England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE UPPER TRIBUNAL
TAX AND CHANCERY CHAMBER
Mr Justice Marcus Smith and Judge Hellier
[2018] UKUT 38 (TCC)

		Royal Courts of Justice Strand, London, WC2A 2LL
		15/05/2019

 


Lord Justice Floyd:

1. In 2009, the respondent, Mr Raymond Tooth, participated in a tax avoidance scheme which was designed to utilise employment-related losses incurred in 2008/09 to relieve his liability to tax on other income ("the Romangate scheme"). He claimed to be entitled to carry back these losses and set them off against income for the 2007/08 year of assessment. The scheme as a whole was defeated by anti-avoidance legislation, but Mr Tooth maintained that his self-assessment for 2007/08, which took account of the disputed losses, should stand as it had not been validly challenged. In 2014, the appellants, Her Majesty's Commissioners for Revenue and Customs ("HMRC"), purported to raise a "discovery assessment" under section 29 of the Taxes Management Act 1970 ("TMA") in respect of income tax which they contended was due for the 2007/08 year of assessment. Mr Tooth successfully appealed from this discovery assessment to the First-tier Tribunal (Tax) ("FTT") (FTT Judge Brooks) and a subsequent appeal to the Upper Tribunal (Tax and Chancery Chamber) ("UT") (Marcus Smith J and Judge Charles Hellier) was dismissed. HMRC now bring this further appeal, with permission granted by Lewison LJ on 16 July 2018.
2. There are two broad issues on this appeal. The first ("the discovery issue") is whether HMRC made a relevant "discovery" about Mr Tooth's self-assessment, and in particular whether HMRC "discover[ed] … that an assessment to tax [was] insufficient" within the meaning of section 29(1)(b) of TMA. The second ("the deliberateness issue") is whether Mr Tooth, or a person acting on his behalf, can be said to have "deliberately" brought about a situation in which "an assessment to tax is or has become insufficient" within sections 29(1)(b) and (4) of TMA. If so, HMRC were not prevented by section 29(3) from raising the assessment. A closely related question is whether HMRC can show, as they must, that they can take advantage of the 20 year time limit for raising an assessment provided by section 36(1A)(a) of TMA, which applies if a "loss of … tax … [is] brought about deliberately" by a person or someone acting on his or her behalf.
3. Section 8 TMA provides that, for the purposes of establishing the amounts in which a person is chargeable to tax for a year of assessment, an officer may require a person to make a return. Subject to exceptions, section 9(1)(a) provides that every return shall include a self-assessment, that is to say:

"an assessment of the amounts in which, on the basis of the information contained in the return and taking into account any relief or allowance a claim for which is included in the return, the person making the return is chargeable to income tax and capital gains tax for the year of assessment".

4. One of the exceptions is that a person is not required to comply with section 9(1) if he makes and delivers his return before 31 October following the year of assessment (section 9(2)). In such circumstances it can be left to HMRC to compute the tax which is payable for the year and make the assessment on the basis of the information contained in the return (section 9(3)). Where this occurs, HMRC's computation is treated as the taxpayer's self-assessment.
5. Section 29 TMA provided at the relevant time (so far as material):

"(1) If an officer of the Board or the Board discover, as regards any person (the taxpayer) and a year of assessment –

(a) that any income which ought to have been assessed to income tax, or chargeable gains which ought to have been assessed to capital gains tax, have not been assessed, or

(b) that an assessment to tax is or has become insufficient, or

(c) that any relief which has been given is or has become excessive,

the officer or, as the case may be, the Board may, subject to subsections (2) and (3) below, make an assessment in the amount, or the further amount, which ought in his or their opinion to be charged in order to make good to the Crown the loss of tax....

(3) Where the taxpayer has made and delivered a return under section 8 or 8A of this Act in respect of the relevant year of assessment, he shall not be assessed under subsection (1) above–

(a) in respect of the year of assessment mentioned in that subsection; and

(b) in the same capacity as that in which he made and delivered the return,

unless one of the two conditions mentioned below is fulfilled.

(4) The first condition is that the situation mentioned in subsection (1) above was brought about carelessly or deliberately by the taxpayer or a person acting on his behalf.

(5) The second condition is that at the time when an officer of the Board –

(a) ceased to be entitled to give notice of his intention to enquire into the taxpayer's return under section 8 or 8A of this Act in respect of the relevant year of assessment; or

(b) informed the taxpayer that he had completed his enquiries into that return,

the officer could not have been reasonably expected, on the basis of the information made available to him before that time, to be aware of the situation mentioned in subsection (1) above."

6. The discovery assessment which HMRC purported to make in the present case was on the basis that an officer had discovered, as regards Mr Tooth and the year of assessment 2007/08, that an assessment to tax was or had become insufficient. HMRC contended that the prohibition in section 29(3) was overcome because the first condition (i.e. that specified in section 29(4)) was satisfied, namely that the insufficiency of the assessment had been brought about deliberately by Mr Tooth or a person acting on his behalf.
7. A "discovery assessment" is not the only way in which HMRC can go behind past assessments to tax or claims to relief. HMRC can launch an enquiry into a claim for relief under Schedule 1A TMA. They can also launch an enquiry into a return under section 9A TMA. Schedule 1A and section 9A TMA are mutually exclusive mechanisms. Each has a prescribed time limit within which HMRC must take action.
8. Under Schedule 1A, HMRC can only enquire into a claim which is not included in a return, and the time limit is, in broad terms, 12 months from the date of the claim. By contrast, where a claim is included in a return made under section 8 or 8A TMA, the enquiry must be under section 9A. Enquiries under section 9A extend to anything contained in the return or required to be contained in the return, including a claim or election included in the return. The time limit for section 9A enquiries is, again in broad terms, 12 months from the filing of the return.
9. Schedule 1A paragraph 3 contains a power for HMRC to amend a claim, but that power is not exercisable during the period when a claim is being enquired into. Section 9ZB contains a power to amend a return, which must be exercised within 9 months of the delivery of the return or a relevant amendment.
10. Where an enquiry is launched into a claim under Schedule 1A, paragraphs 4(1) and (3) of that Schedule provide that HMRC are not required to give effect to the claim until the enquiry is completed.
11. It can therefore be of some importance to decide whether or not a claim is included in a return. This question was considered by the Supreme Court in Revenue and Customs Commissioners v Cotter [2013] UKSC 69; [2013] 1 WLR 3514 ("Cotter"), where it held that a return did not include the entirety of the information set out in the return, but only that information which was submitted for the purpose of establishing the amount to which a person is chargeable to tax for a year of account. Lord Hodge (with whom the other members of the Supreme Court agreed) said this:

"24. Where, as in this case, the taxpayer has included information in his tax return but has left it to the Revenue to calculate the tax which he is due to pay, I think that the Revenue is entitled to treat as irrelevant to that calculation information and claims, which clearly do not as a matter of law affect the tax chargeable and payable in the relevant year of assessment. It is clear from sections 8(1) and 8(1AA) of TMA that the purpose of a tax return is to establish the amounts of income tax and capital gains tax chargeable for a year of assessment and the amount of income tax payable for that year. The Revenue's calculation of the tax due is made on behalf of the taxpayer and is treated as the taxpayer's self-assessment (section 9(3) and (3A) of TMA).

25. The tax return form contains other requests, such as information about student loan repayments (page TR2), the transfer of the unused part of a taxpayer's blind person's allowance (page TR3) or claims for losses in the following tax year (box 3 on page Ai3) which do not affect the income tax chargeable in the tax year which the return form addresses. The word "return" may have a wider meaning in other contexts within TMA. But, in my view, in the context of sections 8(1), 9, 9A and 42(11)(a) of the TMA, a "return" refers to the information in the tax return form which is submitted for "the purpose of establishing the amounts in which a person is chargeable to income tax and capital gains tax" for the relevant year of assessment and "the amount payable by him by way of income tax for that year" (section 8(1) TMA).

