Grupo Hotelero Urvasco SA v Carey Value Added SL [2013] EWHC 1732 (Comm)

Case No 2009 FOLIO 931 & 2009 FOLIO 1692[2013] EWHC 1732 (Comm)
IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
COMMERCIAL COURT
Venue Royal Courts of Justice, Strand, London, WC2A 2LLDate 24/06/2013
Before
MR JUSTICE BLAIR
Between
GRUPO HOTELERO URVASCO S.A.Claimant(1) CAREY VALUE ADDED S.L. (Formerly Losan Hotels World Value Added I S.L.)Defendants
Between
CAREY VALUE ADDED S.L. (Formerly Losan Hotels World Value Added I S.L.)ClaimantGRUPO URVASCO S.A.Defendant
Between
LONDON VALUE ADDED I LIMITED
Mr Duncan McCall QC and Mr Tom Smith (instructed by Hogan Lovells LLP) for Claimant in Folio 931 and the Defendant in Folio 1692.Lord Grabiner QC, Mr Manus McMullan QC, Mr Andrew De Mestre and Mr Douglas Paine (instructed by Mayer Brown LLP) for Defendants in Folio 931 and the Claimant in Folio 1692.Hearing Hearing date: 16 May 2013
JudgmentApproved Costs Judgment
[1]This judgment is consequential to that handed down in the two actions on 26 April 2013 ([2013] EWHC 1039 (Comm), “the judgment”). It deals with the question of costs. The costs are substantial, because the proceedings were substantial, and culminated in a lengthy trial. As in the judgment, I shall refer to Grupo Hotelero Urvasco S.A. as “GHU”, and Carey Value Added S.L. as “Carey”. Each company was in substance the claimant in one action and the defendant in the other. There is no need in this judgment to differentiate between GHU and Grupo Urvasco S.A., or between Carey and London Value Added I Ltd.[2]I need not say anything further about the facts of the dispute. The issues are summarised at the beginning of the judgment in the section headed, “What the case is about”. A significant number of the points raised by Carey at trial were decided in GHU’s favour. However, my overall finding was that GHU was in default under the Loan Agreement as at 6 June 2008, and that Carey was not obliged to make the advance otherwise due on that date, and that its subsequent cancellation of the agreement was lawful. I found therefore that Carey was entitled to succeed on its claim for the sums it advanced to GHU, and on its claim under a guarantee. In the result, Carey obtained judgment for €65.9m (subject to a deduction for sums received under a performance guarantee). GHU’s claims against Carey did not succeed. These were valued in closing submissions at about £71.38m.[3]I should make it clear that the questions for decision do not relate to whether costs were reasonably incurred, which is a precondition of their recovery from the other party. That is a question that arises on the assessment of costs by a costs judge (unless the parties can reach agreement). The points that now arise are(1) whether, although judgment was given for Carey, GHU should be entitled to recover from Carey its costs of those issues on which it was successful,(2) alternatively, whether Carey should recover only a proportion of its costs from GHU, and if so what proportion, and(3) the amount that Carey should be awarded by way of an interim payment on account.

