KOHEN AVOCATS FIRM · PARIS 16
Hungary Lost in Luxembourg: Who Can Still Challenge the EU’s Use of Frozen Russian Billions for Ukraine?

Hassan KOHEN acts in urgent matters, from the police station to the assize court. Initial strategic analysis: €80 incl. VAT, personal reply within 24 hours.

100% confidential · Legal professional privilege · No obligation

Article generated by artificial intelligence, through a process designed and supervised by the firm

Source: French Court of Cassation – Judilibre and Légifrance open data

Paris Bar Police custody, investigation, assize court CNB listing avocat.fr
Maître Hassan KOHEN, avocat au Barreau de Paris
Hassan KOHEN
Attorney at the Paris Bar

Hungary Lost in Luxembourg: Who Can Still Challenge the EU’s Use of Frozen Russian Billions for Ukraine?

Speak to a lawyer 06 89 11 34 45

On 9 September 2026, the General Court of the European Union in Luxembourg dismissed Hungary’s attempt to block the use of profits generated by frozen Russian state assets to buy weapons for Ukraine. The judgment in Case T-457/24, Hungary v European Peace Facility and Council, rejects Budapest’s challenge to the decision that allocated the first instalment of those extraordinary revenues to the Ukrainian Armed Forces. The Court’s answer is blunt: it has no jurisdiction, because the decision reflects political and strategic choices made under the Common Foreign and Security Policy. Reuters, Euronews and the Ukrainian press reported the outcome within hours, and Hungary now has a short window to appeal to the Court of Justice. For an international reader, and especially for any person or company whose funds are frozen in France under the EU sanctions regime, this ruling matters far beyond the Budapest-Brussels quarrel. It confirms that the EU courts will not second-guess the political decision to divert the windfall from immobilised Russian billions to Kyiv, while leaving intact, under strict conditions, the individual remedies of listed persons. This analysis explains why Hungary lost, what the judges actually said, and what paths remain open in Paris and in Luxembourg when assets are frozen.

I. Why did Hungary lose its fight over frozen Russian billions in Luxembourg?

Hungary did not attack the principle of sanctions against Russia. It attacked a procedure: the way the European Peace Facility decided, without Budapest, to spend the interest earned on frozen Russian central bank assets on military aid to Ukraine. To understand why that attack failed, one must follow the chain of decisions from 2021 to June 2024, and then grasp the jurisdictional wall on which the action broke.

A. Can a Member State that refused to vote complain about the vote?

The story starts in 2021, when the Council created the European Peace Facility through Council Decision (CFSP) 2021/509 of 22 March 2021 establishing a European Peace Facility, an off-budget fund designed to finance Union actions with military or defence implications under the Common Foreign and Security Policy. The Facility is run by a committee of Member State representatives, the Facility Committee, and decides, as a rule, by unanimity. That unanimity rule is also the source of the mechanism Hungary used: constructive abstention. Under the second subparagraph of Article 31(1) TEU, a Member State may abstain from a vote and qualify its abstention by making a formal declaration; the decision is still adopted, the abstaining State is not obliged to apply it, but it accepts that the decision commits the Union. Abstention is therefore a way to stand aside without vetoing, and Budapest has used it repeatedly on Ukraine-related military aid.

On 21 May 2024 the Council adopted two linked acts. The first, Council Decision (CFSP) 2024/1470 of 21 May 2024, set the principle: the extraordinary revenues generated by the frozen assets of the Central Bank of Russia would be channelled to support Ukraine. The second, Council Decision (CFSP) 2024/1471 of 21 May 2024, organised the allocation of those amounts through the Peace Facility. Hungary voted in favour of the first decision and constructively abstained on the second. That split is crucial. By endorsing the principle while standing aside from the implementing scheme, Budapest placed itself in a hybrid position: politically associated with the direction of the profits to Ukraine, legally recorded as a non-participant in the assistance measure. The operative core of the second decision is its Article 1, which channels the Facility-bound share of the 2024/1470 revenues into assistance measures supplying military support to the Ukrainian Armed Forces, on top of the voluntary contributions governed by Article 30 of Decision 2021/509.

