Bastionspark LLP, Edgedale LLP, Starbrooke LLP, Hawksbridge LLP v The Commissioners for HM Revenue and Customs: [2016] UKUT 0425 (TCC) [2016] UKUT 0425 (TCC)

UPPER TRIBUNAL
TAX AND CHANCERY CHAMBER
[2016] UKUT 0425 (TCC)Case No UT/2015/0051
Between
(1) BASTIONSPARK LLP (2) EDGEDALE LLP (3) STARBROOKE LLP (4) HAWKSBRIDGE LLPAppellantTHE COMMISSIONERS FOR HER MAJESTY’S REVENUE AND CUSTOMSRespondent
Before
Mr Justice NugeeDate 5 October 2016Category: Tax
[1]This is an appeal by four LLPs (Bastionspark LLP, Edgedale LLP, Starbrooke LLP and Hawskbridge LLP) (“the LLPs”) in relation to costs, and is a further round in the litigation between the “Icebreaker” partnerships and the Commissioners for HM Revenue and Customs (“HMRC”).[2]The First-tier Tribunal (Judge Colin Bishopp and Mr Richard Law) (“the FTT”) heard the substantive proceedings in November 2012. Those proceedings consisted in part of appeals by the LLPs (and a fifth LLP, Acornwood LLP) against various decisions of HMRC disallowing expenditure by them which they claimed to be deductible in the calculation of trading losses; and in part of a joint reference to the FTT of a number of questions by seven individuals who were members of the LLPs or other partnerships. The FTT released its decision on the substantive questions (“the Substantive Decision”) on 7 May 2014: see Acornwood LLP v HMRC [2014] UKFTT 416 (TC). I will have to look in more detail below at what the FTT decided, but in summary they allowed the appeal of all five LLPs to a limited extent but rejected the most significant of the claims made by them, and largely rejected the contentions of the individual referrers in the references. An appeal by the five LLPs against the Substantive Decision was recently dismissed by me sitting in the Upper Tribunal (“the UT”): see Acornwood v HMRC [2016] UKUT 361 (TC). An appeal by the individual referrers against the Substantive Decision is being pursued separately and is currently pending.[3]The question of the costs of the appeal to the FTT was then argued as between the four LLPs and HMRC. Neither Acornwood LLP nor the individual referrers were parties to the costs proceedings, Acornwood because its appeal started before the Special Commissioners and was not allocated to the Complex category on transfer to the FTT, and the individual referrers because they opted out of the costs regime, as they were entitled to under the FTT rules. The argument was heard by Judge Bishopp (sitting in the FTT alone) in November 2014 and he released his decision in January 2015, with an amended version released on 3 March 2015 (“the Costs Decision”). Neither party in fact identified any difference of substance in the amended version. By his decision, Judge Bishopp ordered the LLPs to pay two-thirds of the costs of HMRC.[4]The LLPs applied to the FTT for permission to appeal. By a decision dated 23 March 2015 (and issued to the parties on 14 April 2015) Judge Bishopp declined permission. This decision (“the Permission Decision”) contains some further explanation of the reasons for his decision on costs. Permission was subsequently granted by the UT (Judge Roger Berner) on 2 June 2015. The Substantive Decision[5]The LLPs were set up to exploit various forms of intellectual property rights such as the treatment for a book or music recordings. It is not necessary for the purposes of this appeal to set out the arrangements which the LLPs entered into for this purpose in 3 any detail: a full account can be found in the Substantive Decision of the FTT. I should however summarise the claims to losses which the LLPs made, which can conveniently be done by referring to the simplified example which Mr Jonathan Peacock QC deployed on the appeal against the Substantive Decision, and which is referred to in my decision on that appeal: see [2016] UKUT 361 (TC) at [6]. This simplified example uses a nominal amount of capital subscribed by the members to an LLP of 100. Of this the members contribute 20 out of their own resources and borrow the other 80 from a bank. The LLP pays 5 of the 100 to an Icebreaker company (Icebreaker Management Ltd or Icebreaker Management Services Ltd (“IML” and “IMSL”)) in part for advisory services and in part for administrative services, and the other 95 to a company, which agrees to arrange for the exploitation of the rights by having the relevant product (book or CD or the like) first produced and then marketed. In the case of the relevant LLPs this was Shamrock Solutions Ltd (“Shamrock”). Shamrock agrees to pay the LLP a share of the resulting revenue, and also certain guaranteed sums by way of quarterly payments and a Final Minimum Sum. The Final Minimum Sum (which is payable in various circumstances but could be made payable, and in practice was paid, at the end of four years) matches the amount of the members’ borrowings (so in this example is 80) and the quarterly payments match the members’ interest payments to the lending bank. Shamrock is under an obligation to provide security for these sums which meant that it had to deposit 80 with a bank (in practice the same bank) to procure a letter of credit, usually, although not invariably, using 80 of the 95 for this purpose.[6]The LLPs completed self-assessment tax returns for their first years of operation in which they each claimed to have sustained trading losses based on the whole of the 95 paid to Shamrock and the 5 paid to IML/IMSL as deductible expenditure. HMRC in each case opened an enquiry and in due course issued a closure notice which amended the losses to exclude any part of the 95 or the 5. The LLPs’ appeals to the FTT were against these closure notices.[7]The question for the FTT was therefore whether the payments made by the LLPs to Shamrock and IML/IMSL were properly deductible in calculating the LLPs’ trading losses. The FTT considered that question by reference to four aspects of the payments:(1) The part of the 95 paid to Shamrock which corresponded to the Final Minimum Sum (the 80).(2) The balance of the sum paid to Shamrock (the 15).(3) The part of the 5 paid to IML/IMSL for advisory services.(4) The part of the 5 paid to IML/IMSL for administrative services.[8]The answer that the FTT gave in the Substantive Decision was as follows:(1) The 80 was not deductible. It was not paid wholly and exclusively for the purposes of the LLP’s business (that being the exploitation of intellectual property rights) but for the acquisition of a guaranteed income stream: [270].(2) The 15 was largely but not wholly deductible. Part of it represented the cost 4 of work done by Shamrock in the relevant year and was an allowable expense; part represented pre-payment for future work and was not an allowable expense in the relevant year. The FTT could not determine from the evidence available to them how the overall cost should be apportioned (beyond saying that it was likely that the former would significantly exceed the latter) and left the parties to agree that, or to return for further argument if they could not: [293].(3) The part of the 5 paid as an Advisory Services Fee was not deductible. In the case of Edgedale, Starbooke and Hawksbridge, this was because it was a payment for a package of projects (and hence the consideration for an asset): [317]. In the case of Bastionspark, it was because part of it was payment for a package and part a pre-payment for advice to be given in future years. The FTT did not have the evidence to apportion the fee between the two, but it made no practical difference to the outcome of Bastionspark’s appeal as none of the fee represented an allowable expense in the relevant year: [318].(4) The part of the 5 paid as an Administrative Services Fee was an allowable expense: [319].[9]It can be seen immediately that the LLPs had mixed fortune in their appeals. They succeeded in establishing that part (likely to be significantly the larger part) of the 15, and the whole of the Administrative Services Fee were, contrary to HMRC’s contentions, deductible. But they failed to establish that the 80, or the remainder of the 15, or any part of the Advisory Services Fee, were deductible.[10]Ms McCarthy, who appeared for the LLPs, put before me a table which showed the effect of the FTT’s Substantive Decision in monetary terms. It is not necessary to set it all out, but I will give an example, that of Hawksbridge:(1) Hawksbridge paid £5,188,500 to Shamrock (ie the 95), and sums of £384,681 and £50,000 to IMSL by way of Advisory Services and Administrative Services Fee respectively (ie the 5).(2) It completed a self-assessment return for the tax year ended 5 April 2010 which included these sums as deductible expenses, the £5,188,500 as Cost of Sales, and the fees as General Administrative Costs. The overall loss for the year shown on the return was £5,628,653.(3) By a closure notice dated 14 May 2012 HMRC amended the return by adding back the whole of the Cost of Sales figure of £5,188,500 and £432,099 for General Administrative Costs. (It is not apparent, and I do not think I was told, why the figure of £432,099 is slightly less than the sum of the two fees, which add up to £434,681, but this makes no difference to the questions of principle.) That represented an amendment of £5,620,599 and led to an amended figure for the loss of £8,054.(4) The outcome of the FTT’s Substantive Decision was that (i) the fee paid to Shamrock was allowed to a limited extent; (ii) none of the Advisory Services Fee of £384,681 was allowed; and (iii) the whole of the Administrative Services Fee of £50,000 was allowed. The FTT did not determine how much 5 of the fee paid to Shamrock was allowable. But the Final Minimum Sum payable by Shamrock to Hawksbridge (the 80) was £4,636,600 so the total allowable cannot exceed £5,188,500 – £4,636,600 or £551,900.