Musst Holdings Limited v Astra Asset Management UK Limited & Anor

This is my judgment on an application, dated 24 January 2024 (“the Application”), by the Defendants, Astra Asset Management UK Limited and Astra Asset Management LLP, which I will refer to collectively as “Astra”, save insofar as it may be necessary to distinguish between them. The Application seeks security for costs pursuant to CPR Part 25, rr. 25.13(2)(a) and/or (c), either in a form “substantially the same” as that provided in earlier proceedings between Astra and the Claimant, Musst Holdings Limited (“Musst”); alternatively by way of a payment into Court in the sum of £720,000.

Musst, the respondent, is a BVI company. Musst does not dispute that the gateway for an order for security for costs is in principle satisfied, nor the appropriate level of security. Musst has obtained a Litigation Insurance Policy, with an Inception Date of 15 November 2023 (“the Policy”), Appendix 2 of which is an Anti-avoidance endorsement (“the AAE”).

Astra have raised a number of objections to the Policy and the AAE, and submit that the Policy is simply not suitable given the risks they have identified. The dispute between the parties on the Application is whether the Defendants’ identified objections to the Policy are valid and realistic, or merely theoretical or fanciful, such that the Policy does not provide sufficient protection to Astra in respect of the risk of any costs order that may be made in their favour not being satisfied.

These proceedings are the second set of proceedings between the parties, the first under claim nos. BL-2018-002369 and BL-2019-001483 having been tried over three weeks in April and May 2021 and the subject of a judgment by Freedman J on 17 December 2021, following a three week trial (“the Old Claim”). For present purposes, the precise details of the Old Claim do not matter. In the Old Claim, Musst submitted that, under a written contract entered into in April 2013, Octave Investment Management LLP and Octave Investment Management Ltd (collectively, “Octave”) agreed to pay to it a commission for introducing investors who invested in a “Fund” which followed the “Current Strategy” as defined in that contract (“the Octave Contract”). Musst claimed that it had introduced two such investors.

As subsequently held by Freedman J, Octave had transferred the management of these accounts to Astra LLP, and the Octave Contract was novated. Astra LLP subsequently in turn transferred the management of these accounts to Astra UK, and the Octave Contract was again novated. Astra resisted paying management fees on various bases. In addition, Astra counterclaimed on the basis that Musst’s owner, Mr Siddiqi, had made disparaging comments about Astra’s principal owner, Mr Mathur. Separate defamation proceedings were issued against Mr Siddiqi and another Musst entity. These proceedings, the dispute over the Octave Contract and the associated defamation proceedings, were tried together.

Freedman J found that Musst was entitled to be paid certain management and performance fees by Astra (to which Octave had transferred the management of various accounts to Astra, and the Octave Contract having been novated to Astra). He subsequently ordered a payment on account of US$3,826,952.23 to Musst in a consequentials judgment handed down on 18 March 2022.

Astra appealed, with the Court of Appeal’s permission, on various grounds. On 13 February 2023, the Court of Appeal dismissed the appeal.

In the course of the earlier proceedings between the parties, Astra disclosed an email to Musst in September 2020 which (so Musst says) means that there were further fees and/or revenues payable to it. Musst says that it was too late to amend its existing claim against Astra; instead it issued these separate proceedings on 29 April 2021 (“the New Claim”), serving its Particulars of Claim in the New Claim on 27 August 2021 (that is, before Freedman J’s judgment in the Old Claim).

Astra applied to strike out the New Claim, on the basis that they were res judicata and anyway they were an abuse of process as they could and should have been brought in the Old Claim. Freedman J dismissed these applications in a judgment handed down on 28 February 2023.

Musst served Amended Particulars of Claim on 30 May 2023. Astra served their Defence on 15 June 2023, and Musst its Reply on 14 August 2023. Astra disputes the New Claim on a number of grounds. In summary, Astra says that: (1) Musst is estopped by res judicata from bringing the New Claim; (2) Musst cannot establish any claim that would permit it to avoid the effects of the Limitation Act 1980; and (3) the novation did not extend to the further investments and/or the relevant investments were not made in a Fund or for the Current Strategy as defined.

In the Old Claim, when faced with a demand for security, Musst also obtained an adverse costs insurance policy and suggested that it was adequate to protect Astra’s position. Astra objected to the form of the policy. One of its objections was that the policy did not contain proper anti-avoidance provisions. In the event, Astra issued an application for security for costs in the Old Claim. Following a costs and case management conference in March 2019, Deputy Master Bartlett by Order dated 10 June 2019 ordered that “The form of security is to take the form of a deed of indemnity in substantially the same form as that proffered by the Claimant …”. The Deputy Master also made provision for the removal of a particular clause, clause 5(d) of the deed. The Deputy Master ordered Musst to pay Astra’s costs of and occasioned by the application for security, summarily assessed in the sum of £18,534.89, on the basis that Astra had obtained significantly more by pursuing their application than had been offered to them by Musst prior to the hearing.

