Anthony Outram & Anor v The Commissioners for HMRC [2024] UKUT 00203 (TCC)

[2024] UKUT 00203 (TCC)UT (Tax & Chancery) Case Number: UT/2023/000096
Upper Tribunal
(Tax and Chancery Chamber)
INCOME TAX – discovery assessments – Appellants had claimed losses on tax returns in relation to transactions which had been marketed as tax avoidance – agreed scheme did not achieve intended tax benefits – FTT held inaccuracies in tax returns were deliberate – Appellants appealed – held – no procedural error in review of decision by FTT, and FTT was entitled to make the amendments made to the decision – whilst FTT had purported to apply the correct subjective test when assessing whether conduct was deliberate, reasoning included an objective approach – appeal allowed and remitted to FTT
Venue Hearing venue: The Rolls BuildingHearing Heard on: 1 and 2 May 2024Date Judgment date: 12 July 2024
London
EC4A 1NL
Further written submissions: 3, 10 and 13 May 2024
Before
JUDGE JEANETTE ZAMANJUDGE ASHLEY GREENBANK
Between
ANTHONY OUTRAMAppellantROSS OUTRAMAppellantTHE COMMISSIONERS FOR HIS MAJESTY’SRespondentREVENUE AND CUSTOMSRespondentJeremy Woolf, Counsel, instructed by Barnes Roffe LLP for AppellantsSadiya Choudhury KC, Counsel, instructed by the General Counsel and Solicitor for His Majesty’s Revenue and Customs for RespondentsDECISION

Introduction

[1]The Appellants, Anthony Outram (“AO”) and Ross Outram (“RO”), appealed to the Upper Tribunal against the decision of the First-tier Tribunal (the “FTT”) that there was a deliberate inaccuracy in their self-assessment returns for 2005/06 in which they had claimed losses as self-employed options traders. The Appellants had claimed to set those losses against other income in that year with the remainder being carried back to set off against income in preceding years.[2]By the time of the hearing before the FTT in these appeals, it was accepted by the Appellants that the Appellants were not entitled to the claimed losses. The issues between the parties narrowed further during the course of the FTT hearing, the result of which was that the only issue to be decided by the FTT was whether the extended time limit applied for HMRC to issue discovery assessments, which required HMRC to establish that the loss of tax or excessive relief was brought about deliberately.[3]The FTT’s decision was released on 27 April 2021 and is reported at [2021] UKFTT 126 (TC) (the “Original Decision”). The FTT dismissed the Appellants’ appeals. That decision was subsequently amended by the FTT and the amended decision was released on 25 September 2023 (the “Revised Decision”). One issue in the appeal before us was as to the basis on which those amendments were made, and that is addressed further below as part of the discussion of Ground 2. The Revised Decision was not published. We have attached the Revised Decision as the Appendix hereto. References to numbers in square brackets are to paragraphs of the Revised Decision unless the context indicates otherwise.[4]The Appellants had applied for permission to appeal against the Original Decision. The FTT granted permission to appeal on 25 September 2023.

The scheme

[5]The FTT recorded at [9] that the mechanics of the contractual arrangements and much of the documentation were very similar to that which had been the subject of appeals to different compositions of the FTT in Thomson v HMRC [2018] UKFTT 396 (TC) (“Thomson”) and Sherrington v HMRC [2020] UKFTT 128 (TC) (“Sherrington”), and we refer to these arrangements as the “Montpelier Arrangements”.[6]The Montpelier Arrangements as entered into by the Appellants are described by the FTT in the Revised Decision. However, the concession by the Appellants meant that, in contrast to Thomson and Sherrington, the effectiveness of those arrangements was not part of the appeal. We set out below the description by the FTT in Thomson of the intended operation of the Montpelier Arrangements (footnotes excluded):
“65. We have drawn the following conclusions… : (1) Montpelier presented the Pendulum CFD and surrounding arrangements to its customers as a tax avoidance scheme that, provided it went into Phase Two, would deliver trading losses. Montpelier told users of the scheme that they would first need to “establish a financial trade” before they purchased the Pendulum CFD which was the instrument by which the tax loss would be delivered. (2) The tax avoidance result could be achieved only if a Pendulum CFD entered Phase Two (or subsequent Phases). In that case, it was important that a user of the tax avoidance arrangements should appear to pay a high Designated Issue Value for rights under the Pendulum CFD but that, shortly after entering Phase Two, a Pendulum CFD could be said to have a low value for accounting purposes. So, for example, in Mr Worsfold’s case, the Designated Issue Value of the Pendulum CFD was £300,000 but just 5 or 6 days after it moved into Phase Two, the Pendulum CFD was said to have a value of just £4,653 for accounting purposes. The difference between the high Designated Issue Value and the low accounting value would be the tax loss that would be generated. That was the “GAAP anomaly” to which Mr Gittins referred in his evidence… Indeed it is precisely the basis on which all appellants are claiming the loss that is in dispute. (3) To achieve the result set out at [(2)], the Index Target Levels applicable to Phases Two to Five (and the lengths of Phases Two to Five) in the Pendulum CFD needed to be set at values that meant that, when Pendulum came to make its repurchase offer described at [44] above, it could justifiably offer a low price. Pendulum was not purporting to “value” the Pendulum CFD. However, it was hoped that a low repurchase value offered by a counterparty who was, at least ostensibly, transacting at arm’s length, would justify a low value for accounting purposes. Without such pricing of the Pendulum CFD, the “GAAP anomaly” that Mr Gittins identified could not be achieved, and the desired tax loss could not be generated. (4) If the appellants had had to pay the full Designated Issue Value of the Pendulum CFDs out of their own pockets the steps set out above would have achieved little. For example, Mr Worsfold would have paid £300,000 for a CFD that, a few days later, was, at least according to Pendulum, worth only £4,653. He would have made an economic loss of £295,347 and even if he obtained a tax loss as a result, that would only compensate him for part of his economic loss. (5) For the arrangements to function as a tax avoidance scheme, the arrangements had to produce a tax loss without an economic loss. That was achieved by the Bayridge Loan which meant that the appellants were not themselves funding the entire Designated Issue Value of the Pendulum CFDs out of their own resources. Under the Bayridge Loan, Bayridge funded 95% of the Issue Value of the Pendulum CFD on highly advantageous terms. The Bayridge Loan therefore operated to “ramp up” the amount that the appellants could claim they invested in the Pendulum CFD even though they had not in any economically real sense invested the full Designated Issue Value. (6) Phase One of the Pendulum CFD had two functions. Its first was to act as a smokescreen by enabling the appellants to argue that the Pendulum CFD was not inevitably going to produce a loss. That is why the presentation … speaks in slightly apologetic terms about the possibility that there might be a profit at Phase One. It also explains why Mr Gittins attached significance … to the effect that the Pendulum CFD might not produce a tax loss. Since counterparties had to fund the Initial Margin at Phase One out of their own resources, the second function of Phase One was to ensure that Pendulum would receive the Initial Margin from counterparties which was in the nature of a “fee” payable to Pendulum for the tax avoidance scheme that was offered.”
[7]The claims to loss relief were based on s380 and s381 Income and Corporation Taxes Act 1988:(1) Claims to set-off their losses against current year income were made under s380 (the conditions for which are set out in s384). The combined effect of these provisions is that where a taxpayer incurs a loss in a trade in a particular year, that loss can be set off against other taxable income arising in the same year, or the immediately preceding year, but only where the requirements of s384 are met. Those requirements are that the trade was being carried on a commercial basis and with a view to the realisation of profits (which is deemed to be the case if the trade is carried on so as to afford a reasonable expectation of profit).(2) Claims for loss relief in earlier years were made under s381, which provides that relief is available for losses incurred in the first four tax years of a trade which, if the loss exceeds the income of that year, is applied to the preceding three years starting with the earliest. The loss relief under s381 is only available if the trade was carried on throughout the relevant year on a commercial basis and in such a way that profits could reasonably be expected to be realised in the period in which the loss occurred or shortly thereafter.[8]As set out in further detail below, the FTT found in the present appeals that there were no loans to the Appellants (at [88]). Mr Woolf and Ms Choudhury addressed the relevance of this in their submissions before us, but we nevertheless consider that the above summary is helpful in explaining how the Montpelier Arrangements were intended to operate.

Relevant legislation

[9]Section 29 Taxes Management Act 1970 (“TMA 1970”) sets out the basis on which HMRC may make a discovery assessment.[10]Section 29(1) provides that if an officer of the Board discovers that any income which ought to have been assessed to income tax has not been assessed, that an assessment to tax is or has become insufficient or that any relief which has been given is or has become excessive, the officer may make an assessment. Section 29(3) provides that if an officer makes a discovery and the taxpayer has made and delivered a return for that year of assessment, one of two conditions must be satisfied for HMRC to make a discovery assessment for that year. These alternative conditions are set out in s29(4) and 29(5).[11]The ordinary time limit for the issue of a discovery assessment is four years after the end of the tax year to which the assessment relates (s34 TMA 1970). In a case where the relevant loss of tax has been brought about deliberately by the taxpayer, that time limit is extended to 20 years. The extended time limit is set out in s36(1A) TMA 1970:
“An assessment on a person in a case involving a loss of income tax or capital gains tax - (a) brought about deliberately by the person, … may be made at any time not more than 20 years after the end of the year of assessment to which it relates (subject to any provision of the Taxes Acts allowing a longer period).”
[12]Section 118(7) TMA 1970 provides:
“In this Act references to a loss of tax or a situation brought about deliberately by a person include a loss of tax or a situation that arises as a result of a deliberate inaccuracy in a document given to Her Majesty's Revenue and Customs by or on behalf of that person.”