(5) Ms McCarthy told me that the parties considered that the total allowable to the four LLPs would ultimately amount to between 14.03% and 15% of the amounts disallowed by HMRC. Since the total disallowed in the closure notices for the four LLPs was some £23.8m, this meant that the four LLPs together would succeed in having some £3.3m to £3.6m allowed. Some £820,000 of that was attributable to the Administrative Services Fees, which meant that between £2.5m and £2.75m would be allowed in respect of the fees payable to Shamrock out of a total of about £22m originally claimed, which equated to about 11%-12% of the original claim; and the £820,000 allowed for Administrative Services Fees represented about 47% of the total administrative costs claimed.(6) The five LLPs that took part in the substantive appeals were selected as lead cases, each one representing one tax year from 2005-6 to 2009-10. There were a further 46 LLPs which were “related” cases, 40 of which fell in the same tax years as the four LLPs which were parties to the costs proceedings. If one added those 40 LLPs to the four appellant LLPs, the total losses claimed in the relevant returns were some £315m, and the total that they would between them establish to be deductible would be some £44m to £47m. Ground 1 – who was the successful party?[11]Ms McCarthy advanced two grounds of appeal. Ground 1 was that the FTT erred in law in making a proportionate costs order in favour of HMRC, recognising HMRC as the substantial victor. The FTT should instead, she said, have identified the LLPs as the successful parties.[12]In his Costs Decision, Judge Bishopp said that although all of the relevant circumstances must be taken into account in any individual case, the normal rule in the FTT, when costs-shifting is possible, is the same as in the courts, namely that the unsuccessful party will be ordered to pay the successful party [5]. He accepted that it was necessary to start by identifying the successful party [19]. He then said (at [20]):
“…it does not seem to me, despite Ms McCarthy’s eloquent submissions to the contrary, that there is any real room for doubt that HMRC have been the substantial victors… I agree with Ms McCarthy that it is beside the point that these were, as we decided, tax avoidance schemes; what matters is that the outcome was substantially in HMRC’s favour, whether one measures success in monetary terms, or by reference to the number of issues decided in one way or another.”
At [21] he added:
“Despite what I have said about other issues, the payment equivalent to the final minimum sum was the main battle ground in the appeals, and not 6 merely because in money terms it dwarfed all of the other claims for relief.”
At [24] he expressed his conclusion as follows:
“After taking account of the arguments I have set out above, my conclusions that HMRC were the substantial victors, that nevertheless the appellants achieved something, and that monetary success is not the only yardstick, I have arrived at the view that the appellants should pay two thirds of HMRC’s costs of the appeals, to be assessed on the standard basis by a costs judge of the Senior Courts if not agreed.”
[13]Judge Bishopp gave further reasons for his conclusion in refusing permission to appeal. At [3]-[5] of the Permission Decision he said: “3. The application points out, correctly, that the identification of the successful party is not a matter of discretion, even if in a case of this kind it is a matter requiring the exercise of some judgment. I am not, however, convinced that it is wholly a matter of law, as the application suggests; in my view it is a mixed question of law and fact. Nevertheless, it does not seem to me make any difference which is the correct view on that issue since in my judgment there is no arguable error in the conclusion I reached.[4]The application proceeds from the proposition that since the applicants succeeded in reducing the scale of the amendments of their returns, they must be regarded as the victors regardless of the fact that they clearly lost in respect of other issues. That argument is in my view misconceived. If a claimant with a single cause of action seeks an award of, say, £10,000 but recovers only £2,000 he is, as I agree the authorities show, the victor even though he has recovered much less than the claimed amount. But if, in the same proceedings, he makes a further claim based on a different cause of action for say, £50,000, which the defendant is able to defeat, it would be unrealistic to say that because the claimant has recovered £2,000 of a total claim of £60,000 he is the victor and should recover all his costs. Whether by means of separate orders or a composite order he would be required to meet the defendant’s costs relating to the second cause of action while recovering his own costs relating to the first.[5]In my view this case is much closer to the second example than the first. The applicants were successful, to a greater or lesser degree, in challenging some of the grounds on which the amendments had been made, but unsuccessful in respect of other grounds. It was not a case of a single basis of challenge which was partially successful, but of several challenges, some of which failed. Whatever the yardstick adopted – monetary amounts, number of issues, time taken up at the hearing – it is, in my view unquestionably, a reasonable conclusion open to the tribunal that HMRC were the substantial victors.” The statutory provisions[14]The relevant statutory provisions are as follows. By s. 29(1) of the Tribunals, Courts and Enforcement Act 2007 (“TCEA 2007”), the costs of and incidental to all 7 proceedings in the FTT and UT “shall be in the discretion of the Tribunal in which the proceedings take place”; and by s. 29(2) the relevant Tribunal “shall have full power to determine by whom and to what extent the costs are to be paid”. These provisions are similar to the corresponding provisions applicable to the senior courts and county court which are found in s. 51(1) and (3) of the Senior Courts Act 1981.[15]By s. 29(3) TCEA 2007, s. 29(1) and (2) have effect subject to Tribunal Procedure Rules. The relevant rules for present purposes are the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009, SI 2009/273 (“the FTT Rules”). Rule 10 deals with costs. Rule 10(1) provides that the FTT may only make an order in respect of costs in certain specified cases, one of which (by rule 10(1)(c)) is that the proceedings have been allocated as a Complex case and the taxpayer has not opted for the proceedings to be excluded. The effect of this is to give the taxpayer, but not HMRC, an option to opt out of the costs-shifting regime in a Complex case.[16]There are no other rules dealing with costs, and hence no guidance in the rules as to the exercise of the FTT’s discretion, save for the general provision in rule 2(3) that the FTT must seek to give effect to the overriding objective (which under rule 2(1) is to enable the FTT to deal with cases fairly and justly) when exercising any power under the rules. There is therefore no equivalent of CPR Part 44 which contains general rules about costs, and in particular no equivalent of CPR 44.2(2) under which 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, although the court may make a different order. But although there is no express provision to this effect, it does not seem surprising that if the FTT is to have a discretion over costs, the starting point will usually be that if any order for costs is made at all, it will be that costs should follow the event, that is that the loser will pay the winner. This is what fairness and justice would seem normally to require.[17]Ms McCarthy said that the purpose of giving taxpayers a right to opt out of the costsshifting regime in the FTT in Complex cases was to give both parties certainty as to the costs regime that will apply (cf Atlantic Electronics Ltd v HMRC [2012] UKUT 45 (TCC) at [7] per Warren J), and that taxpayers who do not opt out have a right to expect that if they are successful they will receive their costs. I have no difficulty with that as a general proposition (although I think Warren J was in fact referring to the requirement to opt out within 28 days being designed to make it clear to both parties at an early stage which costs regime was to apply, which is a rather different point); and I agree that although costs are discretionary, it is helpful to litigants if they are reasonably predictable, so far as the circumstances permit. But it is also the case that decisions on costs are “peculiarly fact sensitive” (per Sir Anthony May P in Medway Primary Care Trust v Marcus [2011] EWCA Civ 750 (“Medway”) at [1]).[18]Ms McCarthy referred me by way of analogy to BPP Holdings v HMRC [2016] EWCA Civ 121 (“BPP”) as illustrative of the proposition that the tax tribunals (FTT and UT) apply the same principles as are set out in the CPR even if their rules are not identical. In that case, the FTT had made an order under rule 8(3)(a) of the FTT Rules providing that if HMRC did not provide certain particulars they might be debarred, and on HMRC’s failure to comply had indeed debarred them. In doing so the FTT had taken account of the guidance in the well-known Mitchell case on CPR 3.9 (Mitchell v News Group Newspapers [2013] EWCA Civ 1537) and given significant weight to the two factors listed in CPR 3.9(1), namely(a) the need for 8 litigation to be conducted efficiently and at proportionate cost and(b) the need to enforce compliance with rules, practice directions and orders, despite the fact that the FTT Rules contained no equivalent of CPR 3.9, and that the overriding objective as set out in rule 2 of the FTT Rules did not refer to such factors. Ryder LJ said at [33] that the FTT made no error in doing so; and at [37] that although the FTT Rules were silent on the point, there was nothing in the wording of the FTT Rules that justified a different approach to compliance or the efficient conduct of litigation at proportionate cost, or in the overriding objective as set out in the FTT Rules, that was inconsistent with the general legal policy described in Mitchell.