European Parliament votes to pull hundreds of downstream steel and aluminium goods into the carbon border levy, and Europe’s metals supply chain says the fix is still only half built
BRUSSELS, 19 September 2026 – Europe’s carbon border tariff is about to stop being a raw materials instrument. On 15 September the European Parliament voted by 464 votes to 50, with 159 abstentions, to widen the scope of the Carbon Border Adjustment Mechanism to a long list of finished and semi-finished goods that contain steel and aluminium. Within seventy-two hours the reaction from the industries that will live with the consequences had arrived, and it was not the unqualified celebration that Brussels might have expected.
The European steel distribution federation EUROMETAL said on 18 September that the vote moved in the right direction but left the job unfinished. The European Steel Association, EUROFER, which had backed the motion before the plenary, used the moment to press for tighter anti-circumvention rules. German producers’ body Wirtschaftsvereinigung Stahl called for faster trilogue talks and for a remedy for European exporters who continue to carry domestic carbon costs into markets where their competitors carry none.
For importers, customs brokers and procurement teams outside the European Union, the significance is straightforward and large. Until now, CBAM has applied to a narrow band of commodity inputs: iron and steel, aluminium, cement, fertilisers, electricity and hydrogen. A bolt, a spring, a length of wire, a metal household article or a fabricated component could enter the EU market carrying embedded carbon that no importer ever had to declare or pay for. The Parliament’s position would end that exemption for a very large number of product lines, and would do so at a moment when the European steel market is already being reshaped by a far more restrictive tariff-rate quota regime.
What the Parliament actually approved
The vote concerned the geographic and product scope of CBAM rather than its architecture. The European Commission had adopted a proposal in June 2026 to extend the mechanism to roughly 180 additional steel-intensive and aluminium-intensive downstream products, with application from 2028. Parliament went considerably further. According to analysis published by EUROMETAL on 18 September, the position adopted on 15 September could see the mechanism cover well over 400 downstream products.
The categories named in the plenary debate are instructive because they are ordinary industrial and consumer goods rather than commodity metal. Fasteners, wire, springs and household articles were all cited. These are precisely the product families that European service centres, rerollers and small fabricators compete against, and precisely the families that have been outside the reach of both CBAM and the tariff-rate quotas established under the EU’s Steel Regulation.
The measure now goes into negotiation between Parliament and the member states in the Council. The final scope, the phase-in calendar and the treatment of exporters will all be settled in that process, which means that the number of covered product lines is still a variable rather than a fixed figure. Importers should treat the 400-plus figure as the upper end of a negotiating range, not as a legislated outcome.
Industry reaction: a step, not a solution
Alexander M. Julius, president of EUROMETAL, gave the sharpest assessment. “Today’s vote is a step in the right direction, but it does not yet deliver a level playing field for European industry. The proposed scope remains incomplete, implementation is too slow, and there is still no solution for EU exporters carrying carbon costs when competing globally,” he said in comments reported by EUROMETAL.
Julius went on to frame the problem as one of value chain coverage rather than individual product coverage. “Moreover, CBAM alone cannot offset the broader cost disadvantage faced by European manufacturers due to higher steel prices and regulatory burdens. Europe must protect the entire value chain if it wants to prevent carbon leakage and deindustrialization,” he added.
That formulation matters. It concedes something that European policymakers have been reluctant to say out loud, which is that a border carbon adjustment is a partial answer to a competitiveness gap that has several other causes, including energy prices, permitting timelines and the cumulative compliance load of EU industrial regulation.
EUROFER’s position was supportive but conditional. Axel Eggert, director general of the association, argued before the vote that the mechanism only works if it is strong enough to justify the capital European mills are committing to decarbonisation. “Europe’s steelmakers are investing billions to produce cleaner steel, but they cannot make that transition without a level playing field. MEPs have the opportunity to strengthen CBAM so that investing in low-carbon steel production in Europe makes economic sense,” Eggert said, as reported by EUROMETAL.
He put the downstream logic more plainly in separate comments carried in the same reporting: “Extending CBAM to more steel-intensive products would help ensure that steel produced in Europe and steel contained in imported products compete under comparable carbon conditions.”