Decision of the FTT

[13]We have summarised below the findings and reasoning of the FTT in the Revised Decision.[14]The Appellants claimed losses in their tax returns for 2005/06, which had been prepared and submitted by Barnes Roffe LLP (“BR”):(1) AO claimed losses of £216,273. His tax return showed previous employment with Refco Overseas Limited (“Refco”), and AO confirmed he had traded in oil futures as an employee and then on his own account. He declared sales of contracts for differences (“CFDs”) of £87,850 and purchases of £283,944, of which £200,000 related to a contract between AO and Pendulum Investment Corporation (“Pendulum”). The margin on the other sales, was £3,906. At most, his only transactions in CFDs took place in the space of less than four months but in all probability in the space of less than one month.(2) RO claimed losses of £506,370. His tax return also showed previous employment with Refco, and he confirmed he had worked in oil futures. He declared CFD sales of £327,653 and purchases of £817,865, of which £500,000 related to a contract between RO and Pendulum. There was a discrepancy in the evidence related to these purchases and sales, but in oral evidence RO conceded that the small number of other sales had produced a loss of £155. His only transactions in CFDs took place in the space of less than one month.[15]The mechanics of the arrangements and much of the documentation were very similar to those in Thomson and Sherrington.[16]The contract between each Appellant and Pendulum is a “Pendulum Contract”. Describing the Montpelier Arrangements and the Pendulum Contract, the FTT’s findings included:(1) The arrangements constituted a marketed tax avoidance scheme ([40]).(2) That scheme sought to create an artificial trading loss for tax purposes which the scheme users would be able to set against their general income ([41]).(3) The Pendulum Contract was a simple bet that the FTSE would have moved up or down from its level at the date of the contract by a specified range of points at specified dates in the future ([43]).(4) The Pendulum Contract was documented by a Master Agreement, an Offer to Trade and an Acceptance Confirmation Note (although the Appellants were not able to produce a signed copy of the version of the Master Agreement which governed their Pendulum Contracts) ([44] and [46]).(5) The Pendulum Contract provided for a maximum of five Phases. The counterparty would pay the Initial Margin to Pendulum. If the Designated Index moved (up or down) by an amount greater than the designated swing movement over Phase One, Pendulum would be obliged to pay the “Trade Profit” to the counterparty. If the Pendulum Contract did not terminate, it would move into Phase Two, and Pendulum would serve a Notice of Obligation on the counterparty requiring the counterparty to pay the Margin Call Balance. Phase Two was to last for two years. The Pendulum Contract provided for further Phases ([47]).[17]The Appellants’ contacts with Montpelier and Pendulum were described as follows:(1) RO was looking for tax planning advice from Montpelier ([53]). In his witness statement AO stated his brother had introduced him to Montpelier, but in the hearing he said that Matthew Woolf and his brother had told him about Montpelier ([57]).(2) Montpelier Financial Services Limited (“Finance”) wrote to both Appellants on 28 February 2006 and each Appellant was told “You should consult your accountant or a tax specialist” as Finance were not giving advice on the tax treatment of CFDs. On 1 March 2006 the Appellants signed and returned a letter in which they each confirmed they understood that Finance was not advising them as to tax and that they must seek the advice of an accountant or tax specialist ([59 and 60]).(3) On 3 March 2006 both Appellants attended a meeting with Peter Crawford from Finance and Andrew Simpson from Montpelier Tax Consultants (City) Limited (“City”) ([62]).(4) The Appellants signed various documents at that meeting, including a Professional Services Agreement with Montpelier Tax Planning (Isle of Man) Limited (“MTP”) in which services provided by MTP were taxation advice in respect of the UK tax implications and consequences of the client commencing the trade of purchase and sale of derivative contracts ([65]).[18]The FTT recorded the evidence given by the Appellants about tax advice:
“63. The appellants’ recollection was that they were told that the scheme had been backed by Counsel and that it was legitimate. Anthony Outram said that no detail was given about that Opinion. In cross-examination he conceded that he was aware that Montpelier marketed tax planning and what he called “investments”
. Neither could remember much else that they had been told although Ross Outram conceded that he had been aware that Montpelier marketed tax planning and that they were tax advisers. He said in cross-examination that Montpelier had marketed both tax planning and “trade” at the meeting. He said that it was “very possible” that tax advice had been given at the meeting. He said that he had been told that the fees were built into the cost of the CFD and that a loan was available to fund the Margin Call Balance (“MCB”) if it became payable. In his witness statement Anthony Outram said that he recollected that “…the fees were wrapped-up in the price paid for the CFD contract.” In cross-examination he could not remember if it was included in the initial Margin.”[19]The FTT found that neither Appellant could, or should, have been in doubt that they were dealing with an offshore tax planning company that was not FSA registered. Since neither Appellant sought tax advice from BR, “we can only assume that the only tax advice, if any, that [the Appellants] received” was from Montpelier in the shape of MTP ([66]).[20]The FTT found that neither Appellant was able to produce “anything remotely like a complete set of signed documents” ([49]).[21]The FTT’s findings on the transactions as entered into by the Appellants included:(1) The Appellants received an Offer to Trade from Pendulum ([72] to [77]), and the first phase in the Pendulum Contract was for seven days, with a designated swing movement of up and down 140 points of the close of FTSE trading on the previous day.(2) The Appellants paid the Initial Margin of 7% of the issue value on 16 March 2006; AO paid £14,000 and RO paid £35,000 ([78]).(3) The FTSE did not exceed the designated swing movement at the end of Phase One, so both entered Phase Two. Under the terms of the Master Agreement the Appellants were required to pay the balance of the Designated Issue Value (ie the Margin Call Balance) under the Pendulum Contracts on being served with a Notice of Obligation to pay the Margin Call Balance from Pendulum. Service of those Notices should have triggered a draw-down of the loans ([80]).(4) Pendulum served Notices of Obligation on the Appellants on 27 March 2006; £186,000 for AO and £465,000 for RO ([81]).(5) The Appellants both stated that they signed loan agreements with Mandaconsult AG (“Mandaconsult”). The copies produced to HMRC were undated and signed only by the Appellants ([83]). The loan was not interest bearing but the lender was entitled to a fee (specified as equal to varying levels of profit, payable in the event that profits were made), repayment was due on the 50th anniversary of the agreement or earlier upon specified defaults such as being of unsound mind or bankruptcy but not in the event of death ([84]). In a letter dated 23 June 2014, Mandaconsult informed BR that it had never signed any loan agreements with AO or RO and had never made any payments to AO or RO ([87]).(6) The terms of the proposed loans were wholly uncommercial but, of course, in the event there were no loans ([88]).(7) At no time has either Appellant contacted Pendulum to check if the Margin Call Balance had been paid and if so by whom ([89]).(8) Neither Appellant had contacted Pendulum at the end of the subsequent phases to ascertain whether or not they had made a profit or if there was a different valuation for the contract. Although to be fair, Ross Outram did say that he knew he had not made a profit ([91]).[22]The FTT set out the relevant legislation, recorded the concession by Mr Woolf that the discovery assessments had been validly made and identified that the only question for the FTT was whether or not the Appellants’ behaviour had been deliberate.[23]Having considered some of the documentation from the material seized from an HMRC raid on the offices of Montpelier (including emails and a PowerPoint presentation and accompanying speaking notes dated 10 May 2006), and referred to the decision of the FTT in Sherrington, the FTT concluded that it is unlikely that either Appellant would have had any reason to believe that even if there had been a loan that it would be repayable ([112] to [116]).[24]In the Discussion, the FTT described both Appellants as “less than compelling witnesses”, acknowledging that the events were 16 years ago but both had access to the bundle and “seemed unaware of numerous pertinent matters” ([126]). The FTT’s reasoning then included:(1) The true objective of the Appellants in entering into the Pendulum Contracts was not to make a profit at the end of any Phase but to lose at the end of Phase One so as to create a loss in respect of which they did not bear the full economic cost but which reduced their liability to tax ([137]).(2) The PowerPoint presentation and speaking notes (which we refer to as the “Montpelier Materials”), whilst dated two months after the date of the Appellants’ Pendulum Contracts, accurately represent how Montpelier marketed the arrangements ([138]). The Pendulum Contracts were marketed to the Appellants as a tax avoidance scheme, and the Appellants knew that ([140]).(3) Montpelier’s marketing focused on a two-stage process, establishing a trade and then incurring losses. That part of the message does not appear to have been acted upon by the Appellants. The Appellants did not commence any trade before entering into the Pendulum Contract ([142] to [143]). Both Appellants only entered a very small number of other CFD contracts; and these were after the date of the Pendulum Contracts ([144] and [145]). The CFD contracts other than the Pendulum Contracts were mere window dressing to give the impression of trading ([147]).(4) The FTT did not accept that either Appellant was trading in CFDs and even if they were wrong they were not doing so on a commercial basis with a view to profit. Very few, if any, of the badges of trade are present. If you do not have a trade, as Montpelier made very clear, you cannot relieve any losses ([148] and [149]).(5) There was no loan constituted in any way. The Appellants suffered no economic loss ([157]).[25]The FTT then identified at [158] that the issue is whether the Appellants’ behaviour in submitting tax returns containing the losses was deliberate. We address the FTT’s reasoning thereafter in the context of Ground 3.

Procedural history

[26]The Original Decision was issued to the parties on 27 April 2021. On 7 June 2021, BR submitted the Appellants’ applications for permission to appeal (dated 28 May 2021 for AO and 27 May 2021 for RO) (the “PTA Application”), submitting that the FTT erred in law on three grounds, which broadly correspond to what we have described as Grounds 1, 3 and 4.[27]On 25 September 2023 the FTT issued its decision on the PTA Application (the “PTA Decision”). Permission was granted, but the PTA Decision also set out the procedural history and explained and apologised for the delays:
“2. On 23 August 2021, at my behest, the Tribunal wrote to the parties intimating that I had considered the Applications for Permission to Appeal. I had not had an opportunity to speak to Mr Bell but it was my intention to review the Decisions in terms of Rule 41 the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009 (as amended) ("the Rules”). … 5. I considered in accordance with Rule 40 of the Rules whether to review the Decision and decided that if there was a perceived lack of clarity, which would be an error of law, then I should undertake a review. 6. That being the case, in terms of Rule 41(3) of the Rules, I was required to give the parties the opportunity to make representations in relation to the proposed action. 7. I did so and the draft suggested changes were issued on 2 May 2023. 8. On 25 May 2023, as directed, both parties lodged with each other their submissions on the proposed revisions. … 11. I had said in the decision on the application for leave to appeal that I had had in mind the parameters for review set out in paragraph 45(8) in Vital Nut Co. Limited v HMRC [2017] UKUT 192 (TCC). Barnes Roffe argue that the proposed changes appear to do more than clarify the original reasoning. 12. Their submissions on the proposed revisions were detailed and I have carefully considered them. I had intended to address those but on reflection, I believe that there is a bigger problem. 13. One of the issues that Barnes Roffe raise, fairly, is that the delay in drafting the proposed revisions is a factor because the Tribunal’s memory would inevitably be dimmed by the passage of time. 14. Barnes Roffe suggest that the changes are not made and the original application for leave to appeal be granted unless the Tribunal was minded to allow the appeals in light of the arguments advanced in relation to the proposed revisions or if the Tribunal accepts that there are flaws in the original reasoning that are adverse to the appellants, that should be highlighted. 15. I do not accept that the appeals are to be allowed. Mr Bell and I decided that the appeals were dismissed and we stand by the Decision. 16. As I indicated in the review decision, when I considered in accordance with Rule 40 of the Rules whether to review the Decision I decided that if there was a perceived lack of clarity, which would be an error of law, then I should undertake a review. The review was not undertaken because I (or Mr Bell) accepted that the Decision was flawed and unreasonable as averred by the appellants. 17. Since I did find that there was a possible error in law in that there was a perceived lack of clarity, then the appellant’s original application for leave to appeal should be granted. Accordingly the proposed revisions will not be made.”
[28]Whilst the PTA Decision refers to a draft amended decision of 2 May 2023 (the “May 2023 Draft”) and the representations made thereon, the parties did not provide those documents to us (with Ms Choudhury’s explanation being that this was on the basis that the draft was not a decision that was the subject-matter of the appeal). We accept that this was an appropriate approach.[29]The FTT also sent a separate letter to the parties on 25 September 2023 attaching the Revised Decision. That letter read as follows:
“Please find enclosed decision notice which has been amended under Rule 37 (clerical errors and accidental slips or omissions) Please note that the release date of the decision to you remains 21 July 2021 and the time limit in which you may exercise your right of appeal is unchanged.”
[30]Notwithstanding the terms of that letter, the Revised Decision enclosed by the FTT had a “Release date” at the end of the decision of 25 September 2023 and a note that it was “Amended pursuant to Rule 41 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009 (as amended) on 21 September 2023.”