[19]As appears from Judge Bishopp’s Costs Decision at [5] it was in fact common ground before him that the normal rule in the FTT when costs-shifting was possible was that the unsuccessful party will be ordered to pay the successful party; and Mr Davey QC, who appeared before me for HMRC, did not dispute that where a costs order is made, the general rule is that costs should follow the event. He also accepted that although the CPR do not apply in the tribunals, case law decided in relation to the CPR can be informative; and that a key issue for the FTT in deciding on an appropriate order for costs is that of identifying the successful party in the proceedings. That I accept, and I accept that BPP is illustrative of the general principle that the FTT will look to cases decided under the CPR as helpful guidance, but I would sound a note of caution. Under the CPR the court has to identify the successful party in order to apply (or decide not to apply) the general rule under CPR 44.2, and as appears from the authorities (below) there has been a tendency for courts to seek to identify one or other of the parties as “the successful party” (and the other as “the unsuccessful party”). But it is not obvious, at any rate to me, that the exercise that the FTT is engaged in is necessarily quite the same. No doubt in a case where there is a clear winner and loser, one would normally expect the costs to follow the event in the FTT as in a court. But that is not because any of the rules require this approach but simply because that is likely to be the fair and just outcome and hence in accordance with the overriding objective applicable in the FTT. It by no means follows that in a case where both sides have had some measure of success the FTT has to, or ought to, approach the question of what is fair and just by seeking to identify one or other party as the successful party. I would have thought that what the FTT should be doing is seeking to identify a fair and just outcome, and that that is likely to be one that reflects, by one means or another, the fact that the parties have each been successful in part.[20]Mr Davey reminded me that under s. 11(1) TCEA 2007 an appeal to the UT from a decision of the FTT only lies on a point of law, and that where what is in issue is the exercise of a discretion, an appellate court or tribunal can only interfere where the lower court or tribunal has “exceeded the generous ambit within which reasonable disagreement is possible” (G v G (Minors) [1985] 1 WLR 647 at 652 per Lord Fraser); or where “the judge has either erred in principle in his approach, or has left out of account, or taken into account, some feature that he should, or should not, have considered, or that [the exercise of] his discretion is wholly wrong because the court is forced to the conclusion that he has not balanced the various factors fairly in the scale” (Roache v News Group Newspapers Ltd (1992) [1998] EMLR 161 at 172 per Stuart Smith LJ (“Roache”)); both these passages were cited by Maurice Kay LJ and applied to a trial judge’s exercise of his discretion on costs in Painting v University of Oxford [2005] EWCA Civ 161 (“Painting”) at [12]. 9[21]These are very familiar principles. Ms McCarthy said that they were not in point, as the identification of the successful party is a point of law: see Day v Day [2006] EWCA Civ 415 where Ward LJ described the failure of the trial judge to identify the claimant as the successful party as a “fundamental error of principle”, which entitled the appellate court to intervene. Mr Davey indeed accepted in terms that the identification of the successful party is an issue of principle and if the lower court or tribunal gets it wrong, this is an error of law. I will therefore proceed on this basis, although again sound a note of caution: I agree that if the FTT identifies a party as successful when they are not (or fails to identify them as successful when they are), this is an error of principle which undermines their decision. But it does not necessarily follow that the question of whether a party is successful is always a yes/no, or hard-edged, question to which there is only one right answer. In some cases it may be that the question is rather whether the decision of the FTT is one that was open to it on the facts. I think this is probably what Judge Bishopp meant when he said in the Permission Decision at [3] that the question was a mixed question of law and fact; and why he referred at [5] to his conclusion that HMRC were the substantial victors as a reasonable conclusion that was open to the tribunal. The present case[22]With that introduction one can consider the present case. I will have to look at the authorities in due course, but before doing so I find it helpful to ask who was the successful party as a matter of ordinary understanding. That does not seem to me to give rise to any significant difficulty. As a matter of common sense the answer is self-evidently that neither party was wholly successful but both parties were successful in some respects. So far as the LLPs were concerned, they succeeded in persuading the FTT that HMRC were wrong to have treated the entirety of the fee paid to Shamrock as not deductible; and wrong to disallow the Administrative Services Fee. As Ms McCarthy submitted, these were real successes which cannot be dismissed as de minimis or trivial: they amounted to over £3m being deductible for the four LLPs in question (and over £40m for the LLPs for whom they were the lead cases); and the LLPs needed to bring the appeals to secure this result, as the default position if they had not done so would be that tax would be payable in accordance with HMRC’s position as set out in the closure notices (this is the effect of certain provisions in the Taxes Management Act 1970 (“TMA 1970”) which I need not set out). As she put it, the appeals were necessary to vindicate the LLPs’ legal rights, and the litigation would have been worth it to secure these successes alone. On the other hand, HMRC also had successes. Most significantly, they persuaded the FTT to agree with them that the bulk of the Shamrock fee was not deductible (the whole of the 80 and an undetermined part of the 15); they also succeeded in establishing that the whole of the Advisory Services Fee was not deductible. And just as the LLPs needed to appeal to establish that some of their expenditure was, contrary to HMRC’s contentions, deductible, so HMRC no doubt in practice needed to appear as respondents to the appeal to achieve these successes. (In strict theory even if HMRC had not opposed the appeal, the LLPs would have had to persuade the FTT to allow their appeals, but in practice if HMRC had not opposed them, there would have been at least a substantial risk that the unopposed appeals would have been allowed.)[23]In these circumstances it seems to me an inadequate account of what happened to say that one or other party was the successful party. The reality is that both parties were 10 in part successful, and in part unsuccessful. In seeking to apply the overriding objective of being fair and just, I do not myself see why the FTT needs to identify one overall winner; what one would expect it to do is fashion a costs order that reflected the fact that each side won in some respects but lost in others. Since costs are ultimately a matter of discretion which has to be exercised on the facts of the individual case, and since cases vary infinitely in their facts, I think one should be wary of seeking to be too prescriptive in how the discretion should be exercised or trying to lay down general principles. But what I think can be said is that there would not at first sight appear to be anything wrong in a case like the present in the FTT making a costs order which seeks to reflect the relative success of the parties.[24]In order to do that the FTT obviously has to form a view as to the relative success of each party. This is what Judge Bishopp did. He gave a number of reasons why in his view HMRC’s success was more significant than the LLPs’, or as he put it HMRC were “the substantial victors” and the outcome was “substantially in HMRC’s favour”. One was that this was so if one measured success in financial terms. Judge Bishopp recorded that in purely arithmetical terms HMRC had succeeded to the extent of about 85% to 87% (and the LLPs 13% to 15%). Before me these figures were slightly, but not significantly, refined, the parties’ position being that the LLPs had succeeded to the extent of 14% to 15% of their total claim. Since the case was entirely about how much (if any) of the sums disallowed by HMRC should have been allowed, it does not seem unreasonable to measure the relative success of the parties in this way. The LLPs’s position was that 100% was deductible, HMRC’s position was that 0% was deductible, and the result was that only some 13% to 15% was deductible. This is obviously a good deal closer to HMRC’s position than the LLPs’.[25]Second, Judge Bishopp referred to the number of issues decided one way or another. Again this cannot be faulted as a matter of fact. HMRC won on the issue of the 80 and the Advisory Services Fee, whereas the LLPs only won outright on the Administrative Services Fee, whereas both parties succeeded to some extent on the[26]Ms McCarthy pointed out that dividing up the issues in this way concealed the fact that the 80 and the 15 were not two separate payments, but one single payment of 95 which was claimed to be deductible as “costs of sale”. She said that Judge Bishopp had been wrong to refer in the Permission Decision to the case as if there were two causes of action. There was only one overall issue, namely what the partnership losses were for the year, and in her submission only one ‘cause of action’ (for each LLP), and on that cause of action each LLP had been successful. Alternatively, one might say that there were two causes of action, namely that concerning the cost of sales (the 95) and that concerning the administrative fees (the 5); or even possibly that there were three: the cost of sales, the Administrative Services Fee and the Advisory Services Fee. But there was no basis, she submitted, for saying that the question of deductibility of the 95 raised two separate causes of action. The LLPs had been successful on that claim, even though they had not succeeded on it in its entirety: and it is a common occurrence that a successful party does not succeed on all issues, or to the full extent of its claim.