Guest Post: Too Big To Fail: The EU Steel Industry, Article XXVIII of the GATT 1994 and the EU’s Free Trade Agreements
This is a guest post by trade lawyer Victor Crochet and law professor Weihuan Zhou
Steel has long been a contentious topic in international trade, in part because of the numerous jobs it creates and its relevance to downstream industries including those related to national defense. In the European Union alone, the steel industry employs roughly 300,000 people. Yet the situation of this industry has been dire in recent years as it has faced a confluence of external hardships including the closure of the United States’ market following the imposition of Section 232 measures on steel imports and overcapacity caused largely by State intervention in particular in China.
When the US first imposed the Section 232 measures, the EU attempted to shield its industries from negative impacts by imposing safeguard measures in mid-2018. These measures took the form of tariff-rate quotas on steel imports based on import volumes during the three preceding years increased by 5% combined with an out-of-quota duty of 25%. These measures were successfully challenged by Türkiye at the WTO in 2022, but remained in place following paper implementation of the ruling by the EU.
As per Article 7.3 of the WTO Agreement on Safeguards, these measures are set to expire by next summer. This puts the EU and its steel industry in a difficult position as the factors that first led to the imposition of the safeguard measures in 2018 have only worsened in recent times. The US has further tightened its Section 232 measures earlier this year; global overcapacity in the steel sector has continued to grow; and the EU industry has been further afflicted by high energy costs resulting from Russia’s invasion of Ukraine.
The European Commission recently put forth a legislative proposal to address this situation. This regulation, if adopted, would essentially roll over the safeguard measures by putting in place new TRQs on imports of steel products except that the volume of those quotas would be based on import volumes in 2013 (thus dividing current quota volumes by nearly half) and the out-of-quota duty would increase to 50%.
At face value, this proposal runs counter to Article II:1 of the GATT 1994 which provides that WTO Members cannot impose duties above the bound rates included in their tariff schedules (the EU’s bound rate for steel is 0%) and, possibly, to Article 7.5 of the WTO Agreement on Safeguards which indicates that a new safeguard measure cannot be imposed on the same product for a period equal to the duration of the previous safeguard measure.
To maintain a long-term, WTO-permissible remedy for the steel industry, the new regulation invokes Article XXVIII of the GATT 1994 to renegotiate EU’s bound tariffs on steel products. European Commission officials are claiming that this renders the proposal WTO-compliant.
Titled “Modification of Schedules”, Article XXVIII of the GATT 1994 essentially provides that every three years (with the next period starting in January 2027), a WTO Member may modify its tariff bindings by negotiation and agreement with its main trading partners for the relevant products. The trading partners entitled to the negotiation include those (1) “with which such concession was initially negotiated” (i.e. initial negotiating right holder) or (2) holding a “principal supplying interest”. In addition, the applicant for renegotiation must also consult any other members having “a substantial interest” in the concession concerned.
A notification of intention to enter into negotiations must be lodged from six months to three months before the relevant date. Outside of such periods, a WTO Member wishing to renegotiate its tariff bindings must be authorized to do so by the WTO General Council. In the course of the negotiations, the WTO Members involved should maintain a similar level of trade liberalization as that in place beforehand. This can be done by, for example, lowering tariffs on other goods to compensate the increase in tariffs on the desired products. If such negotiations do not achieve a satisfactory result by the relevant date, the WTO Member wishing to increase its tariffs can go ahead and do so. In response, the main trading partners can increase their own tariff bindings affecting a similar amount of trade on a most favored nation basis upon notification and no earlier than 30 days after that notification, but also not later than six months after the modification to the schedule. There is also WTO jurisprudence where the EU explicitly recognized that Article XXVIII:2, regarding the requirement to try and maintain a similar level of trade liberalization compared to that in place beforehand, establishes a mandatory obligation (Panel Report, EU – Poultry (China), paras. 7.242-7.243). Moreover, the Appellate Body confirmed that a TRQ established as a result of Article XXVIII negotiations must be administered in a non-discriminatory manner in line with Article XIII of the GATT (Appellate Body Report, EC – Poultry, para. 102).
The European Commission’s proposal thus appears dubious in a number of respects under this provision. The proposal has already been tabled and the aim would be for it to enter into force by next summer in order to replace the safeguard measures when they expire. Any increase to bound tariffs under Article XXVIII of the GATT 1994 must, however, follow the procedures just described before entering into force. Given that the next relevant date is in January 2027 and that it is unlikely that the WTO General Council authorizes earlier negotiations as such authorization would have to be given by consensus, the proposal is thus likely to enter into force before the procedures of Article XXVIII of the GATT 1994 can be followed.
