Flanagan v Liontrust Investment Partners LLP & Ors

On 4 December 2015 I heard argument on consequential matters arising from the judgment in this case which I handed down on 24 July 2015, following the trial on liability issues which had taken place between 20 January and 6 February 2015: see Flanagan v Liontrust Investment Partners LLP and Others [2015] EWHC 2171 (Ch), now reported at [2015] Bus LR 1172. The present judgment assumes familiarity with, and should be read as a sequel to, the earlier judgment, to which I will refer as “the main judgment”.

There were four matters on the agenda for the hearing in December 2015:

It is common ground that the validity of the third termination letter depends on whether Mr Flanagan was still a member of the Management Committee in December 2014 when the meeting of the Committee at which he was required to retire as a member of the LLP was convened (on 10 December) and held (on 18 December). If Mr Flanagan was no longer a member of the Management Committee, there is no dispute that the notice given to him in the letter was valid, and his membership of the LLP therefore terminated on 22 June 2015, six months after the letter was delivered to him on 22 December 2014. If, on the other hand, Mr Flanagan was still a member of the Committee, the notice given to him must have been invalid because he was never given notice of the meeting on 18 December at which the decision to remove him was taken and the terms of the letter were approved: see clause 12.6 of the LLP Agreement, which expressly renders void any decision of the Committee taken at a meeting of which proper notice had not been given.

At the trial, Liontrust argued that Mr Flanagan’s membership of the Committee had indeed come to an end before December 2014, for either or both of the following reasons. First, Mr Flanagan’s membership of the Management Committee depended upon his status as an “approved person for the purposes of the FSA Rules” under clause 12.2 of the LLP Agreement, and Mr Flanagan had ceased to be an approved person when Liontrust de-registered him with the FSA on 4 October 2013. Secondly, by a document dated 1 August 2013 Mr Abrol, on behalf of LIS, informed the LLP that LIS consented “to all Individual Members being appointed as Committee Members”, but then went on to list ten members by name (in some cases with another member shown as an alternate), who did not include Mr Flanagan. The argument was that, by means of this document, LIS had effectively removed Mr Flanagan from his previous membership of the Committee.

My provisional view was that each of these points should be determined in Mr Flanagan’s favour, with the result that the third termination letter (like its predecessors) was invalid. I did not, however, express a concluded view, because the third notice had not been dealt with in the pleadings or witness statements, the minutes of the Management Committee meeting on 18 December 2014 were not in the bundle, and I heard only brief argument on the two issues which I have identified. Mr Thompson QC nevertheless submitted, on behalf of Mr Flanagan, that it would be inappropriate for me to revisit my provisional conclusion, because the parties had been content for me to proceed on the basis of the limited material before me, and both sides had addressed the question in their written closing submissions. Mr Thompson said that the court could therefore properly have expressed a final conclusion. That may well be right, but the fact is that I chose to deal with the matter on a provisional basis, and, having done so, I think it would be wrong to preclude Liontrust from trying to persuade me that my initial view was wrong. In the event, Mr Thompson did not press the objection in his oral submissions, nor did he submit that the question was res judicata. I am satisfied, therefore, that it is open to me to revisit my preliminary conclusion.

I will begin with the document of 1 August 2013. It was typed on the headed notepaper of LIS, and addressed to the LLP at 2 Savoy Court, London, WC2. It was signed by Mr Abrol as director, for and on behalf of LIS. The body of the document read as follows:

The document needs to be read in the light of Mr Abrol’s unchallenged evidence in paragraph 14 of his first witness statement, which reads as follows:

Without objection from Mr Flanagan, Liontrust now also relies, by way of corroboration of Mr Abrol’s evidence, on an email sent by Mr Catton on 1 August 2013 to the others members of the LLP, excluding Mr Flanagan, even though it was not included in the trial bundles. The subject of the email was “Corporate Governance Changes”, and so far as material it read as follows:

Finally, it is relevant to have in mind the terms of the initial appointment of all individual members of the LLP as Management Committee members on 8 July 2010: see paragraph 12 of the main judgment. Like the subsequent document of 1 August 2013, this document was typed on the headed notepaper of LIS, addressed to the LLP at 2 Savoy Court, and signed by Mr Abrol as director for and on behalf of LIS. Indeed, one can infer from the common layout and appearance of the two documents, and a common typographical error in the description of the capacity in which Mr Abrol was signing (“For an on behalf of”), that the earlier document of 8 July 2010 was cut and pasted, or otherwise used as the basis for, the later document. The body of the earlier document read as follows:

There can be no doubt, I think, that the document of 8 July 2010 constituted a valid exercise of the power of appointment of Management Committee members contained in clause 12.3.2 of the LLP Agreement, even though the language of “consent” was somewhat inept. The document was a written notice given by LIS to the LLP, it referred to the enabling clause in the LLP Agreement, and it was headed “Appointment of Committee Members”. Furthermore, the “consent” was to all of the individual members being appointed as Committee members. As a matter of construction, it is obvious what the document was intended to achieve, and it complied with the formal requirements of clause 12.3.2.

Building on this point, Mr Machell QC for Liontrust argued that, although more ineptly worded than its predecessor, the purpose and effect of the document of 1 August 2013 is in fact clear once it is placed in its proper context. Its purpose must have been to appoint as members of the Management Committee those persons listed in Mr Catton’s email of the same date, and to terminate the existing appointments of all other individual members of the LLP, Mr Flanagan included. The obvious difficulty with the wording of the document as it stands is the apparent conflict between the consent to “all Individual Members being appointed as Committee Members” on the one hand, and the limited list of names which follows, on the other hand. It is clear that something must have gone wrong with the drafting, and the only way to restore sense to the document, and to fulfil the purpose described in Mr Abrol’s unchallenged evidence and Mr Catton’s email, is to reject the words “all Individual Members” as a mistaken, or unthinking, transposition from the earlier document of 8 July 2010, and to treat the list of names as being the list of those whom LIS wished to appoint. If necessary, harmony between the two parts of the document could be restored by replacing the word “all” with “the following”, or by inserting the words “the following” after “all”. This is the kind of mistake in a written document, submits Mr Machell, which can be corrected as a matter of construction, because it is obvious what was intended, and the substance (if not the precise wording) of the required correction is also clear. And if the argument is sound thus far, he submits, the clear intention to appoint the listed members must by necessary implication carry with it the intention to terminate the Committee membership of all the other individual members.

In the main judgment, I was unpersuaded by the forerunner of this argument. I thought that the reference to “all” individual members had to be read literally, and that the purpose of the list of names was obscure, not least because the LLP Agreement makes no provision for alternates. I said that “on no view can it reasonably be read as removing Mr Flanagan from membership of the Committee, on the assumption that he was then an existing member of it”: see the main judgment at [167]. With the benefit of fuller argument, however, and the contextual material set out above, including in particular Mr Catton’s email of 1 August 2013, I have changed my mind. I accept Mr Machell’s submission that, read in context, the intended purpose and effect of the document are clear, and the reference to “all” individual members being appointed must be rejected as a careless slip which had its origin in the wording of the earlier document. Only in this way can sense be restored, and the document saved from futility.

Nor am I deterred from reaching this conclusion by the fact that clause 12 of the LLP Agreement does not make express provision for the appointment of alternates for members of the Committee. On reflection, I see no good reason why the power to appoint Committee members in clause 12.3.2 should not include power to appoint alternates for the members who are primarily appointed; and, even if that is wrong, the effect would only be to invalidate the appointment of the alternates, not the appointment of the primary members. A related point is that Mr Ions, who heads the list, could not technically have been appointed pursuant to clause 12.3.2, because he was never a member of the LLP. He could, however, be appointed as a representative of LIS pursuant to clause 12.3.1, and the words “LIS Member” in brackets after his name are in my view a sufficient indication that this is what was intended.