The principles

[4]The principles are not in dispute, though some of the relevant CPR provisions have changed with effect from 1 April 2013 pursuant to Lord Justice Jackson’s reforms. (This case does not involve the new provisions as to the proportionality of costs incurred, which apply to proceedings commenced after that date.)[5]In summary, the rule is that the court has a discretion as to whether costs are payable, the amount of those costs and when they are to be paid (CPR 44.2(1)). If the court decides to make an order about costs, the general rule is that the unsuccessful party will be ordered to pay the costs of the successful party (CPR 44.2(2)(a)). However, the court may make a different order (CPR 44.2(2)(b)). In deciding what order to make, the court will have regard to all the circumstances (CPR 44.2(4)), including “… (b) whether a party has succeeded on part of its case, even if that party has not been wholly successful”.[6]CPR 44.2(6) provides that the orders which the court may make include an order that a party must pay: “(a) a proportion of another party’s costs; … (c) costs from and until a certain date only; … (f) costs relating only to a distinct part of the proceedings”.[7]CPR 44.2(7) goes on to provide that, “Before the court considers making an order under paragraph 6(f), it will consider whether it is practicable to make an order under paragraph 6(a) or (c) instead”. In practice, it is (a) and (f) that most often arise for decision, the choice being between what is often called a “percentage” order, or an “issue-based” order for costs.[8]It is to be noted that until 1 April 2013, this rule read, “Where the court would otherwise consider making an order under paragraph (6)(f), it must instead, if practicable, make an order under paragraph 6(a) or (c)”. The former wording pushed the court in a particular direction, whereas the current wording is neutral. However I agree with Carey that it remains the case that if the court is inclined to depart from the general rule and to make an order from the menu in paragraph (6), it should generally favour an order requiring payment of a percentage of costs over an issue-based order.[9]The reasons for this approach are primarily practical, because of the difficulty and ensuing costs involved in an issue-based order when it comes to the assessment stage (Verrechia v Metropolitan Police Comr [2002] 1 WLR 2409 at [115]-[116]). As Jackson J put it in Multiplex Constructions (UK) Ltd v Cleveland Bridge UK Ltd [2008] EWHC 2280 (TCC) at [72], “Where the circumstances of the case require an issue-based costs order, that is what the judge should make. However, the judge should hesitate before doing so, because of the practical difficulties which this causes and because of the steer given by rule 44.3(7)”. The steer has gone, but the practical difficulties remain.[10]After this passage, Jackson J went on to say that, “In many cases the judge can and should reflect the relative success of the parties on different issues by making a proportionate costs order”. This is a reference to a “percentage” order.[11]Finally, the notes to the White Book (2013 ed., vol.1, 44.3.1.17, pp.1315-1317) say at p.1316 that, “The more distinct an issue, and the more that significant costs can be discretely attributed to it, … the more likely an issue based approach is to be taken, especially if the case in question is not a personal injury case”. Though this passage preceded the new rules, I consider that it remains a valid comment.

The parties’ proposed costs orders

[12]Carey’s position is straightforward. It submits that GHU’s claim has failed entirely, and Carey’s claim has succeeded in full. No reduction in its costs is appropriate.[13]GHU on the other hand submits that it should have its costs of three particular issues on which it succeeded relating to the Long Stop Date, MAC and Actos Propios, and that the costs of an adjournment of an earlier trial date should be subject to no order as to costs.[14]On this basis, GHU submits that the appropriate order for costs is as follows:(1) No order as to the costs thrown away by the adjournment of the January 2012 trial.(2) No order in relation to Carey's costs relating to the Long Stop Date, MAC and Actos Propios issues.(3) Carey to pay 50% of GHU’s remaining costs (which represents its estimate of its costs attributable to the Long Stop Date, MAC and Actos Propios issues).(4) GHU to pay Carey's remaining costs (i.e. representing all Carey's costs except (1) and (2) above).(5) Set off of (3) and (4).

The conduct issue

[15]GHU refers to passages in the judgment where I said that the case was notable for the number of issues advanced by the parties at trial, particularly Carey (paragraph 10). With respect to the construction side of the case, it refers to the judgment where I said at paragraph 627: "As Carey says, it is unusual to have a case where there are so many contested breaches, and it is certainly unusual in the Commercial Court. But this, it says, is GHU's choice. In my view, it is rather a consequence of the number of points relied on by Carey ".[16]GHU submits that in fact Carey took a highly unselective approach to the issues which it raised, following a “scorched earth” policy, and seeking to advance all points which were conceivably open to it in the hope that it would succeed on one or more of them.[17]GHU also refers in this regard to the procedural background. I will return to the procedural background when considering GHU’s submissions at to the costs of the adjourned trial.[18]Under CPR 44.2(4)(a), in deciding what order to make about costs, the court will have regard to the conduct of all the parties. The submission made by GHU goes to conduct. As I have said, CPR 44.2(4)(b) deals with success on part of a case, and founds its primary issue-based submission.[19]So far as GHU raises conduct as a separate issue, I do not accept its submission that Carey followed a “scorched earth” policy. As stated in the judgment, I consider that it should have been more selective as regards the points pursued. But as I shall explain later, the case it advanced has to be seen against the nature of the case which it faced. It faced a very large claim for damages in addition to the claim to extinguish its right to recover the advances it made. Against that background it is not surprising that it raised points which it considered properly arguable. The outcome must in my view have costs consequences, but I do not consider that Carey’s conduct arises as a separate matter.

