Primary steel
Made from iron ore.
Iron ore
Pellets
Reduction with clean hydrogen
Direct-reduced iron
Electric-arc furnace
Crude steel
Main cost drivers: clean hydrogen, electricity, pellets, capital cost and plant utilisation.
Steel
How low-carbon steel requirements could affect project costs
In the core scenario, hydrogen-based primary steel costs about 63% more to produce, but raises the modelled cost of aggregate public construction by only 0.28%. The proposed IAA rule concerns low-carbon steel, not Union-origin steel.
Core scenario
From steel production to complete projects
Hydrogen-based primary steel
+63.0%
Aggregate public construction
+0.28%
Representative road bridge
+1.34%
Representative passenger car
+0.28%
Cost increase(relative to conventional steel or a final use built with it)
The proposal requires at least 25% of the steel to be low-carbon. Steel itself is a small share of the total cost of a construction project.
How the 0.28% result is calculated
The aggregate-construction result covers the complete cost of EU public construction: +63% steel production cost × 25% low-carbon steel volume × a 1.8% basic-steel cost share ≈ +0.28%. The Commission's separate central estimate is +0.25%. For a passenger car, applying the same cost assumptions to 100% low-carbon steel gives +1.13%; the displayed +0.28% reflects the proposed 25% volume share.
01 · What the proposal requires
The requirement is proposed; the clean-steel definition is still open
From 1 January 2029At least 25%
of the total steel volume would have to be low-carbon in covered public procurement and qualifying support schemes.
No Union-origin conditionThe qualifying steel may be produced outside the EU.
The downstream scope covers buildings, infrastructure and civil motor vehicles. For support schemes, Member States would apply the requirement to relevant schemes accounting for at least 45% of the national budget allocated to them.
The requirement need not be applied if it would raise the cost of the final product or project by more than 25% in public procurement or 30% in support schemes.
02 · Production pathways
Primary and secondary steel are different cost cases
H₂-DRI-EAF means using hydrogen to turn iron ore into iron, then melting it in an electric-arc furnace. BF-BOF is the conventional blast-furnace and basic-oxygen-furnace route used as the cost reference.
Secondary steel
Produced mainly from recovered steel scrap.
Steel scrap
Sorting and preparation
Electric-arc furnace
Crude steel
Main cost drivers: scrap grade and price, electricity, yield and certification.
Both routes continue from crude steel through casting, rolling and fabrication before becoming products used in buildings, infrastructure or vehicles. Product data and the final emissions classes will determine whether the steel qualifies.
03 · Low-carbon cost context
Modelled H₂-DRI-EAF costs differ substantially across regions
Origin is not an IAA condition for steel, but regional costs still shape where low-carbon production may be competitive. These bars compare one production route on one consistent model basis.
H₂-DRI-EAF production cost
EUR 2022 per tonne of crude steel · excludes domestic carbon prices and transport
Modelled production cost →
The EU marker is the midpoint of a EUR 636–762/t range, shown by the band; the range reflects electricity-price cases of EUR 60–100/MWh. Country values are point estimates, not observed market prices.
04 · Explore the model
Adjust model assumptions
Choose and adjust a primary-steel estimate
Select a fixed source estimate or adjust the premium directly. The detailed inputs below apply only to the Agora estimate.
Moving this slider creates a manual adjustment and removes the previous source attribution.
Agora published range: 54–72% · reconstructed midpoint: +63.0%
Agora inputs and carbon-price sensitivity
Agora publishes matched production costs and direct plant emissions (Scope 1) for both routes before carbon pricing and subsidies. Changing any input here replaces the selected fixed preset with an Agora-based sensitivity.
adjusted premium = (clean cost + 0.01 × carbon price) ÷ (BF-BOF cost + 1.87 × carbon price) − 1
Agora midpoint: USD 2.5/kg delivered clean hydrogen, within its USD 2–3/kg global 2030 assumption in real 2020 dollars. This is not an EU-specific forecast; adjust upward to test higher-cost European cases.
Agora reports USD 472–499/t; the wider control is a sensitivity.
Analytical full-carbon-price sensitivity—not an EU ETS cash-flow calculator.
791.6 USD/tH₂-DRI-EAF incl. 0.0 USD/t carbon cost485.5 USD/tBF-BOF incl. 0.0 USD/t carbon cost
About the carbon control
It starts at zero because Agora’s core costs exclude CO₂ prices. The 0.01 and 1.87 multipliers are Agora’s direct plant-emission (Scope 1) tCO₂/t factors for H₂-DRI-EAF and BF-BOF steel; multiplying them by USD/tCO₂ gives USD/t steel.
This is an analytical sensitivity, not an EU ETS cash-flow model. Emissions from purchased electricity and suppliers (Scope 2 and 3), casting and rolling are excluded.Final-use assumptions
The selected steel premium feeds every result at the volume and material-cost shares shown below.
The proposal sets a minimum of 25% by volume from 1 January 2029.
Sources
Evidence used on this page
Production costs, legal requirements and downstream cost shares are distinct inputs. Each source card states how its evidence is used.
COM(2026) 100
The Annex II steel requirement is low-carbon-only, not a Union-origin rule. Legal qualification will depend on the final emissions classes; a production route is not automatically compliant.
Commission IA 2026
Policy-impact context. The Commission's +0.25% construction figure is a modelling cross-check, not an observed result or an input to the linked calculation.
Agora Industry 2024
Global route-level production-cost comparison in real 2020 USD before carbon pricing and explicit subsidies. The USD 2.5/kg clean-hydrogen default is the midpoint of Agora’s global 2030 range, not an EU-specific forecast. The optional carbon control applies a full price to the stated direct plant emissions (Scope 1); it excludes emissions from purchased electricity and suppliers (Scope 2 and 3), casting and rolling and does not reproduce EU ETS cash flows.
E3G / ITA 2025
Fixed EU and import-configuration scenarios. Their carbon-price treatment is not sufficiently explicit to support adding the Agora carbon-price sensitivity.
IEA 2025
Wide global uncertainty range rather than a single EU project. 95%, the midpoint of the 50–140% range, is used only when explicitly selected, and the full range is kept visible.
Johnson et al. 2025
One consistent H₂-DRI-EAF model. The country bars exclude domestic carbon prices and transport; the EU range reflects electricity-price cases of EUR 60–100/MWh.
VUB 2024
Aggregate EU public-construction estimate. The selected production-cost premium is applied to the 1.8% basic-steel share, with full pass-through assumed.
SCI bridge study 2023
Specific UK highway-bridge case. Applying a generic steel premium to every listed steel product is an upper-bound full-pass-through sensitivity, not a matched plate-price result.
E3G / ITA 2025
The car's steel costs EUR 609.5 (1.15 t × EUR 530/t), or 1.79% of the EUR 34,000 car. The selected steel premium is applied to this share with full pass-through and no retail mark-up. The same car is used on the aluminium page.