I am similarly unpersuaded by Mr Thompson’s argument that the form of words employed in the document does not purport to remove from the Committee anyone who was already appointed. The admissible surrounding circumstances show that the object of the exercise was to rearrange the governance structure of the LLP and to reduce the size of the Management Committee. The power of appointment in clause 12.3.2 includes power to remove members once appointed. In order to give it business efficacy, the document must therefore be construed as including an exercise of the power of removal in relation to those Committee Members not included in the new list. In my judgment the case falls comfortably within the principles relating to implied exercise of a power (here, the power to remove Committee Members) discussed by Scott J (as he then was) in Davis v Richards & Wallington Ltd [1990] 1 WLR 1511 at 1530A to 1531F.

Finally, Mr Thompson argued that it was impermissible to have regard to Mr Catton’s email as part of the surrounding circumstances, because it was never sent to, or seen by, Mr Flanagan. I am satisfied that there is nothing in this point, because the only relevant intention relating to an exercise of the power in clause 12.3.2 is that of the person upon whom the power is conferred, namely LIS. The proposed transaction was unilateral in nature, and although it affected Mr Flanagan, he was not a necessary party to it.

For all these reasons, I consider that my preliminary view on this question was wrong, and the third termination letter achieved its intended effect because in December 2014 Mr Flanagan was no longer a member of the Management Committee, having been removed by the written notice from LIS to the LLP dated 1 August 2013. Mr Flanagan therefore ceased to be a member of the LLP on 22 June 2015.

This conclusion makes it unnecessary for me to consider Liontrust’s alternative argument based on Mr Flanagan’s deregistration with the FSA.

I must now deal with the costs of the action. Given the number and complexity of the issues, it is unsurprising that the costs incurred on each side have been substantial. Ignoring VAT, Mr Flanagan has incurred costs down to 28 September 2015 of £709,294.14, while the respondents (including the individual members of the LLP as well as Liontrust) have incurred costs to 6 February 2015 of £1,599,277.71. As I explained in the main judgment at [33], the individual members of the LLP who were joined as the fourth to twenty-third respondents have throughout been represented by the same solicitors and counsel as Liontrust, so no separate costs schedule has been served in respect of their costs. Accordingly, in this section of my judgment the expression “Liontrust” should where the context admits be read as including all the respondents.

According to Liontrust, there are three main reasons why its costs are more than twice as high as Mr Flanagan’s. First, the substantial burden of disclosure fell on Liontrust. Secondly, Mr Flanagan had no witnesses apart from himself, whereas Liontrust produced witness statements from 23 witnesses. Thirdly, Mr Flanagan’s solicitors (Bolt Burdon) charged him hourly rates significantly below the rates normally recoverable for a case of this scale and complexity. This last point is illustrated by the fact that the hourly rate claimed for the Grade A solicitor acting for Mr Flanagan is £300 to 30 June 2014, and £360 thereafter, whereas the hourly rates claimed for the three most senior solicitors acting for Liontrust range between £500 and £600. These figures, and the schedules as a whole, will of course be subject to detailed assessment in due course, to the extent that I make orders for costs in favour of either side. For present purposes, the important point is that I need to have regard to the great disparity between the costs on each side when considering what orders to make, and whether any netting off of costs might be appropriate.

There is no disagreement between the parties about the basic principles which fall to be applied. Despite their familiarity, I will set out the key provisions of CPR 44.2:

A common starting point in commercial disputes such as the present case is to be found in the principles which Jackson J (as he then was) distilled from the earlier case law in Multiplex Constructions (UK) Ltd v Cleveland Bridge UK Ltd [2008] EWHC 2280 (TCC), [2009] 1 Costs LR 55, at [72], as follows:

A helpful review and commentary on these principles may be found in the White Book, 2015 edition, volume 1, para 44.2.7 at pp 1429-1430.