Issue-based costs

[20]GHU’s main contention is that it is appropriate for the court to take an issue-based approach to costs. The particular issues in respect of which it submits that it should have its costs against Carey are(1) GHU’s ability to complete the development by the Long Stop Date (LSD),(2) material adverse change (MAC) in the financial condition of the relevant companies, and(3) the estoppel-like doctrine of Spanish law called Actos Propios, all of which Carey raised as issues, and on which GHU succeeded.[21]As regards Carey’s Long Stop Date point, GHU submits that this required expert evidence, and extensive factual evidence. It was introduced late into the proceedings. Neither of these points is in issue. GHU also submits that LSD raised discreet and distinct issues, which is in issue.[22]As regards material adverse change, GHU submits that Carey failed on the facts as regards two of the three companies, and in the case of the third failed to establish the necessary representation. Again, this is correct. Again, it submits that MAC was a discreet and distinct issue, which is in dispute.[23]As regards Carey’s Actos Propios argument, GHU submits that expert evidence was required, that the court accepted its contentions (both of which points are correct), and accordingly submits that Carey should bear the costs of this issue as well.[24]To support its case that it should have a costs order in its favour on these issues, GHU submits in particular that each of these issues was a distinct issue in the case, which Carey chose to advance but on which it was unsuccessful. It submits that the costs associated with each issue are far from trivial, and that significant costs can in fact be attributed to each issue. If these issues had not been advanced, the scope of the proceedings (particularly the expert evidence and the length of the trial) would have been much reduced.[25]In this regard, it relies on evidence from the partner in charge of the litigation on its behalf that the total costs incurred by GHU were £10,152,000. Of this, it says that costs of about £4.3 million were attributable to the Long Stop Date, MAC and Actos Propios issues. That figure would of course be subject to assessment, but GHU’s submission is that it has been able to show that the costs of these issues can readily be segregated, so that the practical difficulties referred to in the case law do not arise.[26]In Carey’s evidence in response, it is stated that its total costs are £12,403,475. Fees were not charged or recorded on an issue-based basis, but it has produced calculations which identify costs attributable to financial matters (which include MAC), development matters (which includes LSD), and Spanish law, (which includes Actos Propios). In his oral submissions, Mr McCall QC for GHU suggested that a rough and ready attribution of costs within the Carey figures to the three issues in question produces a total of about £4m, which is similar he says to GHU’s own calculation. I express my conclusions on this issue below, but first deal with the adjourned trial issue.