[27]I agree that one characterisation of the case is that the LLPs challenged HMRC’s contention that the 95 was not deductible and that they were partially successful on that challenge. I also agree that the LLPs did not have, or bring, two separate claims, 11 one in relation to the 80 and another in relation to the 15. What they did was exercise a right of appeal under TMA 1970. The technical position under TMA 1970 is as follows. By s. 12AC(1) an officer of HMRC may enquire into a partnership return by giving notice of enquiry; by s. 28B(1) such an enquiry is completed when an officer states his conclusions in a closure notice, which by s. 28B(2) must make the amendments of the return required to give effect to his conclusions; by s. 31(1)(b) an appeal may be brought against any conclusion stated or amendment made by a closure notice under s. 28B; and by s. 50(6) if on such an appeal the FTT decides that any amounts contained in a partnership statement are excessive, the amounts shall be reduced accordingly. It follows that what each LLP did was appeal against the amendments made by the closure notice, one of those amendments being the amendment to disallow the whole of the 95. In Hawksbridge’s case, for example, the relevant amendment disallowed the whole of the cost of sales of £5,188,500. When Hawksbridge appealed against that, it did not bring two separate appeals, one in relation to the part equivalent to the Final Minimum Sum of £4,636,600, and one in relation to the balance of £551,900; it brought a single appeal against the disallowance of the £5,188,500. In that sense I agree that this was analogous to a single cause of action not two separate ones.[28]But although it is true that the payment to Shamrock by each LLP was a single payment, and the appeal by each LLP against the disallowance of it was a single appeal, it is quite apparent from the way in which the FTT’s Substantive Decision is structured that the arguments in relation to the part of that single payment equal to the Final Minimum Sum (the 80) were very different from those in relation to the balance (the 15), and the two parts of the payment were treated quite separately. That was also the case on the appeal against the Substantive Decision which I heard, although I accept Ms McCarthy’s warning that I should not assume that the way the issues were argued in the UT on appeal necessarily reflected the way they were argued before the FTT, not least because HMRC did not cross-appeal the points on which they lost. Nevertheless it does seem to me from what is apparent from the Substantive Decision of the FTT itself that the FTT was justified in treating these two aspects of the payment (the 80 and the 15) as giving rise to two distinct issues. That seems to me to be the case even though they may technically have arisen on the same claim.[29]I therefore do not accept that there was anything wrong in the FTT saying that HMRC had succeeded on more issues than the LLPs. What however this really shows is that counting up the number of issues is likely in many cases (including this one) to be a rather crude way of measuring relative success. Some issues take up a good deal more time, and cost, than others. Mr Davey told me that the Administrative Services Fee, on which HMRC lost, by his reckoning took no more than 2 hours of the FTT’s time (although the hearing as a whole lasted some 3 weeks, of which some 2½ weeks were devoted to the LLPs’ appeals rather than the references). On the other hand Judge Bishopp accepted that a significant amount of effort was devoted to the issues on which the LLPs succeeded (by which I assume he meant primarily the bulk of the 15), and that a direction which reflected that fact was necessary. I will come back under Ground 2 to the way in which he gave effect to that.[30]Third, Judge Bishopp referred to the fact that the payment equivalent to the Final Minimum Sum (ie the 80) was the main battle ground and not merely because in money terms it dwarfed all of the other claims for relief. A judge who has heard the 12 case is obviously at a distinct advantage in identifying the main battle ground or most significant issue, and it would take cogent reasons to persuade an appellate court or tribunal that he had erred in his assessment. In the present case, there seems to have been ample justification for Judge Bishopp’s view. Mr Davey told me that in the estimation of HMRC’s legal team at least 75% of the time at the hearing and of the parties’ closing submissions were devoted to the question of the 80, and Ms McCarthy did not suggest that was wrong.[31]There is a further consideration (which Judge Bishopp may also have had in mind, although he did not say so expressly). The importance of an issue in litigation cannot always be measured in terms of its financial value, or of the time taken up at the hearing with it. In the present case, another reason why the 80 may have been regarded as the main battle ground was that success or failure on the 80 in practice determined whether the members of the LLPs derived any net benefit from having joined the LLPs at all.[32]The FTT referred in its Substantive Decision to the way in which the Icebreaker partnerships were marketed by reference to their intended tax advantages (at [179]- [194]). In his Costs Decision Judge Bishopp accepted a submission by Ms McCarthy that it was besides the point that the FTT had decided that the schemes were tax avoidance schemes. This I accept: a taxpayer is not to be penalised in costs just because the FTT has characterised the arrangements as a tax avoidance scheme. But the present point is a slightly different one, which is that the deductibility of the 80 was central to the intended benefit of joining the scheme. The FTT referred in their Substantive Decision (at [191]) to a letter written by an IFA to an investor which showed how an investor who was looking to recover tax on income of £500,000 over the previous three years could invest £503,408, contributing 25% (£125,852) in cash, and borrowing the remaining £377,556. He was advised that this should enable him to make a loss claim leading to a tax refund of £186,674 “giving you a positive cash flow of £60,882” and that after four years the partners would have the option to sell the assets of the partnership which could lead to a liability for Capital Gains Tax of £37,756.[33]That illustrates that the intended effect of joining the LLPs was that the member would receive more back in initial tax relief than the cash he had to pay to the LLPs; and that even allowing for any CGT payable on sale of the business after four years, he would be left with a net financial benefit (regardless of whether any significant revenue was derived from the exploitation of the rights). In terms of the simplified example, a taxpayer who had income of 100 and would otherwise have paid tax of 40 on it (leaving him with a net 60), could instead pay 20 in cash to an LLP, borrowing a further 80. If the scheme worked as intended, the LLP would spend the whole 100 in the first year, and the member would be able to claim loss relief on the whole (or nearly the whole) of that 100. That would mean he would not pay tax on his income of 100, so leaving him with a net 80 (after paying 20 to the LLP) instead of a net 60. He would therefore be financially better off by joining the scheme than not, even without allowing for any revenue from the successful exploitation of the LLPs’ rights.[34]But that would of course only work if the whole of the 100, or at any rate the bulk of it, were deductible. In practice that meant that the LLPs had to win on the 80 in order for the scheme to produce a net benefit to the members at all. If the 80 were not deductible, then even if all the remaining 20 were deductible (and the member could 13 duly claim sideways loss relief) the member would be left with taxable income of 80 (on which he would have to pay tax of 32), and so after paying 20 in cash to the LLP would be left with a net 48. In other words far from providing a financial benefit, the effect of participating in the scheme would be to leave the member significantly worse off than if he had never joined (unless, which the FTT regarded as unlikely, the project were so successful as to deliver significant revenue from its exploitation).[35]In that sense too, the deductibility of the 80 can I think fairly be regarded as the key issue, because on that issue turned the question whether the venture was a financial success or a financial failure for its members. In circumstances where the LLPs lost on that issue, it certainly does not look as if they have been the successful parties overall. It looks more like a case where the LLPs have failed in their main objective, even though they have had some success on other issues. If one asked the members of the LLPs after the Substantive Decision whether they had won their appeal, one would not expect them to say that they had. The short answer would be likely to be No; a longer and more complete answer would be that they had won some points, but lost the main issue.[36]It seems to me therefore that on any normal view Judge Bishopp was fully entitled to take the view that HMRC were the substantial victors even though they had not won every point. Indeed to describe the LLPs as being overall the winners would seem in the circumstances to be entirely at odds with the realities and a very curious way of characterising the outcome of the litigation. The authorities[37]I must now consider whether there is anything in the authorities to which I was referred which suggests a different conclusion. The particular question is whether a taxpayer who has achieved something of value by appealing to the FTT, even if it has not achieved all that it sought, is to be regarded as the successful party.