26. In this case, the figures in box 14 on page CG1 and in box 3 on page Ai3 were supplemented by the explanations which Mr Cotter gave of his claim in the boxes requesting "any other information" and "additional information" in the tax return. Those explanations alerted the Revenue to the nature of the claim for relief. It concluded, correctly, that the claim under section 128 of ITA in respect of losses incurred in 2008/09 did not alter the tax chargeable or payable in relation to 2007/08. The Revenue was accordingly entitled and indeed obliged to use Schedule 1A of TMA as the vehicle for its enquiry into the claim (section 42(11)(a)).

27. Matters would have been different if the taxpayer had calculated his liability to income and capital gains tax by requesting and completing the tax calculation summary pages of the tax return. In such circumstances the Revenue would have his assessment that, as a result of the claim, specific sums or no sums were due as the tax chargeable and payable for 2007/08. Such information and self-assessment would in my view fall within a "return" under section 9A of TMA as it would be the taxpayer's assessment of his liability in respect of the relevant tax year. The Revenue could not go behind the taxpayer's self-assessment without either amending the tax return (section 9ZB of TMA) or instituting an enquiry under section 9A of TMA.

28. It follows that a taxpayer may be able to delay the payment of tax by claims which turn out to be unfounded if he completes the assessment by calculating the tax which he is due to pay. Accordingly, the Revenue's interpretation of the expression "return" may not save it from tax avoidance schemes. But what persuades me that the Revenue is right in its interpretation of "return" is that income tax is an annual tax and that disputes about matters which are not relevant to a taxpayer's liability in a particular year should not postpone the finality of that year's assessment."

12. The taxpayer, Mr Cotter, had, like Mr Tooth, made use of the Romangate scheme. The Supreme Court held that a taxpayer who had suffered an employment loss in a later year (year 2) could attribute that loss to an earlier year of assessment (year 1) and obtain relief, but that relief was applied in year 2 by obtaining a reduction in liability to, or a repayment of tax for that year, and would not result in a change in the amount of tax chargeable in year 1.
13. In Cotter HMRC had instigated an enquiry into the claim for losses in year 1 under Schedule 1A TMA, but the taxpayer contended that Section 9A TMA was the appropriate mechanism because the relief was claimed in the return. The Supreme Court held that, because the taxpayer had left HMRC to calculate his tax, the claim to employment loss relief for year 1 did not form part of his return, and so did not affect the computation of tax for that year. Accordingly, HMRC had correctly made use of Schedule 1A in Mr Cotter's case. As will be seen, the present case is not on all fours with the facts of Cotter because Mr Tooth had performed his own self-assessment computation. His case therefore fell within Lord Hodge's remarks in Cotter at paragraph 27.
14. The time limits which apply in the case of a discovery assessment under section 29 TMA are:

(1) save where a loss of tax has been brought about carelessly or deliberately: not more than 4 years after the end of the year of assessment to which the assessment relates (section 34(1) TMA);

(2) where the loss of tax has been brought about carelessly by the taxpayer: not more than 6 years after the end of the year of assessment to which the assessment relates (section 36(1) TMA);

(3) where the loss of tax has been brought about deliberately by the taxpayer: not more than 20 years after the end of the year of assessment to which the assessment relates (section 36(1A) TMA).

15. Section 36(1A) is in the following terms:

"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)."

16. Section 118 TMA is the interpretation section. So far as material, it provides:

"(5) For the purposes of this Act a loss of tax or a situation is brought about carelessly by a person if the person fails to take reasonable care to avoid bringing about that loss or situation.

(6) Where –

(a) information is provided to Her Majesty's Revenue and Customs,

(b) the person who provided the information, or the person on whose behalf the information was provided, discovers some time later that the information was inaccurate, and

(c) that person fails to take reasonable steps to inform Her Majesty's Revenue and Customs,

any loss of tax or situation brought about by the inaccuracy shall be treated for the purposes of this Act as having been brought about carelessly by that person.

(7) 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."

17. HMRC rely on section 118(7) in support of their argument on the deliberateness issue. They say that there was a deliberate inaccuracy in Mr Tooth's return, as a result of which the loss of tax arose.
18. With that introduction, I can turn to the facts which are, to a large extent, summarised uncontroversially in section C of the decision of the UT, and from which I have borrowed heavily below.
19. Towards the end of 2008, Mr. Tooth sought advice from Grunberg & Co Chartered Accountants ("Grunberg") as to how he could legitimately reduce his income tax liability for 2007/08. He was advised of the Romangate scheme promoted by NT Advisors. The Romangate scheme sought to make use of Section 128 of the Income Tax Act 2007 ("ITA 2007") which makes provision for employment loss relief against general income.
20. Mr. Tooth understood, as a result of the advice he was given, that employment-related losses of £1,185,987 would be generated for 2008/09. These losses, he was told, could then be set-off under section 128 ITA 2007 against income recorded in his 2007/08 self-assessment return, thereby reducing his tax liability for that year. He was told that the Romangate scheme had a reasonable prospect of success, and that it had been sanctioned by leading counsel. On this basis, Mr. Tooth decided to utilise the scheme in January 2009.
21. Participants in the Romangate scheme were told how to complete their self-assessment tax return in order to claim the losses accruing under the scheme. Accordingly, in about January 2009, Grunberg began to prepare Mr. Tooth's self-assessment tax return. To do this, they used HMRC-approved software provided by IRIS Software Limited. Following instructions for the Romangate scheme provided by NT Advisors, Grunberg sought to enter the employment-related loss sustained by Mr. Tooth into box 3 on Additional Information page 3 ("page Ai3"). It was not possible, however, to access this box so as to make this entry. Grunberg contacted IRIS about this problem and their engineer confirmed that box 3 on page Ai3 could not be accessed because of a technical issue with the IRIS software. The engineer advised that, to ensure the claim was included in the 2007/08 return the loss should be included on another part of the return and reference made in the "white space" to explain what had been done.
22. Following this advice, Grunberg entered the employment-related loss on the partnership pages of the return (in box 7). This created an additional problem. Because there was no partnership, there was no ten-digit partnership unique taxpayer reference ("UTR") number which was necessary for the electronic submission of the return. Having encountered similar problems previously, where clients had not been allocated a UTR in advance of the self-assessment filing date, Grunberg had used a UTR of "99999-99999" and was able to file the return electronically before the deadline, thus preventing the imposition of a late-filing penalty. This was the course followed – albeit for a different reason – in the case of Mr. Tooth's return, and a UTR of "99999-99999" was entered in the partnership pages of Mr. Tooth's return.
23. The following text was entered at box 19 on page TR6 of Mr. Tooth's return:

"…During the tax year ending 5 April 2009, I sustained an employment related loss for which relief is being claimed now in accordance with section 128 ITA 2007. Please refer to the partnership pages of my return. Full details of this loss will be reported on my 2008/09 tax return in due course."