At the hearing before me, Musst stressed that, in the Old Claim, one of the Astra’s objections was that the policy did not contain an anti-avoidance provisions but relied on a deed of indemnity. Now, when the Policy contains anti-avoidance provisions, Astra submits that any security should be provided by way of a deed of indemnity. That approach, submits Musst, reveals that the objections now being made by Astra are in truth unjustified and being raised for purely tactical purposes. Astra replies that this is not a fair characterisation of its position with regard to the provision of security in the Old Claim. Whilst it is correct that, in December 2018, Astra’s solicitors wrote to complain (among other things) about the absence of an anti-avoidance provision in the ATE policy provided, Astra submits that the correct position in the Old Claim was that the provision of a deed of indemnity was agreed (eventually) in principle, but there was no agreement as to the terms of that deed.

Another somewhat unusual feature of the application for security is that the sum of £180,000 is presently standing in Court to the credit of Musst. These funds represent part of the amount ordered to be paid by Astra to Musst in the Old Claim. Musst submits that, in the event that the Court determines that the Policy provides sufficient protection to Astra such that no order for security for costs is required, the sum of £180,000 may be paid out to Musst. The sum of £180,000 was intended to take Astra to up to hearing of the Costs and Case Management Conference (“the CCMC”), when costs would be budgeted and the matter could be reviewed. The CCMC has not yet taken place.

The courts have disapproved of applications for security being blown up “into a large interlocutory hearing involving great expenditure and time”: Porzelack KG v Porzelack (UK) Ltd [1987] 1 WLR 420, 423E, per Sir Nicholas Browne-Wilkinson V-C. Applications for security for costs ought to be decided on comparatively limited evidence.

One of the features of the litigation between the parties, in both the Old Claim and the New Claim, is an apparent willingness to leave no stone unturned. In a judgment on a specific disclosure application, Chief Master Marsh referred to the parties’ pursuing the proceedings with great vigour and at great cost: [2020] EWHC 1871 (Ch), at [9]. Reflecting this propensity on the part of both parties, a large volume of evidence was filed on the Application, as follows:

By order dated 15 April 2024, I permitted both parties to adduce expert evidence on BVI law.

Jurisdiction to order security is materially to be found in CPR Part 25, rr. 25.13(1) and (2)(c):

“(1)

The court may make an order for security of the costs if:

(a)

it is satisfied, having regard to all the circumstances of the case that it is just to make such an order; and

(b)

(i) one or more the conditions in paragraph (2) applies…

(2)

The conditions are-

(a)

the claimant is-

(i)

resident out of the jurisdiction; but

(ii)

not bound by the 2005 Hague Convention, as defined in section 1(3) of the Civil Jurisdiction and Judgments Act 1982;

…

(c)

the claimant is a company or other body (whether incorporated inside or outside Great Britain) and there is reason to believe that it will be unable to pay the defendant’s costs if ordered to do so …”

Relevant case law

Legal expenses insurance (usually an after the event , or “ATE”, policy) can, in principle, be taken into account on the question whether the court should make an order for security for costs: see Premier Motorauctions Ltd (in liquidation) v Pricewaterhouse Coopers llp [2017] EWCA Civ 1872, especially at [20] – [24]. The ATE policy may be accompanied by a deed of indemnity or an anti-avoidance endorsement. If the application for security is brought against a claimant company, relying on the insolvent or impecunious company gateway, then the existence of an ATE policy may be enough to persuade the court that there is no “reason to believe” that the claimant will be unable to pay the defendant’s costs if ordered to do so. In such a case, the application will be dismissed, unless the defendant can establish any of the other grounds for security listed in r. 25.13(2): see the notes to the White Book at paragraph 25.12.9.

Even at the jurisdictional stage of considering security for costs, defendants are entitled to some assurance that the insurance was not liable to be avoided for misrepresentation or non-disclosure: Premier Motorauctions, especially at [27] and [29]. In Michael Phillips Architects Limited v Riklin [2010] EWHC 834, Akenhead J stressed that the three cases cited to him were not absolutely determinative as to whether ATE insurance can provide adequate or effective security for the defending party’s costs because “… it will depend upon whether the insurance in question actually does provide some secure and effective means of protecting the defendant in circumstances where security for costs should be provided by the claimant”. He added that

“It is necessary where reliance is placed by a claimant on an ATE insurance policy to resist or limit a security for costs application for it to be demonstrated that it actually does provide some security. Put another way, there must not be terms pursuant to which or circumstances in which the insurers can readily but legitimately and contractually avoid liability to pay out for the defendant’s costs.” (at [18]).