Besides this temporal limitation, the EU could technically force the hands of its trading partners in accepting the revised tariffs. Indeed, as mentioned, if negotiation fails, the EU can rely on Article XXVIII:3 of the GATT 1994 to unilaterally amend its bound tariffs. The consequence might be that its main trading partners can increase their own tariffs as per paragraph 3(a) of this provision. However, this is unlikely to happen. The EU’s trading partners could already have increased their tariffs on EU imports as a response to the existing safeguard measures in accordance with Article 8.3 of the WTO Agreement on Safeguards. This is so because these measures were found to be in breach of the agreement in EU - Safeguard Measures on Steel (Turkey) and, in any event, have been in place for more than three years. Yet, while many reserved their rights to do so, it seems that no one actually took action. A response to the use of Article XXVIII:3 of the GATT 1994 by the EU is thus improbable. The timeline to do so is tight and such a response must be applied on a most favored nation basis but can only target tariff bindings initially negotiated with the WTO Member invoking Article XXVIII making it sensitive to implement (while a tariff increase in response to safeguard measures only target the trade with the WTO Member imposing the safeguard measures and can target any trade).
Assuming the procedures of Article XXVIII of the GATT 1994 are followed and the EU can impose TRQs and increase its bound tariffs on out of quota steel imports, there is still another issue with the proposal. Any increase in tariffs resulting from Article XXVIII of the GATT 1994 must still respect the rules of Article I (most favored nation) and Article XIII (non-discriminatory administration of quotas) thereof. The details of the administration of the TRQs under the proposal are not yet fully clear given that Article 4 thereof provides that the allocation of those quotas between countries is to be made in a separate implementing act. However, it already seems clear that this will result in administration contrary to Article XIII of the GATT 1994. Article XIII.2 requires that a WTO Member implementing TRQs achieves a distribution of trade product approaching the shares expected in the absence of such restrictions. Subparagraph (d) clarifies that, if an agreement cannot be reached with WTO Members having a substantial interest, the allocation of quota volumes between supplying countries must be based upon the proportions they supplied during a previous representative period.
However, Article 4 of the proposal indicates that, when deciding how to allocate the quota volume between countries, the European Commission will take into account different factors such as whether the country at hand has a free trade agreement with the EU, concluded an agreement with the EU when negotiating under Article XXVIII of the GATT 1994, is part of an agreement or understanding to address overcapacity as well as the EU’s need for diversification of supply. In other words, it seems that the European Commission does not intend to allocate the quota volumes on an objective basis based on expected unrestricted trade but to use the negotiations to work towards creating a “steel club”. This is made clear in the press release accompanying the proposal where it stated that “the Commission will swiftly engage with affected EU trading partners under the Article XXVIII of the GATT 1994 procedure regarding this change to the EU's WTO tariffs, with a view to offering them country specific allocations.”
Besides these WTO-related hurdles, the EU’s proposal will also run counter to the numerous free trade agreements the EU has in place. All these agreements guarantee tariff free trade in steel with the EU and these commitments would remain regardless of the EU’s reliance on Article XXVIII of the GATT 1994. To avoid violating these commitments, the EU would have limited options. It could of course renegotiate them but that is politically unthinkable and these agreements do not include provisions similar to Article XXVIII of the GATT 1994. The only other alternative would be to rely on the bilateral safeguard provisions in those agreements (which allow one of the parties to raise import duties on imports from the other party above the level committed to under the free trade agreement) separately against each free trading partner. However, the issue here is that these provisions often limit the level of any bilateral safeguard measure to the lesser of either (i) the applied most favored nation duty rate in place when the bilateral safeguard measure is taken or (ii) the applied rate when the agreement entered into force. Given that for steel this latter was 0%, it seems that bilateral safeguard measures would be of limited use in justifying imposing TRQs with out-of-quota duties of 50% on imports from free trading partners.
The EU’s proposal is thus a pandora’s box. Article XXVIII of the GATT 1994 has never been relied on in such a way and such a scale as the one the EU is suggesting to do. For example, in 2011 Ukraine requested to re-negotiate 371 agricultural and non-agricultural tariff lines to better reflect its economic situation. However, due to the systemic implications of such a sweeping modification and the fact that Ukraine’s accession had not been that long ago, it led to a lot of push-back. In fact, most members (including the EU and Korea) were of the view that Article XXVIII procedures were not intended for modifications of such a broad range, and that adequate compensation for these modifications would be impossible for Ukraine to be provide – leading to Ukraine’s withdrawal of the request.
The implications of the EU’s proposal are thus systemic and significant. While the EU apparently seeks to address domestic protectionist pressure within WTO rules and mechanisms, its stated priorities and objectives (e.g. combating overcapacity and pursuing diversification) entail selective engagement and potential discriminatory treatment among trading partners. Doing so would not only breach WTO rules as discussed above but would also provoke retaliatory actions by trading partners including similar approaches to undertaking Article XXVIII renegotiations. These actions would only further undermine the already declining global trade order.
With regard to free trade agreements, the EU’s proposal seems difficult to justify altogether. The EU has already been repeatedly criticized for undermining access to its markets for free trading partners in recent years through its environmental measures such as the Carbon Border Adjustment Mechanism and Regulation on Deforestation-Free Products. Such a blatant disregard for free trade agreements under this new proposal will thus probably further undermine the EU’s efforts to appear as a reliable free trading partner.
Domestically, the question will inevitably come as to why this type of protectionist measure is limited to the steel industry and calls for broadening its scope to other industries will arise. This could hence be the first step in a downward spiral of lawless protectionism in the EU as we have seen emerging in other countries.