[38]Mr Davey referred me to Ensign Tankers (Leasing) Ltd v Stokes (Inspector of Taxes) [1992] 1 AC 655 (“Ensign”). In that case the taxpayer’s position was that it was entitled to a capital allowance of $14m which it claimed to have spent on financing a film. The Revenue’s position was that it was not entitled to any allowance. The House of Lords held that it was entitled to an allowance of $3¼m or 25% of the amount it claimed. When it came to costs, the taxpayer’s counsel submitted that the appeal had been successful in that it had established an entitlement to an allowance of $3¼m, but Lord Templeman rejected this: see at 680D-F. He said that the taxpayer had denied the existence of an allowance limited to $3¼m. The tax avoidance scheme in question had presented the taxpayer with an opportunity to claim from the Revenue a large sum which it had never expended; it was thought to obtain an allowance of $14m for expenditure of only $3¼m; this was a raid on the Treasury; and the raid had failed. The taxpayer was ordered to pay the costs in the House of Lords and below.[39]Mr Davey said that was on all fours with the present case and pointed out that Judge Bishopp’s decision, in only awarding HMRC two-thirds of their costs, was more favourable to the taxpayer than the decision in Ensign. Ms McCarthy said that Ensign was decided before the introduction of the CPR and hence took no account of the more recent case law on the CPR. I agree with Ms McCarthy that despite its authority and its apparent similarity with the present case, Ensign is of limited utility. It is not 14 just that it pre-dates the CPR and so takes no account of the extensive jurisprudence developed on costs under the CPR; it dates from an era where costs were much more commonly ordered to be borne wholly by one side or the other, rather than the more nuanced orders now routinely made, and where both arguments and decisions on costs were usually comparatively brief and without reference to authority. It can no doubt be said that it illustrates that Lord Templeman’s view (and that of the other Lords who agreed with him) was that a taxpayer who had ended up with a 25% allowance had overall failed, but even that I think has to be read against the facts of the particular case. Although it was the Revenue’s primary position in the House of Lords that no allowance was claimable, it was the Revenue that was arguing (by way of crossappeal) that in the alternative only $3¼m was properly allowable. The taxpayer’s position was that this was wrong: see the report of the arguments at 660C-G. It is hardly surprising in those circumstances that Lord Templeman said that the taxpayer had denied the existence of a limited first year allowance. Where a taxpayer is arguing for $14m, and the Revenue for either $0 or $3¼m, and the answer is $3¼m, it does not seem at all surprising that the court regards the taxpayer as having lost. It is an illustration of the fact that decisions on costs are particularly fact sensitive as I have already referred to. I do not therefore accept Mr Davey’s submission that it should in effect be determinative of this appeal.[40]Putting Ensign on one side, Ms McCarthy advanced a carefully structured argument that the FTT had erred in principle in failing to find that the LLPs were the successful parties. This was developed by reference to four propositions, the first of which was that the appellant to the FTT should be treated as the equivalent of a claimant in ordinary civil litigation.[41]She illustrated this with Versteegh Ltd v HMRC [2014] UKFTT 397 (TC) (“Versteegh”), where the FTT (Judge Berner) dealt with the costs of conjoined appeals arising out of a scheme designed to achieve a corporation tax saving and based on an intra-group lending. In order for the scheme to work, one group company (the borrower) needed to be entitled to a deduction for the cost of the borrowing, but without any taxable receipt by either of two other group companies, namely the lending company (the lender) and a company which received shares in lieu of interest (the share recipient). HMRC both denied that the borrower was entitled to the deduction, and claimed that either the lender or the share recipient (but not both) was liable to tax. The FTT allowed an appeal by the borrower, and held that the lender was not, but the share recipient was, liable to tax. In other words the aim of the scheme was to have a deduction but no tax charge; HMRC’s position was that the scheme generated a tax charge but no deduction; and the result was that there was both a deduction and a tax charge. The overall effect therefore was that the scheme failed. Judge Berner decided that HMRC should pay the borrower its costs of its appeal but the lender and share recipient should pay HMRC’s costs of their appeals.[42]Ms McCarthy put this before me as an example where the borrower was entitled to its costs because it had been required to appeal to vindicate its legal rights. That I accept, and I have no difficulty with the proposition that in a straightforward case where a taxpayer appeals to the FTT and the appeal is allowed, the taxpayer should be regarded as in a similar position to a claimant who brings a claim and establishes it. That said, I do not think that Versteegh is of much direct assistance, as (as Ms McCarthy herself accepted) it was not a case of a taxpayer obtaining part of what it 15 was claiming. It was a case of three taxpayers, each with different issues and different arguments, one of which (the borrower) succeeded and was entitled to its costs, whereas the other two (the lender and share-recipient) did not succeed (at least if looked at together rather than separately) and had to pay HMRC’s costs.[43]There are two points which are I think of some interest: first, Judge Berner regarded HMRC as the successful party overall “in real life, having regard to the litigation as a whole, and looking at the position in a realistic and commercially sensible way” (at [14]), on the basis that the scheme failed and the taxpayers did not succeed in their objective. This was so notwithstanding that the appellants overall had succeeded in achieving a better outcome by appealing than if they had not done so. If anything therefore it suggests that one cannot simply identify the overall successful party by asking if the appeal to the FTT has achieved anything of value for the appellants.[44]The second point of interest is the way in which Judge Berner dealt with the costs of the lender’s appeal. Viewed in isolation, the lender had won its appeal; but HMRC had always made it clear that the claims against the lender and the share recipient were alternative claims and would not have pursued the lender if it had been accepted that the share recipient was taxable. In those circumstances, Judge Berner regarded the appeals of the lender and share recipient as a single appeal on which HMRC were successful. That is illustrative of the proposition that decisions on costs are always fact-sensitive and cannot be reduced to a mechanistic exercise.[45]Ms McCarthy’s second proposition was that the LLPs’ appeal was based on a single cause of action. I have already referred to this submission (above), where I concluded that although the LLPs’ appeal against HMRC’s amendment to disallow the 95 was a single claim and analogous to a single cause of action, the FTT was justified in treating the 80 and the 15 as giving rise to two separate issues.[46]Ms McCarthy’s third proposition was that having succeeded in part on their cause of action, the LLPs were entitled to be regarded as the successful party. This is the real question which arises on Ground 1 of their appeal.[47]Ms McCarthy referred to two authorities in support of this proposition. The first was Day v Day. Here the dispute concerned the entitlement to the proceeds of sale of a house. This had been acquired by Mrs Elsie Day, the then tenant, under right to buy legislation at a 60% discount, her son John contributing the amount needed to complete the purchase. Elsie was the legal owner and died leaving the house by will to the Defendant, Philip (John’s son). The house was sold and Philip received the proceeds. The Claimant, Mrs Lillian Day, was John’s widow (and Philip’s mother). Her primary claim was that the common intention of Elsie and John was that the house should be held by Elsie on trust for herself for life and then for John (and so was now held for John’s estate). Philip’s primary contention was that John had intended to make a gift of the purchase price to Elsie, so Elsie was the sole beneficial owner (and the house was therefore hers to leave to Philip). Both sides put forward as their fallback position that the house was held on resulting trust as to 60% for Elsie and 40% for John, based on her contribution by way of qualifying for the discount and his by providing the discounted purchase price. The trial judge (Mr Launcelot Henderson QC) held that there was no solid evidence to displace the resulting trust analysis, and hence that neither side had made out their primary contention. He therefore held that Elsie had held the property on trust as to 60% for herself and 40% 16 for John.[48]When it came to costs, he described the hearing as a draw. He therefore ordered that there be no costs of the hearing. But he accepted that it was improbable that Philip would have agreed to pay his mother anything unless the action had been brought and ordered Philip to pay costs down to a date which he described as when the evidence was complete and trial preparation was nearly complete.[49]The Court of Appeal allowed an appeal by Lillian in relation to costs and ordered Philip to pay all his mother’s costs. Ward LJ said (at [16]) that the question was which of the parties enjoyed success in the litigation, and (at [17]) that:
“in a case like this, the question of who is the unsuccessful party can easily be determined by deciding who has to write the cheque at the end of the case; and there is absolutely no doubt at all that the person who has to put his hand in his pocket and pay up the money that is in dispute is Philip. He failed; his mother succeeded. She succeeded, all the more so because Philip adamantly and persistently refused to pay her a penny piece, notwithstanding his fallback position. So I am in no doubt at all that this case did not end in a draw, but ended in victory for mother.”