Wirtschaftsvereinigung Stahl, the German steel federation, added a procedural demand, saying it was now crucial for Parliament to reach agreement quickly with the Commission and the Council so the regulation can be implemented. German mills are among the most exposed to the export-side problem, since a substantial share of their output serves customers outside the single market.
The context: a market already reshaped by quotas
The CBAM vote did not land in a calm market. It landed in one that has been transformed during 2026 by the EU’s new steel import framework.
Regulation (EU) 2026/1384 was published in the Official Journal on 24 June 2026, entered into force on 25 June and began applying on 1 July. It replaced the previous safeguard architecture with something considerably more restrictive. Tariff-free quota volumes were cut by 47 per cent against 2024 levels, to 18.3 million tonnes a year. The out-of-quota duty was doubled to 50 per cent. The number of separate quotas rose to 30. A melt-and-pour origin requirement, obliging importers to evidence where the steel was actually made rather than where it was last processed, applies from 1 October 2026.
The price effect has been immediate and measurable. Platts, part of S&P Global Commodity Insights, assessed domestic hot-rolled coil in Northern Europe at 730 euros per tonne ex-works Ruhr and Southern European domestic HRC at 725 euros per tonne ex-works Italy in mid-September, both up 110 euros per tonne since the start of the year. Imported HRC was assessed at 585 euros per tonne CIF Antwerp and 580 euros per tonne CIF Southern Europe, both up 85 euros per tonne over the same period. The MEPS Europe Average hot-rolled coil price has risen by more than 16 per cent between January and September.
That is the crux of the downstream grievance. European fabricators are paying materially more for their principal input because the quota regime has restricted their access to imported coil, while their competitors abroad buy at world prices and ship finished articles into the EU market that face neither a quota nor a carbon charge. EUROMETAL’s argument, set out in its 18 September analysis, is that stringent trade defences on upstream steel “shift rather than eliminate import pressure” and in doing so undermine the competitiveness of the downstream supply chain.
The association also makes a carbon leakage argument that cuts against the environmental rationale for the current design. If third-country steelmaking and fabrication is more emissions intensive than the European equivalent, and if finished goods made with that steel enter the EU untaxed, the regime displaces emissions rather than reducing them.
Circumvention through modification
A recurring theme in the September reporting is what market participants describe as subtly modified steel products entering the EU outside the reach of trade defences. MEPS respondents told the research firm that the influx of such products has increased.
The mechanism is familiar to anyone who has worked trade remedies. A product is altered just enough to fall outside the tariff classification covered by a measure, or just enough to qualify as a downstream article rather than a covered semi-finished good. Width adjustments, minor coating changes, marginal additions of alloying elements and light further processing are all standard techniques. Vietnam’s own trade remedies authority is currently working a case built on exactly this pattern, examining whether hot-rolled coil from China wider than 1,880 millimetres and up to 2,300 millimetres was produced to that specification specifically to escape anti-dumping duties imposed on narrower coil.
EUROFER’s call for stronger anti-circumvention rules in the CBAM text is a direct response to this. A carbon border mechanism that covers a bolt but not a bolt assembly, or a wire but not a wire product, simply relocates the entry point. The wider the product list, the fewer such entry points remain, which is why the association and the distribution federation both pushed for maximal scope even though they represent different and sometimes opposed interests within the chain.
Economic impact: who pays and how much
The direct cost of CBAM extension falls on importers of record in the European Union, who must purchase certificates matching the embedded emissions of the goods they bring in, net of any carbon price already paid in the country of production. For commodity steel the calculation is comparatively tractable because emissions intensity per tonne of crude or rolled steel is well documented and increasingly subject to verified reporting.
For downstream goods the calculation becomes considerably harder. A fastener contains a small quantity of steel that may have passed through three or four processing stages in two or three jurisdictions. Establishing the embedded carbon of that item requires either supplier-level data that many small exporters do not hold, or the use of default values, which are typically set conservatively and therefore penalise efficient producers who cannot document their performance.