Before going any further, it is convenient to refer to the Part 36 offers, and other “without prejudice save as to costs” correspondence, in the present case. I begin with the period before proceedings were issued by Mr Flanagan on 5 September 2013:

Following the issue of proceedings, the following developments may be noted:

On the information before me, this was the highest level at which Mr Flanagan pitched his claim. By the time of the pre-trial review in December 2014, the estimated value of his claim which his counsel were prepared to give in open court had reduced to about £8 million: see the main judgment at [30]. It can also be seen that the Part 36 offers on each side could hardly have been further apart. Liontrust never increased its initial offer to settle the case for £50,000 plus a departure package which reflected Mr Flanagan’s minimum entitlement under the LLP Agreement, while the lowest sum that Mr Flanagan was ever prepared to accept was £750,000, soon increased to £800,000, and then withdrawn on 24 July 2014.

Against this background, the first question I have to consider is who is the successful party. Mr Flanagan’s claim to be entitled to a share in the LLP under the default rules has failed in its entirety, but the claim failed only at the last hurdle, namely the question whether the common law doctrine of repudiatory breach is implicitly excluded by the legislative scheme of LLPA 2000, at least in relation to multi-party section 5 agreements: see paragraphs [218] to [243] of the main judgment. Mr Flanagan succeeded on all of the prior steps in the argument, each of which was hotly contested by Liontrust, that is to say:

In these circumstances, the argument for treating Mr Flanagan as the successful party is straightforward. He has recovered approximately £214,000, which is more than four times as much as Liontrust offered in its original Part 36 offer, which was never updated. Mr Flanagan had to go to court to recover this sum, which is a substantial one even in the context of commercial litigation. Furthermore, the authorities show that a claimant who had to go to trial in order to recover a significant sum should still be regarded as the successful party even if he also made a much larger claim which has failed. Counsel for Mr Flanagan refer in this connection to Fox v Foundation Piling Ltd [2011] EWCA Civ 790, [2011] C.P.Rep. 41, (CA) where Jackson LJ (with whom Moore-Bick and Ward LJJ agreed) said at [46] to [47]:

It is also important to note what Jackson LJ went on to say at [62]:

Mr Flanagan also relies on the well-established principle that, in complex litigation, it is almost inevitable that the ultimate winner will not succeed on all the issues in the case. See, for example, H L B Kidsons v Lloyds Underwriters [2007] EWHC 2699 (Comm) at [10] to [11] per Gloster J; Budgen v Andrew Gardner Partnership [2002] EWCA Civ 1125 at [35], per Simon Brown LJ; and Travellers’ Casualty v Sun Life [2006] EWHC 2885, where Clarke J said at [12]:

Liontrust submits, however, that on any realistic appraisal it should be treated as the successful party. It is obvious that the alternative money claim from 5 October 2013 was not the objective which Mr Flanagan pursued when he began the proceedings a month earlier, and it was first introduced by amendment to the petition on 22 September 2014. As such, the claim was a purely contractual one which turned on the terms of the LLP Agreement and the validity of the termination notices. If it had been Mr Flanagan’s only claim, it could have been pursued in relatively simple and self-contained proceedings under CPR Part 7, and would probably have been suitable for determination in the county court. Instead, the real substance of Mr Flanagan’s claim lay in the petition and his ill-fated attempt to obtain what on any reasonable view would have been an unmerited windfall by bringing himself within the default rules. It was this primary case which accounted for the bulk of the expenditure in the litigation, as well as the need to join the individual members; and in view of the extravagant relief which Mr Flanagan sought, Liontrust cannot fairly be criticised for resisting his claim at every step.

Liontrust further submits that the question of who is the successful party should be determined by the application of common sense rather than any mechanical test. It does not follow that a claimant who succeeds in recovering something is necessarily to be treated as the winner. In order to decide the question, it is necessary to investigate the issues and their resolution.