Sir Martin Nourse agreed saying (at [23]):
“It was not a draw; it was a victory for the Claimant because, as the judge found in the next paragraph of his judgment, it was necessary for her to bring the action even if only to succeed on the fallback basis of 40%. She had to go to judgment in order to obtain the sum of £53,181.65 plus interest which the judge awarded her. Had she not done so she would not have obtained that sum.”
[50]Ms McCarthy puts this forward as a clear example of a case where the Court held the claimant to be the successful party because she succeeded in recovering something and would have got nothing if she had not sued. That generated a prima facie entitlement to costs as a starting point (even though she did not recover everything she was seeking). Mr Davey drew attention to the words of Ward LJ “in a case like this”, and when asked what the material distinction was between a case like that and the present case, said that the difference was the complexity of the issues in the present case.[51]Ms McCarthy’s other authority was Fox v Foundation Piling Ltd [2011] EWCA Civ 790 (“Fox”). This was a personal injuries claim. By December 2007 liability (subject to an agreed discount for contributory negligence) had been agreed, so the remaining issue was one of quantum. The claimant claimed damages of £280,000; in September 2008 the defendant made a Part 36 offer of £63,000 (but only worth a net £23,000 or so to the claimant after payment to the CRU); and the case settled after a further offer of £37,500 (worth a net £31,000 or so to the claimant) was made in November 2009 and accepted. On these facts the trial judge had regarded the defendant as the successful party from the time of the 2008 offer, but in the Court of Appeal the defendant conceded that the claimant should be regarded as the successful party as the net sum ultimately recovered by him exceeded that available under the 2008 offer: see at [28] and [51]. The Court of Appeal held that there was on the facts 17 no ground to depart from the starting point that the defendant should pay the claimant’s costs.[52]Ms McCarthy points to this as another case where a claimant only recovered a relatively small proportion of his claim (in that case about 11%) but was nevertheless the successful party and prima facie entitled to his costs unless there was some good reason to deprive him of some of them. Jackson LJ (with whom Moore-Bick and Ward LJJ agreed) said (at [46]) that it was a not uncommon scenario that both parties were over-optimistic in their Part 36 offers, the claimant recovering more than the defendant had offered but less than the claimant had previously offered to accept. In such a case, he says:
“the claimant should normally be regarded as “the successful party” within rule 44.3(2). The claimant has been forced to bring proceedings in order to recover the sum awarded. He has done so and his claim has been vindicated to that extent.”
He went on to say (at [47]-[48]) that the starting point in that situation was that the successful party should recover its costs, that an adjustment may be required to reflect the circumstances, such as to reflect the costs referable to a discrete issue on which the successful party has lost, but that in a personal injury action the fact that the claimant has lost some issues along the way is not normally a reason for depriving the claimant of part of his costs.[53]Mr Davey said that this was a case where the defendant conceded that the claimant was the successful party, and referred me to the judgment of Briggs J in Magical Marking Ltd v Ware & Kay LLP [2013] 4 Costs LR 535 (“Magical Marking”) at [15] where he said that since in Fox the question of who was the successful party became common ground, and did not have to be decided by the Court of Appeal, it could not be taken as detracting from a consistent line of Court of Appeal authority on that issue. Ms McCarthy however submitted that it was tolerably clear from Jackson LJ’s judgment in Fox that he regarded the defendant’s concession as rightly made. On this point, I of course agree with Briggs J in Magical Marking that since the conclusion in Fox that the claimant was the successful party proceeded on a concession, it does not technically form part of the ratio of the Court of Appeal’s decision; but I consider that Ms McCarthy is right that Jackson LJ evidently thought the concession rightly made: see his statement of principle at [46] (cited above) and the way he expresses it at [51] (“Accordingly, as the defendant now accepts, the claimant is the successful party in the litigation.”). As Briggs J says, that cannot detract from earlier Court of Appeal authority (see below), but I do not have any real doubt that Fox does indicate that the claimant in such a case who beats the defendant’s best Part 36 offer is normally to be regarded as the successful party even though his ultimate recovery “fell far short” of the original claim (see at [59]).[54]Fox however was a very straightforward type of case. The claimant had a single claim, that for damages for personal injuries. By the relevant stage of the litigation, liability had been accepted. The only live question was one of quantum, that is what the claimant’s admitted injury was worth in terms of damages. Jackson LJ was evidently concerned to lay down a clear approach to costs in such a case: a claimant who succeeded in recovering more than a defendant had offered should normally be regarded as the successful party, and should normally expect to recover his costs 18 under the general rule: see at [46], [62] and [63].[55]As Ms McCarthy accepted however, there are cases where a claimant, despite recovering something, is not to be regarded as the successful party. I was referred to a fair number of these. In date order they are as follows: (1) Roache This was a libel claim in which the plaintiff sought damages and an injunction against republication. The defendants paid £50,000 into court, which the plaintiff did not accept and at trial the jury awarded him exactly the same sum. He did obtain an injunction at trial, but the Court of Appeal held that the defendants were entitled to their costs from the date of payment in. The case pre-dates the CPR but is regularly cited for certain statements by Sir Thomas Bingham MR. Having said (at 166) that one principle habitually applied was that in the ordinary way costs follow the event, and that in a simple case this raised no problem but more complex cases may arise in which “it is necessary to investigate with some care who is really the winner and who is really the loser”, he reverted to the point at 168 where he said:
“The judge must look closely at the facts of the particular case before him, and ask: who, as a matter of substance and reality, has won? Has the plaintiff won anything of value which he could not have won without fighting the action through to a finish? Has the defendant substantially denied the plaintiff the prize which the plaintiff fought the action to win?”