One month later, on 21 June 2024, the Facility Committee, acting by written procedure, allocated the first instalment of those profits to the Ukrainian Armed Forces and held that Hungary could not take part in that vote, precisely because it had constructively abstained on the May allocation decision. From the Committee’s viewpoint the logic was consistent: a State that formally stood aside from an assistance measure cannot then co-decide its execution, and the 2024/1471 scheme expressly provides that abstaining States are not contributing members for the measures concerned and bear neither the costs nor the related non-contractual liability. Hungary saw the matter in the opposite light. It brought an action before the General Court seeking annulment of the 21 June 2024 decision and of the minutes recording it, arguing that the voting rules had been breached, invoking Article 31(1) TEU and Articles 11(8) and (14) of Decision 2021/509, and adding a broader constitutional plea: the exclusion allegedly offended the Union’s fundamental values, the equality of Member States and the democratic functioning of the Union. The paradox of the case is therefore sharp. Hungary asked judges to review a vote from which it had deliberately absented itself, in the name of principles whose guardian, it argued, must be the Court even inside foreign and security policy.

The sums at stake explain the intensity of the fight. Around 210 billion euros of Russian sovereign assets are immobilised in the Union, mostly as cash balances accumulated on the books of central securities depositories after principal and interest payments to the Central Bank of Russia were prohibited. Those balances, prudently managed, throw off large interest income: EU reporting relayed by the financial press speaks of a fifth tranche of about 1.4 billion euros delivered in early August 2026, bringing total extraordinary revenues directed to Ukraine to roughly 8 billion euros. The Council’s reasoning, set out in the recitals of Decision 2024/1470, is that the European Council had called for decisive progress so that extraordinary revenues held by private entities and stemming directly from immobilised Russian assets could support Ukraine’s recovery and reconstruction, echoing the European Council call for work, in coordination with partners and consistently with contractual obligations and Union and international law, on directing such revenues toward Ukraine recovery and reconstruction. Hungary’s action threatened the legal plumbing of that scheme, not only the first instalment but every future allocation decided the same way. That is why the judgment of 9 September 2026 resonates from Kyiv to Paris: it secures the revenue stream, at least against this type of procedural challenge.

B. Why did the EU judges declare they have no power to decide?

The General Court never reached Hungary’s arguments on voting rules or democratic principles, because it held that it had no jurisdiction to examine them at all. That outcome flows from one of the oldest structural features of EU law: in principle, the Court of Justice of the European Union has no jurisdiction over the Common Foreign and Security Policy or over acts adopted on its basis. The carve-out is written in the last sentence of the second subparagraph of Article 24(1) TEU and in the second paragraph of Article 275 TFEU. Only two narrow exceptions let the EU courts into this field: review of restrictive measures adopted against natural or legal persons, and policing the boundary between CFSP and other Union competences under Article 40 TEU. Everything else, the treaties say, belongs to the Member States acting politically in the Council, not to judges.

Before applying that rule, the Court settled a preliminary question in Hungary’s favour, and practitioners should not miss it. The Peace Facility has legal capacity, so the action directed against an act of its committee is not one of the categories reserved to the Court of Justice for actions by a Member State against the Council; the General Court was therefore the correct court to hear the case. Competent to hear, but not competent to judge: that is the subtle distinction the judgment draws. The Court then recalled that neither treaty exception applied here. Hungary is a Member State, not a listed person challenging restrictive measures against itself, and no Article 40 TEU boundary dispute was at stake. The decisive test came from the Court of Justice’s 2024 ruling in the joined cases KS and KD v Council and Others, C-29/22 P, and Commission v KS and Others, C-44/22 P, paragraph 117, which the press release cites expressly: where a CFSP act is directly related to political or strategic choices, the EU courts must decline jurisdiction, including over its so-called external legality. The official Press Release No 123/26 on the judgment of 9 September 2026 states the holding without ambiguity: the Court disclaims jurisdiction to review the legality of such a decision, including its external legality, because the act is directly tied to political and strategic choices taken within the CFSP..

Applied to the facts, the test was easily met. The contested 21 June 2024 decision did not merely execute technical details; it chose where the money goes and what it buys. In the reasoning of the Court, the contested decision is strategic through and through: it fixes how the money is shared out to finance military equipment for the Ukrainian Armed Forces and selects the assistance measure receiving it. Choosing weapons for the Ukrainian army over any other use of the billions is, in the Court’s eyes, the very definition of a strategic choice, and strategic choices are for the Council, not for judges. The broader pleas about the rule of law, equality of Member States and democratic functioning changed nothing, because those principles, however fundamental, do not create jurisdiction where the treaties exclude it. The judgment is careful to present itself as an unofficial media document that does not bind the Court, but its reasoning follows the KS and KD line faithfully and leaves little room for surprise.