In support of this submission, Liontrust relied on the decision of a two-judge Court of Appeal (Auld and Mummery LJJ) in Islam v Ali [2003] EWCA Civ 612. The appellant, Mrs Ali, was the defendant to a claim by Mr Islam in the Central London County Court for remuneration for his services as a chartered accountant in running the accountancy business of her late husband. The judge found for Mr Islam on his claim, and gave him judgment for approximately £12,750, including interest. He also ordered Mrs Ali to pay Mr Islam’s costs of the action. She then appealed to the Court of Appeal against the costs order alone, on the basis that the judgment sum did not truly represent a win for Mr Islam, “but merely a relatively small balance in his favour between much larger sums in issue in the case”. The appeal succeeded, and the court decided that the proper order to make was no order as to the costs of the action.

In the course of his judgment, Auld LJ said this:

Mummery LJ agreed, saying at [28]:

Viewed in money terms, submits Liontrust, Mr Flanagan has recovered only a very small percentage of what he was seeking (£214,383 is about 2.7% of £8 million, the approximate value of the claim given by Mr Flanagan at the pre-trial review in December 2014). But this is not simply a case where a claimant has pursued a substantial money claim and succeeded in recovering only a small part. Rather, Mr Flanagan has pursued heavy, complex and expensive litigation in an attempt to secure a very substantial economic benefit, but has failed on his primary case, and has secured only a relatively small monetary benefit on an alternative basis added by way of amendment a year after the proceedings had commenced.

In his submissions in reply, Mr Thompson QC submitted that Mr Flanagan’s claim was always ultimately a money claim, since there was no realistic prospect of his remaining an active member of the LLP and the appropriate remedy for his wrongful exclusion was a buy-out order. Moreover, the only reason why his claim for unpaid drawings was not included in the petition as originally presented was that the claim did not arise until the expiry of Mr Flanagan’s initial two year term on 4 October 2013. The amount now recovered is a substantial proportion (about 27%) of the £800,000 for which Mr Flanagan would have been willing to settle the case in accordance with his second Part 36 offer, between 22 November 2013 and its withdrawal on 24 July 2014. By contrast, Liontrust had never increased its original Part 36 offer of £50,000.

Despite the considerable force of Liontrust’s submissions, I have concluded that Mr Flanagan is properly to be regarded as the successful party. I accept his submission that this case has always ultimately been about money, whether through the medium of a buy-out order or (after 4 October 2013) a contractual claim for Mr Flanagan’s fixed share of profits. The amount which Mr Flanagan has now recovered is not insignificant, either in itself (even, I would venture, for a hedge fund manager) or when compared with his two Part 36 offers which remained on the table from 14 August 2013 until 24 July 2014, that is to say from three weeks before the presentation of the petition until well after disclosure had taken place and the parties were in the last stages of preparation for the mediation in early August 2014.

It is true that the contractual claim for unpaid profits had not been pleaded before the Part 36 offer was withdrawn, but in my judgment it must always have been obvious to Liontrust and its advisers that Mr Flanagan would be entitled to these payments if the first and second retirement notices were invalid. Yet Liontrust took no steps to protect its position by increasing its own Part 36 offer. Nor is it a relevant objection, in my view, that Mr Flanagan could have pursued his contractual claim in separate Part 7 proceedings. Since the petition was already on foot, it was obviously sensible to pursue the contractual claim within the petition proceedings, and no objection was taken by Liontrust to the amendments on this score. It follows, however, that the unfair prejudice claim and the contractual claim were thenceforward pursued together, and they cannot logically be disaggregated for the purpose of ascertaining who the successful party is.

Even though Mr Flanagan is to be treated as the successful party, however, there are in my judgment powerful reasons why the court should depart from the general rule in CPR 44.2(2)(a) and not order Liontrust to pay the whole of his costs (subject to detailed assessment).

