What ROCE means

ROE only looks at shareholders’ own money. But most companies run partly on borrowed money too — a term loan, working capital debt. ROCE — Return on Capital Employed — asks the bigger question: how much profit did the business generate on all the capital invested in it, whether that capital came from shareholders or lenders?

Capital Employed = Equity + Borrowings — everyone who’s put long-term money into the business, not just its owners. And because that capital belongs to lenders too (who get paid via interest, before shareholders see anything), ROCE uses EBIT — profit before interest and tax — instead of PAT, so the return isn’t already net of what’s owed to one of the two groups who supplied the capital.

The formula

ROCE (%) = EBIT / Average Capital Employed × 100
Capital Employed = Total Equity + Total Borrowings

As with ROE, we average the opening and closing capital employed rather than using the closing figure alone.

Worked example: Desi Bites Foods, FY25

  ₹ Lakh
EBIT (FY25) 326
Capital employed, start of year (FY24) 1013
Capital employed, end of year (FY25) 1078
Average capital employed 1045.5
ROCE 31.2%

Worked example: Britannia Industries, FY25

From Britannia Industries’ audited consolidated FY25 results (year ended 31 March 2025, filed 8 May 2025). Total equity here includes non-controlling interests, since capital employed is a whole-business measure, not an owners-only one. For illustration only.

  ₹ Crore
EBIT (FY25) 2873.81
Capital employed, start of year (FY24) 6007.23
Capital employed, end of year (FY25) 5606.09
Average capital employed 5806.66
ROCE 49.5%

Notice Britannia’s ROCE (49.5%) is close to, but a little below, its ROE (52.5%) from the previous post. That’s actually a reassuring pattern, not a red flag — it means the strong ROE isn’t being artificially pumped up by heavy borrowing; the company earns a genuinely high return on capital regardless of who supplied it.

🧒 Explain it like I'm 10 (optional — skip if this is already clear)

Imagine two friends open identical lemonade stands. Friend A used ₹1,000 of her own savings. Friend B used ₹500 of his own savings and ₹500 borrowed from his dad. If both stands make the exact same ₹300 profit before paying dad back any interest, judging them only on “return on my own money” makes Friend B look like the better businessman — his ₹500 “earned” 60%, versus Friend A’s ₹1,000 earning 30%. But that’s not because Friend B ran a better stand — it’s because he used less of his own money. ROCE looks at the ₹1,000 total in both stands, so it judges the lemonade-selling skill itself, not who financed it.

Common mistakes

  • Comparing ROCE across companies with very different debt levels without also checking ROE. If ROCE is healthy but ROE is dramatically higher, leverage is doing a lot of the work — worth understanding before assuming the business itself is that efficient.
  • Using EBITDA instead of EBIT in the numerator. EBITDA still includes the benefit of assets that are ageing and losing value — EBIT accounts for that, which is why ROCE specifically uses EBIT.
  • Using closing capital employed instead of average, especially in a year with a large mid-year capital raise or debt repayment — this can distort the ratio significantly in either direction.
  • Treating a high ROCE as a reason to buy at any price. ROCE measures business quality, not value for money — a genuinely excellent business can still be a poor investment if bought at too high a price. That’s a separate question, covered in the Valuation module.

Takeaway: ROCE measures the return a business generates on all the capital invested in it — equity and debt alike — making it a fairer way to judge operating quality than ROE alone, which can be flattered by leverage.