Steel is CBAM’s widest-reaching sector — the rules follow the metal from raw ore all the way to the screws in a shipping crate.

In this article:

• Why CBAM steel scope covers far more than crude steel — down to fasteners and finished structures
• How the BF-BOF and EAF production routes lead to very different carbon costs
• Why steel’s precursor chains are the most complex of any CBAM sector, and how mass loss during cutting and machining inflates emissions per tonne
• When the “bubble” simplification is available — and when it isn’t
• What producers, fabricators, and importers should prioritize right now

If you make, trade, or import anything built from iron or steel, this post is for you — and that’s a bigger group than most people assume. CBAM (the EU’s Carbon Border Adjustment Mechanism) requires importers of certain carbon-intensive goods to buy certificates reflecting the emissions embedded in making them, mirroring the carbon price EU manufacturers already pay under the EU Emissions Trading System. It closes a loophole where imports could undercut EU products simply by skipping the carbon cost. For steel, that requirement reaches deeper into the supply chain than in almost any other CBAM sector.

Steel Has the Widest Product Scope in All of CBAM

Most people hear “CBAM and steel” and picture slabs, coils, or rebar arriving at a European port. That’s a small part of the picture. CBAM’s iron and steel sector is organized into seven “aggregated goods categories,” spanning an extraordinary range of products from raw sintered ore through crude steel to fully finished goods:

• Sintered ore — agglomerated iron ore and pellets (CN 2601)
• Pig iron — the semi-refined output of a blast furnace (CN 7201, part of CN 7205)
• Ferro-alloys — ferro-manganese, ferro-chrome, and ferro-nickel (CN 7202, specific subcategories)
• DRI (Direct Reduced Iron) — iron reduced without melting, an alternative iron source to pig iron (CN 7203)
• Crude steel — ingots and primary semi-finished forms across carbon, stainless, and alloy grades (CN 7206, 7207, 7218, 7224)
• Iron or steel products — the catch-all covering everything downstream, semi-finished and finished (CN 7205, 7208–7217, 7219–7223, 7225–7229, 7301–7311, 7318, 7326)

That last category is where scope really opens up: flat-rolled sheet and coil, bars, rods and wire, structural sections, rails, pipes and tube fittings, fabricated structures like bridge sections and window frames, tanks and reservoirs, and — worth flagging specifically — fasteners: screws, bolts, nuts, washers, and rivets under CN 7318, plus a broad residual category (CN 7326) for other iron or steel articles. Ferrous scrap (CN 7204) is explicitly excluded, as are a handful of specialty ferro-alloys not named in the regulation.

The practical takeaway: if your business imports structural steel, pipe, rail, tanks, or even a container of bolts into the EU, you are very likely a CBAM steel importer, whether or not you think of yourself as being in the “steel business.”

Two Production Routes, Two Very Different Carbon Footprints

Nearly all crude steel in the world is made one of two ways, and CBAM tracks them separately because the emissions gap between them is large.

BF-BOF (Blast Furnace – Basic Oxygen Furnace), often called the integrated route, is the dominant global method. Iron ore is sintered, then fed into a blast furnace along with coke — a carbon-rich fuel that acts as the chemical reducing agent stripping oxygen from the ore. The furnace produces liquid “hot metal” (pig iron), which then goes into a basic oxygen converter along with scrap to make crude steel. Because its core chemistry runs on coke, this route is inherently carbon-intensive — emissions come from fuel combustion, the reduction reaction itself, and the decomposition of limestone used in the process.

EAF (Electric Arc Furnace) starts from a different feedstock entirely: mostly scrap steel, sometimes supplemented with DRI, melted using electricity as the primary energy input plus graphite electrodes. Because post-consumer scrap is treated as carrying zero embedded emissions — its carbon cost was already “spent” in an earlier life — and because electricity and DRI can be produced far more cleanly depending on the power source used, EAF is structurally the lower-carbon route.

A couple of variants exist: a smelting-reduction route can produce pig iron using coal directly instead of coke, and stainless or other high-alloy steels sometimes use a nickel pig iron (NPI) route from nickel ore. DRI itself can be made using natural gas, coal, or — flagged in EU technical guidance as a major decarbonization lever for the industry — hydrogen as the reducing agent.

One detail matters for the math: in the definitive CBAM period, only direct emissions count toward embedded emissions for essentially every steel category except sintered ore. Indirect emissions from purchased electricity aren’t counted, which narrows EAF’s advantage somewhat, since decarbonized electricity is its biggest “clean” lever — but EAF still comes out ahead because it avoids most of the direct fuel and reduction emissions that define the blast furnace route.

BF-BOF runs through a chain of interdependent, coke-fed units; EAF is a shorter, electricity-driven path built mainly on scrap.

The Benchmark Numbers: How Big Is the Gap?

EU technical guidance for the sector publishes worked examples that pull real benchmark figures from the CBAM Free Allocation Adjustment Act. These are the EU’s own default emissions values for the sector, and they make the size of the route gap concrete.

