On 23 October, India’s Commerce Minister Piyush Goyal told an audience in Berlin that the European Union was “inviting disaster” with its carbon tax on steel and its 50% duty on steel imports above quota, warning of infrastructure built “with costlier steel, costlier aluminium, costlier cement, with all the overlay of CBAM and deforestation rules”. Europe’s steelmakers say the opposite: that without the Carbon Border Adjustment Mechanism (CBAM) and the new quotas they cannot survive the transition to clean steel.

Both sides are describing the same problem from different ends. Europe wants green steel that, at European gas, power and hydrogen prices, it cannot yet afford to make. CBAM and the quotas protect the industry that is supposed to make it, but they do so by raising what European factories pay for steel and by shutting out cheaper imports, while doing nothing about the energy prices that are the real obstacle. This piece looks at what “green” and “clean” steel mean, what they cost, how dependent European manufacturing is on gas, what compliance costs exporters, and why a slower, simpler CBAM rollout would serve Europe’s own goals better. It follows this blog’s piece on what CBAM costs Indian exporters per tonne.

Green, clean, low-carbon: the thresholds don’t agree

There is no single definition of green steel. The International Energy Agency’s “near-zero” standard, also used by ResponsibleSteel, allows 0.05 to 0.4 tonnes of CO2 per tonne of crude steel depending on how much scrap is used. Fastmarkets, which publishes the main European price for green steel, counts flat steel as green at up to 0.8 tonnes, including supply-chain emissions. India’s Green Steel Taxonomy, issued by the Ministry of Steel in December 2024, calls steel “green” below 2.2 tonnes per tonne of finished steel, with five stars below 1.6. Berkeley’s India Energy and Climate Center notes that this is loose enough for existing lower-emission Indian routes to qualify already.

Europe’s own rules are still being written. The proposed Industrial Accelerator Act (March 2026) would require at least 25% low-carbon steel in public buildings, infrastructure and vehicles from 2029, but leaves the definition to a later act under the EU ecodesign regulation (ESPR). Draft performance classes for that act, criticised by recyclers and environmental groups in June 2026, would put hot-rolled coil of up to 2.66 tonnes in class B, above the EU blast-furnace average of about 1.9. On those drafts, typical Indian blast-furnace steel at 2.5–2.9 tonnes could sit in the same class as much European steel. CBAM, by contrast, charges Indian steel on a default of 4.7 tonnes in 2026 unless the plant’s own data are verified. The same tonne of steel can therefore be “green” in India, mid-table in an EU product class and heavily penalised at the EU border.

What clean steel costs to make

The estimates vary with place, year and method, but the pattern is consistent. Where hydrogen is expensive, green steel costs far more than blast-furnace steel: Argus’s index for a fully hydrogen-based plant in the EU stood at €1,074 a tonne at the end of 2024, 1.9 times the price of ordinary hot-rolled coil. Where hydrogen and clean power are cheap, the gap narrows to a few percent or disappears; Berkeley’s India Energy and Climate Center expects hydrogen-based steel in India to reach parity with new blast furnaces by 2030, as delivered hydrogen falls toward US$3 a kilogram.

The inputs explain why. A direct-reduction plant uses about 10 GJ of natural gas per tonne of iron, so at the first-half 2026 average Dutch TTF price of €43 a megawatt-hour, the gas alone costs about €120 a tonne; at September’s €80, about €224. The same gas costs about €58 at India’s administered price and €29 at the US Henry Hub price. Switching to hydrogen needs about 54 kg per tonne of finished steel: at the €7–12 a kilogram EUROFER gives for renewable hydrogen in Europe, that is €378–€648 a tonne before anything else is paid for. India’s lowest green hydrogen tender, Rs 279 a kilogram for the Numaligarh refinery in February 2026, implies about €145. Every US$1 a kilogram off the hydrogen price takes roughly US$50–65 off a tonne of steel.