Costs of the adjourned trial

[27]A separate issue is raised by GHU in respect of the adjournment of an earlier trial date. The trial was originally fixed for November 2011 with a 3 week time estimate. It was then re-listed for January 2012. During the course of 2011, Carey was adding construction-based issues to its case, which had previously consisted of alleged financial defaults. The trial estimate increased, and at the PTR both sides appeared to have accepted that the January 2012 date was not practical. The case was re-listed to commence on 22 October 2012 with a time estimate of 9 weeks (this was what in the event happened).[28]GHU submits that its costs thrown away by the adjournment of the trial amount to £900,000, and that no order should be made in relation to these costs. This is because it says that the root cause of the problems which led to the adjournment of the trial was Carey’s insistence that the trial should take place in November 2011 and its resistance to GHU’s proposal of a trial in April or June 2012. This led to the trial being fixed for January 2012, which turned out to be impractical.[29]GHU raises further matters connected with the way in which Carey’s construction case was developed, and a complaint that Carey raised points on the quantum of its damages claim relating to the hotel at the last minute. Accordingly, GHU submits that it would be unfair to lay all the costs thrown away by the adjournment at its door, when it submits that the fault for the adjournment primarily lies at the door of Carey. An appropriate course, it says, would be to make no order as to these costs.[30]Carey disputes GHU’s submissions. It says that the reason that an adjournment was necessary was late disclosure by GHU which was still continuing in December 2011. It says that the late amendments to its case were made because the information had only just become available. GHU likewise made significant changes to its damages claim, and raised at a late stage a case on Spanish law based on materiality. There is no justification, it submits, from departing from the usual order in such circumstances, which is costs in the cause.[31]In my view, each party must bear a degree of responsibility for the adjournment of the trial date. As I said in the judgment, I have had considerable sympathy with GHU’s complaint that during the course of this litigation it has had to deal with an ever increasing number of allegations by Carey in relation to the building development (paragraph 905).[32]However, the broader picture also has to be taken into consideration. Legal proceedings were begun by GHU, not the other way round. Then Carey issued proceedings to recover the loan sums advanced. After it failed to obtain summary judgment under the guarantee, Carey had little alternative but to continue. The damages claim by GHU was a substantial one, the quantification in its opening submissions indicating figures in excess of €100,000,000. Further, if GHU succeeded in relation to the default issue, its case was that none of Carey’s lending was recoverable. It was against that background that Carey explored possible defaults by GHU in relation to the building project, some of which it established.[33]Further, I consider that the fact that Carey was pressing for the matter to come on for trial is not a point that counts against it. It is clear that Burton J was reluctant to contemplate an adjournment, and had the trial date stood, the issues for decision would have been simpler. I consider that the costs of the adjournment are properly treated as costs in the cause.

Discussion and conclusions as to GHU’s application for costs on an issue-based basis

