Cement: EBITDA per tonne and capacity utilisation

Why cement is read per tonne

Cement is about as close to a commodity as a manufactured product gets. One company’s 50 kg bag is much like another’s, it’s too heavy to ship far cheaply, and its price in any region moves with that region’s supply and demand. The costs — power and fuel, freight, limestone — are also counted in tonnes.

So cement analysts don’t start from margins in percent. They start from rupees per tonne: how much the company realised on each tonne, what each tonne cost, and what was left — EBITDA per tonne (earnings before interest, tax, depreciation and amortisation, divided by tonnes sold). Percentage margins move whenever cement prices move, even if nothing about the business changed. Per-tonne profit strips that out.

The second number is capacity utilisation: how full the plants are. A cement plant is a big fixed cost. Once it’s built, every extra tonne it sells carries most of its price straight to profit — the operating leverage mechanism at industrial scale. Steel and aluminium makers are read the same way, per tonne and per unit of capacity.

The formulas

EBITDA per tonne        =  EBITDA  /  tonnes sold

Realisation per tonne   =  net sales  /  tonnes sold
Cost per tonne          =  realisation per tonne − EBITDA per tonne

Capacity utilisation    =  volume produced (or sold)  /  installed capacity
                           (for a quarter, annualise the volume — × 4 —
                            or use a quarter's worth of capacity)

Units:  ₹ crore = 10,000,000 rupees;  MTPA = million tonnes per annum

The whole trick is to keep the numerator and denominator on the same base: the same businesses, the same geography, the same period.

Worked example: UltraTech Cement, FY26

From UltraTech’s results press release of 27 April 2026, covering the quarter and year to 31 March 2026 (consolidated). UltraTech calls EBITDA “PBIDT” — profit before interest, depreciation and tax. Historical, for illustration only.

Item	Q4 FY26	Q4 FY25	FY26	FY25
Net sales, ₹ crore	25,467	22,788	87,384	74,936
PBIDT, ₹ crore	5,688	4,721	17,598	13,302
India grey cement sold, million tonnes	42.41	—	145	—

Step 1: the company’s own number

1. The release reports an operating PBIDT per tonne of ₹1,253 for the March 2026 quarter, 11% higher than a year earlier, and says total cost per tonne fell 2% year on year.

Step 2: rebuild it yourself — and see why it doesn’t match

1. Divide the quarter’s consolidated PBIDT by its India grey cement tonnes: ₹5,688 crore ÷ 42.41 million tonnes ≈ ₹1,341 per tonne.
2. For the full year: ₹17,598 crore ÷ 145 million tonnes ≈ ₹1,214 per tonne.
3. The homemade quarterly figure is about ₹88 higher than the company’s. That’s not an error on either side; the bases differ. The numerator here is all of consolidated PBIDT — white cement and value-added products, overseas plants and, depending on how the company defines PBIDT, other income — while the denominator is only India grey cement tonnes. The company’s figure is labelled “operating”, and the release doesn’t set out exactly which lines and tonnes go into it.

The lesson: use a company’s own per-tonne series to follow its trend, use your own to cross-check, and never compare your version for one company with the reported version for another.

Step 3: how full are the plants?

1. The release gives capacity utilisation of 89% for the quarter.
2. Rebuild it roughly. Domestic grey capacity was 200.1 MTPA after 8.7 MTPA was added in April 2026, so about 191.4 MTPA at 31 March 2026. The quarter’s 42.41 million tonnes × 4 ≈ 169.6 million tonnes a year, and 169.6 ÷ 191.4 ≈ 88.6% — close to the reported figure.
3. Annualising one quarter is rough: construction activity varies through the year, so a strong quarter overstates the year, and the company may measure capacity on a slightly different base (for example, averaging capacity added mid-quarter).

Step 4: what volume and utilisation did to margins

Item	FY26	FY25
Net sales growth	16.6%
PBIDT growth	32.3%
PBIDT margin (PBIDT ÷ net sales)	20.1%	17.8%
Capex, ₹ crore (share of net sales)	9,600 (11%)

1. PBIDT grew about twice as fast as net sales, and the margin rose — the operating-leverage pattern. The release credits “strong volume growth, cost discipline, and the progressive integration of acquired assets”.
2. That last phrase matters: FY26 isn’t a clean like-for-like year. Part of the growth came from integrating acquired plants, so not all of it is organic, and the release doesn’t split the two.
3. Growth in cement is built in steps: the company commissioned 8 MTPA during FY26 and spent ₹9,600 crore on capex — about 11% of net sales, against Britannia’s 2.1% in FY25 (see capex intensity). New capacity usually runs below full rate at first, which pulls utilisation down until demand fills it.

None of this is a view on UltraTech. It’s one year’s release, read per tonne.

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

A bakery has one big oven. Paying for the oven, the rent and the baker costs the same whether it bakes 100 loaves a day or 400.

If the oven is only a quarter full, each loaf has to carry a big share of those costs, and the bakery barely makes money. If the oven is nearly full, each loaf carries only a small share, and most of the price is profit.

“How full is the oven?” is capacity utilisation. “How much profit on each loaf?” is profit per tonne. A cement company is a bakery with a very, very big oven.

Common mistakes

• Mixing bases in a per-tonne number. Consolidated profit divided by one segment’s tonnes (as in Step 2) gives a number that looks precise and isn’t comparable with anything. Match the profit to the tonnes that earned it, or use the company’s own series.
• Comparing per-tonne profit across companies without asking why it differs. Distance to market, the fuel mix, captive power and the regions a company sells in all move cost per tonne. A gap between two companies is a question, not an answer.
• Treating new capacity as new sales. Capacity is the ceiling, not the volume. Utilisation often falls when big capacity arrives, and rising capacity with flat utilisation means more tonnes sold, not a fuller plant.
• Ignoring acquisitions in year-on-year growth. When a company buys plants, volumes and profits jump without any change in the underlying business. Look for the company’s own note on acquired assets before reading growth as organic.

Takeaway: Read a cement company per tonne, not per rupee of sales: profit per tonne strips out price swings, and capacity utilisation shows how hard the fixed plant is working. Keep the numerator and denominator on the same base, or the per-tonne number means nothing.