The Nord Stream war exclusion has survived its first courtroom test, and the ruling reads as a green light for underwriters betting on broad war wording in an era of deniable, state-linked sabotage. On 6 July 2026, the Commercial Court found that Nord Stream AG’s insurers do not have to pay out on a claim that had grown to roughly EUR 580 million ($662 million) over the destruction of three Baltic Sea pipeline strings in 2022. The judgment turns on how loosely a war exclusion can be read when nobody can — or will — name the attacker.
Commercial Court Deals a Blow to Nord Stream AG’s Recovery Bid
The case, formally Nord Stream AG v Lloyd’s Insurance Company SA & Anor, cited as [2026] EWHC 1685 (Comm), was decided in London’s Commercial Court, part of the King’s Bench Division. The judgment, handed down by Dame Clare Moulder DBE on 6 July 2026, is set out in full in the Commercial Court judgment published on the National Archives caselaw database. The defendants named in the proceedings were Lloyd’s Insurance Company S.A. and Arch Insurance (EU) DAC, two of the carriers that had written all-risks property cover over the pipeline infrastructure.
Nord Stream AG’s claim had grown since the dispute first became public. What was reported at roughly EUR 403 million when insurers first signalled they would deny liability — a dispute chronicled in earlier coverage of insurers denying liability on the pipeline claim — had climbed to the roughly EUR 580 million figure argued at trial.
At the center of the ruling is Exclusion 2.i of the policies, which bars cover for loss or damage directly or indirectly occasioned by, happening through, or in consequence of war, among other listed perils. The court held that this wording was engaged by the September 2022 detonations, closing off Nord Stream AG’s principal route to recovery under its all-risks property cover.
A “Loose Causal Connection” Standard Redraws the War-Exclusion Bar
The most consequential part of the judgment is not the outcome but the reasoning that produced it. The court held that the causal language in the war exclusion demands only a loose causal connection between the loss and the state of war, a much looser threshold than the proximate-cause or efficient-cause tests that ordinarily govern insurance causation disputes. That distinction matters because proximate-cause tests tend to isolate the single dominant trigger of a loss, while a loose-causal-connection standard lets a tribunal treat a broader chain of contributing circumstances as sufficient to switch the exclusion on.
Applying that lower bar, the judge reasoned that a sabotage plan said to have been conceived years before execution, but only carried out in the immediate aftermath of Russia’s invasion, still demonstrated that the war was an indirect cause of the attack. Insurers had argued from the outset that the blasts were connected to the Russia-Ukraine war, which began on or about 24 February 2022, and so fell within the policy meanings of war, invasion, hostilities and military power; the court’s causation analysis effectively accepted that framing.
For drafters, the ruling is a signal that English courts are willing to stretch war-exclusion causation language well beyond a narrow, immediate trigger — a reading that favors insurers resisting claims tied to grey-zone or hybrid conflict activity, where the connection between a state of war and a specific act of sabotage is rarely clean or immediate.
Court Sidesteps the Perpetrator Question Entirely
Perhaps the most striking feature of the judgment is what it declines to decide. The court did not attempt to determine who actually carried out the sabotage. Instead, the judge found that the war exclusion applied regardless of which of the possible perpetrators was responsible, reasoning that on any of the plausible scenarios the war would still have been a significant cause of the attack. That finding lets the court sidestep one of the most politically fraught questions hanging over the case, since a German criminal investigation had concluded that the most plausible explanation for the sabotage was a group of Ukrainian divers, a theory that sits alongside competing state-actor narratives that have circulated since the pipelines were struck.
By ruling that attribution is unnecessary to the causation analysis, the Commercial Court has effectively built a template for handling deniable, state-linked attacks on infrastructure: insurers do not need to prove which state or proxy actor pulled the trigger, only that the loss sits within a loose causal chain running back to an existing war or hostilities. That removes a defense strategy — forcing insurers to pin down a specific attacker — that claimants in future grey-zone disputes might otherwise have leaned on.
What the Ruling Means for War and Sabotage Carve-Backs
For marine, energy and political-violence underwriters, the judgment arrives at a moment when war and sabotage carve-backs are already under review across critical-infrastructure books. A loose-causal-connection standard that survives attribution uncertainty gives insurers considerably more confidence that a well-drafted war exclusion will hold even when the facts never definitively identify a state sponsor — precisely the scenario that has made subsea cables, pipelines and offshore energy assets harder to underwrite since 2022.
That confidence is already reshaping how standalone cover is built. Some carriers have moved to separate state-sponsored and war-linked exposure out of general property and cyber wordings altogether, a shift visible in how insurers have begun building standalone cover specifically for state-sponsored attack scenarios rather than relying on ordinary all-risks language to hold up in court. The Nord Stream ruling gives that trend a legal foundation: buyers of critical-infrastructure cover now have reason to seek dedicated sabotage and hostile-act policies rather than general property placements that assume war exclusions will be read narrowly.
The ruling also sharpens a transatlantic contrast in how governments versus private markets absorb attacks on infrastructure tied to geopolitical conflict. In the United States, Congress has continued to lean on a public backstop for terrorism-linked losses, most visibly through the way lawmakers have extended the federal terrorism reinsurance backstop out to 2034. London’s Commercial Court, by contrast, has confirmed that private war exclusions can shut off recovery entirely, leaving European infrastructure owners more exposed to gaps between property, war and terrorism wordings than US counterparts with a state-backed mechanism to fall back on.
Energy asset owners and brokers are likely to respond by pushing harder for explicit sabotage buy-back endorsements that survive a loose-causation test. Underwriters, for their part, have received judicial validation that broad exclusionary language, drafted with the reach tested here, can hold even against a claim exceeding half a billion euros.
Mini-FAQ
How much was Nord Stream AG’s insurance claim worth?
Why did the war exclusion apply without identifying the attacker?
Who were the insurers in the case?
Sources
