How do you calculate CBAM embedded emissions for a steel product?

The Annex IV formula for a complex good is one line. The precursor carries almost all of the answer, and three easy errors take hundreds of tonnes off it.

Julian R. Sterling · from The CBAM Compliance Handbook

For a complex good such as a steel fastener, specific embedded emissions are the direct emissions attributed to the product's own process plus the embedded emissions of the precursors it consumed, divided by the tonnes of product made. On an illustrative plant that emits 2,400 tonnes of CO2e to turn 21,500 tonnes of wire rod at 1.90 into 20,000 tonnes of fasteners, the answer is 2.1625 tonnes of CO2e per tonne, and 94.5 per cent of it comes from the wire rod, not the fastener plant.

Worked in full in The CBAM Compliance Handbook by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →

The formula

Annex IV of Regulation (EU) 2023/956, the CBAM Regulation, defines the specific embedded emissions of a complex good as the attributed emissions of its production process plus the embedded emissions of its relevant precursors, all over the activity level, which is the mass of the good produced. The precursor term is itself a sum: the mass of each precursor consumed times its own specific embedded emissions. For iron and steel, listed in Annex II, only direct emissions count during the definitive period; the emissions of the electricity consumed in the process are not added. For cement and fertilisers, indirect emissions are counted as well.

SEE_g = (AttrEm_g + EE_InpMat) / AL_g

EE_InpMat = Σ M_i × SEE_i, over the relevant precursors consumed

Excel: =(DirectEm+SUMPRODUCT(PrecursorTonnes,PrecursorSEE))/OutputTonnes

The assumptions

A fictional fastener maker outside the EU buys hot-rolled wire rod from a blast furnace producer and exports part of its output to an EU importer. Every figure is illustrative; the supplier's 1.90 is a plausible order of magnitude for the route, not a published value.

Plant data for the reporting period
Input	Value
Fasteners produced (activity level), tonnes	20,000
Attributed direct emissions of the fastener process, tCO2e	2,400
Emissions from electricity consumed, tCO2e (not counted for steel)	1,600
Wire rod consumed, tonnes	21,500
Supplier's actual SEE of the wire rod, tCO2e per tonne	1.90
Fasteners imported into the EU in the year, tonnes	3,000

The calculation, step by step

Start with the precursor, because it is the large number. The plant consumed 21,500 tonnes of wire rod to make 20,000 tonnes of product: 1.075 tonnes in for each tonne out, a 7.5 per cent loss to cutting, heading and scrap. Every tonne consumed carries its embedded emissions, including the ones that end up as offcuts.

EE_InpMat = 21,500 × 1.90 = 40,850 tCO2e

Attributed plus precursor = 2,400 + 40,850 = 43,250 tCO2e

SEE = 43,250 / 20,000 = 2.1625 tCO2e per tonne

Embedded emissions declared = 2.1625 × 3,000 = 6,487.5 tCO2e

Where the 2.1625 comes from
Component	tCO2e	Per tonne of product
Fastener process, direct	2,400	0.12
Wire rod precursor	40,850	2.0425
Specific embedded emissions	43,250	2.1625

The result, and what it costs

The importer declares 6,488 tonnes of embedded emissions on its 3,000 tonnes. That figure is the base for everything that follows: the certificates actually surrendered are reduced by the adjustment for free allocation in the EU ETS and by any carbon price effectively paid in the country of origin, and in the early years of the definitive period the free-allocation adjustment removes most of the obligation. It falls away completely by 2034. At an illustrative certificate price of 80 euros, the full exposure on this volume once free allocation has gone is 519,000 euros, or 173.00 per tonne of fasteners, and the cost to model is that one, not the 2026 bill. How the obligation steps up year by year is worked in how much CBAM will cost per tonne of steel from 2026 to 2034.

The fastener plant's own furnaces account for 0.12 of the 2.1625. A decarbonisation plan that starts at the fastener plant is working on 5.5 per cent of the number; the supplier choice works on the rest.

What if the precursor changes?

Hold the fastener plant constant and change only the wire rod: an electric arc furnace supplier, the current blast furnace supplier, and no supplier data at all, which forces a default value. Default values under Annex IV are country averages increased by a mark-up; the 2.30 here is illustrative.

