The free allocation deduction does not grow with the emissions declared, so every tonne a default adds above actual data is a full certificate, and the mark-up grows each year.
Declaring CBAM default values instead of the supplier's verified actual emissions costs the whole difference in embedded emissions at the full certificate price, because the free allocation deduction is fixed by the product benchmark and does not grow with the figure you declare. On an illustrative 5,000 tonnes of a steel product, a marked-up default of 2.64 tCO2e a tonne against a verified 1.90 costs 277,500 euros more in 2026, 55.50 euros a tonne, at 75 euros a certificate. By 2028, with the mark-up at 30 per cent, the gap is 457,500 euros a year.
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 →
For goods imported from 2026, a declarant may use default values published by the Commission when it cannot obtain actual emissions from the installation. They are set by country of origin and product, and for most goods they carry a mark-up that rises from 10 per cent for 2026 to 20 per cent for 2027 and 30 per cent from 2028, so that defaults are dearer than typical actual data and suppliers have a reason to provide it. Check the current table for your CN code: the values and mark-ups are set in implementing acts and can change.
The intuition many budgets carry is that in 2026 only 2.5 per cent of embedded emissions is charged, so the default can only cost 2.5 per cent of its excess. That is wrong for the reason worked in how the CBAM free allocation adjustment is calculated: the deduction is benchmark times the share retained, so every extra tonne declared is a full certificate.
| Input | Value |
|---|---|
| Imports a year | 5,000 t |
| Verified actual embedded emissions | 1.90 tCO2e/t |
| Country default before mark-up (illustrative) | 2.40 tCO2e/t |
| Mark-up: 2026, 2027, 2028 | 10%; 20%; 30% |
| Product benchmark (illustrative) | 1.37 tCO2e/t |
| Free allocation retained: 2026, 2027, 2028 | 97.5%; 95%; 90% |
| Certificate price, held flat (illustrative) | 75.00 euros |
| Annual cost of obtaining verified data (illustrative) | 25,000 euros |
The benchmark is held the same under both routes so that the effect of the default is isolated. Where the rules attach a different benchmark to goods declared on default values, use that one.
1. Default declared, 2026 = 2.40 × (1 + 10%) = 2.64 tCO2e a tonne
2. Free allocation adjustment = 1.37 × 97.5% = 1.33575 a tonne, the same on both routes
3. Certificates a tonne: actual 1.90 − 1.33575 = 0.56425; default 2.64 − 1.33575 = 1.30425
4. Cost: actual 0.56425 × 5,000 × 75 = 211,594 euros; default 1.30425 × 5,000 × 75 = 489,094 euros
5. Gap = (2.64 − 1.90) × 5,000 × 75 = 277,500 euros; 3,700 extra certificates
Excel: =(Default*(1+MarkUp)-Actual)*Qty*Price
The benchmark and the share retained cancel out of the difference. Whatever the year, the gap is the excess tonnes times the certificate price. On the default the bill is 2.31 times the verified one in 2026.
Of the 277,500 euros, 187,500 is the default itself being higher than this supplier's actual intensity, and 90,000 is the 10 per cent mark-up. An importer whose supplier really does emit at the country default still pays the mark-up for not having the data.
| Year | Default declared | Cost on actual | Cost on default | Gap | Gap a tonne | Gap / data cost |
|---|---|---|---|---|---|---|
| 2026 | 2.64 | 211,594 | 489,094 | 277,500 | 55.50 | 11.1× |
| 2027 | 2.88 | 224,437 | 591,937 | 367,500 | 73.50 | 14.7× |
| 2028 | 3.12 | 250,125 | 707,625 | 457,500 | 91.50 | 18.3× |
| Three years | 1,102,500 |
The gap grows with the mark-up, not with the phase-in. At an illustrative 25,000 euros a year for monitoring, verification and the time it takes to get the data out of the supplier, verified data on this flow pays for itself eleven times over in 2026. It would break even on a flow of about 450 tonnes a year, or at a gap of 0.0667 tCO2e a tonne on 5,000 tonnes.
| Actual, tCO2e/t | Gap, euros a tonne | Gap, euros a year |
|---|---|---|
| 1.60 | 78.00 | 390,000 |
| 1.90 | 55.50 | 277,500 |
| 2.20 | 33.00 | 165,000 |
| 2.40 | 18.00 | 90,000 |
| 2.64 | 0.00 | 0 |
| 2.80 | −12.00 | −60,000 |
The cleaner the supplier, the more the default costs. Only a supplier above the marked-up default, 2.64 here, would be cheaper to declare on defaults, and the mark-up schedule pushes that line higher each year. A supplier that knows its emissions are low has every reason to send the data; one that does not send it is telling you something.
The cash profile, the certificate buffer and what supplier data is worth on the book's own importer are in the free workbook for this case; the embedded emissions themselves are built in how to calculate CBAM embedded emissions for a steel product.
Yes, default values published by the Commission can be used when actual emissions are not available, but for most goods they carry a mark-up rising from 10 per cent for 2026 to 30 per cent from 2028. On an illustrative 5,000 tonnes the default route costs 277,500 euros more than verified data at 1.90 tCO2e a tonne in 2026.
No. The free allocation deduction is the benchmark times the share retained, the same whatever you declare, so the extra tonnes are paid in full. The illustrative 3,700 extra tonnes cost 277,500 euros at 75 euros, not the 6,938 euros that 2.5 per cent of them would suggest.
When the annual cost of getting it is below the excess tonnes times the certificate price. At an illustrative 25,000 euros a year and 75 euros a certificate, the data pays on 5,000 tonnes once the marked-up default exceeds actual by 0.0667 tCO2e a tonne; with a gap of 0.74 it breaks even at about 450 tonnes a year.
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
Reading guide: regulation, compliance and banking → · All 1340 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.