Grounds of appeal and Respondents’ notice

[31]In their Notice of Appeal from the FTT, the Appellants identified four grounds of appeal:(1) Ground 1 – The Revised Decision is inadequately reasoned, failing to provide a clear explanation for why the FTT rejected the arguments that the Appellants had a bona fide belief that they were entitled to make the claims for loss relief because they were told the arrangements were backed by counsel and also because it was never suggested by Montpelier or BR that the making of the claims was inappropriate.(2) Ground 2 – The changes that were made to the Original Decision in the Revised Decision are too extensive and significant to be justified under the slip rule in Rule 37 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009 (the “FTT Rules”) and are of a nature that should not be made under the review power in Rule 41 of the FTT Rules.(3) Ground 3 – The FTT used flawed objective reasoning when deciding that the conduct of the Appellants was deliberate.(4) Ground 4 – The basis on which the FTT found that the Appellants were guilty of deliberate conduct discloses an error of law and aspects of the reasoning appear unreasonable and suggest that there has been a failure to pay proper regard to material considerations.[32]The Appellants submitted that for the reasons relied upon in their grounds, the decision of the FTT should be quashed and remitted to a new Tribunal for redetermination.[33]In their Respondents’ Notice, HMRC submitted that the Revised Decision is properly reasoned, and should not be quashed.[34]HMRC also submitted that Ground 2 appeared to be a new ground of appeal, acknowledging that this ground could not have been included in the PTA Application but stating that the Appellants should apply for permission to amend their grounds of appeal. The Appellants made such an application on 5 December 2023. All parties made written submissions on that application in their written submissions ahead of the substantive hearing before us. We did not request further oral submissions at the hearing and informed the parties that we gave permission for the Appellants to rely on Ground 2.[35]We received detailed written submissions from counsel for all parties. We received further written submissions after the hearing. We are grateful to counsel for their helpful submissions, both in writing and orally, although we have not found it necessary to refer to each point that they raised.[36]We address Ground 2 first (as the Appellants submit that the changes to which they refer should not be taken into account when deciding whether to allow the appeal). We then address Ground 3 as that concerns the application of the test of whether the loss of tax arose as a result of a deliberate inaccuracy in the Appellants’ tax returns and logically should be considered before considering submissions as to the adequacy of reasons given for that decision.[37]The Appellants submitted that the FTT erred in law as the changes that were made in the Revised Decision are too extensive and significant to be justified under the slip rule in Rule 37 and are of a nature that should not be made under the review power in Rule 41, the consequence of which is that we should treat the relevant changes as not having been made when deciding whether to allow the appeal.[38]We set out first the relevant legislation and rules before summarising the parties’ submissions and reaching our conclusions.[39]Section 9(1) Tribunals, Courts and Enforcement Act 2007 (“TCEA 2007”) provides that the FTT may review a decision made by it on a matter in a case. Section 9(2) provides that this power is exercisable of its own initiative or on an application by a person who has a right of appeal in respect of the decision. By s9(3), Tribunal Procedure Rules (which include for this purpose the FTT Rules) may provide that the FTT may not review certain decisions, may provide that review is exercisable only of the FTT’s own initiative, that an application may be made only on specified grounds or that the power to review of its own initiative is exercisable only on specified grounds.[40]Section 9(4) provides:
“Where the First-tier Tribunal has under subsection (1) reviewed a decision, the First-tier Tribunal may in the light of the review do any of the following – (a) correct accidental errors in the decision or in a record of the decision; (b) amend reasons given for the decision; (c) set the decision aside.”
[41]Section 9(5) provides that where the FTT sets a decision aside, the FTT must either(a) re-decide the matter concerned, or(b) refer that matter to the Upper Tribunal.[42]Paragraph 15(1) of Schedule 5 TCEA 2007 provides that rules may make provision for the correction of accidental errors in a decision or record of a decision. Paragraph 15(3) provides that paragraph 15(1) is without prejudice to any power to correct errors or set aside decisions that is exercisable apart from rules made by virtue of this sub-paragraph.[43]The FTT Rules then set out the procedural rules and requirements applicable to the FTT (Tax Chamber).[44]Rule 37 provides:
“37. Clerical mistakes and accidental slips or omissions The Tribunal may at any time correct any clerical mistake or other accidental slip or omission in a decision, direction or any document produced by it, by - (a) sending notification of the amended decision or direction, or a copy of the amended document, to all parties; and (b) making any necessary amendment to any information published in relation to the decision, direction or document.”
[45]Rule 40 (Tribunal’s consideration of application for permission to appeal) provides at Rule 40(1) that on receiving an application for permission to appeal the FTT must first consider, taking into account the overriding objective in Rule 2, whether to review the decision in accordance with Rule 41. Rule 41 provides:
“41. Review of a decision (1) The Tribunal may only undertake a review of a decision – (a) pursuant to rule 40(1) (review on an application for permission to appeal); and (b) if it is satisfied that there was an error of law in the decision. (2) The Tribunal must notify the parties in writing of the outcome of any review, unless the Tribunal decides to take no action following the review. (3) The Tribunal may not take any action in relation to a decision following a review without first giving every party an opportunity to make representations in relation to the proposed action.”
[46]Rule 36 (Interpretation) provides that “review” means the review of a decision by the FTT under s9.

Appellants’ submissions

[47]Mr Woolf submitted that it was unclear whether the FTT had purported to amend the Original Decision under Rule 37 or Rule 41, or even if they had intended the changes to be made at all, drawing attention to:(1) the letter of 25 September 2023 enclosing the Revised Decision referred to it as having been amended under Rule 37;(2) the PTA Decision of the same date says at [17] “the proposed revisions will not be made”, whereas many were, submitting that the inclusion of the changes may itself have been an accidental slip by the FTT; and(3) the end of the Revised Decision refers to Rule 41, and the release date of the Revised Decision had been updated.[48]The Appellants’ submissions related to three of the changes which had been made by the Revised Decision (and Mr Woolf confirmed that these were the only changes which were relied upon as errors of law). Mr Woolf confirmed that all three of these changes had been included in the May 2023 Draft which had been sent to the parties, and the Appellants had made representations on them. Those changes were (with additions being underlined and deletions in square brackets): (1) [116] – The FTT had described parts of the Montpelier Materials, the need for a trade and that a loan is advanced:
“116. In our view, on the balance of probability, given that evidence and Judge Sinfield’s analysis, it is unlikely that either [neither] appellant would have had any reason to believe that even if there had been a loan that it would be repayable.” (2) [143] – At [142] the FTT had set out that the Montpelier Materials focused on a two-stage process, establishing a trade and then incurring losses: “143. Unfortunately for the appellants that part of the message does not appear to have been acted upon [understood] by them. As Judge Richards makes very clear (in relation to an email but the same point is made in the Montpelier slides and speaking notes) at paragraph 51 of Thomson: “This email indicates that Montpelier intended the Pendulum arrangements to function as a device to deliver a trading loss to a user of the scheme but that, before such a loss could be delivered, the user first needed to commence a trade of dealing in derivatives.”
The appellants did not commence any trade first.” (3) [149] – The FTT had referred to the timing of entry into the Pendulum Contracts, made findings as to whether the Appellants had entered into other transactions, and concluded at [148] that they did not accept that either Appellant was trading in CFDs. This was then followed by:
“149. If you do not have a trade, as Montpelier made very clear, you cannot relieve any losses. [In any event, the losses could only be created if there was a loan.]”
[49]Mr Woolf submitted that the changes which had been made to these three paragraphs were not corrections capable of falling within Rule 37. He did accept in his oral submissions that the changes to [116] were capable of being made under Rule 41 if we were to conclude that there had been such a review. However, Mr Woolf submitted that the changes which had been made to [143] and [149] were not appropriate even following a review under Rule 41.[50]Mr Woolf relied on the decision of the Upper Tribunal (Administrative Appeals Chamber) in JS v Secretary of State for Work and Pensions [2013] UKUT 100 (AAC) (“JS”) (with paragraphs of that decision being referred to as JS[x], and a similar approach being adopted for the reference to paragraphs of other decisions to which we subsequently refer). Mr Woolf relied on the Upper Tribunal’s statements that “the power of review must not be used in a way that subverts the appeal process and bypasses the proper function of the Upper Tribunal” (at JS[28]) and that it must not be used “to correct defective reasoning or to provide commentary on the grounds of appeal” (at JS[36]). The Upper Tribunal in JS decided that the safeguards within the power to review a decision were both procedural and substantive, stating at JS[45] that the substantive safeguard is “to interpret ‘amend’” in a way that minimises the risk and apparent dangers inherent in the process and to confine it to cases that properly fulfil the purposes of the provision. So it is limited to cases in which it would be proper to amend the reasons rather than set aside the decision… It covers cases where there is some objective guarantee that the reason have not drifted into justification.”[51]Mr Woolf also relied on Vital Nut Co. Ltd v HMRC [2017] UKUT 192 (TCC) (“Vital Nut”) where the Upper Tribunal adopted the review of relevant authorities in JS and said at Vital Nut[45(9)] that “whilst it is perfectly permissible for the FTT to use the review process to clarify what has already been decided, the FTT should refrain from seeking to justify its decision on other, even better, grounds or from seeking to defend its decision in advance from an attack that is anticipated in an appeal”.[52]Mr Woolf submitted that the changes which had been made by the FTT to [143] and [149] were not appropriate on a review of a decision, emphasising in particular:(1) the change in [143] is not appropriate to be made two years after a hearing, and was made in consequence of contentions in the PTA Application that the FTT was using objective reasoning; and(2) the change in [149] was trying to change the FTT’s reasoning retrospectively, seeking to defeat a challenge based on an error of law.[53]Mr Woolf’s submission was that these two changes should not have been made and that when reaching our decision on the Appellants’ other grounds of appeal, we should reach that decision by reference to the Revised Decision without these two changes having been made.