[34]GHU’s case is that having won on the three issues in question, which were distinct issues in respect of which the costs can be readily identified, it is correct in principle that Carey should pay its costs of these issues.[35]It is to be noted that if this is right, it has a very substantial impact on the proportion of its costs assessed on the standard basis that Carey can eventually expect to recover. In oral argument, Mr McCall QC for GHU estimated that on the basis of Carey’s calculations, it would only recover about 20%- 30% of its recoverable costs.[36]I can deal with a number of Carey’s arguments shortly. This was not in my view an “all or nothing” case (see J Murphy & Sons Ltd v Johnson Precast Ltd [2008] EWHC 3104 (TCC) at [4]). Nor was it a case where alternative bases were pleaded for the same basic claim (F & C Alternative Investments (Holdings) Limited v Barthelemy [2011] EWHC 2807 (Ch) at [21]).[37]In fact, a feature of Carey’s case at trial was that it had two parts, the first part concerning financial defaults, and the second part concerning development defaults relating to the London development. The latter was effectively a construction case added during the course of the litigation (see paragraph 5 of the judgment).[38]Nevertheless, Carey succeeded on both parts of its case, though not on every issue. In my view, for the purposes of GHU’s application the most significant point is that, as Carey submitted, the three issues identified by GHU in respect of which Carey failed were not distinct, but were interwoven and interlinked with other issues in the case upon which it succeeded.[39]This is particularly so as regards Carey’s material adverse change case. Although it failed to establish its case in this respect, a significant part of the debate related to its communications with its creditors, which was evidence of the deterioration in its financial condition in the relevant period. These communications were also relevant in respect of the default which was proved by Carey consisting of beginning negotiations with creditors for the rescheduling of indebtedness by reason of financial difficulties. I consider that the financial difficulties experienced by GHU were central to the case generally, including GHU’s damages claim. One reason that its damages claim failed was my conclusion that GHU could not, for reasons I set out in the judgment, have funded the development to completion.[40]The same point about interlinking applies, albeit to a considerably lesser extent, in relation to the Long Stop Date issue. As Carey says, the progress of the development was at least of some relevance to the “forzado” issue, and the issue about the non-payment of contractors.[41]This was a complex case, and in my view it would be unrealistic to expect the costs judge to master the issues in sufficient detail to understand what costs were properly incurred in dealing with which issues (see Verrechia, ibid, at [115]). Issue-based orders can be an effective and appropriate instrument where the relevant issues are distinct from other issues on which the successful party has succeeded. Where, as in the present case, the relevant issues overlap substantially, the cases indicate that the court should hesitate before making such an order.[42]I further agree with Carey that the order proposed by GHU and set out above could not be readily made. It envisages Carey being ordered to pay 50% of GHU’s costs remaining after the costs of the adjournment are stripped out. Whilst it contends that this would represent its costs attributable to the Long Stop Date, MAC and Actos Propios issues, this is of necessity only an estimate (Carey called it guesswork). A costs judge would have great difficulty, in my view, in applying an order of the kind proposed by GHU, and that alone is sufficient to rule it out.