On the facts he said that the plaintiff’s decision to go ahead with the action could only have been because he wanted to win a larger sum, not because of the injunction, and decided that the defendants were the substantial winners; the other judges agreed with him. (2) Islam v Ali [2003] EWCA Civ 612 The claimant claimed to be entitled to the net profits of an accountancy business which he ran for the defendant, amounting to a further £80,000 over that which he had already received. He failed on the various ways in which he put his claim, the trial judge finding for the defendant that all he was entitled to was a reasonable remuneration, quantified at some £12,000. The trial judge ordered the defendant to pay the claimant’s costs, but the Court of Appeal substituted an order that there be no order as to costs, Auld LJ saying that in reality the defendant was the winner, the claimant losing the case on principle on the main issues of the case (at [23]), and Mummery LJ that there were special circumstances (at [30]). (3) Painting The claimant claimed about £400,000 and recovered £25,000 for personal injuries sustained after falling off a ladder; the Court of Appeal held that the defendant was the real winner, the hearing being concerned overwhelmingly with the question of exaggeration on which the defendant won (at [21]). 19 (4) Pindell Ltd v Airasia Berhad [2010] EWHC 3238 (Comm) (“Pindell”) The claimants claimed damages of over $8m, of which some $7m was in respect of a lost sale. They owed an admitted $2.8m or so to the defendant (which the defendant counterclaimed) but claimed to set off their claim. They succeeded in establishing a claim for about $385,000, but that meant that they had to pay some $2.5m to the defendant on its counterclaim. Tomlinson J held that the defendant was the successful party. The claimants had admittedly established a claim, but that is not what the action was about: the entire driver of the litigation had been their claim for $7m for the lost sale. They had lost on that with the result they had to write the cheque at the end of the case. (5) Medway The claimant claimed that breaches of duty by the defendants had led to the amputation of his leg, the quantum of this claim being originally pleaded at over £700,000 and agreed at £525,000 shortly before trial; the defendants admitted the breaches but denied causation; and the trial judge rejected the claimant’s claim, finding that it was in any event too late to save his leg. There was however a pleaded unspecified claim for pain and suffering, and the claimant’s counsel claimed damages for the additional time for which the claimant had suffered pain before the amputation, for which the claimant was awarded £2,000. The Court of Appeal held that this did not prevent the defendants being the successful party. Various reasons were given by Sir Anthony May P and Tomlinson LJ (in the majority): the claim for pain and suffering was an afterthought and scarcely part of the claim as conducted ([8]); the claimant’s only real claim was for the amputation ([8]); the award of £2,000 was insignificant in the context of the claim and the action as a whole (being about 0.25% of the claim), and in truth a last minute addition to salvage something from an action which the claimant lost ([17]); the subject matter of the £2,000 was not what the action was about, and no rational person would issue proceedings such as these if the recovery was only £2,000 ([17]); the award of £2,000 was irrelevant to the purpose of the action ([47]). (6) Magical Marking The claimant brought a claim for professional negligence, the damages claimed being over £10m, and Briggs J awarded her £28,000. He held that the defendants were the successful parties and ordered the claimant to pay 85% of their costs. There were a number of reasons of which the first was that the main thrust of the claimant’s case wholly failed (at [17]); another was that her recovery represented a tiny fraction (less than 1% of her claim), was introduced as an afterthought and had nothing to do with her main grievance (at [21]).[56]This is not, and does not pretend to be, anything like a comprehensive survey of the relevant cases: a fair number of others are referred to in these authorities and doubtless there are many others which are not. But what is apparent even from this limited selection is that in litigation governed by the CPR, the starting point has been to identify one or other party as the successful party, an exercise of which Tomlinson J said in Pindell (at [4]) “It is surprising how often this is in itself a contentious 20 enquiry”, and which in Medway, as Tomlinson LJ, he described (at [46]) as “this surprisingly elusive process.” I certainly do not mean to cast any doubt on the proposition that this is what the CPR requires, no doubt due to the wording of CPR 44.2(2), as the Court of Appeal has said this on more than one occasion. And I have already said that I accept that Fox shows that a claimant with a straightforward claim for damages, where the only issue is “how much is the claimant’s claim worth?”, is normally entitled to be regarded as the winner if he or she recovers more than the defendant has offered.[57]But the cases also illustrate that as soon as one gets away from that straightforward type of case, there are many different factors that can lead the court to conclude that the person who has won “as a matter of substance and reality” may be the defendant, even though the claimant has recovered something of value. Ms McCarthy’s fourth proposition was in effect that the cases in which this was so had to fall within certain categories – an afterthought, a case of fraudulent exaggeration, a case where the recovery was very small (less than 1% of the claim), or a case where there were claims and counterclaims – and that none of those exceptions applied here. But in truth what the cases to my mind demonstrate is that in a case of any complexity there may be any number of reasons why it is right to regard the defendant as the real winner even if they have not defeated the entirety of the claimant’s claim, and that it would be wrong to try and force such reasons into a fixed number of pigeonholes.[58]What then are the implications for the present case? In the first place, I remain unpersuaded that it is always necessary for the FTT, before deciding a question of costs, to identify one or other as the successful party. The driver for doing so in the CPR cases appears to me to be the wording of CPR 44.2(2), although the practice is no doubt also influenced by the seminal statement of Sir Thomas Bingham MR in Roache that the question is who is really the winner. As I have already pointed out, CPR 44.2(2) does not apply in the FTT, and the only specific requirement under the FTT Rules is the obligation to give effect to the overriding requirement and hence to deal with cases fairly and justly. That does not seem to me to require the FTT in all cases to proceed by characterising either the appellant or the respondent as the successful party to the exclusion of the other; I see no reason why the FTT cannot say that both parties have been to some extent successful but the success of one party is more significant than the success of the other, and I see nothing in the authorities which I have referred to which makes this impermissible.[59]I do not therefore accept the premise of Ms McCarthy’s argument. I have already said that I accept that the LLPs did have some success, that this was not trivial or de minimis, and that they could not have achieved it except by appealing; but I see nothing in this which compels the FTT to characterise the LLPs, to the exclusion of HMRC, as having been the winner in the litigation. For reasons given earlier, HMRC were also successful, and on any normal assessment of the parties’ relative success were more successful (or if this is preferable their success was more significant) than the LLPs. That seems to me to justify Judge Bishopp’s characterisation of HMRC as the substantial victors; and for this purpose it does not seem to me to matter whether the parts of the claim on which HMRC succeeded are or are not to be regarded as part of the same cause or causes of action as those parts on which the LLPs succeeded. That is not, as I read the authorities, a determining criterion. As appears from the cases one consideration is whether the claimant (or here appellant LLPs) has lost on 21 the “main thrust” of their case, and in the present case I consider that Judge Bishopp was fully entitled to regard them as having done so.[60]But suppose I am wrong in this, and the FTT, contrary to my view, is obliged in every case to identify an overall winner. I do not see that this would make any difference. In that case, the circumstances do seem to me to justify Judge Bishopp in regarding HMRC as the overall winner rather than the LLPs, despite the undoubted partial success which the LLPs had. I have effectively already explained why: in litigation of any complexity, the factors which go into an assessment of which party is the overall winner are multifarious, and on the facts of this case, I do not see how it can be said to be erroneous in principle for Judge Bishopp to have concluded that HMRC were in substance the successful party. No doubt, as Day v Day illustrates, if other things are equal, the fact that the claimant has had to take proceedings to recover anything from a defendant is a good reason for regarding the claimant as the winner; but it would to my mind seriously distort the principles by treating this as a rule only to be departed from in certain specified circumstances. The true principle as I understand it from the authorities I have referred to is that the question who is the successful party is to be answered by asking who as a matter of substance and reality has won, and that the factors that can be taken into account in answering that question are not confined to any particular checklist. I find nothing in the authorities which establishes that Judge Bishopp made an error of principle in deciding on the facts of this case that HMRC were the substantial victors.[61]Despite Ms McCarthy’s carefully structured and skilfully presented argument, I therefore reject Ground 1 of the LLPs’ appeal. Ground 2: was the FTT wrong to make a composite order?