In the first place, Mr Flanagan’s primary claim for a buy-out order based on the default rules has failed in its entirety at the crucial last stage of the argument. Moreover, this claim was always an opportunistic one, in the sense that it sought a huge windfall benefit for Mr Flanagan that was wholly at odds with his minimal capital contribution to the business of the LLP and the express terms of the LLP Agreement. I cannot escape the impression that the claim was primarily pursued with the object of putting Liontrust under pressure to settle the case, particularly having regard to the very wide disparity between the value which Mr Flanagan put on the claim (between £8 million and £10.4 million in the six month period before the trial) and the far lower amount for which he would have been willing to settle the case prior to the unsuccessful mediation in August 2014 (£750,000 or £800,000). I recognise that I know nothing about what transpired at the mediation, and that the quantification of Mr Flanagan’s claim was to a large extent dependent on the financial information disclosed to him by Liontrust. Nevertheless, I would be inclined to infer that he always recognised, or at least should have recognised, that he was pursuing a high-risk strategy, designed to elicit a favourable settlement from Liontrust.

Secondly, Mr Flanagan greatly complicated the case by raising and pursuing to the bitter end his allegations of breach of duty in relation to the marketing and selling of the Fund. These allegations encompassed the issues relating to the terms on which he joined the LLP (the alleged collateral contract), what was said to him during the negotiations for the purchase of the Occam business (the alleged misrepresentations) and implied terms, together with the actual marketing and sales activity in relation to the Fund between Mr Flanagan’s admission to the LLP on 4 October 2011 and the closure of the Fund in August 2012. These issues gave rise to a very extensive factual enquiry which covered the whole period from the beginning of the acquisition negotiations in early 2011 until August 2012. Furthermore, Liontrust had to give wide-ranging disclosure on all these issues, and had to adduce detailed evidence from numerous witnesses to rebut Mr Flanagan’s claim. It is worth repeating in this context what I said in paragraph [250] of the main judgment:

The third point concerns the evidence which Mr Flanagan gave in cross-examination about his decision to join Liontrust and the events at the meeting in Christopher’s Bar on 3 October 2011. On the basis of my findings of fact, it is clear that Mr Flanagan had already decided to join Liontrust before the meeting took place, and that his conversation with Mr Ions and Mr Collins at Christopher’s Bar fell far short of anything which could reasonably have been understood as giving rise to either implied terms or actionable misrepresentations. Mr Flanagan is to be commended for the honesty with which he gave his oral evidence at trial, but the unfortunate corollary is that this important part of his claim never had any solid foundation. Indeed, it is difficult to see how it could have been advanced in good faith, given the absence of any documentary support for Mr Flanagan’s version of events, and the fact that his own oral evidence fell so far short of his pleaded case. I am again driven to infer that this was, at best, another attempt to complicate the case and put pressure on Liontrust to settle it. Mr Flanagan must have known that he had no proper case to advance on these matters, but this evidently did not deter him from giving instructions to his lawyers which enabled the relevant claims to be pleaded on his behalf. This is the kind of conduct, submits Liontrust, which can and should attract an award of indemnity costs against Mr Flanagan in relation to this part of the case.

Finally, it needs to be appreciated that, in relation to Mr Flanagan’s allegations of breach of contract in the promotion and marketing of the Fund, nothing less than a repudiatory breach of contract would have sufficed. A breach which sounded only in damages would have been of no use to Mr Flanagan, because he advanced no such claim, quite apart from the obvious difficulties he would have encountered in trying to quantify any loss. The purpose of alleging a repudiatory breach was, of course, to provide an alternative route to unlock the door to the default provisions in the LLP legislation. But that objective could be far more simply achieved by relying upon the circumstances of Mr Flanagan’s exclusion from the business. It is true that even this simpler approach, which is essentially the one which I have upheld, would have involved some investigation of the operation of the Fund and the LLP in and around August 2012, because of Liontrust’s contention that it was justified in placing Mr Flanagan on garden leave, even without a valid resolution of the Management Committee, because there was no work for him to do. Nevertheless, any such enquiry would have been far more limited in scope than that which was necessitated by Mr Flanagan’s misconceived allegations of breach of duty in relation to sales and marketing.