HMRC’s submissions

[54]Ms Choudhury submitted that all of the changes were capable of being made by the FTT under Rule 37 as being in the nature of accidental slips or omissions or clerical mistakes, drawing attention to:(1) [116] – similar reasoning was already present in [156];(2) [143] – this cannot be regarded as anything more than a correction. There was evidence of the Appellants’ actions (or lack of action) in relation to establishing a trade before entering into the Pendulum Contract as noted in [143], and expanded on by reference to each Appellant in [144] and [145]. There was no direct evidence of their knowledge of the requirement to establish a trade given their inability to recall what, if any, advice was given to them by Montpelier; and(3) [149] – the FTT may have considered that the inclusion of the final sentence was confusing as it would require further explanation and therefore decided simply to omit it. The omission falls far short of an attempt to go beyond clarification. Moreover, a similar statement can still be found in the Revised Decision at [171].[55]HMRC acknowledged that it was not entirely free from doubt that the Original Decision was revised under Rule 37, referring to the facts that at the end of the Revised Decision it is said to have been amended under Rule 41 and the release date has been changed to 25 September 2023.[56]In the alternative, Ms Choudhury submitted that the changes were amendments the FTT was entitled to make following a review under Rule 41 and are in accordance with the guidance given in JS and Vital Nut as being clarificatory in nature regarding what had already been decided.