Where a court is considering objections to an ATE policy relied on in the context of an application for security, a court should “… approach such objections with care: in order to amount to a valid objection that an ATE policy does not provide appropriate security, the defendant’s concern must be realistic, not theoretical or fanciful”: per Coulson J (as he then was) in Harlequin Property (SVG) Ltd v Wilkins Kennedy [2015] EWHC 1122 (TCC), at [21(d)].

The decision of Nicholas Vineall KC (sitting as a High Court Judge) in Recovery Partners GB Ltd v Rukadze [2018] 1 WLR 1640, was cited to me as authority for the proposition that on an application for security for costs the Court has wide discretion not only as to whether, and in what sum, such security is to be ordered but also as to the means by which it should be provided. If, on an application for security, two different forms of security would provide equal protection to the defendant, the Court should, all else being equal, order the form which is least onerous to the claimant: see at [17].

In a recent judgment, Saxon Woods Investments Limited v Costa [2023] EWHC 850 (Ch) (“the Saxon Woods case”), Deputy Insolvency and Companies Court Judge Agnello KC, after referring to the earlier authorities on ATE policies and their effect on applications for security, concluded that the court had to form a view on (1) the meaning of the policy, and (2) how readily it may be avoided legitimately and contractually, and (3) the likelihood of circumstances arising which will enable to the policy to be readily, legitimately and contractually avoided: see at [33]. On the particular facts of that case, and on the terms of the policy and the anti-avoidance endorsement at issue in that case, the Deputy Judge held that the policy could stand as full satisfaction for the requirement for security.

Astra accepts that an ATE policy with an anti-avoidance endorsement is, in principle, capable of amounting to sufficient security in some circumstances. However, Counsel for Astra pointed out that in the Saxon Woods case one of the factors which weighed with the Deputy Judge was that a refusal to accept the anti-avoidance endorsement would have led to prejudice to the petitioner and made it unable to pursue its claim as being unable to obtain the cash necessary to pay into Court. In those circumstances, Astra submitted that it was hardly surprising that the Deputy Judge reached the conclusion which she did, given that refusal to allow reliance on the ATE policy would in effect have led to the claim being stifled. I do not think that this is an entirely correct characterisation of the decision in the Saxon Woods case. The Deputy Judge carefully considered the terms of the policy and the endorsement before her, and concluded that the policy provided sufficient protection to the respondent: see at [71] – [72].

Musst in turn referred me to a judgment on funding by the Competition Appeal Tribunal, [2019] CAT 26 UK Trucks Claim Limited v Fiat Chrysler Automobiles N.V. and others (“the UK Trucks Claim case”), where the CAT accepted that an ATE policy with an anti-avoidance endorsement demonstrated that the claimants were in a position to pay the defendant’s costs. Helpfully, the CAT included the full terms of the anti-avoidance endorsement in an Appendix to their decision. However, Counsel for Astra submitted (which I accept) that the CAT cannot really be said to have given a general judicial approval to the form of endorsement found in the UK Trucks Claim case. The respondent parties in that case accepted that the wording of the anti-avoidance endorsement was satisfactory, the challenge being limited to whether the existence of the policy in that case provided sufficient protection from insolvency. In any event, there are certain differences between the form of anti-avoidance endorsement relied on in UK Trucks Claim case and the one before me.

Having set out what seems to me to be the key authorities, every case turns on a careful consideration of the actual terms of the policy relied on in each individual case. I did not understand Musst to argue for a different approach.

I turn to the terms of the Policy. The Policy has a Limit of Indemnity of £900,000 “in respect of Opponent’s Costs only From Policy Commencement Date up to the Conclusion of the Dispute”. The £900,000 figure is intended to cover the entire amount of security for costs requested by Astra, although costs budgeting has yet to occur.

The Policy is written by Litica. Mr Ruffle, who has provided a witness statement on behalf of Musst, is the co-owner and director of Litica. The subscribing insurer is International General Insurance Co (UK) Limited (“the Insurer”). The Insurer is an A rated, UK based insurer.

The Policy is governed by English law, and provides for the exclusive jurisdiction of the English courts, save in respect of any disputes arising out of requests for the Insurer’s Approval (a defined term).

