Provisional safeguard measures on grain-oriented electrical steel reach beyond coils into laminations, cores and even cores already built into transformers, pulling a whole electrification supply chain inside the EU tariff perimeter until February 2027.
BRUSSELS, SEPTEMBER 29, 2026
The European Commission’s provisional safeguard on grain-oriented electrical steel has entered its second full week in force, and the trade community is still working through the scope of a measure that is unusual in both design and reach. Imposed on 18 September 2026, the safeguard applies tariff rate quotas coupled with price thresholds to imports of grain-oriented electrical steel, commonly abbreviated to GOES, and extends that protection down the value chain to laminations, cores and cores already incorporated into transformers.
That downstream extension is the feature that has drawn the most attention from importers, transformer manufacturers and grid operators. It is comparatively rare for an EU safeguard to follow a product into its assembled form. By doing so, the Commission has pre-empted the classic evasion route in which import pressure shifts from the protected raw material to the semi-finished or finished article that contains it.
The measure runs until 26 February 2027 in its provisional form. Definitive measures, if they come, require a qualified majority in the Safeguard Committee, meaning at least 15 member states representing at least 65 percent of the EU population. The investigation itself began in March 2026 and, under the normal timetable, must conclude within nine months, extendable to eleven.
What the quotas actually look like
The provisional measure allocates duty free volumes by origin and product group. For GOES coils, the quota for China is set at 30,694 tonnes and the quota for South Korea at 4,790 tonnes. For laminations and cores, Turkey receives 14,162 tonnes and a residual pool of 2,694 tonnes covers other origins.
Alongside the volume caps, the Commission has layered price thresholds ranging from 2,800 euro to 5,000 euro per tonne, equivalent to roughly 3,214 to 5,740 US dollars. The dual structure is deliberate. A pure quota controls volume but does nothing about price. A pure minimum price controls price but does nothing about volume. Combining the two addresses the two distinct mechanisms through which import pressure injures a domestic producer: the displacement of sales and the suppression of the price at which remaining sales can be made.
Once a quota is exhausted, out of quota imports become liable to the safeguard duty. The practical consequence for importers is that quota administration, which operates on a first come, first served basis within defined quota periods, becomes a live commercial concern. A shipment that clears customs on the last day of an open quota carries a very different cost from the identical shipment clearing two days later.
The evidence behind the case
The Commission’s justification rests on a striking set of import statistics covering 2021 to 2025. Over that period, imports of the products concerned rose 120 percent in absolute terms. Measured against EU production, imports grew 141 percent. Measured against EU consumption, they grew 66 percent.
The gap between those three figures tells the story. Absolute import growth of 120 percent against consumption growth implied by the 66 percent relative figure means imports were not simply riding a demand wave. They were taking share. And the 141 percent figure relative to production means EU producers were losing ground faster still.
The Commission attributes the surge to two reinforcing forces: global overcapacity in electrical steel, concentrated in Asia, and the progressive closure of traditional export markets for that capacity as other jurisdictions have raised their own barriers. When a large exporter loses access to one destination, the tonnes do not disappear. They look for the most open remaining market. For a significant share of global steel capacity, that market has been the European Union.
Why grain-oriented electrical steel matters
GOES is not a commodity steel. It is a highly specialised, silicon alloyed, precisely textured flat product whose grain structure is aligned to minimise core loss when magnetised. It exists for one dominant purpose: to be stacked or wound into the magnetic cores of power and distribution transformers.
Every transformer on every grid in Europe contains it. That makes GOES a strategic input for the entire electrification agenda. Grid reinforcement to accommodate renewable generation, the connection of offshore wind, the build out of electric vehicle charging infrastructure, the electrification of industrial heat and the construction of data centres all consume transformers, and transformers consume GOES.
Europe faced a global transformer shortage through the middle of the decade, with lead times for large power transformers stretching to several years at the peak. Against that backdrop, the Commission’s case is that the EU cannot afford to let its remaining domestic GOES capacity be competed out of existence, because doing so would leave the continent’s grid build out dependent on imported cores from a small number of non EU suppliers.
Eurofer, the European Steel Association, framed the point directly, stating that “Europe cannot expand and strengthen its electricity grids without maintaining a European supply of Grain-Oriented Electrical Steel”. The association welcomed what it described as a balanced approach and made a pointed argument about cost, estimating that the measure adds approximately 0.002 cents per kilowatt hour to grid costs.
That figure is worth examining because it goes to the heart of the trade off. Two thousandths of a cent per kilowatt hour is, on any reading, a small number. Eurofer’s argument is that the security of supply benefit of a domestic GOES industry vastly exceeds a cost increment of that scale. Critics of the measure would counter that the estimate captures only the direct pass through of higher GOES prices into electricity tariffs and not the wider effect of higher transformer prices on the pace and cost of grid investment at a moment when Europe needs that investment to accelerate.