What remains for Budapest is the appeal. As every Curia press release recalls, an appeal lies to the Court of Justice on points of law only, within two months and ten days of notification. An appeal could argue that the Court read the political-choice exception too broadly, or that excluding an abstaining Member State from implementation votes does raise a reviewable question of institutional balance under Article 40 TEU. Those are serious arguments, and the Court of Justice has been slowly expanding judicial protection at the margins of CFSP, as KS and KD itself shows. But the odds are long: the allocation of billions to lethal aid for a belligerent army sits at the hard core of foreign policy, exactly where the treaty drafters wanted judges to stay out. Unless the Court of Justice revisits its own 2024 case law, the dismissal should survive, and the revenue pipeline to Kyiv keeps flowing while the appeal runs, since appeals do not suspend the contested act without a separate interim order.

II. Can you still challenge an asset freeze when your funds are blocked in France?

The Hungary judgment closes one door: no Member State, and no litigant dressed as one, will get the EU courts to annul the political choice of sending Russia-linked windfalls to Ukraine. But it leaves open the doors that matter for individuals, companies and families whose accounts, dividends or real estate are frozen in France. Sanctions practice in Paris follows a two-track logic, and confusing the tracks is the most common and most costly mistake. One track is European: the listing itself, decided in Brussels and reviewable only in Luxembourg. The other track is national: the implementation of the freeze on French territory, handled by the Treasury, the banks and, where needed, the French courts. Each track has its own judge, its own deadlines and its own evidence. Here is how a French practitioner maps them after 9 September 2026.

A. Which court can release frozen money: Paris or Luxembourg?

Start with the freeze you actually feel. In France, EU financial sanctions are applied through the Monetary and Financial Code, whose asset-freeze chapter gives the mechanism its domestic teeth. Article L. 562-3 of the Monetary and Financial Code provides, and I quote: “Le ministre chargé de l’économie peut décider, pour une durée de six mois, renouvelable, le gel des fonds et ressources économiques : 1° Qui appartiennent à, sont possédés, détenus ou contrôlés par des personnes physiques ou morales, ou toute autre entité qui commettent, tentent de commettre, facilitent ou financent des actions sanctionnées ou prohibées par les résolutions adoptées dans le cadre du chapitre VII de la charte des Nations unies ou les actes pris en application de l’article 29 du traité sur l’Union européenne ou de l’article 75 du traité sur le fonctionnement de l’Union européenne, y participent ou qui sont désignées sur le fondement de ces résolutions ou ces actes ;”. A parallel provision, Article L. 562-2, covers terrorism-related freezes decided jointly by the finance and interior ministers, and I quote its opening: “Le ministre chargé de l’économie et le ministre de l’intérieur peuvent décider, conjointement, pour une durée de six mois, renouvelable, le gel des fonds et ressources économiques”. In practice, most Russia-related files in Paris rest directly on EU regulations such as Council Regulation (EU) No 269/2014, which freeze the funds of listed persons across the Union without any French decree, while the Code organises enforcement: banks, insurers, notaries and every person operating in France must obey at once. Article L. 562-4 states, and I quote: “Sont tenus d’appliquer sans délai les mesures de gel et les interdictions de mise à disposition ou d’utilisation prévues au présent chapitre et aux articles L. 712-4 et L. 712-10 et d’en informer immédiatement le ministre chargé de l’économie”. Ministerial freeze decisions, for their part, are opposable from publication: Article L. 562-9 provides, and I quote: “Les décisions des ministres arrêtées en application des articles L. 562-2 , L. 562-2-1 , L. 562-2-2 et L. 562-3 sont publiées par extrait au Journal officiel et sont exécutoires à compter de leur date de publication.” And third parties are not spared: Article L. 562-10 warns, and I quote: “Les mesures prises en vertu du présent chapitre sont opposables à tout tiers qui peut invoquer un droit sur les fonds ou ressources économiques faisant l’objet d’une mesure de gel, y compris lorsque ce droit est né antérieurement auxdites mesures.”

