Return on Equity Explained: The One Ratio That Separates Great FTSE 100 Businesses From the Rest

ROE is net income divided by shareholder equity, expressed as a percentage. That's it.

A company with £100 million of net income and £500 million of shareholder equity has a 20% ROE. For every pound shareholders have invested in the business — through buying shares at IPO, retained earnings left in the company, or additional capital raises — the company generates 20p of annual profit.

The "equity" part is crucial. This isn't return on assets (which includes debt-funded assets) or return on capital (which blends debt and equity). ROE is specifically the return accruing to shareholders — the people who own the residual claim on the business after everyone else gets paid. When a UK investor buys a share inside a stocks and shares ISA, ROE is the closest metric to "what return is this business generating on my capital?"

Book value — shareholder equity on the balance sheet — is the denominator. It equals total assets minus total liabilities. For a company like Unilever, with its portfolio of brands built over a century, book value understates economic reality: the Dove brand isn't on the balance sheet at anything close to its true worth. For a bank like Barclays, book value is dominated by loan portfolios that are marked to market. The same ROE figure means different things depending on what's in the denominator. For tax treatment of dividends received from UK companies, consult the latest HMRC rates and allowances.

The maths is brutal and beautiful. A company that earns 20% on equity and retains all its earnings grows its equity — and therefore its capacity to earn — by 20% per year, every year, without issuing a single new share. That's the compounding engine that built Unilever's century-long track record and RELX's transformation from a print publisher into a data-analytics powerhouse.

RELX, with an ROE of approximately 86%, isn't just profitable — it's in a different league of capital efficiency. The company pivoted from trade magazines to risk analytics and legal databases, businesses where the marginal cost of serving an additional customer approaches zero. That's how you get an ROE that looks almost implausible: you build an asset once — a database of legal precedents, a fraud-detection algorithm — and sell access to it millions of times.

At the other end, Sainsbury's ROE of around 3.9% reflects the brutal economics of UK grocery retail. Supermarkets are asset-heavy (stores, distribution centres, inventory), intensely competitive, and operate on razor-thin margins. A 3.9% return on equity barely exceeds what you'd earn in a savings account. That doesn't mean Sainsbury's is a bad business — it's resilient, cash-generative, and pays a 4.35% dividend — but it does mean the business destroys value if its cost of equity exceeds 3.9%.

For a UK ISA investor, ROE provides a rough ceiling on the long-term return you can expect. If you buy a company at book value and it earns 15% on equity consistently, your investment will compound at roughly 15% — less any dividends paid out. If you pay 3x book value for the same company, your return drops to about 5%. The price you pay relative to book value is a direct tax on the ROE you actually receive. The MoneyHelper guide to investing provides free, impartial guidance for UK investors getting started.

A 36% ROE tells you Unilever is doing something right. It doesn't tell you what. The DuPont formula — named after the American chemical company that pioneered it in the 1920s — breaks ROE into three components:

ROE = Profit Margin × Asset Turnover × Equity Multiplier

• Profit margin (net income ÷ revenue): how much of each pound of sales becomes profit
• Asset turnover (revenue ÷ total assets): how efficiently the company uses its assets to generate sales
• Equity multiplier (total assets ÷ shareholder equity): how much the company relies on debt — higher multiplier means more leverage

Multiply all three, and you get ROE. This decomposition tells you how a company earns its return.

Reckitt Benckiser, with an ROE around 41%, illustrates one path: high margins. Its portfolio — Nurofen, Dettol, Durex — commands premium pricing, producing a profit margin that supermarkets can only dream of. The business doesn't need enormous asset turnover or aggressive leverage; the margins do the work.

NatWest, with an ROE around 14%, takes a completely different route. Bank margins are modest — net interest income minus loan losses — but the equity multiplier is enormous. A bank with £500 billion in assets and £40 billion in equity has a 12.5x equity multiplier. Even a thin margin, multiplied by 12.5x, produces a respectable ROE. The trade-off: high leverage means high fragility. A 2% asset write-down wipes out a quarter of NatWest's equity. The same write-down at Reckitt is an annoyance.

RELX's extraordinary 86% ROE comes from a combination: decent margins on its data products plus extraordinarily high asset turnover. The company's assets are predominantly intangible — databases, algorithms, customer relationships — that produce revenue without consuming capital. There is almost no asset base to speak of relative to the earnings stream.

For an investor, the DuPont breakdown reveals whether a high ROE is sustainable. Margin-driven ROE (Reckitt, Diageo) tends to be durable — brands don't evaporate. Leverage-driven ROE (banks) is cyclical — it works until it doesn't. Asset-turnover-driven ROE (RELX, Experian) is the holy grail but rare, requiring genuine competitive moats that are hard to replicate.

Rolls-Royce reports an ROE of roughly 212%. That's not a typo. It's also not a reason to buy the stock.

Rolls-Royce illustrates the most dangerous trap in ROE analysis: a depleted equity base. After years of write-downs, restructuring charges, and the near-death experience of the pandemic (when wide-body engine flying hours collapsed), the company's book value shrunk dramatically. Net income recovered — the turnaround under the current management has been genuinely impressive — but applied against a tiny equity base, it produces an absurdly high ROE that tells you nothing about business quality.

