Briefing

Court dismisses Nord Stream pipeline insurance claim

The Court has handed down judgment in Nord Stream AG v Lloyd’s Insurance Company SA & Another¹. The judgment will be of interest, not only for its highly topical subject matter, but also due to its discussion of various common insurance law issues (such as construction and causation) in the context of offshore energy insurance. However, it might be said that the outcome of the claim is ultimately unsurprising in many respects.

Background

The Nord Stream pipelines run from Russia, under the Baltic Sea, supplying natural gas to Germany. Nord Stream AG (the claimant) operated one set, Nord Steam 1, with pipelines NS1 Line 1 and NS1 Line 2². On or about 26 September 2022, the pipelines were damaged by explosions occurring approximately 6.5 km apart, rendering them inoperable (the “Explosions/Rupture Damage”). Line 2 was also damaged in a separate place in the form of an indentation (the “Dent”) although the date of that was in dispute³.

Access to the explosion sites was prevented by exclusion zones established by Swedish and Danish authorities. The claimant could not begin subsea inspections until 29 October 2022 (Swedish sector) and 12 November 2022 (Danish sector), and the authorities’ own inspection findings were not made available. The Dent was first identified by a marine contractor employed by the claimant to carry out a seabed disturbance survey on 30 October 2022. It was agreed by the experts that explosives were used that caused the initial piercing of the pipelines and that this was followed by pipe whip causing bending failure. No party has claimed responsibility for the attack.

The claimant was the owner of the pipelines. The defendants were primary and excess layer insurers of the pipelines⁴, and denied cover to the claimant in respect of the damage sustained, on the basis of a policy exclusion in relation to war.

The trial in this matter took place in front of Dame Clare Moulder DBE, over six weeks, hearing substantial expert evidence across a number of fields including geopolitics, the energy insurance market, and materials and explosive science.

Policies

The policies were Operating All Risks policies with wordings derived from the WELCAR 2001 form. They were laid out as a sheet of Policy Declarations, followed by General Conditions, and then three sections, and Section 1 was headed “Property Damage“.

The “Property Damage” section of the policy was subject to an exclusion, Exclusion 2.i, which stated:

“Notwithstanding anything to the contrary herein, this section does not cover loss or damage directly or indirectly occasioned by, happening through, or in consequence of war (whether war be declared or not), invasion, acts of foreign enemies, hostilities, civil war, rebellion, revolution, insurrection, military or usurped power or confiscation or nationalisation or requisition or destruction of or damage to property by or under the order of any government or public or local authority except as otherwise provided in Section I of the Policy.” [emphasis added]

It was common ground that the policies incorporated both the Institute Clauses for Builders’ Risks (“ICBR”) and the Institute War Clauses Builders’ Risks (“IWCBR”).

Judgment

ICBR and IWCBR

As already noted, it was common ground that the policies incorporated the ICBR and the ICWBR. The ICBR makes clear on its face that it only deals with property under construction.

However, the claimant sought to argue that as there was no express insuring clause in the policy wording the parties had chosen “the scheme” of the Institute Clauses to provide the primary insuring clause in the policies. It was said that the Institute clauses should take precedence over Exclusion 2.i, as if the loss came within the cover positively provided by the Institute Clauses then the Exclusion did not apply. The claimant accepted that this was an unconventional way of defining the obligations in the policies, and that it would not normally be expected that the ICBR would cover an offshore pipeline, but this was nonetheless what the parties had agreed.

The defendants accepted that the policies did not contain an express insuring clause, but it was said that the name of the policy and the declaration that it was an “all risks” policy was sufficient to oblige insurers to indemnify on an all risks basis, and it was not necessary for there to be an insuring clause in these circumstances where it is clear what has been agreed. Alternatively, it was said that an insuring clause should be implied. The defendants submitted that any reasonable parties to the policies would understand that the incorporation of ICBR was due to the policies continuing to be based on the WELCAR form, and which might contain provisions not needed once construction had ceased.

The judge found that if it was right that the ICBR was intended to prevail over the rest of the policies, that would require the court to disregard the fact that the ICBR is on its face limited to property under construction. The claimant’s argument that the court should do so on the basis that amendments to standard terms incorporated into policies are generally allowed in order to give effect to the intentions of the parties, was rejected. The judge found that if the parties in this matter had intended to incorporate the ICBR beyond property under construction it would be expected that they would have made more comprehensive amendments to it (the only change being a generic change to substitute “the property insured hereunder” for “vessel“). Absent a case of mistake or rectification (which were not pleaded) it was not clear on what basis the court could make adjustments to the wording, nor what those adjustments should be.

The judge found that although it was clear the parties did not intend to limit the whole of the policies to property under construction, it did not follow that they intended the ICBR should apply without limit. The policies operated without the need to rely on the ICBR and even if the claimant was right that the insuring clause was contained in the ICBR it did not follow that the ICBR terms in general should take precedence over other terms in the policies, and Section I and II of the ICBR terms should simply be disregarded. It was also noted that the claimant largely did not distinguish between the ICBR and the IWCBR, and it was unclear precisely how the claimant extended its arguments to the IWCBR. However, to make the IWCBR applicable the court also had to accept that the pipelines were “launched” once they were installed offshore, and this would involve inviting the court to rewrite the contract. The ICWBR state that it is confined to floating assets, and again it was unclear on what basis the court would rewrite the terms. The judge considered in addition that the evidence of energy insurance market practice, that war risks were confined to floating assets, was relevant background knowledge that supported the defendant’s case.

Drawing this together the court concluded that the ICBR was not the means of providing the primary insuring clause, the ICBR as incorporated was limited to property under construction or repair, the IWCBR as incorporated was limited to floating assets, and the damage to the pipelines did not fall within the scope of the cover granted by the IWCBR.

