Return on Equity (ROE) is one of the most widely used profitability metrics in equity screening — and one of the most frequently misread. A high ROE looks like a signal of quality, but it can just as easily be the product of financial leverage rather than a genuinely strong business. Used correctly, ROE is a powerful quality filter. Used alone, it's a trap.
How ROE is calculated
ROE = Net Income / Shareholders' Equity
If a company earns €15M in net income on €100M of shareholders' equity, ROE is 15%. It answers a simple question: for every euro shareholders have invested in the business, how many cents of profit did the company generate this year?
The DuPont decomposition: why ROE can mislead
The single most important thing to understand about ROE is that it can be broken into three components, known as the DuPont decomposition:
ROE = Net Margin × Asset Turnover × Equity Multiplier
- Net Margin (Net Income / Revenue) — how profitable each euro of sales is
- Asset Turnover (Revenue / Total Assets) — how efficiently the company uses its assets to generate sales
- Equity Multiplier (Total Assets / Shareholders' Equity) — how much leverage the company uses
The first two components — margin and asset efficiency — reflect genuine operating quality. The third — the equity multiplier — is pure financial leverage. A company can boost ROE simply by taking on more debt, without improving the underlying business at all.
Example: Two companies both earn €10M net income.
- Company A has €100M equity and no debt. ROE = 10%.
- Company B has €50M equity and enough debt to fund the same asset base. ROE = 20%.
Company B's ROE looks twice as good — but the improvement came entirely from leverage, not operational quality. If both companies hit a bad year, Company B's higher debt load makes it the riskier holding, despite the flattering ROE.
This is why ROE alone is a dangerous single-metric screen, particularly for banks and highly leveraged sectors where the equity multiplier can be 8-15x — a level where small changes in the multiplier swing ROE dramatically without any change in underlying business quality.
The ROE + Debt/Equity combination
The fix is simple: never screen ROE in isolation. Pair it with a leverage constraint.
Screen: quality without disguised leverage
- ROE > 12%
- Debt/Equity < 1.0 (for non-financials; use CET1 ratio and NPL ratio instead for banks)
- Net margin > 5%
- Positive revenue growth over 3 years
This combination filters for companies generating strong shareholder returns from genuine operating performance and reasonable asset efficiency — not from stacking on debt. It's the quantitative core of quality investing approaches, and closely mirrors how Warren Buffett's screening approach has historically favoured consistently high ROE businesses with conservative balance sheets — See's Candies and Coca-Cola being the textbook examples of high ROE generated by brand and pricing power, not leverage.
ROE vs. ROIC: which is harder to manipulate
ROIC (Return on Invested Capital) is a close cousin of ROE, but structurally more resistant to leverage distortion. ROIC divides after-tax operating profit by total invested capital — equity plus debt — rather than equity alone. Because debt appears in ROIC's denominator as well as being a source of returns, taking on more debt doesn't mechanically inflate ROIC the way it inflates ROE.
The practical implication: when ROE is high but ROIC is much lower, leverage is doing most of the work. When ROE and ROIC are both high and reasonably close together, the business is generating strong returns with limited reliance on debt — a materially higher-quality signal. Comparing the two metrics side by side is one of the fastest sanity checks available in a fundamental screen.
Good ROE thresholds by sector
ROE is not comparable across all sectors without adjustment:
- Banks and financials: ROE of 8-12% is typically considered adequate given regulatory capital constraints; above 15% is strong, but check CET1 ratio alongside it — high bank ROE with a thin capital buffer is a warning sign, not a quality signal.
- Consumer staples and branded goods: ROE of 20%+ is common and often reflects genuine brand pricing power and asset-light business models (low fixed asset intensity relative to revenue).
- Industrials and capital-intensive manufacturing: ROE of 10-15% is typical; above 20% is worth investigating for the leverage or one-off-item explanation before assuming pure quality.
- Utilities: Regulated, asset-heavy, structurally leveraged. ROE of 8-11% is normal; comparing utility ROE to consumer-brand ROE without adjustment is a common screening mistake.
A practical multi-filter European ROE screen
- Region: Europe (all exchanges)
- ROE > 12% — meaningful profitability on shareholder capital
- Debt/Equity < 1.2 — excludes leverage-driven ROE (be more lenient for financials, using CET1 instead)
- Net margin > 5% — minimum genuine profitability
- Market cap > €100M — minimum tradeable liquidity
- Sort by ROE descending, but review the Debt/Equity column alongside every result rather than trusting the ROE ranking alone
This screen typically returns a shorter, higher-conviction list than a P/E or P/B screen alone — ROE combined with a leverage constraint tends to surface genuinely well-run businesses rather than simply cheap or leveraged ones.
Warning signs when reading ROE
Rising ROE with rising debt/equity, flat margins. The improvement is coming from leverage, not operations. Check the trend in Debt/Equity alongside ROE, not just the ROE trend alone.
ROE above 30% for a non-financial, non-asset-light business. Investigate before assuming quality — this level is unusual outside branded consumer goods and software, and often reflects either heavy leverage, a recent one-off gain, or a shrinking equity base (share buybacks or accumulated losses reducing the denominator).
Negative or near-zero equity. ROE becomes mathematically meaningless or wildly distorted when shareholders' equity is very small or negative (common after years of buybacks or losses). Always sanity-check the absolute equity figure before trusting a percentage built on it.
Conclusion
ROE is a genuinely useful profitability filter — but only when read alongside a leverage constraint. The DuPont decomposition is the tool that separates ROE driven by real operating quality (margin and asset efficiency) from ROE driven by financial engineering (the equity multiplier). For European screening specifically, always pair ROE with Debt/Equity (or CET1 for financials) and cross-check against ROIC when the number looks unusually high.
Frequently asked questions
What is a good ROE for a European stock?
ROE above 12-15% is generally considered strong for non-financial European companies, though the right threshold varies significantly by sector — branded consumer goods routinely exceed 20%, while utilities and industrials more typically run 8-15%. Always compare ROE within the same sector, and check it alongside Debt/Equity before treating a high ROE as a quality signal on its own.
Why can a company have high ROE but be a bad investment?
High ROE achieved primarily through financial leverage (a high equity multiplier in the DuPont decomposition) rather than genuine operating profitability (margin and asset turnover) signals a business that looks efficient on paper but carries more balance sheet risk than the ROE figure alone suggests. In a downturn, the same leverage that inflated ROE amplifies losses.
What is the difference between ROE and ROIC?
ROE divides net income by shareholders' equity only, making it sensitive to leverage. ROIC divides after-tax operating profit by total invested capital (equity plus debt), which makes it much harder to inflate simply by adding debt. When ROE is much higher than ROIC for the same company, leverage — not operational quality — explains most of the gap.
How does the DuPont decomposition help with stock screening?
The DuPont decomposition splits ROE into net margin, asset turnover, and the equity multiplier (leverage). Screening for high ROE combined with a reasonable equity multiplier (i.e., not excessively leveraged) filters out companies whose apparent quality is actually a leverage effect, leaving a shortlist where the return is driven by genuine margin and efficiency.
Should ROE thresholds be different for banks?
Yes. Banks are structurally leveraged as part of their business model (taking deposits and lending them out), so the standard Debt/Equity filter doesn't apply meaningfully. For bank screening, pair ROE with the CET1 capital ratio and non-performing loan ratio instead — these measure capital adequacy and asset quality in a way that's actually comparable across banks.
Screen European stocks by ROE → — filter by ROE, Debt/Equity, and net margin across all European exchanges. Free, no account required. Pro at €29/month.