Product SEE and full exposure by precursor source
Wire rod source	Precursor SEE	Product SEE	Embedded on 3,000 t	Cost at 80 euros
Electric arc furnace supplier	0.60	0.7650	2,295.0	183,600
Blast furnace supplier, actual data	1.90	2.1625	6,487.5	519,000
No data, illustrative default	2.30	2.5925	7,777.5	622,200

Each 0.10 on the precursor moves the product by 0.1075, because the yield loss multiplies it: 322.5 tonnes on this import volume. Getting the supplier's actual data instead of the default is worth 1,290 tonnes, or 103,200 euros at full exposure. Switching supplier route is worth 4,192.5 tonnes, or 335,400 euros, more than three times as much.

The common mistakes

Dividing by input instead of output. 43,250 over 21,500 gives 2.0116, 7.0 per cent too low: 453 tonnes under-declared on the import, which is the direction penalties are written for.
Counting the precursor one for one. Using 20,000 tonnes of wire rod instead of the 21,500 consumed gives 2.0200 and drops 428 tonnes. The offcuts were emitted too.
Adding electricity for steel. Including the 1,600 tonnes of indirect emissions overstates the SEE by 0.08, 240 tonnes on the import. Right for cement and fertilisers, wrong for iron and steel.
Leaving out the precursor. The fastener plant alone is 0.12 per tonne, 360 tonnes on the import. It is the error that makes a supplier questionnaire look optional.

Takeaway

SEE = (attributed direct emissions + precursor mass consumed × precursor SEE) / product mass.
Use the precursor consumed, not the precursor in the product: yield loss is embedded too.
For iron, steel and aluminium count direct emissions only; for cement and fertilisers add indirect.
Price the supplier's data and route before the plant's own process: that is where the tonnes are.

The quarterly cash profile, the certificate buffer and a blank model for your own imports are in the free workbook for this book.

Questions readers ask

Are indirect emissions included in CBAM embedded emissions for steel?

No. For iron and steel, aluminium and hydrogen only direct emissions are counted during the definitive period; the emissions of the electricity consumed are not added. In the worked case, adding the 1,600 tonnes of electricity emissions would overstate the SEE by 0.08 tonnes per tonne, or 240 tonnes on 3,000 tonnes imported. Cement and fertilisers do include indirect emissions.

What happens if the supplier does not provide actual emissions data?

The precursor, or the good, falls back on a default value, which Annex IV sets at a country average increased by a mark-up. With an illustrative default of 2.30 for the wire rod instead of the supplier's actual 1.90, the fastener SEE rises from 2.1625 to 2.5925 and the import carries 1,290 more tonnes, 103,200 euros at an illustrative 80 euros once free allocation has gone.

Do you use the precursor in the product or the precursor consumed?

The precursor consumed. The plant used 21,500 tonnes of wire rod to make 20,000 tonnes of fasteners, and the emissions of the offcuts were incurred all the same. Counting 20,000 tonnes one for one gives an SEE of 2.0200 instead of 2.1625 and under-declares 428 tonnes on a 3,000 tonne import.

Read the whole case

This article is one calculation from The CBAM Compliance Handbook. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.

Get the book on Amazon →Free companion files

Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada

Also on this site

How much will CBAM cost per tonne of steel from 2026 to 2034?At 2.0 tonnes of CO2 a tonne and an 80 euro certificate, CBAM costs 4.00 euros a tonne of steel in 2026, 77.60 in 2030 and 160.00 in 2034.
Is it worth staying under the CBAM 50 tonne threshold?Crossing 50 tonnes by 5 brings all 55 into CBAM. In 2026 the certificates cost 210 euros; the fixed compliance cost, 15,000 illustrative, is what makes the cliff.
What a CBAM certificate buffer costs to holdCarry and protection are both linear in the buffer, so the break-even probability is the same at 5 per cent and at 30. And 2027 costs exactly twice 2026.
How do you calculate the false positive rate of AML alerts?Alerts that lead to no suspicion report / total alerts. 12,000 alerts and 240 reports a month is a 98.0% false positive rate, and 53.4 analysts to clear them.
At what odds of procurement does an EDF development grant pay back?A partner's matched spend on an EDF development action is repaid by margin alone only if procurement is 58.7 per cent likely. Hold the prototyping and it is 86.7.
Does an additional tier 1 shortfall reduce headroom to the distribution trigger?Yes. A 40.8 million shortfall plugged with common equity costs 13.6 basis points of buffer: headroom is 2.8432 per cent, not the 2.9796 reported.

Reading guide: regulation, compliance and banking → · All 453 articles →

If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.