That arithmetic is why Europe’s flagship projects have slowed. ArcelorMittal dropped hydrogen plans for Bremen and Eisenhüttenstadt in June 2025, calling even gas-based direct reduction “not competitive”; Salzgitter postponed the investment decision on the second and third phases of its SALCOS project from 2026 to 2028–29, citing a weak hydrogen market; Thyssenkrupp paused its hydrogen tender because bids came in above its assumptions; and Sweden’s Stegra needed a €1.4 billion rescue round in April 2026. EUROFER’s 2026 annual report says most of its 13 planned direct-reduction plants and 20 electric-furnace projects are delayed by energy costs.

Europe’s manufacturing runs on gas it cannot price

Steel is not alone. The European chemical industry association Cefic puts EU gas at about 3.3 times the US price in January–April 2026. EU chemical output in April 2026 was 16% below its 2014–2019 average, with plants running at about 74% of capacity, and a Roland Berger study for Cefic counted about 37 million tonnes of announced capacity closures in 2022–2025, about 9% of Europe’s capacity, half of them blamed on energy costs. BASF closed one of its two ammonia plants at Ludwigshafen in 2023; Yara is cutting about 1 million tonnes of European ammonia output and importing instead. France’s statistics office INSEE found output at the end of 2024 still 24% below mid-2021 in basic iron and steel, 20% below in glass and 15% below in basic chemicals.

The exposure was shown again this year. When QatarEnergy halted LNG output after strikes on its facilities in March 2026, TTF jumped more than 24% in a day; EUROFER blames the further rise in energy prices on “the U.S.-Iranian conflict and the closure of the Hormuz-Strait”. Europe’s plan for green steel runs through gas-based direct reduction first and hydrogen later. As long as both are priced off a gas market this volatile, European green steel will cost more to make than steel almost anywhere else, whatever the carbon price at the border.

Who pays for green steel

So far, few buyers pay much. Fastmarkets’ premium for green flat steel in northern Europe was €100–170 a tonne in January 2026 and €100–150 in March. Mills offer at €200–300, but buyers say €70–150 is what is achieved, and German service centres reported no plans to increase bookings for 2026. Carmakers are the exception: Stegra says customers have agreed to premiums of 20–30%, and Mercedes-Benz, BMW, Volvo, Porsche and Scania are among its buyers. Steel is a small share of a car’s price, so carmakers can pay; builders of bridges, pipes and machinery mostly cannot.

Set against the CBAM numbers in this blog’s previous piece, the green premium is smaller than the penalty for missing data. An Indian blast-furnace mill that verifies its emissions pays about €93 a tonne of CBAM in 2026; one that relies on EU default values pays about €270. Getting verified data is therefore worth about €180 a tonne to an Indian exporter, more than the whole green premium European buyers pay, while the premium itself is open only to steel below about 0.8 tonnes of CO2, which almost no Indian blast-furnace or coal-based plant reaches.

How CBAM and the quotas shut out cheaper steel

In early October 2026 imported hot-rolled coil was offered at about €595 a tonne delivered to Antwerp (Platts, via EUROMETAL), against €731 for EU coil at the mill (Shanghai Metals Market): a gap of about €136. CBAM on verified Indian data closes most of that gap; on default values it turns the import into the dearer option. Above quota, the 50% duty makes the same tonne cost about €892 before any carbon charge. The quotas are already binding: EUROMETAL reported that fourth-quarter 2026 quotas for the category that includes hot-rolled coil were exhausted for “other countries” on 6 October, with India’s own allocation down to 505 tonnes of 8,364. Asian coil offered duty-paid at Antwerp was about US$851–874 a tonne, roughly the EU domestic delivered price. The discount of about €100 a tonne that EU buyers could get on imports in 2025 has largely gone.

