Many, perhaps the majority, of notifications on the basis of the Guidelines on Climate, Environmental Protection and Energy [CEEAG] concern decarbonisation of production processes. In this context, large amounts of State aid have been funnelled to the decarbonisation of steel production. ArcelorMittal has been a major beneficiary. In February 2023, Spain granted EUR 460 million to support ArcelorMittal to decarbonise its steel production in Gijón [SA.104904]. Also in February 2023, Germany funded with EUR 55 million ArcelorMittal’s decarbonisation project of its steelworks in Hamburg [SA.63733]. In June 2023, Belgium granted EUR 280 million to help ArcelorMittal decarbonise its steel production in Ghent [SA.104897]. In July 2023, France granted EUR 850 million to ArcelorMittal’s project for the partial decarbonisation of its steel production in Dunkirk [SA.104903].
Other companies have received similar aid too. For example, Germany provided funding to ThyssenKrupp Steel Europe to decarbonise its steel production processes and accelerating its uptake of renewable hydrogen [SA.105244]. That aid measure was made of two parts: a direct grant of up to EUR 550 million and a conditional payment mechanism of EUR 1.45 billion to support the accelerated phase-in of renewable hydrogen.
This article reviews Commission decision on SA.63733 concerning the German measure that has some unusual features such as that ArcelorMittal is obliged to disseminate the results of the project and its experience.1
The purpose of the German aid to ArcelorMittal Hamburg [AM] was to support a decarbonisation project that involved an innovative use of hydrogen. The project, code-named “H2First”, aimed to establish a fully hydrogen-based “direct reduced iron” [DRI] production on an industrial scale. DRI is the primary raw material for the manufacturing of steel.
The AM group is the largest steel manufacturer in North America, South America and Europe, with approximately 168,000 employees worldwide, and 12 research centres. AM Hamburg was founded 1969 and is one of Germany’s leading producers of long carbon steel products.
The notified measure would provide State aid in the form of a direct grant, amounting to EUR 55 million, that would be disbursed between 2023 and 2026.
By using hydrogen, AM intended to reduce its CO2 emissions by 715 kt over the life of the plant, which was about 15 years. Its existing plant uses natural gas. The output of the H2DRI plant would be 100 kt per year. The H2DRI plant would expand new technology to an industrial scale and would have a demonstration effect for the rest of the industry. This effect was a positive feature in the Commission’s assessment of the compatibility of the aid with the internal market.
Funding gap methodology and WACC
The amount of aid was shown to be less than the funding gap of the project. According to the Commission decision, “(40) the net extra cost (“the funding gap”) of the project is [EUR 100 – 200 million]. The funding gap is defined as the difference between the NPV (discounted net present value, “NPV”) of the H2First project (the factual scenario) and the NPV of the counterfactual scenario (the scenario where no State aid is provided). The NPV is the sum of the discounted future inflows and outflows of cash generated by an investment over its lifetime. The cash flows are discounted at the Weighed Average Cost of Capital (“WACC”).”
“(41) The funding gap calculations have been performed with a WACC of 5.78%, corresponding to the current hurdle rate used by ArcelorMittal Hamburg GmbH.”
Costs of the project
“(42) The NPV of the factual scenario (i.e. of the H2First project) is [EUR -200 – -100 million] and its total costs are [EUR 650 – 750 million].”
“(43) Aid under the present measure is intended to cover up to [40 – 50%] of the investment costs linked to the H2First project. Such costs amount to approx. [EUR 100 – 200 million] and comprise the construction costs of the H2DRI plant (i.e. the costs of the necessary instruments and equipment).”
“(44) Based on the funding gap analysis […], the costs linked to the operation of the H2DRI plant in the period from 2026 to 2040 are projected to be [EUR 500 – 600 million]. The main cost components for the operation H2DRI plant are the input costs of the iron ore, costs of hydrogen, energy and personnel costs. To substantiate the applied assumptions and the projected developments of cost items, internal company documents, market studies and correspondence with a supplier of renewable hydrogen were provided. The assumption about the costs of EU-ETS is supported by an independent prediction provided by a consultancy.”