Construction of Exclusion 2.i

The Judge then went on to consider the correct construction of Exclusion 2.i finding as follows:

There were two separate limbs to the Exclusion so that it would apply if either the War Exclusion applied (“loss or damage directly or indirectly occasioned by, happening through, or in consequence of war (whether war be declared or not), invasion, acts of foreign enemies, hostilities, civil war, rebellion, revolution, insurrection, military or usurped power”) or the Government Act/Order Exclusion applied (“confiscation or nationalisation or requisition or destruction of or damage to property by or under the order of any government or public or local authority”) . The proviso “except as otherwise provided in Section I of the Policy” qualified only limb 2.

The judge considered what casual connection was required by the wording: “directly or indirectly occasioned by, happening through, or in consequence of war”. The claimant accepted that this wording did not mean proximate cause, but submitted the defendant had to show that the war was “an effective cause” of the damage. However, the judge rejected this, finding that the phrase was very broad. The defendants accepted that it was sufficient if the damage had a “significant” causal link to a war. The judge held that the cause needed to be capable of being identified as a contributing factor, and a significant cause was “noticeable” or “specifically accountable” but the test was not that it needed to be of real significance if the claimant was thereby seeking to distinguish between different causes and imply some higher degree of comparative causal contribution.

Applying the above findings to the facts, the court took first the Explosion/Rupture damage and secondly the Dent damage.

Explosion/Rupture Damage

The judge held that there were only three likely perpetrators of the damage to be considered: the US, Russia and the government of Ukraine (or a sub-state actor of Ukraine). Taking each of these possibilities in turn and considering the expert evidence, the court considered whether they had the requisite connection with the war (there being no doubt that the events fell within the extended definition of war in the policy).

The judge found that:

If Ukraine was the perpetrator then that was part of the overall war effort with the motivation of striking a blow against Russia for its February 2022 invasion. If Ukraine carried out the sabotage then the war was a significant cause of its actions. The position would be similar if a sub-state actor of Ukraine carried out the attack.
If Russia was the perpetrator then it would have had no motive for the attack before February 2022, but would have wanted to punish Germany for giving tangible support to Ukraine and/or to coerce it to change its position. The war would therefore have been a significant cause of Russia’s actions.
If the US was the perpetrator then this would have been to threaten Russia about the consequences of continuing the war, and the war would have been a significant cause of its actions.

The claimant submitted that the defendants had not shown a significantly strong connection to the war, but rather that it was the background against which it was carried out. The judge accepted that there was a point at which the causal link would become too remote, however it was not the acts that constituted the war that had to directly cause the damage, they could indirectly cause the damage because the war permitted or encouraged those acts. Although there were gaps in the evidence to establish which perpetrator carried out the attack and how, this was not necessary to determine if there was a sufficient causal connection, The court’s finding was that if any of the possible perpetrators carried out the act then the war would be a significant cause of their actions.

Therefore the Explosion/Rupture damage was excluded by the policy.

The Dent

The next issue for the court was whether the Dent was also caused by an explosion or from another mechanism such as an anchor impact. The court discussed the evidence in relation to this in some detail.

Relevant to this issue, according to the claimant, was the well known proposition put forward in Rhesa Shipping v Edmunds (The Popi M) that Courts do not…..subscribe to Sherlock Holmes’s maxim (“when you have eliminated the impossible whatever remains however improbable must be the truth”): if the admissible evidence leaves a relevant possible conclusion improbable then it is not proven”.

In other words, the claimant argued that where there is thin evidence the right conclusion may be that even the most likely of the possible explanations may not satisfy the test of being more likely than not.

The Popi M was applied by the Court of Appeal in Ide v ATB Sales Ltd, in which it was noted that the Popi M was an unusual case, in which all explanations of the cause of loss of a ship were acknowledged to be highly improbable. The Court of Appeal said that in the vast majority of cases there will be two or more competing explanations as to how an event occurred, which although may be uncommon are not improbable. It is then a permissible train of reasoning for a judge, having eliminated all the causes of loss but one, to ask on the balance of probabilities if that one cause was the cause of the event. It is in cases, such as the Popi M, where there is real uncertainty about the relevant factual background, or where the evidence is highly unsatisfactory, that a process of elimination can result in the least improbable cause and not the probable cause.

Applying it to this matter, the judge found that court was not here in the same position as the Popi M. On the basis of the evidence (including that the damage was close to a welded field joint, it was known that sabotage took place in close proximity, and the Dent resembled the known explosions) the judge was able to determine that it was more likely than not that the Dent was caused by an explosion as part of the same attack that caused the Explosions/Rupture Damage.

Therefore, in conclusion, the Dent damage was also excluded.

Comment

The Court found that there was no cover under the policies in respect of any of the damage claimed. It seems unsurprising on the face of it that the incorporation of standard clauses (here ICBR and IWCBR) could have the effect of nullifying a very clearly drafted exclusion, and the war exclusion in this case was extremely wide (“directly or indirectly occasioned by” war etc). Although there was also an attempt to resurrect arguments from a 1985 House of Lords case (“Popi M“), where there was no obvious cause that could not be ruled out, the facts here were very different, because there was a fairly obvious potential cause which in no way could be ruled out. Although there was an opportunity for the court to create new law in what was undeniably a large loss, the odds would appear to have been stacked against this happening on this occasion. The most obvious cause was accepted, and that cause (war) was clearly excluded.

Footnotes

[2026] EWHC 1685 (Comm)
For our previous briefing about the attacks see here
Nordstream 2 was also damaged, but has a separate owner and this was not part of the claim.
The insurers sued were acting in a representative capacity for all insurers