By Amara Osei

A $16 billion arbitration claim against Luxembourg has coincided, perhaps not accidentally, with Mikhail Fridman's removal from the EU sanctions list on 22 September 2026. The Council delisted the Russian-Israeli businessman weeks after Luxembourg itself reportedly advocated his removal. The Grand Duchy faces its first known investment treaty arbitration. The optics are, to put it mildly, instructive.

What happened, and when?

The EU designated Fridman in February 2022, shortly after Russia's full-scale invasion of Ukraine, freezing his funds and economic resources across member states. The freeze reached assets held through Luxembourg, including his 37.9% stake in the investment group LetterOne.

In 2024, Fridman filed for arbitration under the 1989 Bilateral Investment Treaty between the Belgium–Luxembourg Economic Union and the Soviet Union, seeking approximately $16 billion. His theory is indirect expropriation: Luxembourg's implementation of the EU asset freeze, he argues, interfered with BIT-protected investments.

Fridman has litigated elsewhere. On 10 April 2024, in Fridman v Council (T-304/22), the EU General Court annulled the acts listing him for the period between 28 February 2022 and 15 March 2023. The Court held that material suggesting proximity between Fridman and Vladimir Putin did not sufficiently substantiate the listing criteria (par. 67), nor did it establish support for, or benefit from, Russian decision-makers undermining Ukraine's territorial integrity (pars. 62-64). Later listings survived and remain under challenge in T-296/23 and T-605/24. Fridman also filed a separate arbitration against the Netherlands in March 2025.

Then came 22 September 2026. The Council removed Fridman from the list, and multiple reports link Luxembourg's advocacy to his pending claim. Prime Minister Luc Frieden confirmed his government initially supported continued sanctions but reconsidered after France sought changes to the list, notably the removal of Alisher Usmanov. Frieden explained that Luxembourg acted in the "overriding national interest" to "avert harm to the country". The phrase does considerable work.

Is this 'regulatory chill' — or something narrower?

Commentators have already connected the episode to "regulatory chill": a state abandons, delays or weakens a public interest measure because a measure might trigger an investment claim and its costs. The classic literature concerns environmental regulation and fossil fuel phase-outs.

The Fridman facts may warrant a narrower label — call it administrative chill. Regulatory chill concerns rules of general application. Here, the EU sanctions framework remained untouched; only its application to one individual changed. That nuance matters. Investment claims may affect not only whether states adopt general measures but how those measures are administered in individual cases.

Luxembourg's position inside the EU adds another layer. It could not unilaterally disapply EU sanctions. Instead, it lobbied through the Council process. One member state's exposure to ISDS liability thus influenced negotiations over a collectively maintained policy.

Why would a wealthy state blink?

A $16 billion claim equals roughly 16% of Luxembourg's annual GDP for 2025. The claim is not an award, and an award is not an estimate of liability. Yet even a fraction of that sum would fuel domestic political debate — unwelcome news for a government with diminishing popularity.

The lesson is uncomfortable: capacity to defend a claim and willingness to tolerate the consequences of liability are distinct considerations, including for developed states.

Does delisting end the arbitration?

No. Losses allegedly sustained while assets were frozen may remain compensable if the tribunal establishes jurisdiction and finds a BIT breach. Restoring access may affect the damages assessment without curing every alleged loss, particularly since US and UK sanctions remain in place. Luxembourg stays exposed, and that exposure creates room for bargaining: settlement of a claim could, in principle, be exchanged for sanctions relief. Whether any such arrangement exists here is not publicly known.

What does the sequence teach other sanctioned investors?

Other designated oligarchs can observe the chain: designation, a substantial investment claim, a shift in the respondent state's position, and delisting. Whether or not Luxembourg's motives are ever conclusively established, perception suffices. If the sequence reads as a viable strategy, similar investors may reproduce it — and governments anticipating claims may grow more cautious about renewing designations.

This also dents the old claim that investment arbitration "depoliticises" disputes by channeling investor grievances away from diplomatic confrontation. Fridman's claims against Luxembourg and the Netherlands are formally separate proceedings arising from one EU response to one war. Considered collectively, they feed back into the political conditions under which sanctions are maintained: sanctions generate claims, and claims may reshape sanctions.

The broader risk is a dynamic resembling ransom, with the threat of substantial liability maintained until a respondent state helps secure relief. Settling one claim cheaply may simply advertise the price of the next. That is the plain takeaway.