Revenues of the project
“(45) The revenues stemming from the project would be generated by internal sales of DRI within the ArcelorMittal Group and that the hydrogen-based DRI will be further processed in an electric arc furnace at the beneficiary’s site. The DRI is to be considered a raw material for further processing rather than a marketable steel product, thus its price was calculated according to an established cost-plus approach with a [[…]] mark-up applied to the production costs of iron ore reduction in the existing plant increased by a ‘green premium’ […] The main cost components for the production of DRI in the existing plant are the input costs of the iron ore, costs of natural gas, energy, personnel costs, as well as the costs of EU-ETS. To substantiate the applied assumptions, […], as well as a study regarding the projected developments of demand and supply for natural gas in Europe were provided.”
There would be another source of revenue for AM. “(46) The business case assumes that ArcelorMittal Hamburg GmbH would be able to obtain a ‘green premium’ of EUR [[…]] per tonne of wire rod throughout the projected lifetime of the H2DRI plant, which translates into a premium of EUR [[…]] applied to each tonne of DRI. […] this calculation was based on the proportion of DRI’s content in the overall cost of wire rod, as well as on the beneficiary’s assumptions on the development of supply and demand for green steel substantiated by a market study provided by a consultancy.”
The correct application of the funding gap methodology critically depends on the credibility of the counterfactual scenario that should be based on reasonable assumptions and verifiable data.
In this case, “(47) in the absence of the aid, the beneficiary would not undertake the same investment nor any alternative decarbonisation investment. Instead, in the counterfactual scenario without the aid, ArcelorMittal Hamburg GmbH would continue to use the process of direct reduction of the iron ore based on natural gas. In that scenario, the beneficiary would not engage in the construction of direct reduction plant based exclusively on hydrogen.”
“(48) The beneficiary would not have sufficient incentives to invest in the reduction of its greenhouse gas emissions stemming from the natural-gas-based reduction of iron ore:
(a) Firstly, there are no legally binding targets at national or Union level that would require the closure or conversion of a gas-based direct reduction facility in the foreseeable future.
(b) Secondly, the EU-ETS system is not sufficient to stimulate the stepping-up innovations in production facilities with very long lifetimes. According to Germany, the EU-ETS systems brings about gradual improvements in plant technology through efficiency gains. It does not however sufficiently incentivise technologically burdensome breakthrough innovations. Taking into account current and medium-term expected prices in the EU-ETS, the investment and operating costs for climate-neutral technologies on an industrial scale, as implemented by the present project, remain very high.
(c) Thirdly, the iron ore reduction capacity in the existing plant is sufficient to respond to the current demand and would remain sufficient for the foreseeable future. According to Germany, the H2First project would allow the beneficiary to increase the price of the final
product only to a very limited extent that would not cover the total investment costs (see recital (46)). There is therefore no sufficient market incentive for the project to be executed.”
“(49) On that basis, according to Germany, the NPV of the counterfactual scenario can be assumed to be zero as the most likely counterfactual consists in the beneficiary continuing its business without changes.”
Funding gap results
“(50) The net extra cost of the H2First project may be approximated by the negative NPV of the project in the factual scenario without the aid, i.e. to [EUR 100 – 200 million]. Hence, the notified funding gap is [EUR 100 – 200 million].”
“(51) Based on a schedule of aid instalments provided by the German authorities, the measure will reduce the funding gap to [EUR 80 – 95 million]. In this regard, the German authorities explained that ArcelorMittal Hamburg GmbH will proceed with the H2First project notwithstanding the fact that the aid amount is below the project’s funding gap while seeking additional sources of funding.”
“(52) To reduce the risk of overcompensation in case of unexpected positive developments going beyond the current forecasts, the German authorities committed to put in place a claw-back mechanism covering the entire lifetime of the project.”
“(53) The claw-back mechanism will entail a comparison of the project’s negative and positive cash flows (including the actual State aid disbursements and additional benefits resulting from the project). The basis for such verification will be audited post-tax cash flows that the beneficiary will be required to submit to the national authorities. The claw-back mechanism only applies in case the verification explained leads to a positive value (“surplus”) including the actual State aid disbursements. However, to ensure that the beneficiary has an incentive to deliver its project in an efficient manner, a share of 40% of any potential surplus will remain with the beneficiary, while the remaining 60% will be paid back to the German authorities.’
Germany committed to report regularly to the Commission on the implementation of the claw-back mechanism.