The consequence is that a Paris court can settle disputes about how the freeze is applied in France, for example a bank that over-blocks, a co-owner whose share is caught by a spouse’s listing, or a creditor chasing frozen funds, but no French judge, administrative or civil, can annul the EU listing itself. Only the General Court in Luxembourg can do that, and this is precisely the exception the treaties preserve inside the CFSP no-go zone: actions by listed natural or legal persons against restrictive measures concerning them are reviewable. The Hungary ruling changes nothing on this point; if anything, it underlines it, since the Court distinguished Hungary’s political challenge, which it refused to hear, from the individual-remedy cases that remain admissible. Three recent outcomes illustrate the spectrum. On 3 September 2026 the Court of Justice struck down the freeze imposed on Inter Rao Lietuva, a gavage of hope for applicants who can show factual errors, as analysed in our commentary on the Inter Rao Lietuva ruling. By contrast, a tobacco billionaire and then Roman Abramovich both failed to get delisted in Luxembourg this month, the Courts accepting that the Council enjoys broad discretion and that criteria targeting leading businesspersons in revenue-generating sectors are precise and proportionate, as explained in our analysis of the Kesaev defeat and our analysis of the Abramovich defeat of 9 September 2026. The lesson for a French file is realistic rather than bleak: winning in Luxembourg demands proof, not politics. The standard route runs in four steps. First, file a reasoned request for reconsideration with the Council, attaching evidence that the listing criteria are no longer met, for instance resignation from the relevant functions, divestment, or proof that the sectoral link was wrongly assessed. Second, if the Council maintains the listing at one of its periodic reviews, bring an action for annulment before the General Court within two months of notification or publication, laying out precise pleas: manifest error of assessment, breach of the duty to state reasons, disproportionate interference with property and private life. Third, seek interim relief suspending the freeze where urgency and a prima facie case coincide, for example a company facing imminent insolvency because operating accounts are blocked. Fourth, consider a damages action for unlawful conduct where the listing was gravely and manifestly mishandled. Each step needs documents a Paris team can assemble: corporate registers, audited accounts, employment and residence records, contracts proving the end of the incriminated activity, and every exchange with the bank and the Treasury. For a full French-language guide to the domestic side of these challenges, readers may consult our pillar page on contesting and unblocking frozen assets in France.

Two traps kill otherwise winnable files. The first is delay: the two-month Luxembourg time limit runs fast and is almost never extended, so a file opened in Paris must be aimed at Luxembourg from day one, with translations and powers of attorney ready. The second is attacking the wrong act: asking a French administrative court to suspend an EU regulation wastes months and ends with a declaration of incompetence, while the real fight, over reasons and evidence, should have been lodged in Luxembourg. The Hungary judgment is useful pedagogy here. Budapest attacked the political allocation of someone else’s windfall and was told the courtroom was closed. A listed person attacks their own designation with facts, and for them the courtroom in Luxembourg stays open, provided they arrive on time and armed with proof.

B. How do sanctioned persons and companies live and operate in France while assets stay frozen?

A freeze is not a confiscation, and that distinction is the key to daily survival. Frozen funds still belong to their owner; they are immobilised, and any movement needs prior authorisation. The Hungary case itself rests on an even finer distinction drawn by the Council: the Russian principal, roughly 210 billion euros of sovereign assets, stays frozen and untouched, while only the unexpected interest those balances generate is diverted to Ukraine. The recitals of Decision 2024/1470 explain the theory: since those unexpected revenues need not be returned to the Central Bank of Russia even once the transaction ban ends, they are not sovereign assets, and the protections attached to sovereign assets do not cover them. Whether international lawyers will forever accept that interest on sovereign money is not itself sovereign is a debate for the Hague Academy; for clients in Paris, the operational point is narrower and reassuring. Because the principal is preserved rather than seized, the legal framework remains one of temporary restrictive measures, which the regulations themselves equip with humanitarian and practical safety valves.

