A month after foreign ministers found the outlines of consensus, the bloc still cannot decide whether it has the authority to act. Member states are no longer waiting to find out.

On 13 July, EU foreign ministers gathered in Brussels and did something they rarely manage on questions touching Israel: they converged. Kaja Kallas, the bloc’s foreign policy chief, said afterward that a full settlement trade ban on goods from Israeli settlements in the occupied West Bank had drawn more backing than either of the two alternatives on the table, a system of prohibitive tariffs or an import licensing regime. That should have been the hard part. It was not.

The dispute that has since stalled the file is not about substance. It is about classification, and the classification question has a precise legal address. Does a settlement trade ban belong under Articles 29 TEU and 215 TFEU, the CFSP route for restrictive measures, which requires unanimous agreement among all twenty-seven capitals and bypasses the European Parliament entirely, or does it fall under Article 207 TFEU, the Common Commercial Policy, decided by qualified majority with Parliament co-deciding? The European Commission’s own options paper leans toward the CFSP route, a position Germany has backed, while the Council’s own legal service gave Kallas an oral opinion that qualified majority voting was sufficient because the measure is, at bottom, a trade issue. A group of roughly forty EU law scholars, including HEC Paris professor Alberto Alemanno, wrote an open letter to Kallas, trade commissioner Maroš Šefčovič and Commission president Ursula von der Leyen arguing that Article 207 TFEU was the only defensible basis, comparing the case to existing EU import bans on goods linked to forced labour and conflict minerals, which cleared the same lower bar. A coalition led by France and Sweden, joined by Belgium, Ireland, Luxembourg, the Netherlands and Spain, has pushed the same reading in the Council chamber. Italy and the Czech Republic remain undecided, and their hesitation is doing more to determine the outcome than either legal memo.

This is worth pausing on, because it inverts the usual complaint made against Brussels. The EU is not divided over what to do. Kallas herself said the settlement trade ban was the clear preference. It is divided over which door to walk through to do it, and that argument has now consumed a month, with the next scheduled foreign ministers’ meeting not until October. Diplomats close to the file have raised the prospect that Israeli elections due that same month could scramble the calculation entirely, giving the bloc’s most reluctant members fresh cover to wait longer still.

The Case for Caution Deserves a Hearing

There is a serious version of the argument for treating a settlement trade ban as a unanimity question, and TEP should not wave it away. The Court of Justice has a settled test for cases like this. First developed in the Titanium Dioxide ruling and hardened in the 2008 ECOWAS judgment, it asks not what a measure is called but what it does. Judges look at its centre of gravity, meaning the aim and content of the measure, read together. On that test, a label does not decide the outcome. If the Council can show that a settlement ban’s real purpose is to pressure a specific government over a specific occupation, rather than to regulate ordinary commerce, that purpose alone could push the measure toward the CFSP and its unanimity rule, whatever treaty article the Council cites. Seen this way, Germany’s resistance to unilateral EU action against Israel is not simply domestic politics wearing the language of principle. It is a plausible reading of settled case law, not a stalling tactic dressed up as legal reasoning. The risk cuts both ways, though. Trade policy shaped by geopolitics can easily become trade policy captured by geopolitics, and the same centre-of-gravity logic used here could in principle be turned against other states whose territorial claims are contested, some of them EU member states. Procedural caution, in other words, is not always procrastination.

But the Cost of Non-Decision Is Not Neutral

Yet delay is not a costless holding position, and treating it as one is the flaw in the cautious case. Alberto Alemanno, the HEC Paris law professor, made the sharper point: every month Brussels declines to act is a month the EU continues sustaining trade with what its own institutions and a United Nations inquiry have described as an unlawful occupation, deepening rather than avoiding legal exposure. Indecision is not a neutral holding pattern. It is itself a policy, one that quietly favours the status quo while allowing officials to describe themselves as still deliberating.

The clearer evidence of the cost is what member states are doing while Brussels deliberates. Spain has barred settlement imports since September 2025, citing the International Court of Justice’s July 2024 advisory opinion that states must not enter into economic dealings sustaining the occupation. Ireland’s ban, in force since 2025, is largely symbolic: trade in the goods it covers, mainly fruit, vegetables and timber, totalled under one million euros between 2020 and 2024. The Netherlands’ Council of Ministers approved its own import ban on 22 May, with Prime Minister Rob Jetten framing it as a bar on Dutch economic activity contributing to an unlawful occupation, and the measure is due to take effect in the second half of this year. Belgium’s federal government approved a national ban at its final cabinet meeting before the summer recess, explicitly citing the absence of an EU-wide settlement trade ban as the reason it could no longer wait. Each of these moves is defensible in isolation. Collectively, they describe an EU foreign policy machinery being quietly bypassed by the very capitals that are supposed to constitute it.

A Two-Tier Foreign Policy Is Not a European One

This is the part of the story that should worry Brussels more than the settlement trade ban itself. A common foreign and security policy that member states routinely route around when it moves too slowly is not a common policy in any meaningful sense. It is a forum for coordination among states that have already decided to act, alongside states that have decided to wait, with the difference settled by national politics rather than shared strategy. That pattern predates this file. It will outlast it too, unless Brussels treats the legal classification dispute as the urgent question it actually is, rather than a technicality to be revisited once the political heat has passed.

None of this requires the EU to choose the maximalist option by October. It requires the EU to decide, promptly, whether it is institutionally capable of choosing anything at all on this question without unanimous consent, because that answer will shape far more than the settlement trade ban alone. The bloc’s largest trading partner status with Israel, worth some seventy billion euros annually according to Eurostat-sourced Commission trade data, gives Brussels real weight to bring to bear, if it can agree on how to use it. Every capital that goes it alone while Brussels debates jurisdiction is not undermining the common foreign and security policy. It is quietly replacing it, one unilateral ban at a time, and the CFSP will not survive many more months of being outpaced by the states it was built to coordinate.