24. The following information was entered in the boxes on the partnership pages of the return:

Box 1	Partnership reference number	99999-99999
Box 2	Description of partnership trade or profession	…
Box 5	Date your basis period began	06-04-2007
Box 7	Your share of the partnership's profit or loss	-£1,185,987.00
Box 19	Adjusted loss for 2007-2008...	£1,185,987.00
Box 20	Loss from this tax year set off against other income for 2007-2008	£1,185,987
Box 30	Any other information	During the year ending 5 April 2009, I sustained an employment related loss for which relief is being claimed now in accordance with the provisions of s 128 ITA 2007 (via section 11 ITEPA 2003). I have reported the details of the loss claimed against my other income using box 3 above, which relates to a claim for a partnership Loss from this tax year set-off against other income for 2007–8. However, there is no equivalent box to claim relief now for employment related losses despite the provisions of s 128 ITA 2007. Full details of this loss will be reported on my 2008–09 tax return in due course. The loss arose pursuant to arrangements for which a scheme reference number is required under DOTAS (from AAG at HMRC) – at this time the scheme has not been granted a reference number. When such number is obtained I will report it on my 2008–09 tax return, as that is the year in which the loss arose. I acknowledge that my interpretation of the tax law applicable to the above transactions and the loss (and the manner in which I have reported them) may be at variance with that of HM Revenue & Customs. Further please note that although I have reported (and hereby claim the loss pursuant to section 128 ITA 2007) in box 3 above I wish to make it clear that the deduction I am claiming on my return is not what you would regard as a loss for this tax year set-off against other income from 2007–08 – for all these reasons I assume you will open an enquiry.

25. Grunberg checked this approach with NT Advisors, who confirmed that it was sensible. The return was sent to Mr. Tooth for his approval, Mr. Tooth approved it and the return was filed electronically on 30 January 2009, including a personal computation of the tax due. This showed that Mr Tooth was entitled to a repayment, once the employment-related loss was allowed, of £13,212.
26. On 14 August 2009, HMRC wrote to Mr. Tooth:

"…in respect of your claim to employment losses incurred during 2008-2009 for which you request £914,999 relief be given effect in 2007-08.

This letter is formal notice of HMRC's intention to enquire into that claim under the provisions of Schedule 1A TMA 1970. As a result no effect will be given to the claim at the present time."

27. The letter shows that HMRC had understood that Mr Tooth was not claiming partnership losses, but an employment-related loss. The reference to Schedule 1A TMA suggests that HMRC were investigating a claim, rather than a return. The letter also referred to an announcement made on 1 April 2009 that the 2009 Finance Bill was to include legislation that would have the effect of refusing relief for losses under the Romangate scheme. The proposed legislation referred to in the letter became section 68 of the Finance Act 2009. This inserted a new section 128(5A) ITA which precluded, with retrospective effect, a deduction for an employment loss made in 2008/2009 if that loss was made "as a result of anything done in pursuance of arrangements the main purpose, or one of the main purposes, of which is the avoidance of tax."
28. On 10 December 2009 Mr Tooth filed an amended return which increased the amount of the 2007/08 claim for employment-related losses to £1,210,229.
29. On 15 April 2010 HMRC wrote to Mr Tooth informing him that his 2007/08 return had been "amended … to withdraw the claim for losses in the sum of £1,210,299.00 which arose in the year 2008-2009." The letter went on to say, however, that the officer had used the authority given by Schedule 1A 4(3) TMA to ensure that no effect is given to the claim until the enquiry is completed. It is not entirely clear under what power HMRC were purporting to act when they said they had amended the return, but it is not suggested that the return had in fact been validly amended by them. Neither the written evidence before the FTT, nor the argument before us, touched on HMRC's thinking behind this letter.
30. The letter of 15 April explained that prior to the "amendment" Mr Tooth's statement of account showed a credit of £22.52 in his favour. The amendment resulted in a net liability to HMRC of £498,253.14. The withdrawal of the claim for losses was reflected in a revised statement of account issued by HMRC, dated 19 May 2010.
31. On 7 June 2010 NT Advisors, who had now taken over this aspect of Mr Tooth's tax affairs, wrote to HMRC stating their position that Schedule 1A was not the correct vehicle to investigate the entitlement to the loss relief, because the claim to the losses was made within Mr Tooth's return. HMRC's contrary position was stated in a letter in reply dated 2 August 2010. The letter argued that a claim was not made in a return simply because it was mentioned there, which was all that had happened in the case of Mr Tooth's claim for losses in his 2007/08 return. The use of Schedule 1A for such a "stand alone" claim was therefore appropriate.
32. Having been told by NT Advisors that Mr Tooth would abide by the outcome of current litigation on these issues, HMRC wrote in early 2011 to say that no further action would be taken against Mr Tooth at the present time.
33. The Supreme Court handed down its decision in Cotter on 6 November 2013. On 4 March 2014, in a letter signed by a Mr Scott Webster, HMRC wrote to Mr. Tooth in relation to "Payment of Overdue Tax", stating that there were amounts of tax which had been self-assessed but not paid for 2007/08 of £455,331.98 plus interest of £74,697.24. The letter explained HMRC's view that these amounts were "based on your own self assessments". The letter continued:

"Following a decision of the Court of Appeal in the case of HMRC v. Cotter in February 2012, we have not (until now) been actively seeking to enforce payment by you of your overdue tax and interest arising on it. This is because we considered your circumstances were similar to those of Mr. Cotter and therefore governed by that decision. HMRC has now successfully appealed to the Supreme Court, which reversed the Court of Appeal's decision, and as a result we are now able to enforce payment of the tax and interest that you owe. The Supreme Court's decision is final."

34. On 11 March 2014 Grunberg responded on behalf of Mr. Tooth to say that, in fact, as a result of the decision of the Supreme Court in Cotter, HMRC had incorrectly made an enquiry under Schedule 1A TMA, when (according to Cotter) it should have opened a section 9A enquiry. This followed from paragraph 27 of Lord Hodge's speech, and the fact that Mr Tooth had included a computation of tax with his return. In these circumstances, Grunberg argued that Mr. Tooth's self-assessment calculations must stand.
35. On 23 May 2014, HMRC accepted that Grunberg were correct, and that collection of tax was therefore suspended. A closure notice of the Schedule 1A TMA enquiry was referred to the scheme investigator for consideration. We were told at the hearing that the enquiry remains open.
36. On 28 July 2014 a Mrs K Smith of HMRC wrote to Mr Tooth as follows:

"Dear Mr Tooth

Self Assessment tax return – year ended 5 April 2008

I believe that your return for the above year is inaccurate.

This is because you have claimed a partnership loss which was in fact an employment loss carried back from the year ended 5 April 2009.

What happens now

…

HMRC removed a claim for a partnership loss of £1,210,229 from your 2007–08 return on 14 April 2010. This was done so as to not give effect to a claim as an enquiry into that claim had been opened under Schedule 1a [sic] Taxes Management Act 1970. The Supreme Court decision in Cotter v HMRC makes clear that Schedule 1a did not give HMRC the power to remove this claim under the circumstances.

It is however my intention to make an assessment for that year under the provisions of s 29 TMA 1970. Further s 36(1A) TMA enables HMRC to make:

(1A) 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).

You submitted your tax return for the year to 5 April 2008 on 30 January 2009. You included on a separate partnership page, with the UTR 99999 99999, a claim for your share of the partnership loss of £1,210,229. This was in fact employment losses carried back from 2008–09. It is my view that your actions in making this claim were deliberate.

As this claim has already been removed from your return, I do not intend to make any further amendments."

37. Given that this letter expresses Mrs Smith's, and thus HMRC's, intention at the end of July 2014 to issue a section 29 (i.e. a discovery) assessment, it would suggest that the relevant "discovery" had been made. It refers again to amendments having been made to Mr Tooth's return.
38. In response, by letter dated 12 August 2014, Grunberg disputed that there had been a deliberate act that had resulted in a loss of income tax.
39. The HMRC "Technical Lead" on the case was Mr. Nigel Williams (who has since retired). His role was to decide whether an assessment should be issued and provide guidance on the assessing provisions. Mr. Christopher March also worked as a Technical Lead for HMRC's response to the Romangate schemes, having taken over from a Mr Clarke in April 2014 and being succeeded by Mr Williams in August 2014. Mr March gave evidence before the FTT. In paragraph 8 of his witness statement, Mr March said:

"In October 2014, Mr. Williams reviewed Mr. Tooth's file and concluded, in line with mine and Mr Clarke's thinking before him, that he had discovered a loss of tax as a result of the Appellant deliberately filling out his return in the manner that he did."