The downstream extension and why it is contentious
The inclusion of laminations, cores and cores already fitted into transformers is the measure’s most consequential design choice.
In a conventional safeguard limited to coils, an importer facing quota exhaustion has an obvious workaround. Rather than importing the steel, import the next thing along the chain. Have the coil slit, cut and stacked into laminations or assembled into a finished core outside the European Union, then import the core. The core is a different product under a different customs classification, and the safeguard does not touch it.
The Commission has closed that route at the outset. Eurofer noted explicitly that the measure protects transformer manufacturers and prevents downstream import pressure from simply shifting.
The contention arises because the extension captures European transformer makers who buy imported cores as a legitimate part of their manufacturing strategy, not as an evasion tactic. A transformer manufacturer in central Europe that has sourced cores from a Turkish supplier for a decade now finds that supply inside a quota regime. If the Turkish quota of 14,162 tonnes is consumed early in a quota period, that manufacturer faces either a duty or a supply gap.
The inclusion of cores already incorporated into transformers pushes further still, reaching a product that is unmistakably a finished capital good. Establishing the GOES content of an assembled transformer for customs purposes is not a trivial exercise, and the documentary burden it places on importers is real.
Reactions across the chain
Steelmakers have been the clearest supporters. The European producer base for GOES is narrow, and the association’s position that a domestic supply of the product is a precondition for the electrification build out has found receptive ears in Brussels, where security of supply arguments have carried increasing weight since 2022.
Transformer manufacturers occupy a more ambiguous position. They benefit from the protection of the finished core against imports. They pay for the protection of the coil. Where a manufacturer is vertically integrated and produces its own cores from purchased GOES, the balance is unfavourable. Where a manufacturer buys cores, the extension of the measure to cores is a direct cost.
Grid operators and utilities are the ultimate payers. They are also, in most member states, regulated entities whose ability to pass cost increases to consumers is constrained by regulatory settlements negotiated years in advance. A transformer cost increase arriving mid settlement is absorbed by the operator until the next review.
Exporters in China, South Korea and Turkey face the most immediate commercial disruption. The Chinese GOES quota of 30,694 tonnes is the largest single allocation, reflecting historic import volumes, but it is a cap where none previously existed. Korean exporters, with a quota of 4,790 tonnes, face a tighter constraint in absolute terms. Turkish suppliers, whose position is concentrated in laminations and cores rather than coils, must now manage a 14,162 tonne ceiling on a trade that has grown rapidly.
Economic impact analysis
The measure’s economic effect operates through three channels.
The first is price. Tariff rate quotas with price thresholds function as a floor under the domestic price. Below the threshold, imports become uneconomic regardless of quota position. European GOES producers can therefore price up toward the threshold without fear of being undercut, recovering margin that the Commission found had been suppressed.
The second is volume allocation. Quotas do not eliminate imports. They ration them. The rationing is administered by customs on a first come, first served basis, which rewards importers with the working capital and logistics capability to front load shipments into the opening of each quota period. Smaller importers, and those serving just in time manufacturing schedules, are structurally disadvantaged by that dynamic.
The third is investment signalling. A provisional safeguard running to February 2027 is, by itself, too short to underwrite a capital investment decision in electrical steel capacity, where a new line represents a commitment measured in hundreds of millions of euro and several years of construction. The signal that matters is whether definitive measures follow. If the Safeguard Committee delivers the qualified majority required, European producers will have a multi year protected window, and the investment case changes materially.
Quantifying the cost to European buyers requires assumptions the Commission has not published. At a price threshold midpoint of roughly 3,900 euro per tonne, and with total quota volumes across the categories of roughly 52,000 tonnes, the directly rationed trade is modest in tonnage terms. The leverage comes from the fact that GOES is a small cost component of a transformer but an indispensable one. Constraining it constrains everything downstream.
Implications for global importers, exporters and supply chains
For importers, the first task is scope determination. Any business importing electrical steel, laminations, cores or transformers into the European Union needs to establish with certainty whether its products fall inside the measure. The extension to cores within transformers means that companies which have never considered themselves steel importers may now hold safeguard exposure.
The second task is quota monitoring. The Commission publishes quota balances, and importers who track them can time shipments to avoid arriving after exhaustion. Those who do not track them will discover the position at the point of clearance, when the options are to pay the duty or to leave the goods in temporary storage.
For exporters, the measure is a template. The design used here, a quota plus price threshold applied simultaneously to a raw material and its downstream derivatives, is available to the Commission in any case where it concludes that import pressure would otherwise migrate along the chain. Exporters of any input where the EU retains a domestic industry and where global overcapacity exists should assume the approach can be replicated.
