Return on Capital: The One Quality Number

Picture two shops on the same street. Both clear ₹10 lakh in profit a year — identical bottom line. But the first owner put in ₹50 lakh of his own money and a bank loan to build his; the second built hers for ₹25 lakh. Same profit. Half the capital. Walk into both and they look equally successful. Look at the books and they are not even close.

The second shop is twice the business. She earns ₹40 of profit for every ₹100 she put to work; he earns ₹20. If she wants to double her profit, she opens one more shop. If he wants to double his, he has to raise twice as much money as she does — from a bank, or from you, by selling more shares. That gap, hidden behind an identical profit line, is the single most important quality test in investing. It has a name: return on capital.

What return on capital actually measures

Strip the jargon and it is one question: for every rupee a business puts to work, how many paise of profit does it earn back each year? That is the whole idea. A high number means the business is a good machine — it turns capital into profit efficiently. A low number means it is a hungry machine — it needs a mountain of money to produce a given profit.

You will meet two versions of this on screeners, and the difference matters:

• ROE — return on equity. Net profit divided by shareholders' money (share capital plus reserves). It answers: how well does the business do for its owners specifically?
• ROCE — return on capital employed. Profit measured against all the long-term money in the business — owners' equity plus borrowed money. It answers: how good is the underlying business, regardless of who funded it?

Hold on to that second one. ROCE is the honest number, because — as you'll see — ROE can be quietly inflated with debt while the business underneath stays mediocre.

Warren Buffett said it plainly in his 1979 letter to shareholders: "The primary test of managerial economic performance is the achievement of a high earnings rate on equity capital employed (without undue leverage, accounting gimmickry, etc.) and not the achievement of consistent gains in earnings per share" (Berkshire Hathaway, 1979 Chairman's Letter). Read the words in the brackets twice. Without undue leverage. He is warning you about the trap before we even get there.

Two real shops: Page Industries and NTPC

Let's replace the imaginary shops with two companies you know, using the same arithmetic. (Neither is a recommendation to buy or sell — they are here to teach the number.)

Take Page Industries — the company behind Jockey innerwear and Speedo swimwear in India. In FY25 it earned about ₹729 crore of net profit. The capital it employed to do that — share capital, reserves and all its borrowings added together — was roughly ₹1,669 crore (Inve data, 2026). Divide one by the other and Page earned about 44 paise of profit for every rupee of capital in the business. That is a phenomenal machine.

Now take NTPC, India's largest power generator. It is a much bigger company — in FY25 it earned about ₹19,649 crore of net profit, roughly twenty-seven times Page's (Inve data, 2026). On the bottom line, NTPC dwarfs Page.

But look at what it took. To earn that profit, NTPC employed roughly ₹3,47,872 crore of capital — equity plus borrowings (Inve data, 2026). So NTPC earned about 5.6 paise of profit per rupee of capital. Page earns roughly eight times more profit per rupee deployed.

Sit with the two raw numbers side by side. NTPC earns about twenty-seven times Page's profit — and to do it, it carries about two hundred times Page's capital. That is the two-shops story, at the scale of the National Stock Exchange. One business prints profit off a small, light base. The other is enormous, essential, and capital-hungry: every extra unit of electricity needs another power plant, another few thousand crore of steel and turbines and debt.

This is not a knock on NTPC. Somebody has to build the plants, and the country could not run without them. It is simply the point: size is not quality. The big number on the profit line told you nothing about how hard the business had to work for it.

The DuPont mirage: when ROE lies

Here is where most beginners get fooled, so slow down.

A company can make its ROE — the owners'-return number — look healthy by borrowing heavily, even when the underlying business is ordinary. More debt means less owner equity funding the same profit, which mechanically lifts profit-divided-by-equity. The decades-old DuPont framework, taught in every finance class, exists precisely to crack ROE open and show how much of it is real operating quality versus how much is just leverage. (You don't need the formula; you need the instinct.)

Power Grid shows the mirage in one company. In FY25 it earned about ₹15,522 crore. Measured against owners' equity alone, that is an ROE of roughly 16.8% — a number that looks genuinely good on a screener (Inve data, 2026). A beginner stops there and ticks "quality business."

But about 59% of Power Grid's long-term capital is borrowed money (Inve data, 2026). Measure the same profit against all the capital employed — the ROCE, the honest number — and the return drops to about 6.9%. The 16.8% wasn't a lie exactly; it was leverage doing a lot of the talking. The business itself earns a modest single-digit return on the capital it runs; debt magnifies what's left for owners.

NTPC tells the same tale: an ROE of about 12.2% that shrinks to a ROCE of about 5.6% once you count the borrowings too (Inve data, 2026). The owners'-return number flatters the business by roughly double.

This is the one habit to build: when ROE looks great, check how much of it is debt. A high return earned on the company's own money, with little borrowing, is quality. A high return manufactured by piling on debt is a borrowed costume. Page's near-44% return on capital is real — only about a sixth of its capital is borrowed. Power Grid's headline 16.8% is mostly leverage. Same kind of number, opposite meaning.

Why durable, not just high, is the word that matters

A high return for one year can be luck — a good monsoon, a commodity spike, a one-off. What separates a genuinely great business is a return on capital that stays high, year after year, because something protects it: a brand people pay up for, a distribution network rivals can't copy, a cost advantage. Buffett's word for that protection is a moat. Without one, high returns attract competitors who pile in, add capital, and compete the return back down to ordinary.

You can even see the rough shape of this across whole sectors. The RBI's study of 3,902 listed non-financial companies found FY25 operating margins of about 14.2% in manufacturing versus about 21.9% in IT services and 22.1% in non-IT services (RBI, "Performance of Private Corporate Business Sector during 2024-25," 26 June 2025). Asset-light service businesses tend to keep more of every rupee of sales than capital-heavy factories do. The sector you fish in tilts the odds before you pick a single stock.

And durability is exactly where reading the business beats reading the chart. A capital-heavy company lives and dies by its capex plans — and so does its concall. Look at what management actually talks about: NTPC and Power Grid spend their earnings calls walking analysts through capacity targets and capex schedules — how many more thousand crore, how many more gigawatts, by when. Asian Paints, Pidilite and Page Industries spend theirs defending margins. The difference in what management chooses to discuss tells you which kind of business you're holding. Tracking whether those capex and margin commitments actually land, quarter after quarter, across a whole portfolio, is the job Inve's Promise Tracker does so you don't have to read every transcript by hand.

Where this can fool you

One honest caveat, because return on capital is powerful enough to be dangerous when used alone.

A sky-high return number does not automatically make a stock a good buy. If everyone already knows a business is excellent, they bid the price up, and you can pay so much for that quality that you wait years to make money — owning a wonderful business at a foolish price is its own way to lose. (That second question — what's a fair price — is the job of a margin of safety.) Return on capital tells you whether the business is good. It says nothing about whether the price is fair. Those are two separate questions, and this article only answers the first.

And the number can mislead at the edges. A young company pouring money into a new factory will show a depressed return while the plant is being built and earns nothing yet — punishing it for investing in its future. A company can also flatter the ratio for a year by starving itself of investment. So read several years, not one, and ask why the number is what it is.

The two shops are still the cleanest way to hold all of this in your head. Before you ever look at a price, ask the owner's question: of every rupee this business puts to work, how much comes back — and will it keep coming back? Get that right, and you've already sorted the great businesses from the merely big ones, which is more than most people on the screen ever bother to do.

Frequently asked questions