Discussion and conclusions

[57]We have set out above the powers of the FTT under Rule 37 and Rule 41 of the FTT Rules. The powers of the FTT under these rules are different procedurally and substantively:(1) There is no restriction on when the FTT can exercise the power under Rule 37. The FTT can amend a decision under Rule 37 of its own initiative or following application or notice by the parties. Rule 37 is not limited to cases where there has been an application for permission to appeal and the FTT is satisfied that the decision contains an error of law. There is no procedural requirement as to giving notice to the parties before making any changes. The restriction on the FTT’s power to amend a decision under Rule 37 is one of substance – the power is to “correct any clerical mistake or other accidental slip or omission”.(2) A review under Rule 41 may only be undertaken following an application for permission to appeal and if the FTT is satisfied that there was an error of law in the decision. Where the FTT undertakes a review of a decision, it may not take any action in relation to that decision without first giving every party an opportunity to make representations in relation to the proposed action. Where Rule 41 applies, the FTT has broad powers as set out in s9(4) TCEA 2007.[58]Here, following initial consideration of the PTA Application, the FTT informed the parties that it would be reviewing the Original Decision. The changes which were proposed to be made (which included those subsequently made to [116], [143] and [149] as well as others) were sent to the parties on 2 May 2023 and the parties provided representations thereon. The Revised Decision was subsequently released. The issues which have arisen are:(1) whether the Revised Decision was the result of a review of the Original Decision under Rule 41 or correction of accidental slips under Rule 37; and(2) whether the changes made were ones that the FTT was entitled to make under the relevant process.[59]At the hearing, the panel raised with the parties whether there was a further issue, namely that some of the changes which had been made in the Revised Decision had not been provided to the parties in draft in advance of the Revised Decision being released. Ms Choudhury and Mr Woolf both confirmed that the Revised Decision did include some such changes, and Ms Choudhury provided us with a list of those changes at the beginning of the second day of the hearing.[60]There were 17 changes which had been made by the Revised Decision which had not been proposed in the May 2023 Draft and on which the parties had not therefore been given the opportunity to make representations (the “17 changes”). The vast majority of the 17 changes were, on any view, corrections – correcting a typo in a date, adding punctuation, changing singular to plural. One of the 17 changes was, however, arguably more than a correction. In the Revised Decision, [153] reads as follows:
“His argument was that that could be relied upon to evidence the fact that there must have been some sort of verbal loan. There is absolutely no evidence to that effect. As can be seen from paragraphs 83 to 87 above, both appellants signed loan agreements copies of which were produced to HMRC but it was only when HMRC instigated enquiries in 2014 that it transpired that Mandaconsult AG had never signed the agreements. The witness statements of both appellants refer to the loans and both said that they assumed that Mandaconsult AG had executed the loan agreements. There is absolutely no reference to any verbal loan; indeed both state that they proceeded on the basis that the loan agreements were key to the arrangements and existed. In oral evidence Anthony Outram said that he would not have entered into the Pendulum Contract without the loan and Ross Outram said that it was the existence of the loan that made it attractive and the existence of the loan was the “deciding factor”.”
[61]The underlined text above shows the changes which had been made to the Original Decision (by way of addition). Most of this had been proposed in the May 2023 Draft. However, the closing phrase “the existence of the loan was the “deciding factor”” had not been included in the May 2023 Draft.[62]Mr Woolf and Ms Choudhury both confirmed at the hearing that their position was that the 17 changes (including the change to [153]) were the correction of clerical mistakes or accidental slips or omissions which the FTT had power to make under Rule 37.[63]Having considered the PTA Decision, the FTT’s letter of 25 September 2023 and the Revised Decision, we have concluded that the Original Decision was reviewed by the FTT under Rule 41 and not Rule 37. We recognise that the FTT’s communications with the parties created unfortunate and unnecessary confusion on this point:(1) the contents of the letter of 25 September 2023 are inexplicable, including the reference not only to Rule 37 but to the release date of the decision remaining as 21 July 2021 – the Original Decision was released on 27 April 2021, and the release date of the Revised Decision, which was attached to the letter, had been changed on the final page to 25 September 2023; and(2) the PTA Decision, having recounted the background and in particular the submissions received from BR on behalf of the Appellants, stated at [17] thereof that the Appellants’ original application for leave to appeal should be granted and ended with “Accordingly the proposed revisions will not be made.” Yet the version of the Revised Decision which was released to the parties did include some, but not all, of the revisions which had been proposed in the May 2023 Draft.[64]We describe this confusion as unnecessary as the FTT had clearly informed the parties that it was proposing to review the Original Decision and subsequently sent them in draft the changes it proposed to make, and on which they made representations. The process being followed was that in Rule 41. Irrespective of confusion caused by other communications, we answer the question whether the Revised Decision was amended under Rule 37 or Rule 41 by reference to the Revised Decision itself – that document was released to the parties by the FTT and is the decision notice containing the FTT’s written findings and reasons for the decision. The Revised Decision states expressly that it was amended pursuant to Rule 41 of the FTT Rules on 21 September 2023, and has an amended release date of 25 September 2023. We are satisfied that the Revised Decision was, as it says, amended following a review under Rule 41.[65]This conclusion does raise a question as to the 17 changes which were made and were not included in the May 2023 Draft. We are satisfied that 16 of these could have been made by the FTT under Rule 37, but no such exercise was undertaken. They could also have been made following a review under Rule 41 (being corrections of accidental errors within s9(4)(a)) but the FTT did not comply with the requirements of Rule 41(3) in respect of such changes. The additional changes to [153] seem to us to amount to more than the correction of accidental errors; it is an additional finding that not only was the loan attractive to RO but also it was the “deciding factor”. This change could similarly have been made following a review under Rule 41 (as amending the reasons given for the decision), but the FTT did not comply with the requirements of Rule 41(3). These are procedural errors of law by the FTT in the approach it adopted to the review of its decision. However, we are mindful of the position taken by the parties in respect of these changes and that 16 of the 17 changes could have been made under Rule 37 and that this would not have required the FTT to give the parties the opportunity to make representations in relation to them. We conclude that these procedural errors are not material errors of law.[66]The parties’ positions were different in respect of the three changes set out at [48] above. We need to decide whether these were changes which the FTT was entitled to make when exercising its power under s9(4) in accordance with Rule 41.[67]Mr Woolf and Ms Choudhury confirmed that the changes to [116], [143] and [149] were included in the May 2023 Draft. Accordingly, the FTT had complied with the requirements of Rule 41(3), namely that the FTT may not take any action in relation to a decision following a review without first giving every party an opportunity to make representations in relation to the proposed action.[68]Mr Woolf accepted that the FTT had power to make the changes which were made to [116] on a review of its decision under Rule 41 (whereas he had taken the position that the FTT could not make these changes under Rule 37, if we were to conclude that that was the basis relied upon for the changes). We do not consider those changes any further.[69]Mr Woolf’s submissions focused on the changes which were made to [143] and [149], which we have set out above. The parties disagreed as to the significance of these changes. Ms Choudhury submitted that these changes could, in any event, have been made under Rule 37 or following a review relying on s9(4)(a). We do not accept that submission. They are both substantive changes to the FTT’s reasoning:(1) The change of language in [143] from the message (from Montpelier) about trading “does not appear to have been understood by them” to “does not appear to have been acted upon by them” is significant. It is part of the FTT’s findings as to what the Appellants did or did not do, and is relied upon by the FTT (albeit without express cross-reference) in its reasons for concluding that the conduct was deliberate.(2) The deletion in [149] of “In any event, the losses could only be created if there was a loan” initially appears very significant, particularly in the light of the Appellants’ submissions on Grounds 1 and 4 which included that the FTT’s findings in relation to the loans revealed, in the Appellants’ submission, that the FTT had failed to understand the arrangements and the basis on which the losses were expected to be claimed. However, we recognise that at [171] of the Revised Decision the FTT states “Crucially the appellants do not and never did have any liability to repay a purported loan. Therefore they did not incur expenditure and they incurred no losses that were capable of being relieved.” This makes substantially the same point as that which had been made by the deleted language and we consider that this reduces the significance of the deletion in [149].[70]As the changes to both of these paragraphs do amend the reasoning of the FTT (albeit that we regard the changes to [143] as more significant than the deletion in [149]), we have considered whether there is any restriction as to the type of changes which can be made by the FTT following a review under Rule 41.[71]“Review” is defined by Rule 36 as meaning the review of a decision by the FTT under s9 TCEA 2007, and s9(4) provides that in the light of a review the FTT may (a) correct accidental errors in the decision or in a record of the decision, (b) amend reasons given for the decision, or (c) set the decision aside.[72]At the outset we record that we consider that a straightforward, natural reading of s9(4) and Rule 41 does not, expressly or impliedly, restrict the type of changes that may be made on a review. Rule 41 contains procedural protections, in that the FTT may only review a decision following an application for permission to appeal and if it is satisfied that there was an error of law, and must give the parties the opportunity to make representations in relation to the proposed actions. However, there seems to us to be no restriction set out in these provisions as to the substance of the changes – not only is there no limiting language in the meaning of “review”, but also we consider it counterintuitive that the FTT would be permitted to set aside its decision and re-make it (ie change its mind completely), yet not be permitted to amend its reasons for the decision, not only by explaining further the initial reasoning but also potentially by including additional reasons.[73]We are not persuaded that we are bound by the authorities to reach a different conclusion. We have carefully considered the decision of the Upper Tribunal in JS, which was addressing a factual situation in which there had been procedural errors in the appeal process, and the authorities to which it referred, which were addressing the powers of the courts where there is no equivalent to the power to review a decision.[74]In JS, a claimant’s entitlement to disability living allowance was removed. Her appeal to the FTT was dismissed. That appeal was heard by a fee-paid judge who dismissed the appeal and (following an application) subsequently provided written reasons and then, following an application for permission to appeal, provided additional reasons. The application for permission to appeal was then referred to a salaried district tribunal judge with the amended statement of reasons, who made a decision on that application.[75]The claimant applied for permission to appeal on the ground that the amended reasons were not validly made. The Upper Tribunal identified procedural irregularities:(1) The appeal was heard by a single judge whereas the relevant Practice Statement of the Senior President of Tribunals on the Composition of Tribunals provides that a disability living allowance appeal must be decided by a panel consisting of a judge and two members, and one member must be a medical practitioner and the other must have a disability qualification.(2) The amended reasons were provided by the fee-paid judge who heard the appeal. That Practice Statement provides that the exercise of the power of review under s9 must be carried out by a salaried tribunal judge. The Upper Tribunal considered that amended reasons could not properly be written by a salaried judge who was not a member of the original panel, with the result that power is given to a salaried judge who may not be in a position to implement it. The Upper Tribunal considered that the solution to this is that if the salaried judge considers it may be appropriate to amend the reasons, the proper course is to invite the hearing judge to prepare such reasons as are consistent with the tribunal’s reasoning at the time of its decision, and the salaried judge must then decide whether they satisfy the criterion of being amended reasons.(3) The powers under the rules of procedure must be exercised fairly and justly; this means that they must be exercised transparently. Here, the district tribunal judge did not give the parties an opportunity to make representations – this was said to be inappropriate and unfortunate (at JS[9]). The Upper Tribunal did not comment on the hearing judge’s failure to provide the parties with the opportunity to comment on the amended reasons.[76]It was against this background that the Upper Tribunal then considered the purpose of the review power, stating at JS[28] that the self-evident purpose is to allow the FTT to avoid the need for an appeal to the Upper Tribunal in the case of clear errors, and that this is to the benefit of the parties and the Upper Tribunal. We agree. The Upper Tribunal then set out at JS[29] that there is an issue of balance between inadequate reasons that can appropriately be amended and those for which the only proper course of action is to set aside the decision. In its decision, the Upper Tribunal set out at JS[40] that the purpose of amended reasons is the same as the purpose of the original reasons – to show how the tribunal made its decision. They must be the reasons that led the tribunal to decide as it did, not a later attempt to rationalise the decision, and can only properly be written by the presiding judge or, exceptionally, another member of the panel.[77]The Upper Tribunal drew parallels with some of the authorities addressing the exercise by the courts of their discretionary powers. We consider those in turn:(1) At JS[28] the Upper Tribunal referred to the decision of the Supreme Court in In the matter of L and B (Children) [2013] UKSC 8 (“L and B”) at L and B[17] and [19], stating that “the Supreme Court has recently emphasised that the integrity of the appeal process should not be subverted by diverting matters to an alternative process”. In L and B, Baroness Hale, in a judgment with which the other members of the court agreed, identified the issue in that case as being whether and in what circumstances a judge who has announced her decision is entitled to change her mind. In an oral judgement on 15 December 2011, the “preliminary outline judgment approved by the court” had concluded that the father was the perpetrator of non-accidental injuries to a child. In a written “perfected judgment” on 15 February 2012 she expanded upon the earlier judgment but reached a different conclusion, stating “I am unable to find to the requisite standard which of the parents it was…It could have been either of them who injured [child] and that is my finding”. Baroness Hale said a judge is entitled to reverse his decision at any time before his order is drawn up and perfected. There is no jurisdiction to change one’s mind thereafter unless the court has an express power to vary its own previous order. The proper route of challenge is by appeal. The judge did have power to change her mind, and the question was whether she should have exercised it. Baroness Hale set out at L and B[27] that the judge’s overriding objective must be to deal with the case justly, and a relevant factor must be whether any party has acted upon the decision to his detriment. Referring to examples of cases where it might be just to revisit, these were said to be only examples, and a carefully considered change of mind can be sufficient; every case is going to depend upon its particular circumstances. The court ordered that the father’s appeal be allowed; the welfare hearing should proceed on the basis of the findings in the judgement of 15 February 2012. However, L and B is a case in which there was no separate review process. Baroness Hale expressly allowed for cases where the court (or tribunal) has an express power to vary its order. The review procedure under Rule 41 is a part of such a process, and ordinarily the FTT is not subverting the appeals process by exercising this power (as it is designed to avoid unnecessary appeals). In any event, the FTT has to exercise the power to undertake a review of a decision in the light of the overriding objective and should be mindful of circumstances in which exercise of the power might disrupt the progress of an appeal.(2) At JS[34] the Upper Tribunal stated that the common law or inherent power and the decisions on its exercise form part of the background against which, and by analogy provide guidance on how, s9 TCEA 2007 is to be interpreted and applied. Those decisions “make clear that the power to give additional reasons is only to be exercised exceptionally and with safeguards”. They referred to Mummery LJ’s speech in Woodhouse School v Webster [2009] ICR 818 (“Woodhouse”) (in particular Woodhouse[26] to [28] of that speech). The Employment Tribunal (“ET”) had been divided as to why an employee had resigned. One lay member thought there was no constructive dismissal; the majority took a different view of the evidence. The Employment Appeal Tribunal (“EAT”) made an order that the ET be asked to answer certain questions in relation to its written reasons, and was asked to give its answers by reference to its notes of evidence. Mummery LJ had emphasised at Woodhouse[25] the importance of taking care to observe the limits of the exceptional Burns/Barke procedure, which is available where the EAT considers that there is possibly an inadequacy in the ET’s reasons for its decision. The EAT may, before it finally decides the appeal, refer specific questions to the ET at the preliminary hearing of the appeal, requesting it to clarify or supplement its reasons where no reasons were given or where the reasons were inadequate. He said it is not desirable for the ET to do more than answer the request – it should not, eg, advance arguments in defence of its decision. Mummery LJ’s concerns were twofold: the EAT should identify correctly the point on which the ET’s reasons may be inadequate; and having been asked questions, the ET chairman went further than the questions required and further than was justified. We note that in JS the Upper Tribunal said at JS[35] this reasoning is equally applicable to the review power under s9. However, the reasoning in Woodhouse was based on common law principles, whereas the power of review is granted by statute, and is embodied in the relevant Tribunal rules.(3) At JS[36] the Upper Tribunal recorded that one of the limits on the power to supplement reasons is that it must not be used to correct defective reasoning or to provide a commentary on the grounds of appeal, and considered this is equally applicable to the review power. They referred to the decision of the Court of Appeal in Brewer v Mann [2012] EWCA Civ 246 (“Brewer”). In Brewer, the Court of Appeal said at Brewer[31] “where a judge has received no request from the parties to reconsider his judgment or add to his reasons, and has not demonstrated the need in conscience to revisit his judgment, but on the contrary has received grounds of appeal and an application for permission to appeal on the basis of the alleged inadequacies of his judgment, then it would be most unwise for him to rewrite his judgment (other than purely editorially) and it would take the most extraordinary reasons, if any, to justify such a course on his part”. This guidance, although expressed as provisional, is given in forceful language. But the point remains that this was given against the background that there was no statutory power to review a decision.(4) At JS[27] and [41] the Upper Tribunal referred to Mummery LJ’s speech in Space Airconditioning plc v Guy [2012] EWCA Civ 1664. Mummery LJ stated that that the judgment should be an accurate record of the judge’s findings and of the reasons for the decision; and before a judge corrected a judgment, the judge should give both sides an opportunity to make submissions on whether there is a valid objection. The starting-point of Mummery LJ (in a speech with which the other members of the court agreed) was that CPR Part 52.11(3) sets out that the appeal court will allow an appeal where the decision of the lower court was either (a) wrong or (b) unjust because of a serious procedural error or other irregularity in the proceedings in the lower court. In that case, judgment had been given and an order made and the judgment contained what was described as a plainly wrong finding of fact. After the judgment had been handed down the judge had acknowledged that the finding was wrong but declined to correct it as a typographical slip and refused permission to appeal. The appellant’s submissions included that the erroneous finding meant it had not had a fair trial – the wrong finding gave rise to a real doubt as to whether the judge appreciated the importance of the confidentiality factor central to its claim. The judge misunderstood much of the crucial evidence going to the heart of its case, and the error had a knock-on effect as it coloured the judge’s assessment of the evidence on other issues. Mummery LJ did not accept the respondent’s suggestion that the error was typographical only. One reason given was that the judge herself had acknowledged the error but did not say it was typographical. He allowed the appeal on the ground that the decision appealed was either wrong or it was unjust as a result of an irregularity. The retention of the erroneous finding was an “irregularity in the proceedings” which makes the decision an unjust one. We note that Mummery LJ had said that before the correction is made the judge should obviously give both sides an opportunity to make submission on whether there is a valid objection to a proposed amendment. That requirement is embedded in Rule 41(3).[78]In JS a theme in the reasoning of the Upper Tribunal (eg at JS[35], [36], [40] and [41]) was that the power of review carries a “risk and so an apparent danger” of seeking to defend the Tribunal’s decision, and may drift into responding to a representative’s criticisms and that “amend” must be interpreted in a way that minimises the risks; this led the Upper Tribunal to conclude that it is “limited to cases in which it would be proper to amend the reasons rather than set aside the decision” (at JS[45]). However, we consider that if a Tribunal does, upon review, significantly change the reasoning in the decision, the protections in place are procedural (the requirement in Rule 41(3) that a Tribunal not take any action without first giving every party an opportunity to make representations) and substantive (as a party may apply for permission to appeal within 56 days of the release of notification of the amended decision following a review). There is no need to go further and limit the nature of the amendments which may be made.[79]We therefore respectfully disagree with the Upper Tribunal’s decision in JS that there is a category of “impermissible amendments” (at JS[50]) that cannot be made following a review. We would consider that, provided the FTT has received an application for permission to appeal, identified an error of law, followed the procedure required by Rule 41 and is acting in accordance with the overriding objective, the FTT may amend the decision however is required to record its reasons, and this may include not just clarifying any ambiguity, but also setting out reasons that had not previously been recorded.[80]Mr Woolf also relied on the decision of the Upper Tribunal in Vital Nut, in which the Upper Tribunal had considered whether the FTT was entitled to make the revisions to the decision which it made under Rule 41. In that case, they saw “nothing objectionable” in the review that was carried out. In the course of its decision, the Upper Tribunal had referred at Vital Nut[45] to the statutory basis in s9 TCEA 2007, and said that Rule 41 (once the “gateway” conditions are met) does not constrain the FTT in terms of the sort of review it undertakes, but stressed at Vital Nut[45(4)] and [45(7)] that this does not mean that the FTT is entirely unfettered. They then set out that they adopted the review of relevant authorities in JS and (at Vital Nut[45(9)]) that “whilst it is perfectly permissible for the FTT to use the review process to clarify what has already been decided, the FTT should refrain from seeking to justify its decision on other, even better, grounds or from seeking to defend its decision in advance from an attack that is anticipated in an appeal”.[81]We acknowledge that, although a decision of the Upper Tribunal is not binding on a later Upper Tribunal (see Raftopoulou v HMRC [2018] STC 988 at [24]), as a tribunal of coordinate jurisdiction the later tribunal will normally follow the decision of the earlier one unless it is convinced that the earlier decision is wrong (see Gilchrist v HMRC [2014] STC 1713 at [94]). On this issue, we are satisfied that the decisions in JS and Vital Nut are wrong, and so we will not follow them.[82]We have therefore concluded that there is no restriction on the substance of the amendments which may be made following a review in accordance with Rule 41, and the changes made by the FTT to [143] and [149] were properly made in accordance with the procedure set out by Rule 41.[83]Ground 2 is dismissed. It follows that the remaining grounds on which the Appellants appeal are assessed by reference to the Revised Decision.