For supply chain planners, the broader lesson concerns the growing tendency of trade measures to follow products through processing stages. Traditional tariff engineering, which relocates a stage of manufacture to change a customs classification and escape a duty, is becoming less reliable as authorities write measures that anticipate it. The EU’s parallel introduction of a melt and pour requirement in its steel overcapacity regulation, effective 1 October 2026, points the same way: the relevant question is increasingly where a material originated, not where it was last processed.
The GOES safeguard also illustrates a tension that will define European trade policy for the rest of the decade. The bloc has committed to an electrification programme of historic scale, and that programme requires enormous volumes of equipment at a price that keeps the energy transition affordable. It has simultaneously committed to preserving strategic industrial capacity on European soil. Those two commitments pull in opposite directions whenever the cheapest equipment comes from outside the bloc.
The Commission’s answer, visible in this measure and in the steel regulation that preceded it, is that strategic capacity comes first and that the cost, expressed by Eurofer as two thousandths of a cent per kilowatt hour, is a price worth paying. Whether member states agree will be tested when the Safeguard Committee votes on definitive measures, and the qualified majority threshold means that a bloc of states prioritising cheap grid investment over domestic steel could, in principle, stop it.
Until then, the provisional measure stands, the quotas are open, and every importer of electrical steel into the European Union is operating under a ceiling that did not exist six weeks ago.
The quota mechanics importers must master
Understanding how EU tariff rate quotas are administered is now a commercial competence rather than a customs technicality.
Quotas under an EU safeguard are allocated on a first come, first served basis by the Commission’s customs services, drawn down as declarations for release into free circulation are accepted by national customs authorities. Balances are published and updated, and the rate of drawdown is visible to anyone who monitors it.
The behavioural consequence is a rush at the opening of each quota period. Importers front load shipments to secure duty free volume, which creates a concentration of arrivals, pressure on port capacity and a spike in demand for short term storage. Once the quota nears exhaustion, arrivals slow sharply because no importer wishes to be the consignment that crosses the threshold and attracts the duty.
That pattern imposes a working capital cost. Buying ahead of requirement to secure quota means financing inventory and carrying it. For importers of grain-oriented electrical steel, which is a high value product at roughly 2,800 to 5,000 euro per tonne, that carrying cost is material.
There is also a distributional effect. The first come, first served mechanism systematically favours large importers with the balance sheet to buy early and the warehousing to hold material. Smaller importers, and manufacturers operating lean inventory models, are pushed toward the end of the quota period when the risk of exhaustion is highest.
Price thresholds and the question of enforcement
The price threshold element of the measure, running from 2,800 to 5,000 euro per tonne depending on product group, introduces a second compliance dimension.
A price threshold works by disapplying the duty free treatment for imports priced below the stated level. The mechanism is only as reliable as the customs valuation on which it depends, and customs valuation of related party transactions has always been the weak point of price based trade measures. Where an exporter sells to its own European subsidiary, the invoice price is a transfer price, and establishing whether it reflects an arm’s length value requires documentation that goes well beyond the commercial invoice.
Importers should expect scrutiny of valuation in exactly those cases. Transfer pricing documentation prepared for corporate tax purposes may not answer the questions customs authorities ask, because the two regimes test different things. Businesses importing GOES through related party structures should review whether their valuation methodology withstands a customs audit.
The transformer bottleneck in context
The measure lands on a market that has been supply constrained for several years.
European demand for transformers has been driven by three simultaneous forces: the replacement of ageing grid assets installed in the mid twentieth century, the connection of renewable generation that requires new transmission and distribution infrastructure in locations where none previously existed, and the growth of large new loads including data centres, electrolysers and electrified industrial processes.
Lead times for large power transformers extended dramatically as that demand arrived faster than manufacturing capacity could expand. Distribution transformers, produced in far greater volume, saw similar though less extreme pressure.
Into that environment, the Commission has introduced a measure that constrains the supply of a critical input and of the imported cores that some European transformer makers rely on. The Commission’s judgement is that the alternative, allowing the European GOES industry to be competed out of existence, would produce a far worse constraint within a few years. That judgement is defensible, but it carries a near term cost that falls on grid operators at precisely the moment Europe is trying to accelerate grid investment.
The vote that decides the outcome
Everything about the measure’s long term significance turns on the Safeguard Committee vote on definitive measures.
The qualified majority threshold, 15 member states representing at least 65 percent of the EU population, is a demanding standard. It requires the support of most of the large member states, and the interests of the large member states are not aligned on this file.
Germany, France, Poland and Italy host both steel production and heavy electrical equipment manufacturing, and their positions will depend on which constituency proves more persuasive domestically. Member states without domestic GOES production but with large grid investment programmes have an interest in cheap imported material. Member states hosting transformer assembly have an interest in the downstream protection but not the upstream cost.
Provisional measures expire on 26 February 2027. If no qualified majority emerges before then, the measure lapses and imports revert to unrestricted entry. That deadline, rather than the investigation timetable, is the date that importers, exporters and European producers are all now planning around.