Reinforcement of Loss of Chance as the Standard Basis for Assessing Loss in Insurance Brokers E&O; Norman Hay v Marsh Ltd.

The Court of Appeal’s judgment in Norman Hay v Marsh Ltd (2025) is a significant recent development in the area of insurance broker negligence, particularly as applied to liability policies. The decision reinforces the notion that, in a claim against a broker for failing to place adequate liability cover, the policyholder claimant does not need to show that it would have been held liable to the third party in order to recover. Instead, the matter should be viewed in the round and assessed by reference to loss of chance principles.

Factual background

An employee of a company within the Norman Hay group, whilst on a business trip in the US, hired a car but declined to purchase appropriate insurance. Tragically, the employee then died in a crash where he was at fault. Others were also either killed or seriously injured.

One of the injured parties sued the underlying employer. Norman Hay stepped in and settled the claim for US $5.5 million.

Norman Hay then sued its own broker (Marsh) for negligence in failing to advise on or procure suitable “non owned auto” liability insurance that would have covered the employee regardless of whether or not he had opted for the appropriate cover whilst in the US.

Marsh denied liability on numerous grounds, including the fact that there was no pleaded or established actual liability to the third party on the part of Norman Hay. They argued, did Norman Jay settle the underlying claim unnecessarily, on the basis that they themselves owed the third party no duty?

Legal issues and the Court’s approach

Marsh applied for summary strike out or reverse summary judgment, their key point being that a liability insurer would only pay out where the insured is actually held liable; and, in the absence of a clear finding of liability in this case, the claim against the broker should fail.

At first instance, the judge accepted that a standard liability policy would typically only respond to actual liability. However, when looking at things through the lens of a brokers negligence claim, it ultimately felt inappropriate to give a summary view based on a hypothetical. Instead, what was needed was a full appraisal as to whether or not the policy would have responded in this case had it been tested. The Court therefore declined to award summary judgment and instead ordered that the matter should proceed to a full trial on all issues of fact, causation and quantum. Marsh appealed.

On appeal, the Court of Appeal affirmed the approach of the first instance court. Lord Justice Males (with whom the rest of the panel agreed) held:

• The distinction between a claim under a liability policy and a claim against a broker is that, whilst a policy may strictly require actual liability to respond, a claim against the broker is about what would have happened, and whether an insurer with full knowledge and a good defence would (realistically) have paid had they been approached.
• The question of whether there would have been valid coverage (or the insurer would have resisted) is a question of fact, appropriate for full trial but inappropriate for summary judgment.

Key takeaways

This judgment evidences that brokers negligence claims will not slavishly follow the boundaries of a liability policy’s coverage. Whilst a liability policy contractually requires actual liability in order to respond, a claim against a broker in respect of a failure to obtain (the correct) cover will be judged on a counterfactual: would the policy have responded, had it been called upon?

Whilst Marsh’s application for summary judgment may have failed in this case, it is important to note that the Court did not rule out the possibility of a future dismissal of the claim. It simply said that summary disposal was not appropriate.

Ultimately, this judgment reinforces the fact that broker negligence cases in this type of scenario cannot be short-circuited by deploying technicalities. Evaluating liability and quantum will instead need a more substantive approach, and the value of any claim will most likely end up being determined on the basis of ‘loss of chance’. As readers will know, this approach potentially allows for (often significant) recoveries by policyholders against brokers, even when the chance of the underlying policy responding can be shown to be quite low. In view of this, those facing such claims would be advised to consider early settlement as an option (no matter how unfair that approach may feel), for fear that even a low percentage lost opportunity could sound in substantial damages if tested at trial.

For further questions, or if you require assistance in relation to broker negligence claims, please contact Joe Bryant.