The UK's £9.69 CBAM Steel Rate: A Small Number That Hides a Much Larger Pivot to Actual Emissions Reporting

HMRC's 16 October 2026 package of CBAM guidance has been read by much of the trade press as a pricing story: an illustrative iron and steel sectoral rate of £9.69/tCO2e, built from a Q3 2026 UK ETS average of £58.15/tCO2e and a 2027 free-allocation phase-out factor of 97.5%. That framing is misleading in one important respect. The rate is a mechanism that will reprice every quarter; the default emissions values published alongside it, and the rules governing how they may be combined with actual data, are what will determine whether an importer pays a rounding error or a meaningful cost per tonne.

What the rate actually tells you

The calculation chain is straightforward once laid out. The CBAM rate is the ETS price multiplied by the share of a sector's emissions that is not covered by free allocation, adjusted for the phase-out:

Rate = ETS price × [1 − (baseline free allocation % × phase-out factor)]

HMRC published the cement sector's baseline free allocation (87.1%) but not steel's. Working backwards from the published £9.69 is therefore the only route to the steel figure — and it is an estimate, not an HMRC disclosure:

The implied ~85.5% is internally consistent: the same formula reproduces cement's published £8.78 (calculated £8.77, rounding). Two sensitivities follow directly, and both matter more than the level itself:

• Each £1/tCO2e move in the quarterly ETS average shifts the steel rate by about £0.17/tCO2e.
• Each percentage point of baseline free allocation shifts it by about £0.57/tCO2e — nearly three times as much. The sub-installation list and free allocation data are still "subject to further checks" by the UK ETS Authority, so the number that travels is the one most likely to be revised.

HMRC's own framing is emphatic: the figure is illustrative, creates no liability and is not a forecast. Binding Q1 2027 rates will be published before the end of 2026 and updated quarterly thereafter.

The number that matters: cost per tonne of steel

Multiplying the rate by the published default emissions values gives the per-tonne exposure when an importer cannot or does not use actual data. The default values below are as reported by SteelOrbis; the £/t figures are my own arithmetic, and are indicative of scale rather than a tax computation (carbon-price relief for origin-country carbon paid, precursor treatment and the precise legal mechanics sit in the Finance Act 2026 and the CBAM Calculation Regulations).

The pattern is the analytically useful part, and it contains one genuine anomaly. DRI at 1.149 tCO2e/t looks low for a product whose dominant route (gas-based shaft furnace) typically embeds well over 2 tCO2e per tonne before any downstream processing; it is closer to what the credit side of a DRI-EAF chain would show on a scope-1-plus-2 basis once electricity is largely decarbonised, and to trade-weighted averages dominated by low-carbon origins. Importers of DRI should not assume the default flatters them — it may, and it may not survive to 2029.

The semi-finished and flat-product values cluster in a 2.6–3.6 tCO2e/t band. That band is what a trade-weighted, EU-excluded average of UK import origins looks like in practice: it implicitly credits the low-carbon share of supply rather than penalising the marginal tonne. Two structural decisions drive this:

1. EU data is excluded. HMRC's stated rationale is that defaults should reflect the emissions of imports most likely to rely on them — and, separately, that a future UK–EU ETS linkage would exempt EU-origin goods from UK CBAM anyway. The linkage commitment is the more consequential signal: it makes the EU exclusion a temporary design choice rather than a permanent carve-out, and it means any linkage deal changes both the default arithmetic and the competitive position of EU-origin steel.
2. A flat 10% markup applies to all sectors except fertilisers, justified as ensuring default values give "no advantage" over actual reporting. This is a behavioural lever, not a revenue one: it is calibrated to make verification worth the cost.

The rules that will actually drive behaviour

HMRC's guidance on combining data sources is where the real incentives sit, and it is asymmetric by design:

• Default values may be used for precursor goods while actual data is used for the downstream conversion process.
• Actual data may not be used for precursors while a default is used for the finished product.
• A return filed on default values cannot later be amended with actual data.

The logic is defensible — a default final-product value already embeds an average precursor, so allowing real precursor data would double-count the benefit — but the practical effect is to push importers toward a single early decision: either build verified actual-emissions data for the whole chain, or accept the default and its 10% markup for the year. There is no partial-credit route and no retrospective correction. For a mill selling into the UK, that turns emissions verification from a compliance cost into a commercial prerequisite for competitive pricing.

What to watch before January

The 2027–28 regime is a two-year transition. A single default applies per CN code regardless of origin until 2029, when country-specific defaults arrive and the markup is reviewed — the point at which origin becomes a genuine cost differentiator in the UK market. Between now and then:

• The binding Q1 2027 rates (due before end-2026) will move with the Q4 2026 ETS average and with any revision to the steel baseline free allocation. Watch the baseline, not the ETS price: it carries roughly three times the sensitivity per unit.
• The UK–EU ETS linkage is the single largest variable. It would remove EU-origin goods from CBAM scope entirely, shrinking the default-value sample further and reshaping the relative position of EU versus third-country suppliers.
• Registration opens 1 January 2028, a full year after the mechanism takes effect — a window in which liability accrues but administrative enforcement does not yet bite. Importers should treat 2027 as the year to establish verification capability, not to test the default route.

At £9.69/tCO2e, the sectoral rate is small enough that the default route looks tolerable in 2027. The £25–35 per tonne of flat and long product it implies, combined with a one-way door on amending returns, is what should drive the decision.