[62]Ground 2 is that the FTT erred in making what Judge Bishopp described as a “composite order”. Instead the FTT should have awarded HMRC a proportion of its costs, and the LLPs a proportion of theirs, rather than netting the two off to produce a single figure of two-thirds.[63]I can deal with this much more briefly. I have already said that where both parties were in part successful and in part unsuccessful, one would expect the FTT to fashion a costs order that reflected the fact that each side won in some respects but lost in others. There are in principle various ways in which this could be done. If A and B have both had some success but overall B is the more successful party, one way would be to require A to pay B the costs of the issues on which B won and B to pay A the costs of the issues on which A won. That might sound attractive in theory, but has the drawback that it requires apportionment of costs between different issues, which has practical difficulties. That is no doubt why CPR 44.2(7) provides that before making an order relating to a distinct part of the proceedings (which would I think include an order for the costs of a particular issue), the court will consider whether it is practicable to make an order under CPR 44.2(6)(a) or (c) instead, (a) being an order that a party pay a proportion of another party’s costs. A second method would indeed be to require A to pay B a proportion of B’s costs to reflect the fact that B has overall been the more successful party but has not been wholly successful. A third method would be to require A to pay B a percentage of B’s costs, but require B to pay A a percentage of A’s costs. 22[64]There is nothing prescribed in the FTT Rules which requires the FTT to adopt one method rather than another, which means that the choice between them (assuming that the FTT thinks it an appropriate case to make an order of this type at all) is a matter for the discretion of the FTT. In the present case Judge Bishopp adopted the second method. I do not think this can be characterised as erroneous in principle. Indeed Ms McCarthy accepted that if Judge Bishopp had said that HMRC were the successful party but that he was reducing the costs they were entitled to recover from 100% to two-thirds to reflect the fact that they were not wholly successful, she would be stuck with that and unable to challenge it.[65]Her point, articulated very clearly in reply, was that Judge Bishopp had not used the language of reducing HMRC’s entitlement to costs or deducting a part of it. Instead he had used the language of making a “composite order”. That she said indicated that what he was doing was notionally awarding HMRC a proportion of their costs (x%) and the LLPs a proportion of their costs (y%) and then netting off the one against the other to produce the two-thirds figure which represented (x-y)%. That was, she said, erroneous in principle because it assumed that the recoverable costs on each side were similar. But the FTT had no information that would entitle it to conclude that the costs on each side would be the same because it had dispensed with costs schedules; indeed the LLPs’ costs (as Ms McCarthy had pointed out to the FTT) were inevitably going to be significantly higher than HMRC’s because the LLPs were the taxpayers and appellants and so bore the cost of disclosure and witness statements, and also the cost of preparation of bundles.[66]This submission therefore rests on the proposition that what the FTT was doing was awarding HMRC x% of its costs and the LLPs y% of their costs, and then netting off the two figures to produce (x-y)%, making the erroneous assumption that the overall level of recoverable costs was about the same. If this is indeed what Judge Bishopp did, then as a matter of arithmetic x% must be 83.33% and y% must be 16.67% to produce a net two-thirds, and Ms McCarthy submitted that there should be substituted for the FTT’s order an order that the LLPs should pay HMRC 83% of HMRC’s costs, and HMRC should pay the LLPs 17% of their costs.[67]What Judge Bishopp said about this aspect of the order was as follows. First in the Costs Decision he said: “16. If I were minded to direct that HMRC should recover only some of their costs, Mr Davey said, I could draw some help from what the Upper Tribunal said in Revenue and Customs Commissioners v Marks & Spencer plc [2010] UKUT 296 (TCC)….[17]The Upper Tribunal went on to observe that the CPR allowed a court to make various directions, some in favour of one party and some in favour of the other, but in practice the balance of convenience dictated that the same result should be achieved by a single direction, in favour of the ultimate receiving party, of a proportion of its costs. In this case, looking at the matter in money terms, HMRC had succeeded to the extent (depending on the precise arithmetic) of about 87%, as well as on the numerical majority of the issues, and they should receive 87% of their costs, offset by a direction in the appellant’s favour for the 13% of the amounts claimed in respect of which they had succeeded, or in round figures 75%. I interpose that the 23 figures of 87% and 13% were not agreed; Ms McCarthy’s position was that the appellant had done rather better, and that the respective figures were 85% and 15%. If I were to adopt this approach a direction in favour of HMRC somewhere between 70% and 75% or thereabouts would be appropriate.…[21]I do not, however, take the view that success should be measured only in monetary terms, and that I should simply direct the appellants to pay 70% or 75% of HMRC’s costs. The appellants are correct to argue that a significant amount of effort was devoted to those issues on which they succeeded, and in my view a direction which reflects that fact is necessary.…” He then concluded in [24] in the passage which I have cited above (at paragraph 12).[68]In his Permission Decision he added to this as follows:
“7. The second ground, coupled with an invitation in the alternative to review the decision in accordance with rule 40, is that, perhaps by oversight, I failed to direct that HMRC should pay one third of the applicants’ costs. There was no oversight; the award I made was designed to reflect the extent to which HMRC and the applicants, respectively, had succeeded and failed, by reference to values and issues, and to make a composite award. Indeed, the applicants accepted that, on the basis of values alone, HMRC had succeeded to the extent of 85% and they to the extent of 15%, suggesting an award of 70% in HMRC’s favour. In fact I took other matters into account in reaching the two thirds award that I made. The assessment of an appropriate costs direction is, in the absence of an error of principle, a classic example of an exercise of judicial discretion with which a superior court or tribunal will not interfere, and I see no prospect of such interference in the case.”
[69]It seems to me that there is here no warrant for concluding that Judge Bishopp reached his overall figure of two-thirds by starting with an award of 83% in favour of HMRC and 17% in favour of the LLPs, and netting the two off. It is I think fairly clear from the way Judge Bishopp refers to it that the approach which he referred to in the Costs Decision at [17] was something which was being urged on him by Mr Davey. That approach would indeed involve subtracting (or netting off) 13% from 87% to leave “in round figures” 75% (on Mr Davey’s figures) or 15% from 85% to leave 70% on the rival figures put forward by Ms McCarthy. It can fairly be said that underlying such an approach is an (unspoken) assumption that the recoverable costs on each side would be about the same – and indeed in the Marks & Spencer case the UT said (at [23]) that there was no reason to think that the costs of each side assessed on the standard basis would not be broadly similar. But it does not seem to me that Judge Bishopp adopted this approach. In the Costs Decision he said that monetary success was not the only yardstick; in the Permission Decision he said that he took other matters into account. There is nothing here which suggests that he started with a figure for a percentage one way and a percentage the other and netted them off: what it suggests is that, having taken the view that HMRC were the substantial victors but had not succeeded on every point, he assessed a fair and just outcome to be that they should receive not 100% of their costs but two-thirds of their costs. 24[70]It can perhaps be inferred from the way he expressed himself that he thought this might be a better outcome for the LLPs than taking a strictly mathematical approach which reflected the parties’ success measured simply in monetary terms (which he said in the Permission Decision would suggest an award of 70% in favour of HMRC); and that underlying that thought was an assumption that if he had adopted that approach netting off would have been appropriate. But it seems clear to me that this is not the approach he adopted. Instead, as he says, he took a range of factors and came up with a single overall figure which he considered fairly reflected the parties’ relative successes. As I have already indicated Ms McCarthy did not suggest that there was anything wrong in principle with a court or tribunal making such an order; and once this is accepted, it is impossible to suggest (and Ms McCarthy did not try to) that the actual figure selected of two-thirds was outside the generous ambit of Judge Bishopp’s discretion.[71]It is true that he referred to this as a “composite direction”. That was in the Permission Decision where he said at [3]:
“The first ground of appeal advanced is said to be that I should not have made a proportionate costs direction (ie I should have made one direction in favour of each party rather than a composite direction)…”
But I think that all he means here by a proportionate or composite direction is a single direction that A pay a proportion of B’s costs rather than a direction that A pay some of B’s costs and B pay some of A’s. That does not necessarily involve arriving at the single figure by a process of netting off, and for the reasons I have given it does not seem to me that Judge Bishopp did adopt such a process in this case.[72]In these circumstances it seems to me that this ground of appeal must also fail. Conclusion[73]For the reasons I have given above I dismiss this appeal. MR JUSTICE NUGEE RELEASE DATE: 5 OCTOBER 2016

Cited in 6 later judgments