A professional indemnity insurer was liable under a non-party costs order to pay half the claimants’ costs despite having limited its liability to its insured by aggregating claims under the policy and despite having ceded control over the underlying litigation to the insured. An insurer who relinquishes control of litigation cannot expect immunity from costs liability, particularly if it is aware that there was never any reasonable prospect of successfully defending the claim. The decision shows that even when an insurer has limited its liability to its insured, it remains potentially on the hook for costs: Various Claimants v Giambrone & Law & ors [2019] EWHC 34 (QB)

Non-party costs orders: a reminder

The court’s jurisdiction to award non-party costs orders (NPCOs) derives from its general power on costs in section 51 Senior Courts Act 1981 (section 51) which states that the court has power to determine “by whom and to what extent the costs are to be paid”.1 Section 51 is supplemented by Rule 46.2 Civil Procedure Rules which refers to the section 51 power “to make a costs order in favour of or against a person who is not a party to proceedings”. Although such orders are “exceptional”, this means no more than “outside the ordinary run of cases where parties pursue or defend claims for their own benefit and at their own expense”.2 The ultimate question is whether it is just to make such an order in all the circumstances. The discretion will not usually be exercised against “pure funders”. If, however, the non-party not only funds but controls or benefits from the proceedings, justice requires that it must pay the successful party’s costs since the non-party is “not so much facilitating access to justice as…gaining access to justice for his own purpose”.3

The underlying litigation involved a class action in which negligence claims (the claims) were brought by a number of claimants (the claimants) against Giambrone & Law (Giambrone), a law firm with professional indemnity insurance with AIG (Europe) Limited (AIG). The litigation was fiercely contested at an 18-day trial, despite Giambrone having little confidence in the success of major issues and despite many other, similar claims having been settled before trial. The claimants succeeded on all counts and substantial costs were awarded against Giambrone which it failed to satisfy, leading to the claimants’ application against AIG.

Agreement to aggregate claims

Two years before trial, following a dispute between AIG and Giambrone as to the right of AIG to aggregate claims, the parties entered into an agreement (the Agreement) to settle their dispute. It included two principal terms: (i) cover for the claimants’ damages would be limited to GBP3 million (of which around GBP1.89 million had already been paid out on claims which were settled); and (ii) under clause 2.4, AIG was obliged to fund Giambrone’s defence costs, subject to a right to withdraw funding if it reasonably considered that there was “no realistic prospect of defending the [claims]” (clause 2.4). At this point, AIG was aware of the number of claims and, in view of its dispute with Giambrone, this was an incentive for it to enter into the Agreement. AIG alleged that the Agreement prevented it exercising sole, or even dominant, control, over the way the defence was conducted and that an NPCO should not be made against it.

Was the Agreement relevant to the exercise of the court’s discretion?

The court accepted that, as a result of the Agreement, Giambrone effectively controlled the defence to the litigation. Regarding clause 2.4, AIG could not have thought (whether reasonably or otherwise), that there was a realistic prospect of successfully defending the claims given that they, and their merits, were materially the same as previous cases which had settled. It was also clear that AIG knew “the defence was much more likely to fail than succeed”. A conscious decision was made by AIG not to withdraw funding.

Whilst the Agreement was “commercially sensible” it could not protect AIG from section 51 simply because Giambrone had controlled the litigation. AIG had failed to control Giambrone’s conduct of the defence, either by entering into the Agreement, under which AIG had ceded control of the litigation, or because it had failed to invoke clause 2.4. Given AIG’s awareness of the prospects of success and the tenacity with which Giambrone would protect its interests, it was foreseeable that large sums of money would be expended. In these circumstances, AIG could not expect immunity from costs liability, particularly since the prospects of success were poor. The court also relied onthe principle of reciprocity derived from Travelers Insurance v XYZ4 (another indemnity insurance case) in which the Court of Appeal held that any party who benefits from litigation can be made to bear the burden of costs. The court relied on the benefit to AIG of the Agreement in that it resolved any doubts in relation to the validity of the aggregation arguments; had Giambrone’s defence been successful, AIG’s right to aggregate would never have been tested. Whilst it was not for AIG’s sole benefit, it did derive some material benefit from it.

The court did not accept AIG’s argument that the claimants’ failure to engage with AIG’s efforts to settle the underlying litigation should prevent it exercising its discretion: it was not possible to discern from the “without prejudice” correspondence which party was responsible for the failure of the negotiations in a way which would affect the application. The court also did not accept that the fact the claimants had known and maintained the claims, knowing the terms of the Agreement, was relevant. They had made clear early on their intention to make a section 51 application; AIG had therefore taken the risk that the Agreement might not have the impact on the application that it subsequently alleged.


AIG argued that its funding had not caused the claimants to incur more costs, as it could be inferred that Giambrone had the funds to pursue its defence, since it had funded an appeal to both the Court of Appeal and the Supreme Court. The court disagreed: an appeal usually costs significantly less than a trial. Furthermore, had Giambrone funded the litigation itself, it would have been much more circumspect about its potential exposure to an NPCO. AIG’s funding did therefore materially increase the claimants’ costs.

Quantification of claimants’ costs incurred as a result of AIG’s funding

Every possible point was taken by Giambrone at trial which would probably have been avoided if AIG had not provided funding and had exercised proper control. The court concluded the claimants had broadly spent twice as much as they would otherwise have done and ordered AIG to pay half their costs.


This case illustrates the courts’ increasing willingness to grant NPCOs, both against insurers as well as against funders of corporate litigation such as directors, shareholders or parent companies, who either control or benefit from proceedings. Whilst commonly made against funders of claims, orders can also be made against defence funders. The availability of NPCOs should always be borne in mind, both where an unsuccessful opponent appears not good for the money and where a party is considering funding litigation.

For insurers who agree to hand over control of the litigation to their insured, whilst continuing to fund it, they should include a clear proviso allowing funding to be withdrawn if the prospects of success fall below a certain level. They should also regularly reassess the likely prospects of success; even if they do not withdraw funding, evidence that they considered whether the claim was defensible will be helpful in resisting any subsequent section 51 application. Where the agreement limits insurers’ liability to the insured, funding the defence of uninsured claims still leaves them at risk on costs even for those claims which exceed the indemnity limit.