Where the world’s steel prices sit

Global price benchmarks show how far apart the protected and open markets have moved. In the US, behind a 50% tariff, Nucor’s list price for hot-rolled coil was US$1,255 a short ton on 19 October and spot prices were US$1,220–1,300 a short ton (Kallanish via GMK Center), about US$1,345–1,433 a metric tonne and, according to Fastmarkets, 52% up on a year earlier. EU coil at €731 ex-works is about US$841. India’s domestic benchmark, assessed by BigMint at Rs 62,000 a tonne ex-Mumbai in September (excluding GST), its highest since June 2022, is about US$653; Chinese coil for export was offered at US$492 a tonne FOB on 16 October (Mysteel). SteelBenchmarker’s mid-April 2026 report, the latest found, showed the same ranking: US$1,133 in the US, US$840 in western Europe, US$495 for world export and US$416 ex-works in China.

The EU now sits between the open world market and the closed US one, about US$188 a tonne above Indian domestic coil and US$349 above Chinese export offers. CBAM and the quotas move it toward the US model, where domestic users pay well above world prices. Benchmarks with longer histories, such as MEPS’s monthly regional averages, which cover markets accounting for about 70% of world consumption, are sold by subscription and could not be checked for this piece. worldsteel’s benchmarking systems, despite the name, compare plants’ CO2, energy and other performance, not prices; its CO2 data collection, which non-members can join, is the basis of the ISO 14404 standard for measuring a steel plant’s emissions, and is one of the tools Indian mills can use to prepare the plant data CBAM verification needs.

The people paying are Europe’s own manufacturers, and they have said so.

The German Environment Agency’s result is the striking one: replacing free allowances with CBAM barely changes how much production leaks abroad, because the leakage moves downstream to products that CBAM does not cover or covers poorly. Higher steel prices in Europe make imported machinery and components, made from cheaper steel elsewhere, more competitive against EU-made goods. A trade publication headline this year put it bluntly: CBAM “could offer less protection to steelmakers than free ETS allowances”.

The cost of compliance falls on the smallest firms

For an Indian exporter, compliance means measuring emissions plant by plant, having them checked by a verifier accredited in the EU, and passing the data to the EU importer every year. Vendors quote Rs 5–15 lakh (about €5,000–15,000) for verifying one Indian installation; experts put total compliance and certification at Rs 50 lakh to Rs 1 crore per company, and small fastener and steel exporters report paying about Rs 2 lakh a year to consultants. The Indian Steel Association says small firms “will not be able to afford” EU-accredited auditors, legal advisers warn of a shortage of verifiers in India, and at least ten consignments were held at EU ports in January 2026 over incomplete declarations. The government has promised to bear the “entire cost of certification” for micro and small enterprises; no scheme has been notified.

The EU has already simplified CBAM for its own businesses. Its 2025 omnibus regulation exempted importers bringing in less than 50 tonnes a year, removing up to 80% of affected EU importers from the system, with a net saving the Council estimated at €1.19 billion; it also cut the share of certificates importers must hold each quarter from 80% to 50%. The burden on exporters, who must still produce verified installation data for every tonne, was not reduced.

India’s green-steel position

India starts from a high-carbon base. Coal-based rotary kilns make 70–80% of its direct-reduced iron; it accounts for more than 60% of the blast-furnace capacity being built worldwide, according to Global Energy Monitor; and LNG cost about nine times as much as domestic coal in 2023-24, according to IEEFA, so gas-based steel is not a cheap bridge. But its cheapest decarbonisation route may be cheaper than Europe’s. Green hydrogen tenders have fallen to Rs 279 a kilogram, green ammonia auctions to Rs 55.75 a kilogram, and a 20-tonne-a-day hydrogen direct-reduction pilot led by CSIR-IMMT is due by 2029. AM/NS India aims to have 70% of its output rated green under the taxonomy by 2027; Tata Steel has commissioned a 0.75 Mt scrap-based electric furnace plant at Ludhiana. A green public-procurement mandate (20% three-star, 5% four-star and 1–2% five-star steel in large government purchases) is still a draft.