40. On 23 October 2014, Mr Williams issued an instruction, by email, to issue a discovery assessment. This task was allocated to Mr. Ian Anders. In an email, dated 23 October 2014, Mr. Anders sought clarification from Mr. Williams as follows:

"From looking through the file my understanding is that this [i.e. the discovery assessment] is to replace the amendment made in April 2010 under S[chedule]1A, as the Cotter case concluded that we could not use S[chedule]1A?"

41. The email continued:

"If this is correct should remove [sic] the informal standovers of the amounts on SA relating to the S[chedule] 1A amendment and reverse the amendments made to the Self Assessment in 2010?

Finally, am I correct in thinking the discovery assessment will be the same as the S[chedule] 1A amendment ie assessment of £475,498.37 additional tax resulting from removal of the loss of £1,210,229?"

42. This email is explicit in referring to HMRC's understanding that they had amended, or at least purported to amend, Mr Tooth's self-assessment in April 2010, and enquires whether the amendment should be "reversed". Mr Williams responded:

"… What you suggest seems right. Certainly we will be making a discovery assessment to replace the Sch 1A amendment following the Cotter decision, and in the same figures.

Cancelling the Sch 1A amendments and associated stand-overs makes sense, and I assume this is what has been done in earlier cases."

43. On 24 October 2014 Mr Anders wrote to Grunberg to say that the assessment under section 29 TMA would be issued "shortly". That letter continued stating that Mr. Tooth:

"… attempted to obtain immediate relief for the loss carry back to year 1 by knowingly and deliberately making entries in his 2007–08 tax return to the effect that the loss was a partnership loss of the current year. This was nothing to do with "technical software issues" as you suggest in your letter. The claim could, and should, have been made outside the return, where the existence or not of 'appropriate boxes' would have had no relevance. The amounts claimed were clearly not appropriate to be entered into these boxes, and the notes submitted with the return confirm that your client was fully aware of this. The only possible conclusion is that there was a deliberate failure to report something correctly on his tax return in an attempt to make him liable for less tax than would otherwise be the case.

In my opinion, this conduct falls squarely within Schedule [sic] 36(1A) as involving a loss of income tax or capital gains tax brought about deliberately by the person."

44. The allegation of a "deliberate failure to report something correctly on his tax return in an attempt to make him liable for less tax" is, understandably in the light of the history which I have related, not one which is pursued. The discovery assessment in the sum of £475,489.20 was issued on 24 October 2014. On 11 November 2014 Grunberg, acting on behalf of Mr Tooth, launched an appeal against the discovery assessment. On 19 November 2014 HMRC responded. The letter set out the relevant entries on Mr Tooth's 2007/08 tax return. It pointed out that:

"The losses purported to have been incurred as a result of the arrangements were Employment Losses for 2008/09. They were not, as acknowledged by entries above, either 2007/08 Partnership Losses or 2007/08 Income losses."

45. As to what constituted the relevant discovery, HMRC wrote:

"The conclusion the officer reached in this case is that the assessment to tax on your 2007/08 tax return was insufficient. She reached this conclusion when she became aware that following the decision in Cotter v Revenue & Customs [2013] UKSC 69, an enquiry under Schedule 1a Taxes Management Act 1970 was not the appropriate mechanism to enquire into the claim which gave rise to the insufficiency."

46. The officer who reached this conclusion is not identified, but given the use of "she" it is possible it refers to Mrs Smith. In any event there is no evidence touching on how this conclusion was arrived at. Turning to "deliberateness" the letter went on to assert that, based on Cotter, the claim for employment-related losses, could only impact the tax payable in 2008/9 not 2007/08. It continued:

"The amounts claimed were clearly not appropriate to be entered into these boxes, and the notes submitted with the return confirm that you were fully aware of this. The only possible conclusion is there was a deliberate failure to report something correctly on your tax return in an attempt to make you liable to less tax than would otherwise be the case".

47. An allegation in that form is, again, no longer pursued by HMRC.
48. Grunberg requested a statutory review of this decision. The decision to uphold the assessment was issued on 20 March 2015 following a review. The reasoning on the discovery point was as follows:

"Section 29(1) is a subjective test. The question is whether it was reasonable for the officer, based on the information before him, to come to the conclusion that there was an insufficiency in relation to the 2007-08 return. It is clear from the evidence before the officer at the time of making the assessment that the officer was justified in newly concluding that an assessment was required for 2007-08 to make good a loss of tax."

49. Mr Tooth appealed to the FTT on 17 April 2015. In their statement of case dated 9 October 2015, HMRC pleaded at paragraph 43.1 that the relevant section 29 discovery was correctly made because "an Officer of HMRC newly concluded, following the decision in Cotter, that an enquiry under Schedule 1A TMA was not the appropriate mechanism to make good the insufficiency in the Appellant's 2007-2008 tax return." In paragraph 46 they claimed that "In October 2014, an officer of the Board properly raised a discovery assessment after concluding that the Schedule 1A enquiry had not rectified the insufficiency in Mr Tooth's 2007-2008 self-assessment, following a clarification of the law in Cotter."
50. In their skeleton argument before the FTT HMRC said at [54]:

"In October 2014, an officer of the Board properly raised a discovery assessment after concluding that there was an insufficiency in the 2007-2008 return (because of the incorrect inclusion of a partnership loss figure). That was the "discovery". The officer also realised that HMRC's position was not protected by the existing enquiry, which is why a further assessment was required. In other words, HMRC discovered that the assessment was and would remain insufficient despite previous attempts to address this."

51. The skeleton argument went on to explain, for the avoidance of doubt, that HMRC were not saying that the discovery of the true legal position following Cotter (which confirmed HMRC's previous view) was sufficient in itself.

The decision of the FTT

52. The FTT allowed Mr Tooth's appeal against the discovery assessment on the basis that, although HMRC had made a discovery within section 29(1) TMA, TMA section 29(4) was not satisfied because neither Mr Tooth nor any person acting on his behalf had done anything which deliberately brought about an insufficiency of tax.
53. The FTT Judge's reasoning on the discovery issue is at paragraphs 43 to 46 of his decision. The judge first records the fact that the discovery relied upon by HMRC is of an insufficiency of income tax declared in the 2007/08 self-assessment tax return, and the argument that, as this was first identified in HMRC's letter of 14 August 2009, an officer could not have newly discovered it in 2014. However, "given the low threshold necessary for there to be a discovery" he found that "Mr Williams did make a discovery of an insufficiency of tax in October 2014 as confirmed by the emails of 23 October 2014".

The decision of the UT

54. The UT considered first the issue of deliberateness. They considered that there was no "inaccuracy" in Mr Tooth's 2007/08 return within section 118(7) TMA. Further, and if that was wrong, any inaccuracy was not deliberate. The UT did not find it necessary to reach a conclusion on whether, if there was any deliberate inaccuracy, the insufficiency and any loss of tax, were brought about by that deliberate inaccuracy.
55. The UT went on to consider the discovery issue. HMRC's position had been that the FTT's conclusion, that there had been a discovery in October 2014, was a finding of fact which the UT should not interfere with. At paragraph 86 of their decision, the UT concluded that such findings as were made by the FTT in [43] to [46] of its decision were insufficient to justify the conclusion that there was a discovery within section 29(1). They went on to consider other findings made by the FTT in earlier passages of its decision, and asked themselves whether the discovery in question was the discovery that Mr Tooth was guilty of deliberate inaccuracy, as suggested in the letter HMRC wrote to Mr Tooth on 28 July 2014 and subsequently. This suggestion appears, however, to have been dismissed by the UT at [88(5)]:

"(5) … The "discovery", in this case, is that an assessment to tax is or has become insufficient. Deliberation, negligence or other questions are not relevant to whether there is a discovery. Here, HMRC discovered the insufficiency in 2009. It was incumbent upon HMRC, at that stage, to decide what to do consequent upon this discovery."