Ground 3 – ftt used objective reasoning when concluding inaccuracies were deliberate

[84]The Appellants submitted that the FTT erred in law by using objective reasoning when concluding that the inaccuracy in the Appellants’ returns was deliberate.[85]We have summarised the Revised Decision of the FTT in some detail above. Having expressed agreement at [163] with the meaning of “deliberate inaccuracy” set out in Auxilium Project Management v HMRC [2016] UKFTT 249 (TC) (“Auxilium”), the Revised Decision then sets out its reasoning and conclusion (which we set out in full here):
“164. We have already established that we find that:- (a) The appellants had knowingly taken part in a tax avoidance scheme; (b) They knew that that scheme involved the need to have a trade and thereafter create losses, whereas the reality was that there was no trade; (c) They knew that the losses would be created by a purported loan on uncommercial terms and that that loan would not be repayable on the elapse of 50 years; (d) They knew that their other CFD contracts were not even modest but were in fact minimal in size and occurred within a very short period and after the meeting on 3 March 2006; (e) The only information given to BR was the extent of the alleged trade; (f) They did not seek tax advice from BR at any stage in relation to the Pendulum Contract. 165. Therefore we find that each of the appellants knew at the time of filing their respective SATRs that they were not carrying on a trade which entitled them to make a claim for loss relief. 166. We were wholly unconvinced by the appellants’ argument that they were entitled to rely on Montpelier for everything and that they needed to check almost nothing because Montpelier was FSA registered and had a Counsel’s Opinion which they had not seen. Had they asked to see it, as they were entitled to do, as Judge Richards says at paragraph 48 of Thomson (which is quoted with approval by Judge Sinfield at para 56 of Sherrington) that Opinion made it clear that Counsel was not endorsing the scheme. We have underlined the key words. “He also sought tax advice from UK tax counsel, Mr Shipwright, on the tax consequences for investors. Mr Shipwright’s advice included an analysis of the general law, and HMRC practice on what amounted to the carrying on of a trade on a commercial basis with a view to profit. However, that analysis was generic: Mr Shipwright was not purporting to advise as to whether any particular taxpayer met this requirement and he noted that the question was ultimately a question of fact that depended on what a taxpayer actually did.” 167. As can be seen their primary dealings were in fact with Andrew Simpson of City and their Professional Services Agreements were with MTP, neither of which were FSA registered. The documentation made it abundantly clear that there was no FSA protection and nor was there protection in the Seychelles. They had both requested a sophisticated investors certificate, which was provided by Montpelier and which took them out of FSA protection. 168. Their glib assertions that they were not men with an eye for detail and that they had therefore not read the documents in detail, do not assist them. By any standard, purported indebtedness of £185,000 and £465,000 is not insignificant for individuals of relatively limited means. 169. The fact that neither of them knew whether they had accepted Pendulum’s offer to re-purchase the CFDs and their failure to check what had happened in subsequent years, points to a disregard of anything other than the losses which they had set out to achieve. 170. This is a self-assessment system. The taxpayer is ultimately responsible for submitting an accurate SATR. 171. This was not a question of the appellants turning a blind eye. They did not ask questions or read documents because they knew precisely what they were doing. They were trying to create a significant loss and thereafter make substantial claims for repayment of tax. Crucially the appellants do not and never did have any liability to repay a purported loan. Therefore they did not incur expenditure and they incurred no losses that were capable of being relieved. Furthermore they had not been carrying on a trade on a commercial basis with a view to making a profit. 172. In submitting their SATRs we find that they acted deliberately with a view to claiming non-existent losses. Their intention was to mislead HMRC and obtain repayments of tax. 173. For all these reasons the appeals are dismissed.”

Appellants’ submissions

[86]Mr Woolf submitted that whilst the FTT had correctly directed itself as to the relevant test, namely that in Auxilium, it had then failed to apply that test correctly. His submissions included that the reasoning at [164] to [171] of the Revised Decision explained why the Montpelier Arrangements were not effective, but does not explain why the FTT considered that the Appellants knew that the scheme was not effective (and that therefore they knew their returns were inaccurate), and that the generalised explanations set out therein would apply to any participant in the scheme.[87]In particular, Mr Woolf submitted that:(1) [164] simply recites the underlying facts which are not a matter of dispute, and, at [164(b)] the FTT imputed to the Appellants its own conclusion (which it had reached at [148]) that there was no trade;(2) the finding at [165] that the Appellants knew they were not carrying on a trade which entitled them to make a claim for loss relief does not follow from the findings in [164]. The FTT had referred to the Montpelier Materials but then glossed over significant comments made therein about what amounts to a trade that would have made it understandable that the Appellants should consider they were entitled to relief. The relevant consideration was what the Appellants appreciated amounted to a trade giving them an entitlement to a deduction;(3) the Appellants’ reliance (referred to at [166]) on Montpelier must be assessed by reference to the advice from Montpelier, which included that the scheme was backed by counsel, and the Montpelier Materials. Any lack of diligence, eg not asking for a copy of counsel’s opinion, may be an indicator of carelessness but not actual knowledge. The reference to the contents of that opinion, which the Appellants had not seen, does not explain why the FTT concluded they had actual knowledge;(4) the findings in [168], which referred to “glib assertions that they were not men with an eye for detail”, may similarly be indicators of carelessness but not knowledge;(5) the findings at [168] and [169] are consistent with the Appellants being prepared to bear any losses because of the hoped-for tax reliefs, and are consistent with them knowing that they were engaging in a tax avoidance scheme;(6) the language of [171] echoes the decision of the FTT in Sacutia Healthcare Ltd v HMRC [2018] UKFTT 699 (TC) (“Sacutia”). The FTT had drawn this case to the parties’ attention during the hearing. Mr Woolf submitted that Sacutia had incorrectly taken an objective approach to determining knowledge; and(7) the FTT said at [171] that the Appellants were trying to create a loss and “crucially” do not and never did have any liability to repay a purported loan. This does not address the fact that the Appellants signed loan agreements and thought there was a loan. Mandaconsult did not execute the loan agreements and claimed to have no knowledge of the agreements, but Montpelier did not tell the Appellants of this. There is no explanation for why the finding on the absence of a loan justifies concluding that the conduct was deliberate. The only possible explanation might be if the FTT had thought that there could be no tax loss in the absence of a loan, but that is incorrect in law.[88]Mr Woolf had expanded upon this final submission in the context of his submissions on Grounds 1 and 4. Mr Woolf submitted that the FTT’s emphasis of its finding that the loans had not in fact been made and that the Appellants therefore had no liability to repay any loans not only failed to address the Appellants’ belief that there were loans, but also indicated that the FTT failed to appreciate the way in which the tax loss arose. The liability to a lender was not an essential requirement to obtain the intended tax benefits – the Appellants had a liability under the Pendulum Contracts and were entitled to claim a loss on that basis. Mr Woolf submitted that the Revised Decision at [116], [154] and [156], as well as its reasoning at [171], suggest that the FTT had considered that a loss can only be claimed if there is a payment by the lender to satisfy the liability under the Pendulum Contracts, and this incorrectly formed part of the FTT’s reasoning on deliberate conduct.

