Jean-Luc Demarty / Aug 2025
Maroš Šefčovič, the European Commissioner for Trade and Economic Security. Photo: European Union 2025
The trade agreement concluded last week between the EU and the United States reveals no good or bad surprises.
After more than three weeks of waiting, the trade agreement between the United States and the EU has finally been published. I would remind you that I was in favour of massive retaliation in April, which would have forced Donald Trump to back down, as demonstrated by his retreat on April 9 in the face of the beginning of a crisis in the US Treasury bond market. This would have resulted in a better agreement, not necessarily dramatically better, but without appearing to give in to Donald Trump's extortion and coercion. It is hard to see why the EU would not have been able to resist like China, India, and Brazil.
The strategic choice made by the Heads of State and Government and implemented by Ursula von der Leyen, who also favoured the deal, consisted of postponing retaliation in order to negotiate first. This necessarily led to negotiations based on Donald Trump's principles and the conclusion of an unbalanced agreement. This choice is explained by the rejection of the risk of even a short trade war by both European leaders and business leaders, and by the fear that a trade war would push Trump towards abandoning Ukraine. In my opinion, while these arguments can be understood, they seem weak, particularly in light of Donald Trump's likely abandonment of Ukraine, despite last week’s show in Washington.
The agreement reached respects the spirit and letter of the oral Turnberry agreement. Its main philosophy is to swallow Donald Trump's 10% tariffs in exchange for proper access to sectors subject to prohibitive national security tariffs (cars and auto parts, pharmaceuticals, semiconductors, wood, and others likely to come). This objective is clearly achieved in the form of a 15% cap on all tariffs except the 50% on steel and aluminum. This is roughly equivalent to 10% additional tariffs, taking into account the 4.8% of legal US MFN tariffs on average. It is even more favorable for MFN tariffs above 5%. Thus, dairy tariffs in the United States are around 15% and will remain so, while the United Kingdom will have to pay 25%. Our cheese exports will remain unchanged and will be at an advantage over our competitors.
Furthermore, the EU obtained a waiver with the application of low or zero MFN tariffs for aircraft and their spare parts, generic medicines and their ingredients, chemical precursors, and natural resources unavailable in the United States, such as cork. There are two disappointments: the absence of tariff quotas for steel and aluminum, and the absence of waivers for wines and spirits, although the 15% ceiling is not a significant barrier to trade for the latter. However, these tariffs were not explicitly included in the Turnberry agreement.
Of course, this agreement remains fundamentally unbalanced. In return, the EU zeroed all its industrial tariffs and granted tariff quotas, which have yet to be negotiated, for certain non-sensitive agricultural products: nuts, dairy products, fresh and processed fruits and vegetables, processed food products, seeds, soybean oil, pork, and bison meat. The list seems long but is insignificant. Tariffs on industrial goods are more of a nuisance than anything else. For agricultural products, these are low-duty products already massively imported from the United States, such as nuts and soybean oil, or products for which the United States is uncompetitive, such as fruits and vegetables, processed food products, and most dairy products, or products for which demand is virtually nonexistent, such as bison meat. Particular vigilance will be required for pork. Of course, it will be necessary to continue to ban pork fattened with ractopamine, a type of hormone widely used in the United States. The experience of the agreement with Canada shows that nothing enters Europe with the maintenance of this ban, which renders North American pork uncompetitive in Europe.
The $750 billion purchase of oil, gas, and nuclear energy products by 2028 is maintained, but in an ambiguous form: the EU intends to procure... On the other hand, the $40 billion purchase of semiconductors for data centres is a purchasing intention. The €600 billion investment in strategic sectors is presented as mere anticipations by European companies. It is clear that there are few firm commitments, except perhaps on semiconductors, for which they will not be difficult to meet given the scale of the needs. Similarly, the EU's planning of additional arms purchases from the United States is almost tautological given the enormous rearmament needs that cannot be met by European industry alone.
Some might be tempted to consider that the EU has made concessions on its regulatory sovereignty. In reality, this is stating the obvious, such as considering the risk of deforestation in the United States as negligible, or increasing the de minimis threshold under the Carbon Border Adjustment Mechanism, a gas plant that is ineffective in any case, or reviewing the CSRD and CS3D directives, green bureaucratic monstrosities already undergoing very insufficient simplification. The only significant commitment is not to apply network usage fees. Aside from this last element, the EU can be considered to have fully protected its regulatory sovereignty in all areas, from food to digital technology.
Two favourable elements are worth noting in the automotive sector. If the European Commission is able to make the necessary legislative proposals before the end of August, the 15% tariff will be applied retroactively to August 1st. Furthermore, both parties intend to practice mutual recognition of their standards in the automotive sector. This was one of the strategic objectives of the negotiation of the major TTIP free trade agreement between 2013 and 2016, which failed. No progress has ever been made on this point due to sabotage by US regulatory agencies. This is a major issue about which I remain skeptical, given that it is only a declaration of intent. I fear that the same causes will produce the same effects.
From a legal perspective, this agreement will not take the form of a formal international agreement. These are merely unilateral commitments that will be implemented by each party according to its own procedures. For the United States, these will be executive orders. For the EU, these will be customs regulations under the exclusive jurisdiction of the EU, without the intervention of the European Parliament, requiring only a qualified majority of member states, the achievement of which is almost certainly assured.
This agreement represents a violation of the cardinal principle of most-favored-nation treatment and the rules governing free trade agreements. However, it appears the European Commission intends to present this agreement as a step towards a free trade agreement under Article XXIV of the GATT. This is hardly credible, but it maintains appearances. It is likely that other countries that have concluded an agreement with the United States will do the same.
So this agreement remains deeply unbalanced due to the strategy chosen by the leaders of the EU and its member states. In my opinion, another strategy was possible. However, within the chosen framework, the result obtained is close to the least bad possible. It is now stabilized and much less vulnerable to Donald Trump's whims. We will see in practice if the price paid is worth it in terms of Donald Trump's support for Ukraine. I doubt it. We can regret Ursula von Leyen's exaggeratedly positive communication about this agreement, which has been ongoing since Turnberry. Ultimately, it remains a geopolitical disaster for the EU and its image in the world. This could encourage China and others to practice coercion against the EU. It could also weaken the EU's hand in existing free trade agreements and especially in those still to be concluded, such as with India.
This article originally appeared in French on 22 August 2025 on the Atlantico website