This is the compliance cost that will bite hardest outside Europe. Exporters in Turkey, India, Vietnam, Korea, Taiwan and China who currently ship finished metal goods into the EU without any carbon reporting obligation will need to build emissions accounting systems for products they may sell in small volumes and at thin margins. Some will conclude the European market is not worth the overhead. That is a market access effect that will not appear in any tariff schedule.
There is also a second-order effect on European importers’ working capital. Certificate purchase obligations, combined with the quarterly rhythm of the tariff-rate quotas, favour buyers with balance sheet depth. EUROMETAL’s September analysis is explicit on this point: large distributors and service centres that can fund port warehousing are able to clear imported material through customs at the start of each quota period, and can fund inventory ahead of anticipated price rises. Smaller independents cannot. The regime is, in practice, a consolidating force in European metals distribution.
The near-term price outlook offers some relief. MEPS reported in September that high inventories and soft demand will temper further increases in the coming months, and that some European distributors are selling material at prices significantly below current mill offers as they convert stock into cash before year end. That is a destocking signal rather than a demand signal, and it suggests the 2026 price surge has run ahead of underlying consumption.
Implications for importers and exporters
For companies shipping metal-containing goods into the European Union, four practical conclusions follow from this week’s developments.
First, product classification work needs to start now, not in 2028. The negotiating range between the Commission’s roughly 180 products and Parliament’s 400-plus means that any exporter of a fabricated metal article should assume a meaningful probability of coverage and should map its EU-bound tariff lines against both lists. Where a product sits near a classification boundary, that boundary is likely to be contested and is likely to move.
Second, emissions data is becoming a commercial asset. Exporters who can produce verified, product-level embedded carbon figures will pay less than those who fall back on default values. Building that capability takes a reporting cycle or two, which means starting in 2026 rather than 2027. Suppliers who can hand a European customer an auditable carbon declaration will win business on that basis alone.
Third, the melt-and-pour requirement that applies from 1 October 2026 under the Steel Regulation is a separate and more immediate compliance event. It obliges importers to evidence the country where the steel was melted and poured rather than the country of last substantial transformation. Supply chains routed through a processing hub to change nominal origin will not survive that test. Any importer relying on such a route should have documentation in place before the month end.
Fourth, the export side remains unresolved and is worth watching. Both EUROMETAL and Wirtschaftsvereinigung Stahl have pressed for a mechanism to relieve EU exporters of domestic carbon costs when they compete in third markets. No such mechanism exists in the current text. If one emerges from the trilogue, it will alter the competitive position of European fabricators in markets from the Gulf to Southeast Asia, and it will very likely attract scrutiny at the World Trade Organization on subsidy grounds.
The wider pattern
Viewed from outside Europe, the CBAM extension is one element of a broader consolidation of the European market behind a layered wall: anti-dumping duties on specific origins, a hard tariff-rate quota with a 50 per cent out-of-quota rate, an origin-tracing requirement, and now a carbon charge reaching progressively further down the value chain. Each instrument addresses a different perceived gap, and each closes an avenue that exporters had used to work around the previous one.
The pattern is not unique to Europe. Indonesia opened an anti-dumping investigation into Chinese galvanised steel on 15 September. Australia’s Anti-Dumping Commission has a continuation inquiry running on zinc-coated steel from India, Malaysia and Vietnam. Vietnam is finalising an anti-circumvention case against Chinese hot-rolled coil. The global steel trade is being renationalised measure by measure, and the carbon dimension is simply the newest layer.
What distinguishes the European approach is its ambition to price the externality rather than merely restrict the volume. Whether that ambition survives contact with the Council, and whether the resulting instrument is administrable for the small exporters it will capture, are the questions the trilogue will answer over the coming months.
For now, the message from Brussels to the world’s metal fabricators is that the European market is still open, but the price of entry is rising and the paperwork is about to get considerably harder.
The administrative question nobody has answered
Beneath the politics sits an administrative problem that has received far less attention than it deserves, and which may ultimately determine whether the extended mechanism functions at all.
Calculating embedded emissions in a tonne of hot-rolled coil is a bounded exercise. There are a limited number of production routes, the energy inputs are large and metered, and the sector has been reporting emissions under the EU Emissions Trading System for nearly two decades. The methodology is contested at the margins but the arithmetic is tractable.