GHU’s alternative case for a percentage order

[43]GHU’s submissions mainly concerned its application for costs on an issue-based basis, but its alternative case is that the court should award Carey a percentage only of its costs. In argument, it indicated that its success on the MAC, LSD and Actos Propios issues meant that Carey should only be entitled to recover 20-30% of its costs.In that regard, it relied on Mears Ltd v Leeds City Council [2011] EWHC 2694 (TCC), a case where the recoverable costs of the successful party were reduced by 65%.[44]Carey submitted that this was a case in which it was the clear overall winner, and that no reduction is appropriate because it succeeded on all the key parts of its claim. In that regard, it relied on Murphy v Johnson, ibid.[45]Percentage deductions in other cases are of little help because the decision is entirely dependent on the particular facts. In approaching the issue in this case, it is therefore necessary to say something more about the three issues upon which GHU places particular reliance.[46]In respect of Actos Propios, the costs were relatively small, and in any event by the time of trial Spanish law experts would have had to be called since GHU raised its own case under Spanish law as regards materiality.[47]As regards MAC, most of the argument at trial concerned the financial condition of GU. Carey proved to my satisfaction that there had indeed been a material adverse change in that company’s financial condition. It did not succeed, because it could not prove a representation on the relevant date.[48]However, the position is not as straightforward as it seems. I explained the background in some detail in a judgment given in relation to an application by Carey to amend its pleadings, and I shall not repeat it here. In summary, having put Carey to proof in its pleadings, GHU asserted at the beginning of the trial that no representation had been pleaded. Carey responded in its written submissions in opening. In closing submissions, GHU advanced a positive argument that the representation case failed in any event because of the terms of the relevant documentation. There followed amendments by Carey, followed by GHU.[49]So although it is correct to say that I found in GHU’s favour on the MAC point, the argument on which it succeeded was articulated late. This puts the outcome on this issue in a rather different perspective. Whilst I do not accept Carey’s submission that it was the overall winner on this issue, it is correct to say that it did succeed on the matters which were the subject of the factual enquiry at trial.[50]The issues as regards the Long Stop Date are however more straightforward in my view. This point was added by a late amendment by Carey in February 2012. It resulted in the calling of programming experts, and required extensive investigation at trial as to the factual state of the development at various times, both through the witnesses and the documents. In the event, I rejected Carey’s case on the facts, as well as on the application of the relevant contractual provisions.[51]Carey contends that because I held that there was a gap in funding which GHU could not have filled, GHU could not have funded the development to completion in any event. However, as GHU pointed out, the LSD part of Carey’s case of necessity assumed that the contractors would be paid, since otherwise the work would not have gone on. Had funding been the only issue as regards completion, there would have been no need to go into the programming aspect of the case. I do not think that this was in dispute at trial.[52]It is correct, as Carey point out, that there is no automatic rule requiring reduction of a successful party’s costs if it loses on one or more issue (HLB Kidsons v Lloyd’s Underwriters [2007] EWHC 2699 (Com) at [11]).[53]However, I am in no doubt that the costs order must reflect in a substantial way Carey’s failure to make good its Long Stop Date argument. This is supported by the history of the point. When it was raised, GHU objected to Carey’s application for permission to call planning and programming expert evidence. In giving permission, Eder J expressly stated that it would be open to the trial judge to order that Carey pay the costs of the exercise. He said:
"I should make clear for the record that if, at the end of the day, this evidence is of no utility or even if it is of utility, the points are decided in favour of GHU, then GHU will of course be able to recover all of its costs in this context at least. All of that will be a matter for the trial judge, and nothing that I say should in any way restrict or dictate to the trial judge what he (or she) should do. As I say, it would be entirely a matter for the trial judge. But, of course, one possibility is that he might order indemnity costs in favour of GHU in relation to those matters."
[54]For reasons already given, it is not practicable to order that Carey pay the costs of the exercise. I am however satisfied that the costs order must properly reflect Carey’s failure to make good its case on this issue.[55]I also consider that it should reflect at least to an extent the outcome on the MAC and Actos Propios issues. The question therefore is as to the extent of the percentage reduction that I should order in respect of Carey’s costs.[56]In my view, a reduction of up to 80% as advocated by GHU would bear no relation to the overall outcome of the case. As I have pointed out already, Carey had in effect no option but to fight on a battleground of GHU’s choosing (see Pindell Limited v Air Asia Berhad [2010] EWHC 3238 (Com) at [11]). Carey was the overall winner. It succeeded on both its claim, and on GHU’s claim, and in principle is entitled to recover its costs. For the reasons I have given, I consider that the LSD issue is an exception, and that there should be a substantial percentage deduction primarily to reflect GHU’s success on this issue. Taking into account the other issues as well, I consider that a proper deduction is 25%.