56. At [89] the UT concluded:

"89. The burden of showing that the requirements of section 29 TMA are met is on HMRC. We consider that there is no sufficient basis – given the facts found by the FTT – to justify the conclusion that there was, properly speaking, a discovery. Had this been the only point in issue, we would have allowed the appeal, and remitted the matter for further evidence and argument to the FTT. As it is, given our conclusions on the question of deliberate inaccuracy, this course is unnecessary."

The appeal

The discovery issue

57. On this appeal, Ms McCarthy QC, who appeared for HMRC, argued that the relevant discovery was made in the light of Grunberg's letter of 11 March 2014 ("the Grunberg letter") which explained to HMRC that Mr Tooth's case was, in fact, an example of the situation described by Lord Hodge in Cotter at [27]. She suggested that it is clear from the documents that until that point HMRC believed that Mr Tooth's self-assessment was correct. Whilst a claim for relief appeared on the face of the return, the self-assessment should nevertheless be interpreted by not taking account of the claim for employment-related losses, as HMRC believed (wrongly in the light of Cotter [27]) that this was a claim made outside the return. So much was clear from HMRC's letter of 4 March 2014, which stated that tax of £455,331.98 was due "based on your own self-assessments".
58. Ms McCarthy went on to submit that in October 2014 Mr Williams reviewed the file in the light of the Grunberg letter and concluded that Mr Tooth's self-assessment was understated. Thus, Mr Williams made the relevant discovery and acted on it promptly. She submitted that the UT should have analysed the "discovery" in this way, and was not obliged to remit the matter to the FTT, although it was entitled to do so.
59. Mr Ghosh QC, who appeared for Mr Tooth, submitted that the factual case now being advanced by HMRC was an entirely new one, and was unsupported by any evidence. The UT had been correct to hold that the FTT's conclusion on the discovery issue was unsupported by its findings of fact. Indeed, it was unsupported by the evidence before the FTT. The UT was wrong to indicate that (if it had been necessary) it would remit the matter to the FTT, because the burden of proof on the discovery issue had been on HMRC and they had simply failed to prove their case.
60. Both parties accepted that the legal approach to whether there is a "discovery" is correctly set out in this first passage from the decision of the UT in Charlton & others v RCC [2012] UKUT 770 (TCC); [2013] STC 866 at [37], where the tribunal said:

"37. In our judgment, no new information, of fact or law, is required for there to be a discovery. All that is required is that it has newly appeared to an officer, acting honestly and reasonably, that there is an insufficiency in an assessment. That can be for any reason, including a change of view, change of opinion, or correction of an oversight."

The UT continued in a second passage:

"The requirement for newness does not relate to the reason for the conclusion reached by the officer but to the conclusion itself. If an officer has concluded that a discovery assessment should be issued, but for some reason the assessment is not made within a reasonable period after that conclusion is reached, it might, depending on the circumstances, be the case that the conclusion would lose its essential newness by the time of the actual assessment."

61. I agree with the UT's approach in both passages. The requirement for the conclusion to have "newly appeared" is implicit in the statutory language "discover". The discovery must be of one of the matters set out in (a) to (c) of section 29(1). In the present case the officer must have newly discovered that an assessment to tax is insufficient. It is his or her new conclusion that the assessment is insufficient which can trigger a discovery assessment. A discovery assessment is not validly triggered because the officer has found a new reason for contending that an assessment is insufficient, or because he or she has decided to invoke a different mechanism for addressing an insufficiency in an assessment which he or she has previously concluded is present.
62. In the present case, the assessment which must have been discovered by the officer to be insufficient was Mr Tooth's self-assessment for 2007/08 which he filed with his return on 30 January 2009 (subsequently amended by him thereafter). It was his "statement of the amounts in which, on the basis of the information contained in the return and taking into account any relief or allowance a claim for which is included in the return, the person making the return is chargeable to income tax and capital gains tax for the year of assessment" as required by section 9(1)(a) TMA. Mr Tooth's self-assessment was that, after taking into account the employment-related losses, he was entitled to a net repayment of £13,212.00.
63. HMRC's pleaded case before the FTT does not tackle directly the need to show a new conclusion that the self-assessment was insufficient. It merely asserted, first, that "an Officer of HMRC newly concluded, following the decision in Cotter, that an enquiry under Schedule 1A TMA was not the appropriate mechanism to make good the insufficiency in the Appellant's 2007-2008 tax return." This is a statement about the officer's views on the appropriate mechanism to challenge the insufficiency. It does not state that the conclusion that the self-assessment was insufficient was itself new. Rather, it suggests that the officer's previous view had also been that the self-assessment was insufficient.
64. HMRC's pleading in the FTT also asserted that "in October 2014, an officer of the Board properly raised a discovery assessment after concluding that the Schedule 1A enquiry had not rectified the insufficiency in Mr Tooth's 2007-2008 self-assessment, following a clarification of the law in Cotter." (emphasis supplied). That again suggests that the officer's view was that there was an insufficiency prior to his new conclusion, and that HMRC had used the wrong mechanism to challenge it.
65. HMRC's skeleton argument before the FTT asserted that "in October 2014, an officer of the Board properly raised a discovery assessment after concluding that there was an insufficiency in the 2007-2008 return (because of the incorrect inclusion of a partnership loss figure). That was the "discovery"." Significantly, it was not asserted that the conclusion that there was an insufficiency was a new one. To the extent that this passage relies on the bracketed reason as somehow rendering the conclusion new, then the assertion is flawed, as shown by the passage quoted from Charlton in paragraph 60 above. In any event the Revenue had known from the outset that no partnership loss was claimed.
66. The passage in HMRC's skeleton continued "The officer also realised that HMRC's position was not protected by the existing enquiry, which is why a further assessment was required. In other words, HMRC discovered that the assessment was and would remain insufficient despite previous attempts to address this." This appears to assert that the discovery assessment was required because the Schedule 1A enquiry into the insufficiency in the assessment had proved ineffective. It certainly contains no suggestion that the officer previously considered Mr Tooth's self-assessment to be sufficient, thus rendering the conclusion to the contrary a new one.
67. There is no finding in the FTT's decision that the officer's conclusion that the assessment was insufficient was new. The email exchanges within HMRC, to which the FTT judge referred, and to which I will return in more detail below, do not establish this. The UT was therefore plainly right to hold that the findings made by the FTT were inadequate in law to support the conclusion it reached. It is not clear to me from the decision why the FTT judge thought that the officer had newly discovered that the assessment was insufficient.
68. I am however equally clear that it was wrong for the UT to decide (in the event that it had not decided the deliberateness issue in favour of the taxpayer as well) that it would have allowed the appeal and remitted the matter to the FTT for "further evidence and argument". Given that HMRC had failed even to plead or argue a coherent case of "discovery", it seems to me that there could be no proper basis for allowing HMRC to have another go.
69. That is the background for the argument presented with great clarity and tenacity by Ms McCarthy on this appeal, namely that it can be clearly seen from the documents before the tribunals below that HMRC made a discovery that Mr Tooth's self-assessment was insufficient when they received the Grunberg letter in March 2014, having previously believed the self-assessment to be correct. Her argument does tackle, albeit for the first time and in the absence of a properly pleaded case, the need to enquire into HMRC's thinking about the sufficiency or otherwise of Mr Tooth's self-assessment prior to the asserted new conclusion. She submits that this can be done by an examination of the documents, and without the need for further evidence.
70. It seems to me that the argument suffers, at the outset, from a number of difficulties, not least that the case now advanced is inconsistent with the suggestions in HMRC's pleadings and skeleton argument before the FTT that HMRC had always appreciated that there was an insufficiency in Mr Tooth's self-assessment tax return. Moreover, as Mr Ghosh points out, the suggestion that the Grunberg letter was what triggered the realisation that there was an insufficiency in the assessment is not only novel, but unsupported by any evidence or finding of fact. Mr Ghosh submitted, without contradiction, that no witness before the FTT had even mentioned the Grunberg letter. If the Grunberg letter played a decisive role, one would expect to see it so mentioned. Yet further, until this appeal, it has been said consistently by HMRC that the discovery was made by Mr Williams when he reviewed the file in October 2014. Yet the Grunberg letter was received by HMRC in March 2014, and had been considered by HMRC by May. It is plain that officers other than Mr Williams, such as Mrs Smith, had decided to issue a discovery assessment by July. Mrs Smith was not called as a witness, but the new case puts the focus on her and throws into considerable doubt what role, if any, was played by Mr Williams. Finally, the correspondence is replete with references to the suggestion that HMRC was purporting to use Schedule 1A to amend Mr Tooth's return or assessment to remove the claim for losses, making it clear that HMRC was well aware throughout that the assessment as it stood was insufficient.
71. If one uses the documents which were before the tribunals and which are before us to trace HMRC's thinking concerning the sufficiency of Mr Tooth's assessment, the following chronological picture emerges:

(i) on receipt of Mr Tooth's 2007/08 tax return, HMRC saw that Mr Tooth was claiming immediate relief for employment-related losses incurred in 2008/09. This is clear from HMRC's letter of 14 August 2009 (see [26] above).

(ii) In April 2010 HMRC stated that it purported to have amended Mr Tooth's self-assessment tax return so as to withdraw the claim for these losses. It is this purported amendment which HMRC subsequently referred to internally as "the Schedule 1A amendments" (see [29] and [40-41] above). It is at least a possible view of this letter that HMRC were fully aware of an insufficiency in the assessment, and were purporting to use the powers under Schedule 1A to tackle this insufficiency. HMRC's ability to use Schedule 1A in this way was, however, promptly challenged on behalf of the taxpayer.

(iii) The debate as to what was the correct mechanism for enquiring into Mr Tooth's claim/return ensued, with both parties awaiting the outcome of the litigation which resulted in the decision of the Supreme Court in Cotter. HMRC must have clearly understood that the question of whether they had successfully challenged Mr Tooth's self-assessment was in dispute.

(iv) On 4 March 2014, following the result of Cotter, Mr Webster, on behalf of HMRC, wrote claiming the overdue tax, and saying that the sums due were "based on" Mr Tooth's self-assessments (see [33] above). This might be seen as a contention that the insufficiency in the return had been successfully addressed by the Schedule 1A amendments, but it was plainly wrong. There was no evidence from Mr Webster as to why he thought that Cotter had this effect when Mr Tooth had made the claim in his return.

(v) On 11 March 2014, the Grunberg letter pointed out Mr Webster's error, and that the sums being claimed as outstanding tax by HMRC were incorrect.

(vi) On 19 May 2014, Grunberg wrote again, chasing a response from HMRC. Amongst other things, the letter asserted that HMRC had not amended the return under section 9ZB, and that therefore the original self-assessment must stand.

(vii) On 23 May 2014, HMRC confirmed their agreement that Mr Tooth's circumstances were "similar" to those set out by Lord Hodge in paragraph 27 of Cotter (see [35] above). This must mean that HMRC accepted that Mr Tooth had made the claim in his return and that his return had not been amended. It therefore remained insufficient.

(viii) On 28 July 2014, HMRC, through Mrs Smith, announced their intention to raise a discovery assessment (see [36] above). The letter refers to the Schedule 1A amendments and asserts that the Supreme Court in Cotter had made clear that "Schedule 1a did not give HMRC the power to remove this claim". This was a recognition that the attempt to use Schedule 1A to address the insufficiency had proved unsuccessful. There was no evidence from Mrs Smith that the conclusion that there was an insufficiency was new.

(x) On 23 October 2014, the email exchange between Mr Anders and Mr Williams took place. The email exchange shows that HMRC had decided to abandon their attempt to address the insufficiency using Schedule 1A and to issue a discovery assessment instead.

(xi) On 24 October 2014 the discovery assessment was issued. The assessment said "we have found that there is additional tax due that was not previously shown on your tax return. It is now too late for us to amend your tax assessment so this assessment allows us to collect additional tax." Of course, HMRC had contended from the outset that there was additional tax due which was not shown on Mr Tooth's tax return. This conclusion was not new.

72. The history shows that HMRC first thought they had powers under Schedule 1A to address the insufficiency. Those powers were the subject of immediate challenge. When the decision in Cotter was handed down HMRC appear to have considered, albeit briefly, that they had successfully tackled the insufficiency using Schedule 1A, but there is no evidence or finding to elucidate why that erroneous conclusion was reached. They rapidly accepted that this was not correct, and then sought to address the very same insufficiency through the use of a discovery assessment. The documents do not support the assertion that HMRC believed Mr Tooth's self-assessment to be correct until receipt of the Grunberg letter. There was no discovery of an insufficiency either on receipt of the Grunberg letter in March, or on the review of the file in October 2014. Accordingly the review of the documents set out above does not assist HMRC. But in any event, it is not an appropriate exercise to be undertaken in this court. When a question arises whether a discovery has been made, it is incumbent on HMRC to make out its case and obtain appropriate findings of fact from the FTT.
73. In my judgment, HMRC did not establish, on the basis of the documents or otherwise, that they had made a valid discovery assessment. I would dismiss HMRC's appeal on this issue and decline to remit the matter to the FTT.

The deliberateness issue

74. It follows that it is not necessary to decide the deliberateness issue, but we did hear full argument on it and I will therefore go on to consider it.
75. The deliberateness issue sub-divides into three sub-issues: (i) whether there was an inaccuracy in the return; (ii) whether that inaccuracy was deliberate; and (iii) whether the deliberate inaccuracy resulted in an insufficiency in the assessment or a loss of tax.

Was there an inaccuracy in Mr Tooth's return?

76. The UT further sub-divided their consideration of this issue into whether there was an inaccuracy in the treatment of the employment loss, and whether there was an inaccuracy in the insertion of the employment loss in the partnership pages of the return. It is only the latter on which HMRC now rely. As to this, the UT held at [57] to [59]:

"57. Viewing, solely, the entry into the partnership pages of the Return, these pages were clearly inaccurate. Figures were inserted into those pages that had nothing to do with partnership, and everything to do with the employment loss being claimed by Mr. Tooth. There was, in short, a complete mismatch between the data that these pages were intended to contain and the data that was, in fact, inserted by Mr. Tooth.

58. Once again, however, it is necessary to consider the overall context:

(1) As has been described in paragraph 31 above, the FTT found as a fact that it was not possible to complete the Return in the manner Mr. Tooth intended. The Return – an[d] electronic document – was produced by software that was HMRC approved.

(2) In these circumstances, given the nature of the return that Mr. Tooth wanted to file, the information regarding the employment loss had to be set out somewhere in the Return, and the partnership pages were used.

(3) An explanation was provided, in the Return, of what Mr. Tooth had done.