This is the negative-equity trap. When shareholder equity is very low — or even negative, as with InterContinental Hotels Group — ROE becomes mathematically meaningless. IHG has negative book equity because it's returned so much capital to shareholders through buybacks that its balance sheet equity is negative. A company buying back shares at high prices destroys book value, which mechanically inflates ROE even if the underlying business hasn't improved.

The second trap is excessive leverage. A company that doubles its debt while keeping net income flat will see its equity multiplier double and its ROE soar — but the business hasn't become more profitable, it's just become more dangerous. This is why you should never look at ROE in isolation. The DuPont breakdown exposes leverage-driven ROE for what it is: a risk story disguised as a quality story.

The third trap is one-off gains. If a company sells a division and books a £500 million gain, that flows through net income and inflates ROE for one year. The following year, when the gain disappears and the business is smaller, ROE collapses. Always check whether reported ROE is boosted by exceptional items — the FCA requires UK-listed companies to disclose exceptional items separately in their financial statements — UK companies report these separately under IFRS, so you can strip them out.

ROE connects directly to two other numbers every investor should understand: the P/E ratio and the dividend payout.

A company's sustainable growth rate — the rate at which it can grow without issuing new shares or increasing leverage — is: Sustainable Growth = ROE × (1 − Payout Ratio)

If Unilever earns 36% on equity and pays out 60% of earnings as dividends (retaining 40%), its sustainable growth rate is roughly 14.4% (36% × 0.4). That's the rate at which the business can compound internally — no acquisitions, no share issues, just reinvesting retained earnings at the existing ROE.

This formula is why high-ROE businesses command higher P/E ratios. A company with 30% ROE and a 50% payout ratio grows at 15% organically, every year, without needing external capital. A company with 8% ROE and the same payout grows at 4%. All else equal, the first company deserves to trade at a much higher multiple — and does. RELX trades at 22x earnings; Sainsbury's at 18x. The gap would be wider if Sainsbury's didn't pay a large dividend.

This also explains why share buybacks are so powerful at high-ROE companies. When RELX buys back £1 of its own shares, it's retiring equity that was earning 86p a year. If it borrows at 5% to fund the buyback, it replaces 5p of interest cost with 86p of avoided equity returns — an 81p annual gain per pound bought back. The same trade at a low-ROE utility is far less impactful. Buybacks create value only when the ROE exceeds the after-tax cost of borrowing — and preferably by a wide margin.

For the UK ISA investor, the practical takeaway: when you see a company trading at a high P/E, don't just ask "is it expensive?" Ask "is the ROE high enough to justify it?" A 30x P/E on a 50% ROE business that retains half its earnings is arguably cheap. A 12x P/E on a 5% ROE business that pays everything out is arguably expensive. The P/E is meaningless without the ROE.

You don't need a spreadsheet. The data is in the annual report of every UK-listed company.

Net income is on the income statement — usually labelled "Profit for the year attributable to equity shareholders." Shareholder equity is on the balance sheet — "Total equity attributable to owners of the parent." Divide the first by the second, multiply by 100, and you have ROE.

Most FTSE 100 companies also report ROE directly in their key performance indicators section. But — and this matters — there is no standardised definition under IFRS. Some companies use "adjusted" earnings that exclude exceptional items. Some use average equity over the year rather than year-end equity. Some adjust for goodwill and intangibles. Always check the footnotes to see what's actually being measured. The FCA Handbook sets out the disclosure requirements for UK-listed companies under the Disclosure Guidance and Transparency Rules.

The FCA requires UK-listed companies to publish annual reports within four months of year-end. Most also publish half-year results with abbreviated financials. For real-time data, financial websites calculate ROE using trailing twelve-month figures — useful for tracking, but always cross-check against the statutory accounts at least once a year.

For investors who prefer funds to individual stocks: fund factsheets rarely report ROE directly. But you can find it on Morningstar or the fund manager's website, usually under portfolio characteristics. A fund with a weighted-average ROE above 20% owns fundamentally different companies from one averaging 8%. That single number tells you more about the portfolio's character than three pages of the manager's commentary.

ROE doesn't stand alone. It's the quality-control check that follows the valuation work.

Start with the P/E ratio to see what you're paying. Then check the balance sheet to understand what's in the equity base — negative equity or excessive intangibles should set off alarm bells. Use free cash flow to verify that reported earnings — the numerator of ROE — are backed by actual cash generation. Look at dividend yield to see what portion of that return you're actually receiving in your pocket. And verify with EPS and dividend cover that the payout is sustainable.

Used together, these six fundamentals give you a framework that would have kept you out of most of the FTSE 100's biggest disasters — and pointed you toward its quiet compounders. A company with a reasonable P/E, a strong balance sheet, double-digit ROE, genuine free cash flow, and a well-covered dividend isn't a guarantee of outperformance. But it's a dramatically better starting point than buying what's in the news.

For more on how these metrics interact in practice, see our investing hub and our growing collection of fundamentals explainers — all written for UK investors who want to understand what they own.

This article is for informational purposes only and does not constitute financial advice. You should seek independent financial advice before making any investment decisions. Past performance and historical ROE figures are not reliable indicators of future results. The companies mentioned are examples for educational purposes and do not constitute buy or sell recommendations.