin et al.: Operating and financing disaggregation can improve profitability forecasts under a component-forecasting approach. The article does not extend that finding into a guaranteed return prediction.
- Curtis, Lewis-Western, and Toynbee: Historical-cost asset measurement can bias DuPont ratios, especially asset turnover, when asset age differs.
Method: We checked each load-bearing figure against the official dataset, issuer filing, index provider, or publisher page. No backtest or portfolio-return model is used because the article’s claim is diagnostic and accounting-based. Sector figures are dated snapshots, not forecasts.
Limits: ROIC definitions vary, accounting policies differ, and aggregate sector data does not describe every constituent. Private companies, financial firms, REITs, and international issuers may require different benchmarks.
Damodaran ROE dataset · AutoZone 10-Q · S&P DJI buyback report · Esplin et al. · Curtis et al.
AI-assisted editorial process: AI tools were used to organize source checks and identify internal inconsistencies. Danny Hwang reviewed the final reasoning, figures, source alignment, and wording against the cited primary or publisher-hosted materials. See the editorial policy.
Financial Disclosure
This article is for educational and informational purposes only. It is not personalized investment advice or a recommendation to buy, hold, or sell AutoZone or any other security.
Accounting ratios are starting points, not substitutes for reviewing complete financial statements, valuation, risk tolerance, time horizon, and portfolio context.
Editorial review process for this article
1. Existing-article forensic audit · The original HTML was reviewed for claim provenance, calculation logic, class compatibility, and reader fatigue. 2026-07-18
2. Primary-source verification · Damodaran, SEC, S&P DJI, Berkshire Hathaway, and publisher-hosted academic records were checked. 2026-07-18
3. Human final review · Metric-fusion claims, repeated summary blocks, unsupported wealth projections, stale AutoZone data, and mechanical prose were removed or rewritten. 2026-07-18
Update history
-
v2.0
2026-07-18
MAJOR CORRECTIONRemoved the unsupported $84,232 wealth projection and the claim that a 14.44% aggregate market ROE proves a 15% screen is not rare. Updated AutoZone to its May 9, 2026 filing, corrected the negative-equity explanation, and rebuilt the article around four denominator checks.
-
v1.0
2026-04-18
PUBLISHOriginal publication.
Educational quantitative analysis based on published data. Not investment, tax, or legal advice. Consult a licensed professional before acting on any calculation. About TheFinSense.