Knowledge dissemination by the beneficiary
A rather unusual condition attached the aid by the German authorities was that AM would have to disseminate the knowledge it would acquire from the H2First project. Dissemination would be achieved through knowledge sharing at technical conferences and through publications in specialist journals. For results protected by IPRs, AM would have to issue licences on FRAND conditions.
Compatibility of the aid with the internal market
Germany agreed that the measure constituted State aid. Therefore, the Commission dealt with the presence of aid in a few brief and concise statements.
The compatibility of the aid was assessed on the basis of section 4.1 of the CEEAG, in conjunction with the principles in section 3, which apply to all State aid measures that fall within the scope of the CEEAG. Section 4.1 lays down the conditions that need to be satisfied by aid granted for the reduction and removal of greenhouse gas emissions including through support for renewable energy and energy efficiency.
First compatibility condition
In paragraphs 74 – 93 of the decision, the Commission found that the measure satisfied the “positive” condition for the compatibility of the aid with the internal market; i.e.
1) the aid must contribute to the development of an economic activity;
2) the aid must have an incentive effect; and
3) the aid may not breach any “relevant” provision of EU law.
With respect to the incentive effect of the aid, it is worth noting that, as explained by the Commission, in the environmental field it has a double meaning. “(78) An incentive effect occurs when the aid induces the beneficiary to change its behaviour towards the development of an additional or more environmentally-friendly economic activity, and if this change in behaviour would not otherwise occur without the aid”. That is, the aid must be necessary for a project to go ahead and the project must cause lower environmental degradation than in a situation without the aid. This requires a comparison of the factual and counterfactual scenarios [point 28 of the CEEAG].
The factual scenario in this case was the construction and operation of the H2DRI plant with a deficit of EUR 100-200 million.
The counterfactual scenario was no alternative investment, but a continuation of production at the AM’s existing plant. “(82) The Commission considers that the counterfactual scenario submitted by the beneficiary is credible, in that, in the absence of the aid, the beneficiary would not have sufficient incentives to engage in the supported project nor in any other projects aimed at reducing or avoiding greenhouse gas emissions linked to the reduction of iron ore. Therefore, in line with point 52 CEEAG, the Commission considers that the NPV of the counterfactual scenario can be approximated to zero.” Moreover, “(85) it appears that the market cannot, on its own, provide sufficient incentives to invest in the reduction of emissions in the value chain of steel production, in particular by replacing the natural gas in direct reduction of the iron ore with renewable hydrogen and thus marketing the green steel at a higher price that would allows covering the necessary investment costs. This is because the market for sustainable steel products and possible demand for steel produced using renewable hydrogen are still on a nascent stage of development and, especially, the premium paid for green steel is not expected to cover the higher production costs.”
Although the Commission did not explicitly do so in this part of the decision, it could have added that in the factual scenario there would be a reduction of CO2 emissions by 715 kt.
With respect to the meaning of “relevant” EU law, the Commission decision cited the judgment in case C-594/18 P, Austria v Commission (Hinkley Point C), paragraph 44. That paragraph states that “State aid which contravenes provisions or general principles of EU law cannot be declared compatible with the internal market”. That judgment in turn cites another landmark judgment [C-390/06, Nuova Agricast, paras 50-51] in which the Court referred to “State aid, certain of the conditions of which contravene the general principles of [EU] law, such as the principle of equal treatment, cannot be declared by the Commission to be compatible with the [internal] market”. As can be seen, in neither judgment is there a mention of “relevant” law.
The Commission decision on the present measure also referred to point 33 of the CEEAG, according to which “if the supported activity, or the aid measure or the conditions attached to it, including its financing method when it forms an integral part of the measure, entail a violation of relevant Union law, the aid cannot be declared compatible with the internal market. This may be the case, for instance, where the aid is subject to clauses conditioning it directly or indirectly on the origin of products or equipment, such as requirements for the beneficiary to purchase domestically-produced products.” Point 33 explains the nature of the infringement, but it does not really clarify what the “relevant” law may be. Commission guidelines must comply with the case law. Since in the “Austria” and “Nuova Agricast” cases the Court did not limit the law that may be infringed, it must be inferred that the “relevant” law is any provision in the primary or secondary legislation that may be contravened by the aid measure. I think that what point 33 of the CEEAG tells us is that the contravention must be direct, not indirect or implicit. That is, the aided activity or the terms on the basis of which is the aid is disbursed must infringe a provision of EU law. For example, if the transportation of the output of the aided activity to a port causes road congestion and if that effect would not be part of a mandatory environmental impact evaluation it would be an indirect effect that perhaps the Commission would ignore in its compatibility assessment.