Payment on account

[57]The last issue concerns Carey’s application for a payment on account of costs, in other words an interim payment by GHU pending the final resolution of the amount of its entitlement to costs.[58]GHU’s submissions were relatively limited in this respect, and there was no dispute as to the applicable principles. Since 1 April 2013, the rule as to payment on account of costs has changed. The relevant rule in CPR 44.2(8) now provides:
“(8) Where the court orders a party to pay costs subject to detailed assessment, it will order that party to pay a reasonable sum on account of costs, unless there is a good reason not to do so.”
[59]This is a change from the previous wording under CPR 44.3(8), which provided that, “Where the court has ordered a party to pay costs, it may order an amount to be paid on account before the costs are assessed” (emphasis added).[60]Blakemore v Cummings [2010] 1 WLR 983 is the leading case under the previous wording. At [23], Elias LJ said that “… it is an important consideration that a party should not be kept out of the moneys which will almost certainly be demonstrated to be due longer than is necessary”. On the then wording of the rule, he added that this did not have the status of a presumption, but was “simply a factor which one would expect, in the normal way, to carry significant weight with the judge”.[61]I agree with Carey that the new wording in CPR 44.2(8) represents a subtle but important change, effectively creating a presumption in favour of such an order, rather than it simply being normal practice as was formerly the case. An order is not automatic, since as Jackson LJ noted in his final report on his proposed reforms, this would be going too far; “For example, there may be doubt about the right to further costs or there may be a real prospect of a set off at a later stage” (Review of Civil Litigation Costs: Final Report (December 2009), para 5.10 (p.459)). However, absent such good reasons an order should be made.[62]When considering what else could amount to a "good reason", I agree with Carey that the underlying principle behind ordering a payment on account of costs remains as identified by Jacob J in Mars UK Ltd v Teknowledge Ltd [1999] 2 Costs LR 44:
"Where a party has won and has got an order for costs the only reason that he does not get the money straightaway is because of the need for a detailed assessment. Nobody knows how much it should be. If the detailed assessment were carried out instantly he would get the order instantly. So the successful party is entitled to the money. In principle he ought to get it as soon as possible. It does not seem to me to be a good reason for keeping him out of some of his costs that you need time to work out the total amount. A payment of some lesser amount which he will almost certainly collect is a closer approximation to justice. So I hold that where a party is successful the court should on a rough and ready basis also normally order an amount to be paid on account, the amount being a lesser sum than the likely full amount."
[63]The fact that, as in the present case, the amounts are large does not invalidate this principle. In Beach v Smirnov [2007] EWHC 3499 (QB) referring to “the important principle that justice requires that the claimant should not be kept out of costs clearly due to him”, Ouseley J said at [11]:
"The principle is that the claimant is entitled to something by way of costs and he should be paid it without delay. The fact that there may be difficulties of assessment - large sums involved and great disputes - does not absolve the judge from the need to consider whether the justice that comes from not keeping someone out of the money to which he is entitled, can only be achieved at too high a price to a defendant, putting him in a position where he has paid more than due. … Justice requires that a sum of costs be paid, provided there can be a reasonable assessment of the sum that is very likely to be awarded. The fact there are large disputes and large sums goes to the way in which that is calculated. It has no bearing, in my view, on the principle."
[64]Nevertheless, it is important to keep in mind the right of the paying party to have the bills of the receiving party carefully scrutinised, and have an assessment if necessary. A payment on account has to be calibrated so as to do justice as between both parties. A judge fixing the amount at the end of a trial is likely to be in the best position to do this. Where figures are available, there may be merit in comparing the costs of each of the parties, bearing in mind that had it been successful, the recoverability of the paying party’s costs would also have been subject to assessment.[65]There cannot be any doubt that a payment on account should be ordered in the present case, and the only question is how much. According to its evidence, the costs incurred by Carey were £12,403,475. Carey seeks an interim payment of 50% of its claimed costs, which it submits represents the minimum amount it can expect to receive on detailed assessment. Alternatively, on a reasonable assessment approach it is lower than what is likely to be awarded. A payment of 50% is £6,201,737.50, which is what it seeks.[66]GHU attacked the amount of Carey’s costs on various grounds, particularly in respect of counsels’ fees, and submitted that much less than £12,403,475 was likely to be awarded on an assessment. It submitted that great caution should be exercised in this regard.[67]I consider that the right approach at the payment on account stage is to proceed on the basis that on an assessment Carey may recover considerably less than its claimed costs. Some limited assistance is obtained by comparing its costs with those of GHU, which according to its evidence were £10,152,000. There is obviously a considerable difference between the figures, although it is fair to say that in proportionate terms, it is not a massive difference.[68]I am content to base the calculation on the percentage claimed by Carey. 50% seems to me to be a reasonable percentage for a payment on account. The question then is to find the appropriate starting figure. Taking Carey’s estimate of its costs as £12m, and allowing for the 25% proportionate reduction I have ordered, produces a figure of £9m, 50% of which is £4.5m. By way of comparison, taking GHU’s estimate of its costs as £10m, and allowing for the 25% proportionate reduction I have ordered, produces a figure of £7.5m, 50% of which is £3.75m.[69]There is authority that the court should adopt a conservative approach, and that appears to me to be right in the present case. Accordingly, I will order GHU to make a payment of £4m on account of Carey’s costs. This is a substantial sum, but the reality is that it is likely to be much less than the amount of Carey’s ultimate entitlement. I will hear any submissions that GHU wishes to make on the mechanics and timing of the payment.

Cited in 2 later judgments