59. Given that the use of the "wrong" pages was effectively forced upon Mr. Tooth, but that he gave an explanation of his approach to dealing with the problem, we cannot accept that the Return - considered as a whole - was inaccurate."

77. Ms McCarthy submitted that the UT was wrong to suggest that the reasons for the inaccuracy were relevant to whether an inaccuracy existed. In any event, the reason given, that Mr Tooth was obliged to include the inaccuracy in his return, was factually incorrect. He could have used a paper return as opposed to an electronic one and filled in page Ai3 as he had been advised to do, or alternatively made a free-standing claim. She also attacked the UT's reliance on reading the return as a whole. Section 118(7) referred to an inaccuracy "in a document". The inaccuracy did not go away if some other part of the document referred to it and sought to explain it away.
78. Mr Ghosh submitted that the tax return, read as a whole, does not contain an inaccuracy. HMRC had understood perfectly well what Mr Tooth was saying in the document, as their letter of 14 August 2009 made clear. No reader of Mr Tooth's return could have taken Mr Tooth to be purporting to have incurred a partnership loss. HMRC's case depended on construing terms in a document without regard to their context. The UT had correctly recognised this at [45]:

"45. When considering whether there was an inaccuracy in this document, so defined and whether that inaccuracy was deliberate, it is necessary to consider the document as a whole. It would, in our judgment, be entirely wrong to "cherry-pick" one instance of inaccuracy in an entry in a document, ignoring a correction or explanation elsewhere in that document, and assert that (provided the correction was ignored) there was indeed an inaccuracy. The question of accuracy – just as the question of deliberateness – is a matter of context."

79. I agree with HMRC that Mr Tooth's reasons for including his employment losses in the wrong box are irrelevant to the hard-edged question of whether or not there was an inaccuracy.
80. Nevertheless, I agree with the UT that, in order to determine whether there is an inaccuracy in a document it is necessary, as in all exercises of interpretation, to read the document as a whole. The question is whether the document, understood as a whole, conveys inaccurate information to HMRC. I agree with the UT's analysis that Mr Tooth's return, read as a whole, does not contain any inaccuracy.
81. I recognise that section 118(7) does not refer, as it could have done, to "an inaccurate document", but to "an inaccuracy in a document". Ms McCarthy is right to draw attention to this as a possible indicator that her piecemeal approach to construction is correct, but it is no more than that. I do not regard it as sufficient to displace what I regard as the normal approach to construction of documents, which is to read their individual parts in the context of the document as a whole.
82. It is also correct to say that an individual inaccuracy in Mr Tooth's, or indeed any, tax return will feed mechanically into the computation of the amount of the tax liability, whether this occurs in a computer program used by the taxpayer, or by the manual use of HMRC's assisted tax computation pages. So the effect of an inaccuracy is not necessarily cured by pointing out elsewhere in the return that the inaccurately completed box is intended to convey some different meaning. But section 118(7) is not solely concerned with tax return forms. It applies to any document given to HMRC. So it would be wrong in my view to ascribe to the draughtsman an intention for the sub-section to apply to any isolated inaccuracy, however clearly explained it may be by the context, merely because of this special attribute of returns.
83. I therefore agree with the UT that there was no inaccuracy in Mr Tooth's return merely because he included his employment-related losses in the wrong box, given that he fully explained what he had done elsewhere in the return.

Was the inaccuracy "deliberate"?

84. For these purposes it is necessary to assume, contrary to my conclusion, that there was an inaccuracy in Mr Tooth's return by the inclusion of the employment-related losses on the partnership pages. Ms McCarthy argued that this inaccuracy was deliberate, in that Mr Tooth knew that these were not partnership losses, but employment-related losses.
85. Mr Ghosh supported the UT and FTT in the conclusions they reached on this issue. He submitted that, despite section 118(7), sections 29(4) and 36(1A)(a) require an intention to bring about a loss of tax. He submitted that the scheme of the time limits (four years in the ordinary case, six years for carelessness and 20 years for deliberate) meant that Parliament intended to link the applicable limitation period to the degree of culpability. Further, he submitted that this conclusion was supported by the Explanatory Notes to the Finance Bill 2008, which explained that:

"New subsection (7) provides that where a loss of tax or a situation is brought about by a deliberate inaccuracy in a document, it is to be treated as brought about deliberately. This is to ensure that in cases where a penalty is due for deliberate inaccuracy under paragraph 3 of Schedule 24 to FA 2007, the corresponding increase in time limit also occurs…"

86. The deliberateness requirements of section 29(4) and 36(1A)(a) require HMRC to prove that the taxpayer intended to bring about a particular fiscal result. In the case of section 29(4) it is an intention to bring about a situation in which an assessment to tax is insufficient, and in the case of section 36(1A)(a) it is an intention to bring about a loss of tax. I agree with HMRC's contention, however, that section 118(7) is a deeming provision which means that HMRC can establish the relevant intention by showing that there was a deliberate inaccuracy in a document given to HMRC by or on behalf of the taxpayer, and that the loss of tax followed "as a result of" the deliberate inaccuracy. That is no more than what the language of the statute conveys. It follows that the enquiry about the taxpayer's intention stops once it is established there is a deliberate inaccuracy in a document. Thereafter one enquires into whether the loss of tax or other situation occurred as a result of the inaccuracy. That is simply a question of factual causation.
87. I did not find the Explanatory Notes to the relevant Finance Bill on which Mr Ghosh relied to be of assistance in reaching any other conclusion. The first sentence of the Notes confirms, as Ms McCarthy pointed out, that the section is a deeming provision, because it uses the language "is to be treated as …".
88. The requirement for deliberateness occurs in the different contexts of section 29 and section 36. In section 29(4) it operates as a pre-condition (along with carelessness) for HMRC to be able to raise a discovery assessment. Its obvious purpose is to restrict the availability of a discovery assessment (as opposed to the other mechanisms for enquiring into a taxpayer's return) to cases where there is some blameworthy conduct on the part of the taxpayer. Section 29(5) extends the availability of a discovery assessment to certain cases where HMRC could not be expected to be aware of the situation (e.g. the insufficiency of tax) on the basis of the information available to them within the time period for launching an enquiry. These are cases, therefore, where HMRC is blameless in not raising the assessment earlier, but do not depend on proving any blameworthy conduct by the taxpayer.
89. The triggers for the 20 year time limit identified in section 36(1A)(a) to (d) also do not include a consistent requirement of blameworthy conduct by the taxpayer. Sub-paragraph (b) includes a failure by a person who is chargeable to income tax for any year of assessment and who has not delivered a return of his profits, gains or income for that year to give notice that he is so chargeable. The failure is not required to be negligent or deliberate. Such a failure could occur, for example, as a result of incorrect advice.
90. In the light of these considerations, I do not regard it as surprising that, as a result of the expanded meaning given to the sub-sections by section 118(7), conduct which is overall not blameworthy is brought within the definition.
91. The FTT appears to have considered whether the inaccuracy was deliberate solely by reference to section 29(4), asking itself at [57] whether the taxpayer had deliberately brought about the insufficiency of the assessment. I agree with Ms McCarthy that this was an error of law. The FTT should have asked whether the inaccuracy was deliberate, and then, separately, whether this resulted in fact in the insufficiency of the assessment.
92. The UT held at [46] that section 118(7) did not:

"remove[s] from section 29(4) the requirement that "the situation mentioned [in section 29(1) TMA] was brought about" by a deliberate inaccuracy in a document given to HMRC…"

93. That is correct, in the sense that section 118(7) still requires factual causation of the "situation", although it is important to keep in mind that it is not necessary to show that the taxpayer intended to bring about the situation (or in section 36(1A)(a) the loss of tax). The UT's reasoning as to why there was no deliberate inaccuracy is in [66]:

"The mere insertion of a figure into a document that is inaccurate may be a deliberate act, but it is not, necessarily, a deliberate inaccuracy. In this case, we do not consider that the inaccuracies alleged by HMRC can be said to be deliberate because Mr Tooth took steps to draw the (putative) inaccuracies to the attention of HMRC."