Pursuant to clause 4.1, the Insurer may cancel the Policy upon 30 days written notice in certain circumstances. Clause 4.2 sets out what should happen in the event of termination of cover. If the Insurer cancels the Policy pursuant to clause 4.1, and save as set out further in clause 4.2, the Insurer will not pay the Opponent’s Costs incurred by the Opponent after the date of cancellation, and the Insured remains liable to pay the Premium. However

“[F]or the avoidance of doubt, subject to the other terms of this Policy, the Insurer shall indemnify the Insured for the Insured Liability for: (i) any Opponent’s Costs which were incurred by the Opponent on or before the date of cancellation of the Policy; and (ii) any Opponent’s Costs which were incurred by the Opponent in the assessment of the Opponent’s Costs which falls within (i), even though those costs are themselves incurred after the date of cancellation of the Policy. This applies irrespective of whether or not the order for Opponent’s Costs is made after the date of cancellation.”

Clause 5 deals with payment of the Premium by the Insured.

Clause 7.1 provides that claims made under the Policy should be made as soon as practicable and no later than 5 working days after the Insured becomes legally obligated to pay Opponent’s Costs.

Clause 10 sets out the duty of fair presentation. However, pursuant to clause 10.2, the Insurer waives its right to rescind, cancel or avoid the Policy for any reason other than the fraudulent or deliberate breach of the duty of fair presentation of the risk to the Insurer.

Clause 11 deals with fraudulent claims and provides as follows:

“If the Insured or anyone acting on its behalf makes a fraudulent claim under this Policy, the Insurer:

11.1.1

is not liable to pay the claim;

11.1.2

may recover any part of the claim already paid from the Insured; and

11.1.3

may by notice to the Insured treat this Policy as having been terminated with effect from the time of the first fraudulent act, in which case the Insurer is not liable to the Insured in respect of a relevant event occurring after that time and may retain any elements of the Premium which have already been paid.”

Appendix 2 to the Policy is the AAE. The wording in the AAE is similar to, but not identical with, that considered in the UK Trucks Claims case. I was told in submission that insurers have taken the wording found in the UK Trucks Claims case as an “industry standard”. The key provisions of the AAE are as follows:

The terms of the AAE are such that it essentially forms part of the Policy, and it needs to be considered, not in isolation, but as forming part of the Policy.

Astra has raised a host of reasons why it is said that the Policy (including the AAE) proffered by Musst does not offer sufficient security to Astra in the circumstances. Having considered the evidence and the parties’ submissions, I have come to the view that most of Astra’s objections are either unjustified, or have, following the issue of the Application, now been sufficiently addressed. The key points are as follows.

First, Astra submits that the Policy is a “highly technical document”, not written in plain English, which is difficult to navigate and construe. It is said that this complexity and deficiency render the Policy inappropriate as security generally. This criticism goes much too far. Whilst it is true that the Policy is a technical document, I do not accept that it is difficult to either navigate or construe. Almost any legal document, when subjected to close scrutiny, could be drafted better or more clearly. The overall effect is perfectly clear. By way of illustration of the criticisms, Astra submitted that it was unclear who “the Opponent” in the AAE was. I am surprised that Astra took that point. There is no difficulty with this. The Opponent is defined in the definitions section of the Policy, which refers to Schedule 2, which make it plain that the Opponent is Astra UK and Astra LLP.

Second, Astra submits that it is unclear what terms of the Policy are in place, or indeed whether the Policy is in effect at all. This is to ignore the evidence of Mr Ruffle, the co-owner and director of Litica since May 2019, who says in terms that the Policy is in place. I see no reason to reject this evidence. Moreover, there is a letter dated 24 May 2024 from IGI, the Insurer. This letter is headed Policy Number LIT/2023/041, which matches the policy number given in the Policy, and states that the Insured is “Musst Holdings Limited”. The letter confirms that the Insurer is “… the sole insurer in respect of the above captioned litigation insurance policy with policy number LIT/2023/041 (“the Policy”)”. The letter then expressly states that “… the Policy is active and that coverage is being is being provided as set out herein.” It is signed by Robert Horner, who gives his position as “Underwriter, Legal Expenses”. Astra’s expressed concern here about whether the Policy is in force seems to me entirely fanciful.

Third, Astra complains that, whilst it was originally envisaged that there would be three insurers, at some point this was changed to having just one (namely the Insurer, IGI). Mr Ruffle’s evidence on this point is that Litica have delegated authority to offer terms underwritten by a number of leading insurers without referral “and in this case, prior to the most recent terms offered it was decided that the most suitable option for the claimant …” was to have one insurer, IGI. Astra did not suggest that there was anything wrong with the Insurer per se. This is hardly surprising, given the Insurer’s status as an A rated UK insurer. Instead, Counsel for Astra suggested that the concern was that Litica might at some future stage “change to a less suitable insurer”. That concern appears to me fanciful. In effect, it is an insinuation that Litica would attempt somehow to undermine the existing security so as to risk the protection otherwise afforded to Astra. There is no evidence to justify that fear.