Those valves are the derogations written into Regulation (EU) No 269/2014 and its sister texts. In France, the competent authority is the Directorate General of the Treasury, which can authorise, case by case, the release of frozen funds for basic needs such as food, housing, medical care and insurance, for reasonable legal fees, for the payment of prior contractual obligations, or for compliance with earlier judicial or administrative decisions. Each request is a mini-case: identity and listing reference, bank statements, invoices, medical certificates, fee agreements, proof that the payment does not make funds available to another listed person. Approvals come with conditions, often payment directly to the supplier rather than to the listed person, and refusals can themselves be challenged before the French administrative courts, because there the judge reviews a French refusal, not the EU listing. Companies face a parallel track: keep the frozen entity alive without breaching the prohibition on making funds available. In practice that means segregated frozen accounts, continued filings and tax compliance, staff paid through authorised releases where possible, and no restructuring that could be read as circumvention, since Union law requires Member States to punish violations with effective, proportionate and dissuasive penalties, and the French enforcement chain, from Tracfin reports to criminal prosecution, takes evasion seriously. Banks, for their part, apply the Code without waiting: they freeze first and ask questions later, exactly as Article L. 562-4 orders, and any client who discovers a block should demand the written reason, identify whether the legal basis is an EU regulation number or a ministerial order, and preserve the paper trail for both the Treasury request and any Luxembourg action.

Two further French provisions complete the picture for internationally mobile clients. First, the net of obliged entities is wide: Article L. 561-2 of the Monetary and Financial Code provides, and I quote: “Sont assujettis aux obligations prévues par les dispositions des sections 2 à 7 du présent chapitre”, a list that starts with credit institutions, payment institutions and their branches, which is why a freeze decided in Brussels bites within hours at every French bank counter. Second, real estate does not escape: Article L. 562-8 allows freeze decisions to be published against the property itself, and I quote: “peuvent, à la demande du ministre chargé de l’économie, être publiées au fichier immobilier ou au livre foncier dont dépend le bien immobilier appartenant à la personne dont les fonds et ressources économiques sont gelés”. A Paris apartment or Riviera villa can therefore carry the freeze in the land register, blocking any sale long before a buyer appears.

A practical checklist, drawn from files handled in Paris, helps prioritise. First, map every frozen asset and its exact legal basis, because different bases mean different judges and different forms. Second, secure the derogations that keep life and business running: housing, treatment, payroll where eligible, and counsel fees, since an unrepresented listed person rarely wins in Luxembourg. Third, build the delisting file in parallel, not afterwards: updated corporate documents, proof of severed links with the sanctioned sector or State, independent audits, and sworn statements where useful. Fourth, calendar the Council’s review cycles and the two-month annulment deadline so that a maintained listing triggers immediate action rather than surprise. Fifth, audit the group structure for contagion, because ownership and control criteria can pull subsidiaries, spouses’ joint accounts and business partners into the freeze, and each of them needs its own strategy. Sixth, communicate carefully: public statements denying everything can resurface as exhibits, while a dignified, documented, procedure-focused posture plays better before the Treasury and the Court. None of this guarantees delisting in a climate where, as September 2026 shows, Luxembourg mostly upholds listings and only strikes down the demonstrably unfounded, but it transforms a passive freeze into a managed file with measurable milestones. And where the windfall debate touches your own assets, for instance securities held through a depository whose interest income is being skimmed toward the Facility, ask counsel early whether you, unlike Hungary, have standing as a directly concerned person for the slice that affects you. That question is genuinely unsettled, and the first well-built case on it could matter far beyond its own facts.

Conclusion

The judgment of 9 September 2026 settles the institutional quarrel and leaves the human files untouched. Hungary wanted judges to police how the Union spends someone else’s interest income on a war; the General Court answered that spending choices of a military nature are strategic, that strategic choices under the Common Foreign and Security Policy belong to the Council, and that the treaties give judges no key to that room, even when the pleas invoke the rule of law and the equality of Member States. Expect an appeal, but do not expect the money to stop flowing to Kyiv in the meantime. For listed persons and their families, the message is the mirror image. The courtroom that closed for Budapest stays open for you, on different terms: not politics but proof, not abstract principles but statements of reasons, bank records and severed links, within a strict two-month clock. September 2026, with Inter Rao Lietuva’s victory on one side and the Kesaev, Abramovich and Hungary defeats on the other, draws the map clearly. Luxembourg rewards the file that demonstrates a concrete error and punishes the file that relitigates the war. If your assets are frozen in France, start in Paris with compliance and Treasury authorisations, aim at Luxembourg with evidence, and treat every deadline as fatal. That is how a freeze, which is temporary by design, can end.

📄 Official circular