What ROE means

Margins (gross, EBITDA, net) all measure profit against revenue. ROE — Return on Equity — asks a different question: how much profit did the company generate on the money its own shareholders put in and left in the business (equity, from the balance sheet)?

It’s the number that answers “for every ₹100 that belongs to shareholders, how much did the company earn this year?” — which is a more direct read on how well a company uses your money specifically than any revenue-based margin.

The formula

ROE (%) = PAT / Average Equity × 100

We use the average of opening and closing equity (not just the closing balance) because equity changes over the year — profit gets added, dividends get paid out — and averaging gives a fairer sense of what capital was actually at work across the year.

Worked example: Desi Bites Foods, FY25

  ₹ Lakh
PAT (FY25) 209
Equity, start of year (FY24 closing) 553
Equity, end of year (FY25 closing) 678
Average equity 615.5
ROE 34.0%

Worked example: Britannia Industries, FY25

From Britannia Industries’ audited consolidated FY25 results (year ended 31 March 2025, filed 8 May 2025). Equity here is equity attributable to owners of the company (excluding non-controlling interests), matched against PAT attributable to owners. For illustration only.

  ₹ Crore
PAT (FY25) 2178.73
Equity, start of year (FY24 closing) 3941.52
Equity, end of year (FY25 closing) 4355.72
Average equity 4148.62
ROE 52.5%

A 52.5% ROE is high, even for a well-run FMCG business — a large part of the reason is that Britannia doesn’t need much shareholder capital relative to its profit (a strong, asset-light, brand-led business). That’s worth remembering for the next post: a high ROE alone doesn’t tell you why it’s high, and one common reason — leverage — deserves real scrutiny.

Common mistakes

  • Not checking what’s driving a high ROE. A company can boost ROE two ways: earning more profit (good), or shrinking its equity base through debt-funded buybacks or heavy borrowing instead of using shareholder capital (not automatically good — it means more of the company’s capital is other people’s money, which carries its own risk). ROE alone can’t tell these apart — that’s exactly what ROCE (Return on Capital Employed), the subject of the next post, is for.
  • Using closing equity instead of average. If a company raised a large chunk of fresh equity partway through the year, closing-equity ROE understates true returns on the capital that was actually deployed for most of the year — average smooths this out.
  • Comparing ROE across industries with different capital needs. A capital-heavy business (steel, cement) and a capital-light one (FMCG, services) will show structurally different ROE even if both are equally well managed.
  • Treating one strong year as proof of quality. A single year’s ROE can be inflated by a one-off gain in that year’s PAT — check the trend over several years, not one data point, before drawing conclusions.

Takeaway: ROE tells you how hard shareholders’ own money is working, but it doesn’t tell you how that return was achieved — a business earning a high ROE mostly through debt looks very different, and carries different risk, from one earning it through genuinely efficient operations.