Astra has also raised a concern that the conditions precedent to the AAE have not come into force. Mr Ruffle has confirmed that they are in force. Astra says that this does not suffice, given that it is unclear what authority Mr Ruffle has to speak on behalf of the Insurer, and that in any event the recent letter from the Insurer does not mention the conditions precedent, merely stating that the Policy is active and coverage is being provided. There is an element of unreality to Astra’s submissions on this point. If the conditions precedent had not come into effect, then the Insurer would not be able to state that the Policy was active and coverage was being provided. This is a bad point.

The next expressed concern is perhaps a more substantive one. Astra submits that the existing terms of the AAE do not clearly provide a right to make any direct claim under the Policy. Paragraph 4 of the AAE provides as follows:

“The parties to this Policy agree that irrespective of any other provisions of the Policy the terms of this Endorsement are intended to benefit the Opponent and may be enforced by the Opponent directly pursuant to the provisions of the Contracts (Rights of Third Parties) Act 1999. No other third party is entitled to the benefit of or to enforce any term of the Policy (including this Endorsement) under any provision of the Contracts (Rights of Third Parties) Act 1999 or otherwise.”

Astra’s expressed concern is that the AAE itself does not give a right to make or control any claim under the Policy, and the Policy itself excludes third party rights at clause 7. I have to say that Astra’s concern seems to me to be based on a strained construction of paragraph 4, because paragraph 4 states plainly “[T]he parties to this Policy agree that irrespective of any other provisions of the Policy …” (emphasis added). However, it is perhaps arguable that paragraph 4 could be made clearer to address Astra’s concern. Before the hearing, Musst indicated in correspondence that whilst it considered that a direct right to claim under the AAE is already present, it was prepared to offer further wording to clarify the position. In a letter dated 17 January 2024, Musst’s solicitors suggested that additional wording could be added to the AAE. The further wording proffered was “… andthe Insurer irrevocably agrees to pay to the Opponent any claims made by the Opponent within 21 days of the Opponent writing to Litica to make such request”. This offer of additional wording was again confirmed by Counsel for Musst in the course of the hearing before me. This, coupled with the wording already found at paragraph 1, in particular the reference to “this Endorsement …. may be enforced by the Opponent directly”, appears to me to put the matter beyond any reasonable doubt. Astra does not require Musst’s cooperation to bring a claim. Astra has a right to make a direct claim under the AAE against the Insurer.

Astra has indicated that any lack of a direct claim against the Insurer “might not have mattered, or not as much, had Musst been an English company”, because in that case Astra would be able to rely on the Third Parties (Rights against Insurers) Act 2010 (“the 2010 Act”). However, Musst is a BVI company. Astra contends that (1) in the absence of any information at all as to Musst’s financial position, the Court should proceed on the basis that there is in effect a “presumption of insolvency” and (2) based on the evidence of its expert on BVI law, Mr Hall Taylor KC, there is no legislation in place in BVI that provides direct rights to protect the holders of foreign insurance policies. As to those two contentions:

It is not possible to decide between the two experts as to whether the 1949 BVI Act does, or does not, apply to insurance contracts governed by UK law. Mr Hall Taylor KC provides no reasoning to support his view that “the 1949 Act only extends to insurance contracts made under BVI law.” There is nothing in the wording of the statute itself which states that. Ultimately, however, it seems to me that it does not matter, because (as I have explained above) the AAE creates a direct cause of action between Astra and the Insurer. Accordingly, in the event of the Musst’s hypothetical insolvency, Astra would not need to rely on the 1949 BVI Act to enforce its right in the BVI, because Astra would have a direct right of enforcement against the Insurer. This could and no doubt would be litigated in the High Court of England and Wales.

However, this conclusion, that the terms of the AAE and the Policy would give rise to a direct claim by Astra against the Insurer as a matter of English law, raises a further question which Mr Hall Taylor KC also addresses. He does not express a view as to the construction of the proposed endorsement. He indicates that if it is correct that the endorsement gives rise to a direct claim by Astra against the Insurer, then “[I]t is possible that the proposed amendment to the endorsement which provides for claims to be paid directly to [Astra’s] law firm could be challenged on behalf of other creditors on the basis that it constitutes a reviewable transaction under BVI law” (Report, paragraph 31). Mr Hall Taylor KC refers to two English law principles which he says are part of BVI law and which could “potentially apply”:

These two principles have been described as sub-rules of the general principle that parties cannot contract out of the insolvency legislation: see per Lord Collins in Perpetual Trustee Co Ltd v BNY Ltd, at [1]. In effect, they are rules of public policy. Lord Collins went on to explain that although there is some overlap, they are aimed at different mischiefs, citing Goode, “Perpetual Trustee and Flip clauses in Swap Transactions” (2011) 127 LQR 1, 3 – 4. Lord Collins stated that