If clean steel can be made more cheaply where power and hydrogen are cheap, CBAM’s long-run effect could be to pull green steelmaking toward places like India, the Gulf and Australia rather than to keep it in Europe. Whether it does depends on whether the rules let a verified low-emission Indian tonne into the EU on terms that reflect its emissions, rather than a default value or a quota exhausted in the first week of the quarter.

The case for a softer rollout

The argument for relaxing how CBAM is implemented, not for abandoning it, rests on four points from the evidence above.

• It does not fix Europe’s real constraint. European green steel is held back by gas, power and hydrogen prices that CBAM does not touch. A border charge that raises EU steel prices without lowering EU energy costs protects existing capacity but does not build the new plants.
• It penalises missing data more than emissions. The 2026 gap between default and verified charges on Indian coil, about €177 a tonne, is larger than the green premium Europe pays. For small exporters the cost is mostly paperwork and verifier capacity, not carbon.
• Its costs land on European manufacturers. Steel users estimate €5–9 billion a year in extra tariff costs from the quotas alone, and the German Environment Agency finds CBAM barely reduces leakage.
• The EU has already relaxed it for itself. The 50-tonne exemption and the July 2026 proposal to slow the phase-out of free allocation (COM(2026) 616) ease the burden on EU firms; the exporter side has had no equivalent relief.

The counter-arguments are real. EUROFER says CBAM already “falls short” and wants it widened to downstream goods and tightened against circumvention, and relief for one country could open a route for steel melted elsewhere to enter through it. CBAM must also stay origin-neutral to remain defensible at the WTO, so any relaxation is more defensible if it applies to all exporters meeting objective conditions than if it is a special deal for India; the India–EU trade agreement’s clause promising India any flexibility granted to others points the same way.

What doesn’t follow from any of this

None of this shows that CBAM is protectionism in disguise or that Europe should drop it; the carbon price it puts on imports is the same one EU producers will face as free allocation ends in 2034. Nor does it show that Indian steel is clean: most of it is not, and the taxonomy’s 2.2-tonne “green” threshold is well above any international near-zero standard. The cost estimates in this piece come from different methods and years, several are modelled rather than observed, and the gas and hydrogen arithmetic uses round numbers. Most sources were read through search summaries; see the notes below.

What would fix this

• The European Commission should allow verification by ISO 14065-accredited bodies recognised by a partner’s national accreditation body, such as India’s NABCB, for 2026–2027 imports, while EU accreditation of third-country verifiers catches up. The India–EU technical dialogue under the trade agreement is the place to agree it, on terms open to all countries that meet the same conditions.
• The Commission should hold the default-value mark-up at 10% for 2027, instead of raising it to 20%, for installations that have submitted unverified monitoring data, and publish route-specific defaults (blast furnace, gas DRI, coal DRI, scrap EAF) so that an Indian scrap or gas-based plant is not charged as a blast furnace by default.
• For downstream steel products made by small firms, the Commission should accept the steel mill’s verified precursor emissions plus a standard processing allowance, so that a fastener or forging maker does not need its own installation-level verification to avoid the default.
• The extension of CBAM to about 180 downstream products from 2028 should start with a reporting-only period of at least two years, as the original goods had from 2023 to 2025.
• The EU should tie CBAM’s pace to progress on its own energy prices: the EUROFER target of €50 a megawatt-hour for total industrial energy costs by 2030 is a better test of whether European green steel is viable than any border charge.
• India should notify the scheme to pay certification costs for small exporters, notify the green public-procurement mandate, and agree with the EU how a carbon price paid in India would be deducted, so that Indian revenue stays in India; the options are set out in this blog’s CBAM piece.

These are proposals from this blog, not recommendations any agency has adopted, except where a body is cited.

Sources and caveats

About this article: Researched, written and edited by Umashankar Triplicane Dwarakanathan, with AI research assistance; every figure is meant to trace to the primary source cited. See the Editorial Policy for how sourcing, AI use and corrections work.