Having identified these reasons why the court should depart from the general rule that costs follow the event, the next question is what adjustments should be made to reflect the issues on which Liontrust succeeded, as well as Mr Flanagan’s unreasonable conduct in raising and pursuing the sales and marketing issues, and (more generally) what I would regard as the high-risk litigation strategy which he chose to adopt in pursuing his claim for a buy-out order based on the default rules. Both sides agree that an issue-based analysis is needed in order to help answer this question, but they differ widely about the scope and outcome of the relevant enquiry.

Counsel for Mr Flanagan submit that the only adjustment which should be made against him relates to genuinely separate issues on which he was wholly unsuccessful. They identify those issues as being (a) Mr Flanagan’s claim in relation to the marketing of the Fund, and (b) the legal issues whether, on the true construction of LLPA 2000, the common law doctrine of repudiatory breach applies to section 5 agreements.

They also submit that account should be taken in Mr Flanagan’s favour of Liontrust’s conduct in fabricating the minutes of a non-existent Management Committee meeting on 20 August 2012: see the main judgment at [198] to [208], where I held that this reckless attempt to give the appearance of due process strongly reinforced the repudiatory nature of Liontrust’s breach of contract, and made it impossible for me to conclude that the LLP was acting in good faith when it purported to place Mr Flanagan on garden leave.

Counsel for Mr Flanagan now submit that the matter goes further than that, because the misleading nature of the minutes did not come to light until after the mediation in August 2014. Liontrust therefore went into the mediation allowing Mr Flanagan to believe that the fabricated minutes were genuine, whereas the true position should have been clearly explained when disclosure was given. The mediation was therefore held on a false basis, and as Liontrust knew this to be the case it should pay all the costs of that wasted exercise.

On the question of how Mr Flanagan’s total costs should be allocated between the different issues, the methodology and calculations put forward on Mr Flanagan’s behalf lead to the conclusion that 32% of his costs relate to the issues on which he lost, and 68% to issues on which he was successful, or which are common issues. It is therefore submitted that a fair result would be an order which gave him approximately 68% of his costs, and did not require him to pay any part of Liontrust’s costs.

Liontrust submits, by contrast, that Mr Flanagan should not only bear his own costs in relation to the issues on which he lost, but should also pay Liontrust’s costs of those issues, and should pay Liontrust’s costs of the sales and marketing issues on the indemnity basis. With regard to the allocation of costs to different issues, Liontrust submits that the issues can be grouped under three broad headings:

I do not consider it necessary for the purposes of this judgment to explain and comment on the rival methodologies and calculations designed to attribute each party’s costs to the relevant issues. That is not because I found this material unhelpful – quite the contrary – but rather because the question does not lend itself to precise analysis, given the number and nature of the assumptions which have to be made, and because my ultimate decision depends on my assessment of a large number of factors which cannot be precisely calibrated. It is enough to say that I am grateful for, and have been assisted by, the material supplied on each side; and although there were inevitably points of difference and disagreement, it is reassuring that to a substantial extent the rival approaches had much in common, and did not lead to widely divergent results.

It is now time for me to state my conclusions.

In the first place, I consider that Mr Flanagan’s status as the successful party means that he should recover a substantial proportion of his costs. I think, however, that 68% is significantly too high a proportion of his costs for Liontrust to have to pay, bearing in mind the very limited nature of his success, and the cumulative force of the criticisms which I have made of the way in which he formulated and pursued his case. In my judgment the fair result is that Mr Flanagan should received 50% of his costs, to be assessed on the standard basis.