Second compatibility condition
In paragraphs 94-15 of the decision, the Commission examined compliance with the “negative” condition of State aid [meaning no affectation of trade that would be contrary to the common interest]; i.e.
1) there must be a need for state intervention;
2) the aid must be proportional;
3) the aid must be transparent;
4) the aid must be appropriate; and
5) avoidance of undue negative effects on trade and competition.
With respect to the need for state intervention, since the NPV was negative to the tune of EUR 100-200 million, the Commission concluded that no rational operator would have made the investment without aid.
With respect to the proportionality of the aid, public assistance must be the minimum necessary. But since this measure involved individual aid, there was a problem. “(100) Points 103 and 104 CEEAG state that aid for reducing greenhouse gas emissions should, in general, be granted through a competitive bidding process, that, in principle, should be open to all
eligible beneficiaries to enable a cost effective allocation of aid and reduce competition distortions.”
“(101) The Commission notes that the present measure falls under the exception from the requirement to allocate aid and determine the aid level through a competitive bidding process set out in point 107(a) CEEAG on the grounds that there is insufficient potential supply or number of potential bidders to ensure competition. By demonstrating that the measure can be successfully implemented only at a site where the infrastructure to process the DRI further is already available and that such infrastructure is currently available in Germany only at a beneficiary’s site (recital (31)) Germany demonstrated that it is not possible to increase competition by a competitive bidding process.”
“(102) Therefore, as set out in point 51 CEEAG, in the absence of a competitive bidding process, the Commission has assessed the proportionality of the aid in question by comparing the profitability of the factual and counterfactual scenarios, based on the quantification of costs and revenues submitted by the Member State.”
Paragraph 106 of the decision summarises the relevant figures, as follows:
Total costs: EUR 650-750 million.
Funding gap [NPV]: EUR 100-200 million.
Nominal aid amount: EUR 50 million.
NPV of aid amount: EUR 25-35 million.
It follows that in real terms, the aid intensity was 12.5%-35% of the FG, which represented the net extra costs of the project. Since it did not exceed the FG, the Commission concluded that it was proportional.
An important question that is never fully answered in Commission decisions is why aid recipients agree to go ahead with projects when the aid falls far short of the FG. In this case, AM would stand to lose EUR 65-175 million, in real terms. Perhaps the knowledge it would acquire would give it a first-mover advantage that could also enhance the competitiveness of its other plants, despite the fact that it would have to disseminate the results of the project. Given that AM has received State aid from other Member States for similar projects, it is a bit puzzling that the Commission did not examine possible connections with those other projects. In other Commission decisions, especially on large regional investment projects, it is mentioned that aid recipients reap qualitative benefits [e.g. closeness to the market, good relations with distributors] or want to avoid certain risks [e.g. political uncertainty in foreign markets] that are not quantified in funding gap calculations.
With respect to avoidance of undue negative effects of the aid on competition and trade, the Commission noted that the aid would be limited to no more than 10 years [point 70 of the CEEAG], that emissions would be reduced and there would be no increased consumption of fossil fuels [points 116, 117 & 127 of the CEEAG], and that the production of DRI would not distort market signals as it would be fully consumed by AM’s steel operations at a mark-up [point 123 of the CEEAG]. But as mentioned above, the possible beneficial impact on AM’s other projects in Europe and the cumulative repercussions on the market were not examined. Instead, the Commission concluded that the aided project would not lead to creating or
strengthening of market dominance as its output would be very small in relation to other DRI plants [point 132 of the CEEAG].
Given that the aid was not granted through a competitive process, Germany complied with point 133 of the CEEAG by requiring AM to disseminate its know-how to accelerate the roll-out of the successfully demonstrated technology.
For all the above reasons, Commission concluded that aid avoided undue negative effects on competition and trade.
As stated in point 93 of the CEEAG, “(124) the Commission presumes, provided all other compatibility conditions are met, the appropriateness of aid for achieving decarbonisation goals, in particular to the EU climate target”.
Consequently, the Commission found that the positive effects outweighed the negative effects of the aid.