HMRC’s submissions

[89]Ms Choudhury submitted that the FTT made it clear that it was applying the subjective test in Auxilium, and a tribunal can be expected to have been seeking faithfully to apply the legal principles expressly identified, and to have done so unless the contrary is clear from the language of the decision (citing DPP Law v Greenberg [2021] EWCA Civ 672 (“Greenberg”)).[90]Whilst the Appellants contend that the FTT’s comments have clear echoes of the FTT’s decision in Sacutia, Sacutia is not referred to in the Revised Decision. The closest reference is that in [168] to the Appellants not being “men with an eye for detail”. The taxpayer in Sacutia had argued that it ought not to be liable for a penalty on the grounds that its director was not a “detail man” (at Sacutia[57]). The FTT had rejected this as a basis for escaping the penalty. Moreover, the FTT in Sacutia had referred to Auxilium and said that some measure of knowledge was required for an inaccuracy to be made deliberately, but it considered that a taxpayer had the required “knowledge” for an inaccuracy to be deliberate where the taxpayer knew that they should take steps to check the accuracy before information is submitted or relied upon and did not do so. The “objective reasoning” in Sacutia was in the context of considering whether the appellant in that appeal had acted carelessly.[91]In any event, none of the examples referred to by the Appellants indicate that the FTT had lost sight of the Auxilium test, whether due to Sacutia or otherwise:(1) The findings in [164] and [165] are sufficient. In particular, the Appellants had experience of trading in oil futures. There is no basis on which the Appellants could have said they were trading; this is not credible.(2) The Appellants’ criticisms of these findings are based on evidence which was not before the FTT (eg, the advice Montpelier was said to have given regarding the need to establish a trade beforehand); evidence which is cherry-picked (eg, one statement in the notes for one slide out of the complete set of notes and slides before the FTT); or evidence which was specifically rejected by the FTT (they expected a tax loss but would have been happy to generate a profit).(3) As to [166], there was no evidence of what tax advice was given by MTP, why the Appellants chose not to look at the counsel’s opinion or what they had been told about it. HMRC submitted the FTT was entitled to take into account, when determining knowledge, that the Appellants had not seen the opinion even though they could have asked for a copy, and, if they had seen the opinion, it would have been clear that the scheme was not backed by counsel.(4) At [168] the FTT rejected the argument that not having an eye for detail was a sufficient reason for not asking questions or reading documents.[92]The FTT’s mode of reasoning leading to its conclusions was not “impermissibl[y] objective”: rather, the FTT was fulfilling its duty to give sufficient reasons for its decision by explaining in detail the process by which it had come to its conclusion. The FTT had referred to Auxilium, summarised the findings it had already made at [164], stated its conclusion at [165] and then based on this conclusion as to knowledge the FTT went on to conclude at [171] and [172] that the behaviour had been deliberate.[93]Ms Choudhury submitted that, at most, the Revised Decision shows that the FTT considered that the true overall reason that the Appellants’ behaviour deviated in several instances from what might have been reasonably expected of individuals in their position was because they knew that their tax returns were inaccurate as a result of claiming a loss to which they were not entitled.[94]Responding to Mr Woolf’s submissions in relation to the absence of any loans, Ms Choudhury submitted that the Montpelier Arrangements were premised on the Appellants claiming a loss which was artificial, or not an economic loss. In that situation, it didn’t matter whether such a loss arose from a loan or from the terms of the Pendulum Contracts. The existence or absence of the loan was significant because it was key to how the Montpelier Arrangements were designed and intended to work, and (as the FTT found at [153]) was important to the Appellants.[95]In their written submissions before the hearing, HMRC submitted that the findings at [166] and [168] are consistent with the Appellants having “blind-eye knowledge” that their returns were inaccurate, referring to CPR Commercials Ltd v HMRC [2023] UKUT 00061 (TCC) (“CPR”), where the Upper Tribunal stated at CPR[23]:
“In our view, where a taxpayer suspects that a document contained an inaccuracy but deliberately and without good reason chooses not to confirm the true position before submitting the document to HMRC then the inaccuracy is deliberate on the part of the taxpayer. If it were otherwise then a person who believed there was a high probability that their return contained errors but chose not to investigate would never be subject to a deliberate penalty. However, the suspicion must be more than merely fanciful. …”
[96]HMRC submitted that if we were to conclude that the Appellants had established that the FTT had made an error of law, we should go on to consider whether the Revised Decision could nevertheless be upheld on the basis that its findings were consistent with the Appellants having blind-eye knowledge.