94. I approach this second sub-issue on the assumption (contrary to my conclusion on the first sub-issue) that there is an inaccuracy in the document. If there is no inaccuracy then there is no deliberate inaccuracy either. If there is an inaccuracy, however, that must be because it is incorrect to construe the tax return as a whole, and correct to focus on the individual inaccuracy on the partnership pages of the return. The incorrect insertion of the employment losses in the boxes reserved for partnership losses was, viewed in this way, a deliberate inaccuracy. Whilst it is no longer suggested that Mr Tooth and his advisers were, by this means, deliberately seeking a reduction in his liability to tax, the inaccuracy was, on any view, deliberate. I agree with HMRC that Mr Tooth cannot escape from this conclusion by the suggestion, accepted by the UT, that he was forced to enter his employment losses in this way. There were other means by which he could communicate his loss claim to HMRC without including inaccuracies in his return, if that is what they were.
95. At [67] the UT explained why it upheld the decision of the FTT. They said:

"We consider that the FTT did not err in finding that Mr Tooth had not acted deliberately. There is no evidence of any intent on the part of Mr Tooth to bring about an insufficient assessment of tax or give HMRC a deliberately inaccurate document." (original emphasis).

96. The first part of this passage relies on the absence of any evidence of an intent to bring about an insufficient assessment of tax, a point no longer pursued by HMRC. The second part of the passage suggests that there is no deliberately inaccurate document. On the assumptions I am making, I am forced to disagree. If (contrary to my view) there was an inaccuracy in the document, it was the inclusion of the loss in the wrong box, and that was deliberate.
97. I therefore disagree with the UT's conclusion on this sub-issue.

Did the inaccuracy result in a loss of tax/insufficiency of assessment?

98. Ms McCarthy submits that the inclusion of the employment losses in the wrong box resulted in the insufficiency in the assessment, because the partnership loss is forced into the tax computation by the software utilised to complete the self-assessment return. Likewise, the inclusion of the losses in the wrong box resulted in a loss of tax.
99. Mr Ghosh submitted that nothing arose as a result of the inaccuracy, since HMRC understood that the partnership pages were not intended to represent an actual partnership loss.
100. Neither the FTT nor the UT reached a conclusion on this issue.
101. I think HMRC are correct on this issue. Once the focus is on the inaccuracy in the partnership pages, it is clear on the facts that the resultant calculations performed by the software will force the losses into the calculation of the tax due. Had Mr Tooth adopted a course not involving the inaccuracy, such as referring to the claim on page Ai3, the losses would not have been reflected in the calculation, and therefore in Mr Tooth's self-assessment. Moreover, in those circumstances, Mr Tooth's situation would have been in all material respects the same as Mr Cotter's, and HMRC would have been able to collect the tax via a Schedule 1A enquiry. In my judgment that is a sufficient causal connection for the court to conclude that the "situation" or "the loss of tax" occurred as a result of the inaccuracy.
102. Had it been necessary to do so, I would have reached a conclusion on the deliberateness issue which is in accord with that reached by the UT and the FTT.

Conclusion and postscript

103. For the reasons I have given, Mr Tooth is entitled to succeed on the discovery issue. I would therefore dismiss HMRC's appeal.
104. I should add that after this judgment was circulated in draft to the parties in the usual way our attention was drawn by HMRC to the recent decision of the Supreme Court in Derry v HMRC [2019] UKSC 19, but it did not appear to me to affect either of the issues which we had to decide.

Lord Justice Males:

105. I agree that this appeal must be dismissed on the discovery issue for the reasons given by Floyd LJ. However, while it does not affect the outcome of this appeal, I have reached a different conclusion from him on the deliberateness issue.
106. The main provision dealing with deliberateness is TMA section 29(4). This refers to the position when an insufficiency of assessment "was brought about … deliberately by the taxpayer or a person acting on his behalf". That was not the case here. However, the meaning of "deliberately" in this context is extended by what is in effect a deeming provision contained in section 118(7):

"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."

107. There are, therefore, three sub-issues: (1) was there an inaccuracy in the return? (2) was the inaccuracy deliberate? and (3) did the inaccuracy bring about (i.e. cause) the assessment to be insufficient?
108. It is important to say at the outset that there is no question of Mr Tooth or his advisers having acted dishonestly or even reprehensibly. They sought to make clear what was being claimed, stated in terms that they expected HMRC to challenge the claim, and did not intend to mislead anyone. Nor did they in fact mislead anyone. However, that does not in itself answer the three questions identified above, to which I now turn.

Was there an inaccuracy in the return?

109. Whether there was an inaccuracy in the return depends on whether the right approach is to read the document as a whole. If it is, I would agree with Floyd LJ that the document as a whole can be regarded as accurate. However, I do not consider that this is the right approach.
110. The statutory question is whether there is an inaccuracy "in" a document given to HMRC. It is a perfectly standard use of language to say that there is an inaccuracy in one part of a document which is corrected in another part; or that despite an inaccuracy in one part of a document, the document as a whole is not misleading. In my judgment that is the position here. As the UT found, the entry of figures into the partnership pages of the return meant that these pages were inaccurate, even if the true position was apparent on a fair reading of the document as a whole.
111. Usually, if a document contains an inaccuracy but read as a whole is not misleading because the reader to whom it is addressed is able to understand the true position, no harm will be done. More specifically, the inaccuracy will have no causative effect. But that is a matter of causation. It does not mean that there is no inaccuracy in the document. This is, as Floyd LJ describes it at [79], a hard-edged question.
112. I would hold, therefore, that there was an inaccuracy in the return.

Was the inaccuracy deliberate?

113. For the reasons given by Floyd LJ, I agree that the inaccuracy (which he assumes for this part of his judgment but which I have concluded does exist) was deliberate.

Did the inaccuracy bring about (i.e. cause) the assessment to be insufficient?

114. Again I agree with the reasoning of Floyd LJ on this point. So far as human readers of Mr Tooth's tax return were concerned, it appears that nobody was misled. Although there was some confusion on the part of HMRC as to the correct method for challenging what Mr Tooth was claiming, it appears that nobody actually thought (for example) that he was claiming a partnership loss, or that he was claiming to be a partner in a firm which had been allocated a UTR of 99999-99999.
115. However, the return (using that term in the narrow sense explained by Lord Hodge in Cotter at [25] quoted at [11] above) was not addressed solely to human readers. Because the entry of a figure for a partnership loss in the relevant box fed automatically into the computation of Mr Tooth's tax liability for the relevant year, this entry did as a matter of fact cause his self-assessment to be insufficient. That is all that is required for the purpose of section 118(7).

Lord Justice Patten:

116. I agree with Floyd LJ that the appeal should be dismissed for the reasons which he gives in relation to the discovery issue.
117. On the deliberateness issue and, in particular, as to whether there was an inaccuracy in the return for the purposes of s.118(7) TMA, I take the same view as Males LJ. The fact that the document read as a whole neutralises any inaccuracy in part of the document does not prevent that being an inaccuracy "in" the document. This is a technical condition for the application of the extended definition of what is meant by "deliberately" in s.29(4) and should be given its straightforward literal meaning. The taxpayer's argument that the document when read as a whole did not mislead is adequately accommodated as part of the wider issue as to whether the inaccuracy in the document brought about an insufficiency of assessment.