“The anti-deprivation rule is aimed at attempts to withdraw assets on bankruptcy or liquidation or administration, thereby reducing the value of the insolvent estate to the detriment of creditors. The pari passu rule reflects the principle that statutory provisions for pro rata distribution may not be excluded by a contract which gives one creditor more than its proper share.” (at [1])

In Perpetual Trustee Co Ltd v BNY Ltd, the attempt to invoke the anti-deprivation principle failed in the context of a complex commercial swap counterparty priority agreement. The argument failed at first instance, in the Court of Appeal, and in the Supreme Court. The Supreme Court held that the court was to look at the substance of the agreement, rather than its form, and ask whether the purpose and effect of the relevant provision amounted to an illegitimate intent to evade the bankruptcy law or had a legitimate commercial basis: see, for example, at [78] – [79], [102] – [103], [108] – [109] (per Lord Collins), [122] and [132] (per Lord Walker), [148] – [152] (per Lord Mance).

Similarly, the submission that the pari passu principle served to invalidate an agreement entered into by the parties in good faith for genuine commercial reasons whereby monies standing in a “special account” were held on trust by the defendant failed in Carreras v Freeman Matthews.

In a further authority cited to me, Revenue and Customs Commissioners v Football League Ltd [2021] EWHC 1372 (Ch), attempted reliance on both the anti-deprivation and the pari passu rules to challenge corporate articles of association again failed. The court there made the point that those principles did not constitute broader anti-avoidance rules permitting the court to rewrite contract and property rights so as to create in the company’s favour an asset which it did not have.

In his report Mr Hall Taylor KC acknowledges that he is not aware of any authority under BVI law which deals with the application of these principles in the context of an ATE policy. Mr Thorp does not address the issue, because he is of the view that the 1949 BVI Act has the effect that, should an adverse costs award be made in favour of Astra, the Insurer is liable to pay out the insured sum to Astra even if Musst has become insolvent. Nevertheless, Astra submitted that it would be exposed to a risk that a liquidator appointed over Musst might seek to challenge the validity of the Policy and accompanying endorsement, litigation with a liquidator to establish the position would take further time and incur further expense, and it would not be fair for Astra to run that risk.

Astra also drew my attention to Giaquinto v ITI Capital Ltd [2023] EWHC 2467 (KB), a decision of Master Stevens, where Astra said that “a comparable point arose”. I do not agree that Giaquinto v ITI has any real relevance to the issue I am considering. In that case, the claimant sought to vary an order that it provide security in a particular way for the defendant’s costs. The application failed, principally because the court held that there was no material change of circumstances, there was delay on the part of the claimant in making the application, the claimant was in essence trying to reargue the earlier hearing on the basis of authorities which it could and ought to have relied on earlier, and the interests of finality firmly pointed towards refusing the relief sought. It is true that the second claimant in that case was incorporated in the Isle of Man, and the defendant (among other things) was concerned that the 2010 Act would not protect their position: see at [87]. But there was little real analysis by the Court of the point: see at [88] – [89], at [93] – [94], and [99]. In any event, the insurer in that case was also based abroad, which added another layer of complexity to the arguments.

Standing back, it is important for the court to assess the extent of the risk that Astra is facing. The important point is that, were Musst to enter an insolvency procedure in the BVI, Astra would be able to enforce their rights under the Policy and the AAE against the Insurer in the courts of England and Wales. Were the arrangement to be challenged by an insolvency practitioner in the BVI (which is the possible scenario outlined by Mr Hall Taylor KC) and if the Insurer were subsequently held to have deprived the insolvent estate of the fruits of the Policy, I accept the submission of Counsel for Musst that this is a risk that would be borne by the Insurer.

In any event, and even if I were wrong on that, in my view the risk that the Policy and the endorsement could be successfully challenged by any liquidator of Musst on the basis of either the anti-deprivation or pari passu principles appears completely remote, having regard to the authorities cited to me. It is most unlikely that any court, looking at the purpose and effect of the Policy and the AAE, would conclude that this was an illegitimate intent to evade the bankruptcy law. It does not create an asset which belongs to Musst which would be removed on an insolvency. That it not what this insurance policy is intended to do. It is fair to characterize Astra’s concern on this point as theoretical or fanciful.

Astra expressed a further concern related to the termination provisions in the Policy (clause 4) and the AAE, paragraph 5. Astra points out, correctly, that the Policy may be cancelled by the Insurer under clause 4 in a number of “plausible situations that are out of Astra’s control”. It was said that there was no protection for Astra at all “including retrospectively” if there was termination by consent, or under clause 4.3 of the Policy. Astra then accepts that “these cancellation rights might not matter as much, if the AAE contradicted their effect, so as to ensure the continuation of cover”. But Astra submits that even the AAE does not provide sufficient protection, because (a) paragraph 1 of the AAE only says that the Policy is “non-voidable”, but not “non-cancellable” and (b) the rights of termination are kept alive expressly at AAE, paragraph 5, the drafting of which contains “inadequate and confusing notice requirements, that are potentially very short (if not after the event)”.