Discussion and conclusions

[97]At [158] to [159] the FTT identified that the issue was whether or not the Appellants’ behaviour in submitting the tax returns was deliberate. They expressed agreement with the statement of the FTT in Auxilium where at Auxilium[63] the FTT said in relation to “deliberate inaccuracy” in Schedule 24 Finance Act 2007:
“In our view, a deliberate inaccuracy occurs when a taxpayer knowingly provides HMRC with a document that contains an error with the intention that HMRC should rely upon it as an accurate document. This is a subjective test. The question is not whether a reasonable taxpayer might have made the same error or even whether this taxpayer failed to take all reasonable steps to ensure that the return was accurate. It is a question of the knowledge and intention of the particular taxpayer at the time.”
[98]In CF Booth Ltd v HMRC [2022] UKUT 217 (TCC) (“CFB”) the Upper Tribunal has, at CFB[37], subsequently agreed with these comments of the FTT in Auxilium.[99]It was common ground before us that the FTT had set out the correct legal test, and had not misdirected itself. The Appellants submitted that the FTT erred in law in the application of this (subjective) test, by applying what they submitted was objective reasoning.[100]In Greenberg, Popplewell LJ (with whom the other members of the court agreed) set out (at Greenberg[57]) the principles which govern the approach of an appellate tribunal or court to the reasons given by, in that case, an employment tribunal. These included:(1) the decision “must be read fairly and as a whole, without focusing merely on individual phrases or passages in isolation, and without being hypercritical”;(2) a tribunal “is not required to identify all the evidence relied on in reaching its conclusions of fact…Nor is it required to express every step of its reasoning in any greater degree of detail than that necessary to be Meek-compliant”; and(3) it is not legitimate for an appellate court or tribunal to reason that a failure by an employment tribunal to refer to evidence means that it did not exist, or that a failure to refer to it means that it was not taken into account in reaching the conclusions expressed in the decision. What is out of sight in the language of the decision is not to be presumed to be non-existent or out of mind.[101]The reference to “Meek-compliant” is a reference to the decision of the Court of Appeal in Meek v Birmingham City Council [1987] IRLR 250 in which Bingham LJ set out the fundamental criteria that must be met by the decision of a tribunal as follows: “It has on a number of occasions been made plain that the decision of [a Tribunal] is not required to be an elaborate formalistic product of refined legal draftsmanship, but it must contain an outline of the story which has given rise to the complaint and a summary of the Tribunal's basic factual conclusions and a statement of the reasons which have led them to reach the conclusion which they do on those basic facts. The parties are entitled to be told why they have won or lost. There should be sufficient account of the facts and of the reasoning to enable [an appellate tribunal] or, on further appeal, this court to see whether any question of law arises…”.[102]Significantly, having set out these principles, Popplewell LJ then stated in Greenberg:
“58. Moreover, where a tribunal has correctly stated the legal principles to be applied, an appellate tribunal or court should, in my view, be slow to conclude that it has not applied those principles, and should generally do so only where it is clear from the language used that a different principle has been applied to the facts found. Tribunals sometimes make errors, having stated the principles correctly but slipping up in their application, as the case law demonstrates; but if the correct principles were in the tribunal’s mind, as demonstrated by their being identified in the express terms of the decision, the tribunal can be expected to have been seeking faithfully to apply them, and to have done so unless the contrary is clear from the language of its decision. This presumption ought to be all the stronger where, as in the present case, the decision is by an experienced specialist tribunal applying very familiar principles whose application forms a significant part of its day to day judicial workload.”
[103]This guidance as to the approach to be taken by an appellate tribunal to a challenge based on the FTT having erred in its application of a subjective test when it reached its conclusion that the Appellants knew that their returns were inaccurate emphasises the high bar which must be met by the Appellants, and warns us as the appellate tribunal of the dangers to avoid when considering these submissions, including in particular the need to avoid being hypercritical or focusing on individual phrases or passages in isolation.[104]We have considered carefully the entirety of the Revised Decision when considering whether the FTT’s reasoning involves an error of law. We have not, however, addressed Mr Woolf’s submissions which were made by reference to the reasoning of the FTT in Sacutia – we do not regard it as helpful or relevant when determining this appeal to express any conclusions as to the application of the Auxilium test in Sacutia, a decision which was not the subject-matter of the appeal before us.[105]When considering the FTT’s reasoning and conclusion that the loss of income tax was brought about deliberately by the Appellants, ie that the Appellants knowingly provided HMRC with tax returns which contained an error with the intention that HMRC should rely upon them as accurate, we remind ourselves that in the present context the relevant error was the claim by the Appellants for trading losses in their tax returns for 2005/06. We have described the Montpelier Arrangements under the heading “The scheme” and summarised the FTT’s findings under the heading “Decision of the FTT”. We comment on two aspects at this stage:(1) For the Montpelier Arrangements to deliver the intended tax benefits, participants needed to be trading. This is clear from the summary in Thomson at [65(1)]. It was emphasised by the FTT in the Revised Decision, and the summary of the Montpelier Materials at [142] makes clear that the marketing focused on a two-stage process – establishing a trade and then incurring losses. As a matter of law, there would be no need for the trade to be established first, but this is not how the scheme was marketed.(2) The Montpelier Arrangements had to produce a tax loss without an economic loss. This is explained in Thomson at [65(5)], and was achieved by the Margin Call Balance being funded by a loan (there, from Bayridge, for the Appellants, intended to be from Mandaconsult). The FTT found that there was no loan to the Appellants. This raises a question of fact as to the outstanding amount apparently due under the Pendulum Contracts, but we recognise that the absence of the loan would not of itself necessarily render the scheme ineffective – a trader would be claiming a loss arising from the fall in value of the Pendulum Contract, relying on the value implied by the amount of the repurchase offer.[106]We have set out [164] to [173] of the Revised Decision in full above. These paragraphs need to be considered in their entirety, and as part of the whole decision.[107]When reading this part of the Revised Decision there are two significant features which are relevant when assessing Mr Woolf’s challenges:(1) As already identified, the FTT had expressed its agreement with the subjective test for a deliberate inaccuracy as set out in Auxilium. That appears at [163] of the Revised Decision, and includes the relevant paragraph from the decision in Auxilium. It appears immediately before the FTT’s consideration of the application of that test.(2) The finding which the FTT reaches at [165], namely “Therefore we find that each of the appellants knew at the time of filing their respective SATRs that they were not carrying on a trade which entitled them to make a claim for loss relief”, is absolutely clear.[108][164] summarises six findings which are said to have already been made by the FTT, and then leads to the finding at [165] that “Therefore we find that each of the appellants knew at the time of filing their respective SATRs that they were not carrying on a trade which entitled them to make a claim for loss relief.” Ms Choudhury submitted that the findings at [164] were sufficient to reach this conclusion, whereas Mr Woolf submitted that the findings do no more than record that this was a marketed tax avoidance scheme, the Appellants knew this, and they did not seek advice from BR; he submitted that the key finding at [164(b)] that “they knew that that scheme involved the need to have a trade and thereafter create losses, whereas the reality was that there was no trade” was based not on a finding of knowledge of the Appellants but imputing the FTT’s own conclusions to the Appellants.[109]We recognise that Mr Woolf’s submissions did invite us to undertake the very type of critique which Greenberg reminds us is not appropriate. That is particularly significant here as [164] is stated to be a summary of earlier findings which have been made by the FTT (“We have already established that we find…”) and should not be analysed in isolation.[110]By this point in the Revised Decision, the FTT had:(1) made findings as to the transactions which had been entered into, including that the terms of the proposed loan were wholly uncommercial but that in the event there were no loans (at [88]);(2) assessed the evidence which was before it – this included not only observations on the documentary evidence (including at [49] that one of the “major problems” was that neither Appellant had been able to produce anything remotely like a complete set of signed documents), but also an assessment of the Appellants as witnesses (describing them as “less than compelling witnesses” at [126]);(3) described their contact with both BR and the various Montpelier entities, including that any tax advice they obtained was from MTP, and the Appellants’ recollections of what they had been told (at [63]);(4) reached conclusions as to the “true objective” of the Appellants in entering into the Pendulum Contracts being to lose at the end of Phase One so as to create a loss in respect of which they did not bear the full economic cost but which reduced their liability to tax (at [137]);(5) described and made findings in relation to the Montpelier Materials, concluding that they accurately represent how Montpelier marketed the arrangements (at [138]); and(6) reached the conclusion that the Appellants did not commence any trade before entering the Pendulum Contracts, and that the CFD contracts other than the Pendulum Contracts were “mere window dressing to give the impression of trading” (at [147]), that very few, if any, of the badges of trade were present and that if you do not have a trade, as Montpelier made very clear, you cannot relieve any losses (at [149]).[111]However, having conducted this exercise, there are then two potential problems which emerge from the FTT’s reasoning, and it is those that we assess further below, remaining mindful of the guidance set out in Greenberg:(1) the FTT appears to rely on its own conclusions as to the absence of a trade when concluding that the Appellants had actual subjective knowledge of the same; and(2) the FTT’s emphasis on the Appellants having failed to check anything is reminiscent of an approach to carelessness rather than deliberate, or perhaps blind-eye knowledge (which had not been pleaded before the FTT).[112]The FTT, when reaching its conclusions as to whether the Appellants knew that their tax returns were inaccurate, ie that they knew they were not entitled to the claimed losses, did not need to cross-refer back to each of the findings which it had made on which it relied. It did, however, need to explain its reasons for concluding that the subjective test had been met. It is in this regard that we consider that [164(b)] potentially discloses an error of law.[113]The FTT states that they have already established that they find that “They knew that that scheme involved the need to have a trade and thereafter create losses, whereas the reality was that there was no trade” (at [164(b)]). The reference to “whereas the reality was that there was no trade” appears to refer back to the FTT’s own conclusion on the absence of a trade, rather than being a subjective reference to the Appellants’ knowledge. On its own, such a concern could be no more than an example of the very real dangers which arise when scrutinising each phrase of a decision too closely. It could have been entirely plausible that the FTT intended to say something along the lines, eg, that “whereas it was obvious to anyone that there was no trade and we infer that the Appellants must also have known this”. The difficulty is, and the reason why we consider that the FTT did indeed intend to refer to its own conclusions at this point, is that the FTT’s own conclusion as to the absence of a trade is the only relevant finding that had been made as to what amounts to a trade.[114]Explaining this further, it is clear from the opening words of [164] that the FTT was not seeking to make further findings of fact at this point in the Revised Decision. It was simply referring to some of the findings it had already made. The Revised Decision does refer to the other transactions entered into by the Appellants as being “mere window dressing” (at [147]), and we were taken to the Montpelier Materials, the speaking notes forming part of which include both a statement that a one-off transaction may amount to a trade but also the recommendation that an individual trades daily. The FTT accepted that this is how the Montpelier Arrangements were marketed to the Appellants. However, there is no finding as to what the Appellants understood (however they came by that understanding) to amount to trading – eg, whether they were told that they should undertake “some” unspecified number of further transactions, or that “one or two” would be sufficient, or referring this back to their previous background transacting in oil futures and what they “must have known” based on their own experience. There is, similarly, no finding or inference earlier in the Revised Decision that the Appellants must therefore have known that the number of transactions that they entered into did not amount to trading. This is significant in the context of a subjective test and whether the Appellants knew that they were not trading and therefore knew that they were not entitled to claim the losses. We do recognise that the FTT did then reach a clear conclusion at [165] when it found that the Appellants knew that they were not carrying on a trade, but that is reached by reference to the findings at [164] and potentially contaminated by the reasoning at [164(b)].[115]We concluded that this was then exacerbated by the reasoning which followed from [166] to the conclusion at [172] that “In submitting their SATRs we find that they acted deliberately with a view to claiming non-existent losses”.[116]In this part of the reasoning, the FTT refers to the Appellants’ reliance on Montpelier, their failure to check anything, the contents of counsel’s opinion (acknowledging that the Appellants had not asked to see it), that there was no FSA protection, the Appellants’ assertions that they were not men with an eye for detail, that they did not know whether they had accepted Pendulum’s repurchase offer, and their failure to check what had happened in subsequent years (which we infer is a reference to the loans).[117]We agree with Mr Woolf that the FTT is here assessing whether the Appellants took reasonable steps to ensure their returns were accurate, and is applying an objective test of the kind that the FTT in Auxilium had warned against. We have considered two counter-arguments against this:(1) We did consider whether this part of the Revised Decision could be read in a way which does not prejudice any earlier conclusions reached, potentially by reading the FTT as having effectively reached its conclusion at [164] and [165] (although we have already identified an error in that approach) and having then moved on to consider and dismiss other factors which might otherwise be used to explain the Appellants’ conduct, eg that they failed to take reasonable care, did not ask to see and therefore did not read counsel’s opinion, and did not read the documents. The difficulty is that the Revised Decision itself does not support such an approach, and Ms Choudhury did not seek to persuade us that it should be read in this way even when invited to do so by the panel at the hearing.(2) [171] has echoes of a more subjective approach, addressing the FTT’s conclusions as to why the Appellants did not ask questions or read documents. However, the answer given by the FTT is that this was because “they knew precisely what they were doing. They were trying to create a significant loss and thereafter make substantial claims for repayment of tax.” This reference to the Appellants’ knowledge is thus confined to their knowledge that the Montpelier Arrangements were a marketed tax avoidance scheme; it does not address their knowledge as to the lack of a trade.[118]Whilst we have remained mindful of the need to be slow to conclude that where, as here, the FTT has correctly stated the legal principles, it has not applied those principles, we do reach that conclusion here. We have concluded that this is clear from the language used by the FTT when it was assessing the knowledge and intention of the Appellants. We reach this conclusion based on the language used in [164] (both the reference to the findings already having been made and to the FTT’s assessment that the reality was there was no trade), the absence of findings as to the Appellants’ understanding of what amounts to a trade, the absence of inferences or conclusions from this as to what the Appellants then knew about their own level of activity and whether that met their understanding of a trade, and the reference to factors which relate to the Appellants’ lack of reasonable care rather than their knowledge. This is a material error of law.

Grounds 1 and 4

[119]The Appellants submitted that the FTT had made further errors of law. There was considerable overlap between the Appellants’ submissions on Grounds 1 and 4 and we had the benefit of full written and oral submissions from both parties. However, in view of our decision on Ground 3 it is not necessary to reach a decision on these further grounds and we do not do so.

Disposition

[120]We set aside the Revised Decision made by the FTT pursuant to s12(2)(a) TCEA 2007 because it contained a material error of law.[121]We have considered whether to remit the case to the FTT for its reconsideration or to re-make the decision.[122]In their Notice of Appeal from the FTT the Appellants submitted that the Revised Decision should be quashed and remitted to a new tribunal for redetermination. Mr Woolf maintained this position in his written submissions before us. HMRC submitted in their written submissions that if we were to conclude that the Appellants had established that the FTT had made an error of law, we should go on to consider whether the Revised Decision could nevertheless be upheld on the basis that its findings were consistent with the Appellants having blind-eye knowledge, ie inviting us to re-make the decision.[123]We have significant reservations about remitting the appeal. The losses claimed by the Appellants were stated to relate to the period from 18 April 2005 to 31 March 2006. The meeting between the Appellants and Montpelier took place in March 2006. The Revised Decision acknowledges that the matters occurred 16 years ago and records at [126] that both Appellants repeatedly said that they could not remember what had happened. The position is unlikely to have improved since the hearing before the FTT in March 2021.[124]However, we have concluded that it is necessary in accordance with the overriding objective for us to remit the case to the FTT for a new hearing of the appeal:(1) we reject HMRC’s submission that we could decide the appeal on the basis of blind-eye knowledge – this had not been pleaded by HMRC in their Statement of Case before the FTT and appears to have been first raised by HMRC in their written submissions before this hearing. It would in this situation be completely unfair to the Appellants to make any such decision; and(2) we are not able to re-make the decision ourselves – we are not able to re-make the decision on the basis of the facts as found by the FTT and, having not heard the evidence or indeed been taken to any transcripts of the hearing before the FTT (if indeed such transcripts exist), we are not in a position to make any additional findings of fact.[125]Accordingly, we remit the case to the FTT to be re-heard by a differently constituted panel. JUDGE JEANETTE ZAMAN JUDGE ASHLEY GREENBANK UPPER TRIBUNAL JUDGES Release date: 15 July 2024

Cited in 1 later judgment