Even the “hardest” EAF case here — stainless steel, which needs carbon-intensive ferro-chrome and ferro-nickel inputs — still comes in noticeably below the BF-BOF benchmark. Plain carbon EAF steel, which skips those alloy inputs, would be expected to show an even wider gap. Route choice is, quite literally, a major driver of your CBAM cost exposure.

Steel is one of the largest product categories moving through CBAM by trade volume, given how many finished goods it covers. We won’t speculate here about which exporting countries are most exposed — that depends on trade data this guidance doesn’t contain — but we’ve covered country-specific exposure in our regional posts, including CBAM in India and CBAM in Austria, both significant steel-producing markets.

Why Steel’s Precursor Chains Are the Hardest in CBAM

Every CBAM sector deals with “precursors” — inputs that are themselves CBAM-covered goods, whose embedded emissions must be traced and added into the final product’s footprint. Steel has this problem worse than any other sector, because its production chain can stack up to four levels deep: sintered ore feeds ferro-alloys, pig iron, or DRI; those feed crude steel; crude steel feeds semi-finished and finished iron or steel products, everything from flat sheet to a finished screw.

A product category can be its own precursor

Here’s a wrinkle that trips people up: a plant that buys steel rod and turns it into screws and nuts is making a product in the exact same CBAM category as its own input (both fall under CN 7318, “iron or steel products”). The rod is still legally a precursor, and its embedded emissions still have to be added to the fastener maker’s own process emissions, even though input and output sit in the same bucket on paper.

Cutting and machining mathematically inflates emissions per tonne

This is the part that catches fabricators off guard. When steel rod, bar, or sheet is cut, drilled, forged, or machined into a finished part, a meaningful share of the input mass ends up as scrap. EU guidance gives a direct example: if 20% of a rod’s mass is cut away making screws, it takes 100 tonnes of rod to produce 80 tonnes of finished screws.

That mass loss matters because of how the calculation works. Precursor emissions get carried into the final good using a “mass ratio” factor — how much precursor input it took to make one tonne of finished output. When cutting waste is high, that ratio rises above 1, so the same total precursor emissions get spread across a smaller final tonnage, concentrating the embedded emissions per tonne of finished product. The cutting step itself isn’t counted as adding new emissions — but the mass loss it causes still pushes up the specific emissions figure per tonne. A fabricator can do everything right on its own process and still watch its per-tonne emissions number climb, purely because of how much material ends up as swarf on the shop floor.

Emissions accumulate at every tier — and shrinking output mass at the finishing stage concentrates them further, tonne for tonne.

Some rules that ease the burden — and one that doesn’t

A few rules help. If a precursor originates in the EU or an exempted country, its embedded emissions count as zero downstream. And where a plant produces two or more goods in the chain — say, pig iron and crude steel at the same integrated site — with no intermediates ever sold externally, it can define one single “bubble” production process covering the whole chain, sharply simplifying monitoring.

That simplification has a hard boundary: it only applies when no intermediate product leaves the installation for sale. The moment any crude steel, pig iron, or DRI is sold externally, the operator must track production processes and precursor flows separately — a materially heavier compliance lift.

What Producers and Importers Should Do Now

Steelmakers: know which route your product comes from, and if you’re on BF-BOF, know you’re starting from a structurally higher benchmark. That doesn’t mean switching routes overnight — it means being deliberate about scrap sourcing and process efficiency, since verified emissions below the default benchmark are a real cost advantage.

Fabricators and machine shops: precursor data collection is your single highest-leverage compliance task, not a side detail. Given how many downstream products CBAM covers, nearly every fabricator has a purchased precursor whose upstream emissions and route need documenting, tier by tier, back toward the melt shop.

Importers: map your supply chain by CN code and determine, product by product, whether suppliers can provide verified actual data or whether you’ll rely on EU defaults. Defaults run conservative and generally higher than what an efficient EAF producer could show with real data — pushing suppliers toward verified reporting is directly in your financial interest.

How Sprih Helps

Steel is the clearest example of why CBAM compliance is really a supply-chain data problem. With up to four tiers of precursors, route-dependent benchmarks, and mass-loss math that can shift your numbers even when your own process hasn’t changed, tracking this by spreadsheet becomes unmanageable fast — especially for fabricators, structural contractors, and fastener makers several tiers removed from the melt shop.

Sprih’s core product, SustainSense, indexes data from over 120,000 companies and 400,000+ sustainability reports — AI infrastructure for climate, in the same way Bloomberg is infrastructure for finance. We work with enterprise clients across 21+ countries, including major metals and industrial manufacturers, and we’re building CBAM-specific tooling to help importers and their suppliers calculate and manage embedded-emissions data across exactly this kind of multi-tier precursor chain. If your steel supply chain runs several layers deep, that’s precisely the problem we built this for.

Sprih is an AI-native sustainability intelligence company helping enterprises turn climate and compliance data into competitive advantage.