The first objection has been addressed. Musst indicated that it was willing to add the words “non-cancellable” in paragraph 1 of the AAE. As to Astra’s other expressed concerns regarding termination:

Overall, I accept Musst’s submissions that despite the right granted to the Insurer, in certain circumstances which are outside Astra’s control, to terminate or cancel the Policy, Astra is suitably protected, subject to the one point I have set out above at paragraph 55(4). However, the right way to deal with that one point, rather than refusing to allow the Policy to stand as security, is to order that some or all of the £180,000 standing in Court should remain there to cover the (possibly fairly unlikely) situation where Astra’s costs of applying back to the Court following notice of termination having been given are not covered. The parties may address me further on this point at the hearing to deal with consequential matters arising from this judgment.

Musst submitted that the real thrust of Astra’s case on this point would be that an ATE policy can only provide sufficient protection if it provides that the insurer will directly indemnify a defendant for adverse costs for the entirety of the costs of the proceedings, even if the insurer legitimately terminates the policy at some earlier stage. I accept that the case law does not go as far as that. To accept Astra’s submission on this point would be, in effect, significantly to reduce the situations in which an ATE policy could ever be accepted as an alternative form of security.

Astra also submitted that “there is obviously a real risk that an insurer will look to challenge para 1 of the AAE if it is found that the policy was procured by some form of fraud.” Paragraph 1 of the AAE provides that the Policy is non-voidable, and Musst has confirmed that it is prepared to add the words “and non-cancellable”. Musst submits that paragraph 1 expressly confirms that the Policy is “non-voidable” and “any claim will be honoured in full [subject to the limit of the indemnity in the Policy] irrespective of any exclusions or any provisions of the Policy or of the general law …” (emphases added). Musst submits that the AAE is “comprehensive, covering any and all exclusions, as it says in terms”.

There are two answers to Astra’s concerns about the lack of express mention of “fraud” in paragraph 1 of the AAE. The first is that the Court should consider the risk of the Insurer being able to terminate the Policy for fraud. It is true that the Policy itself contains a right to terminate for fraud, in clause 10 (duty of fair presentation) and clause 11 (fraudulent claims). It is also true that Musst has not disclosed the placing information provided to the Insurer. However, Musst’s claim against Astra is not one where any allegations of dishonesty have been raised against Musst. The proceedings do not involve questions of the honesty of Musst or those natural persons acting for it, as is sometimes the case. For example, see Premier Motorauctions v Pricewaterhouse Coopers, where although the claim was being brought by the liquidators of the claimant, and there was no reason to doubt their honesty or professionalism, the claimant’s case relied on the evidence of a Mr Elliott; if his evidence were rejected, the claimant would lose: see at [26 – [27]. In this case, Astra has not put forward any reasons for thinking that there is any particular reason to believe that the Insurer would be able to rescind the Policy for fraud.

The second answer involves consideration of the authorities. Astra, citing HIH Casualty and General Insurance Ltd v Chase Manhattan Bank [2003] 1 All ER (Comm) 349, submits that for a provision to be effective in excluding the ordinary consequences of fraud by another party to the agreement “such intention must be expressed in clear and unmistakable terms on the face of the contract … General words, however comprehensive the legal analyst might find them to be, will not serve: the language used must be such as will alert a commercial party to the extraordinary bargain he is invited to make” (per Lord Bingham, at [16], [68]) and “… in the absence of words which expressly refer to dishonesty, it goes without saying that underlying the contractual arrangements of the parties there will be a common assumption that the persons involved will behave honestly” (per Lord Hoffman, at [95]). Gloster LJ, in Holyoake v Candey [2017] EWCA Civ 92; [2018] Ch 297, at [95], made a similar point: “Thus clear and specific wording is required to exclude remedies arising from dishonesty or fraud, on the assumption that it is, in principle, possible to do so.”

In response, Musst relies on the Saxon Woods case. The Deputy Judge there concluded that, in her view, the relevant clause in the endorsement did extend to “reckless and fraudulent non-disclosure entitling the insurer to avoid the policy”: at [64] - [65]. The relevant clause was in substance virtually identical to that found in the AAE, paragraph 1: see the quote in the Saxon Woods case, at [56]. The Deputy Judge considered both HIH Casualty v Chase Manhattan and Holyoake v Candey. The Deputy Judge relied on a line of authorities relating to settlement agreements in concluding that, even though the AAE in the case before her did not use the word “fraud”, nevertheless it was obviously intended to cover fraud. The Deputy Judge referred to another part of Gloster LJ’s judgment in Holyoake v Candey to reject the notion that as a matter of public policy such clauses cannot exclude fraud: see at [64] – [73].