Secondly, I consider that Mr Flanagan should not only bear his own costs, but should also pay a large proportion of Liontrust’s costs of the issues on which he lost, that is to say (adopting Liontrust’s broad classification) Issues B and C. I appreciate that an order in this form will lead to Mr Flanagan paying Liontrust substantially more in costs than he receives from Liontrust, but this is the logical consequence of (a) the issue-based approach to costs which this case in my judgment calls for, and (b) the fact that (subject to detailed assessment) Liontrust’s costs are more than twice as high as Mr Flanagan’s. There is certainly no principle of law that the successful party cannot end up, on an issue-based approach, paying more than he receives: see Summit Properties Ltd v Pitman [2001] EWCA Civ 2020, especially at [25] and [27] per Chadwick LJ. The present case is a highly unusual one in many respects, and in my view the fair outcome on costs is indeed one where Mr Flanagan will end up as the paying party.

Thirdly, however, I think that Liontrust should recover significantly less than the 77% of its costs for which it contends. Whatever the defects of Mr Flanagan’s own claims and his conduct of the litigation, he was in my view badly treated by Liontrust once Liontrust had decided to part company with him, and his complaints of unlawful exclusion were brushed aside with dismissive arrogance. This arrogance was compounded by the truly disgraceful episode of the fabrication of the Management Committee minutes, rightly castigated by Mr Thompson as conduct unworthy of an entity regulated by the FSA in a group headed by a Plc with premium listing on the London Stock Exchange. Looking at the matter in the round, and taking account of the issues relating to conduct on both sides which I have mentioned in this judgment, my conclusion is that Mr Flanagan should pay Liontrust 60% of its costs, assessed on the standard basis. I would add that Mr Flanagan may perhaps consider himself fortunate that I have narrowly decided not to order assessment of Liontrust’s costs of Issue B on the indemnity basis, but have preferred to take account of the issues relating to conduct on both sides in fixing the percentage of Liontrust’s costs which he must pay.

Pausing here, I have now decided that Liontrust should pay 50% of Mr Flanagan’s costs, and Mr Flanagan should pay 60% of Liontrust’s costs. Taking the figures set out in paragraph [18] above, and ignoring VAT, this would result in a payment of approximately £354,647 by Liontrust to Mr Flanagan, and approximately £959,567 by Mr Flanagan to Liontrust. The difference between those two figures is £604,920, which represents about 37.8% of Liontrust’s total costs. In principle, therefore, it would be possible to net off the two cross-payments and instead order Mr Flanagan to pay 37.8% of Liontrust’s costs. Nevertheless, given the very large disparity between the costs schedules of the two sides I do not think I can safely assume that detailed assessment would have the same proportional impact on each side’s costs, and I therefore think it is safer to leave my order in the form of proportional costs orders in each direction, leaving it to the parties to agree on an appropriate netting off if they are able to do so.

The final question is whether I should order an interim payment on account of costs by Mr Flanagan to Liontrust. By virtue of CPR rule 44.2(8), such an order should be made “unless there is good reason not to do so”. Neither side submitted that there was any good reason not to make an interim order, and I am satisfied that it would be appropriate to do so. For this purpose, I can take as my starting point the net sum of £604,920 in Liontrust’s favour mentioned above. Subject to any further submissions which may be made when this judgment is handed down, I would propose to order Mr Flanagan to make an interim payment to Liontrust of £300,000.

I should add a word in relation to VAT. In their skeleton argument, counsel for Liontrust say that the respondents’ are unable to recover all the VAT payable by them on the costs that they have incurred. There are two reasons for this. First, LIS and LAM make no supplies which are subject to VAT, and are therefore unable to reclaim any input tax. Secondly, the LLP makes both taxable and non-taxable supplies, and is able to recover only a proportion of its input tax. It is therefore submitted that, if the court orders Mr Flanagan to pay costs to Liontrust, the payment should include any VAT that has not been reclaimed or is not reclaimable. I have not heard argument on this point, nor do I know what proportion of the LLP’s VAT is said to be unrecoverable. In the circumstances, I do not think it would be appropriate to add anything to the interim payment on account of VAT, unless the parties are able to agree on a minimum percentage of the relevant VAT which will be irrecoverable.

A number of points on the form of the order to be made were briefly discussed at the hearing, and I hope will be capable of agreement before this judgment is handed down.