Calculating embedded emissions in a box of screws is a different problem entirely. The steel in those screws may have been melted in one country, hot-rolled in a second, drawn into wire in a third and formed, heat-treated and plated in a fourth. Each stage has its own emissions profile, its own electricity mix and its own data availability. The exporter of record, who is typically the final processor, may have no visibility of the upstream stages and no commercial leverage to obtain it.
The European system’s answer to missing data is default values. Defaults are deliberately set at conservative levels, meaning they assume a high-emissions production route, because a default that flattered the importer would create an incentive never to report. The practical consequence is that a small exporter who cannot document its supply chain pays a penalty rate regardless of how clean its actual production is.
This creates a regressive effect that runs contrary to the mechanism’s stated purpose. A large multinational with integrated reporting systems and dedicated compliance staff will produce verified figures and pay accordingly. A small producer in Turkey, Malaysia or Tunisia with genuinely low emissions but no reporting infrastructure will pay the default. The instrument will therefore tend to consolidate European import supply into the hands of larger exporters, which is a competition outcome rather than a climate outcome.
Industry submissions during the legislative process raised this point repeatedly. The Parliament’s text does not resolve it, and the trilogue will need to decide whether to build a simplified compliance route for small consignments and small exporters, or to accept that the mechanism will function as a de facto barrier to smaller foreign suppliers.
Trade law exposure
There is also a legal dimension that international exporters will be watching.
The Carbon Border Adjustment Mechanism has always attracted scrutiny under World Trade Organization rules, principally on national treatment and most-favoured-nation grounds. The European Union’s defence has been that the mechanism mirrors a domestic carbon cost rather than imposing an additional one, and that it therefore equalises rather than discriminates. That argument is strongest where the imported product is directly comparable to a domestic product covered by the Emissions Trading System.
Extending coverage to downstream articles complicates that defence. A fastener made in Europe is not itself covered by the Emissions Trading System; only the steel inside it is, and only at the point of steelmaking. Charging an imported fastener for its embedded carbon therefore requires the European Union to argue that the domestic equivalent bears an indirect cost passed through the supply chain. That is a coherent argument in economics and a harder one in trade law, because the pass-through is neither measured nor guaranteed.
The dispute settlement environment reduces the practical risk. The World Trade Organization’s Appellate Body remains unable to hear appeals, and a panel report against the European Union could be appealed into a void unless the complainant were a party to the interim appeal arbitration arrangement. That does not mean no challenge will come. It means any challenge is more likely to be pursued politically, through retaliatory measures or negotiated carve-outs, than through binding adjudication.
Several affected exporting nations have already raised the mechanism in WTO committees. India has been among the most vocal, arguing that unilateral climate-linked trade measures place a disproportionate adjustment burden on developing economies. That objection will grow louder as the product scope widens, and it intersects awkwardly with the European Union’s simultaneous effort to conclude and ratify a free trade agreement with India.
What to watch over the next six months
Three markers will tell exporters how this resolves.
The first is the trilogue outcome on scope. If the final product list lands above 300 items, the mechanism becomes a genuine value chain instrument and importers of fabricated metal goods need full compliance programmes. If it lands near the Commission’s original 180, the downstream gap persists and the political pressure that produced the September vote will return.
The second is the implementation date. The Commission proposed 2028. Industry has called the timeline too slow. Any acceleration compresses the window in which exporters can build emissions accounting capability, and would be the single most disruptive element of the final text.
The third is whether an export-side mechanism appears. Both EUROMETAL and Wirtschaftsvereinigung Stahl have demanded one. If the trilogue produces some form of relief for European exporters carrying domestic carbon costs into third markets, that changes competitive dynamics in every market where European fabricators compete, and it will almost certainly draw a formal challenge on subsidy grounds.
None of these will be settled quickly. Trilogue negotiations on contested industrial files routinely run for six to twelve months, and the Council has member states on both sides of the argument. Exporters should plan for uncertainty through 2027 and for a compliance obligation arriving somewhere between 2028 and 2030.
What is no longer uncertain is the direction. The European Union has decided that its climate policy will be enforced at the border, and that the border will be drawn progressively further down the value chain. Everything else is timing.