Without in terms stating that the Deputy Judge’s conclusion was wrong, Astra submitted to me that there remained considerable uncertainty about the correctness of the Deputy Judge’s reasoning and conclusion, in light of what was said in HIH Casualty v Chase Manhattan and Holyoake v Candey. Accordingly, Astra submitted there was “clearly a material chance” that the AAE would be found not to exclude fraud, and that there continued to be a material risk of setting aside the AAE for reckless or fraudulent non-disclosure, and that Astra should not have to bear that risk.

It seems to me that I ought to follow the decision of the Deputy Judge, unless I am of the view that it is wrong. Respectfully, I consider that her reasoning was correct. I do not intend to summarise all the steps in the Deputy Judge’s reasoning, but the following appear to me to be the key points, which apply just as much in the case before me:

Finally, Astra suggested that because paragraph 1 of the AAE refers to the right to recover payments not only from the “Insured”, but also from “the Claimant”, the reference to “the Claimant” could also be construed to mean Astra. That seems to me plainly wrong, in circumstances where Astra is defined as the Opponent, and the Policy and the AAE consistently refers to the Opponent. Reading the Policy and the AAE as a whole, the reference to “the Claimant” must be to Musst, despite Musst also being at times referred to as “the Insured”. I do not accept that the word “Claimant” can realistically be construed as referring to Astra.

Astra had a further overarching point, which I should consider. I have already mentioned that in the Old Claim, Astra objected (at least initially) to the proposed ATE policy as security on the ground that it did not contain an anti-avoidance provision. On this application, Astra asks why a deed of indemnity could not be provided and the first head of relief sought in the Application is that Musst should provide security “in a form substantially the same as that proffered” in the Old Claim. Mr Ruffle explains that at the time, that is, in 2018 – 2019, deeds of indemnity were typical instruments in the context of security for costs applications, “but they have not been used regularly by the market for approximately 3 / 4 years”. He goes on to say that deeds of indemnity have not been used for a variety of reasons “including the fact that deeds of indemnity are considerably more difficult to execute in comparison with AAEs (i.e. they typically require a greater number of signatures from often board level directors or large corporate entities, which can make signing protracted and inefficient).” Astra attacked this explanation as making little sense, and submitted that no proper reason had been put forward as to why a deed of indemnity could not have been proffered, instead of the AAE.

I do not think this matters. The question for the Court is not whether a deed of indemnity is easier to put in place than an AAE, or the other way round. This is not something than the Court can decide, absent perhaps expert evidence as to market practice, which would be wholly disproportionate on an application for security for costs. Musst has obtained a letter, dated 5 June 2024 from “the Judge”, a well-known litigation insurance broker, which states that “[I]n practice, AAEs are far simpler to negotiate, arrange and execute compared to Deeds (which will often require a Director of the insuring entity to sign which presents some practical issues given the scope of the potential insuring entities) …”. I place little weight on this letter, which was obtained after the close of the exchange of evidence. Instead, I accept the simpler point made by Musst, which is that it is not a question of what the defendant might prefer; rather, the real question for the Court is whether the Policy and AAE provide adequate security in respect of Musst’s potential costs liability. I am satisfied that it does.

I have sought in this judgment to deal with each of the objections raised by Astra. It is also necessary to take a step back and look at position in the round. As Coulson J made plain, the Court should evaluate whether there is a real, as opposed to theoretical or fanciful, risk, that the Policy may be set aside and Astra left unprotected. Ultimately, given the presence of the AAE, including the additional amendments which Musst and the Insurer were prepared to make, I am satisfied that the Policy and the AAE together provide adequate security for Astra’s costs.

It seems to me that, for the reasons given in this judgment, I should dismiss Astra’s application. However, I will hear from the parties at the hearing to deal with consequential matters as to the precise form of order that should be made.

I would add this. Part of Musst’s submission was that the Court should on Astra’s application make an order permitting payment out to Musst of the £180,000 standing in Court to its credit. I do not consider that I should make such an order, at any rate before dealing with costs budgeting. There is an upcoming CCMC listed in these proceedings. That seems to me the right time to consider whether I should allow payment out of some or all of the amount presently standing to Musst’s credit, including possibly retaining some or all of the £180,000 in order to deal with any costs that might be incurred by Astra in applying back to Court for further security in the event that the Policy was terminated: see the discussion above at paragraph [55(4)] of this Judgment. Subject to hearing from the parties, the upcoming CCMC would also be a suitable time to deal with any other consequential matters arising from this judgment.